tuberculosis case study journal

Get new issue alerts Get alerts
Submit a Manuscript

Secondary Logo

Journal logo.

Colleague's E-mail is Invalid

Your message has been successfully sent to your colleague.

Save my selection

Role of families in tuberculosis care

A case study.

Samal, Janmejaya; Dehury, Ranjit Kumar 1

International Union Against Tuberculosis and Lung Disease (The Union), Pune, Maharashtra, India

1 Department of Healthcare Management, Goa Institute of Management, Panaji, Goa, India

Address for correspondence: Dr. Janmejaya Samal, C/O - Mr. Bijaya Ketan Samal, At - Pansapalli, PO - Bangarada, Via - Gangapur, District - Ganjam, Pin - 761 123, Odisha, India. E-mail: [email protected]

This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.

Dear Editor,

In case of tuberculosis (TB), being a chronic infectious disease, the role of the family in care and support becomes very important. Moreover, the family plays an important part in maintenance of the optimum level of health as well as in the dynamics of the disease.[ 1 ] The families and the individual family members strongly influence the health-seeking behavior of patients. Health-seeking behavior allows the patients to choose their preferred healthcare destination and the time of seeking help for TB treatment as per their own wish. In most of the Indian communities, it has been observed that the first point of contact by a chest symptomatic/cough symptomatic is a private health facility. However, this case study is about the health-seeking behavior of a TB patient who was reported with high patient compliance and undergoing treatment at a public sector health facility. Poor health-seeking behavior has been reported by various studies, however, this particular lady, with the support of her family, got good patient compliance and is now in the continuation phase of treatment as per the Revised National Tuberculosis Control Program (RNTCP). This case study delineates about the perspectives of her health-seeking behavior and the story behind her good patient compliance and the role of her family in this endeavor.

A semi-structured interview schedule consisting of 10 questions was used to interview the patient in-depth. The questions were pertinent to the demographic profile of the patient, the time of TB detection, reaction of her family to TB being detected, role of family members in TB care, significance of family in TB care, role of health system in TB care, and above all, her role in spreading the message in the community where she resides.

Reaction of the family toward the TB patient (response of families to a social pathology)

In many developing and underprivileged communities, TB patients face dire consequences once detected with TB. They face various barriers in day-to-day life, as well as isolation and rejection from their respective families and communities.[ 2 ] In this particular case study, the respondent had to face similar consequences in her family after she got detected with TB. As she says;

"They were not happy with the test results, especially my mother-in-law and she started abusing my maternal family, my parents and blamed that I have contracted this infection from my maternal home and now going to kill everybody in her family."

Furthermore, gender plays a significant role in TB care, as TB is associated with social stigma. Studies reveal that women prefer home remedies at the onset of symptoms.[ 3 , 4 ] This is primarily due to the fear of getting treated from designated public sector health facilities which may reveal the truth of them as being the TB patient leading to social isolation. A study in Russia reported “female gender” as a significant predictor of multidrug-resistant tuberculosis (MDR-TB).[ 5 ] In the Indian context, harassment by in-laws, difficulty in getting married, or dismissal from work were reported as major barriers for women to get appropriate treatment.[ 6 ]

Perspective of family strength in TB care

TB being a chronic infection requiring long term treatment, the role of families cannot be neglected. Beginning from the infection, manifestation of signs and symptoms, health-seeking behavior, and outcome of treatment, the family plays a pivotal role.[ 7 ] In this particular case study, the respondent has also understood the importance of the strength of the family in her treatment adherence and outcome as well. As she says.

"This is important sir, as without the support of my family members I could not have taken the medicines properly. They have also helped me accompanying to the hospital for regular checkups and sputum examination. Family support makes me feel better and I always feel nothing is going to happen to me as my husband is with me."

Concerns and support of families are required for treatment adherence, quality of care, and treatment completion in case of TB patients.[ 8 ]

Contribution of family strength in TB care

The contribution of families toward TB care can be twofold; support and care. In a similar study conducted in the Pune district of Maharashtra, India, the study participants defined care and support to be rendered by their family members and as expected by them toward TB care. Respondents defined good support and care in terms of helping them in routine activities, monetary help, emotional and moral support, and motivation to complete treatment. The respondents added that the support could be measured in terms of accompanying the patient for treatment, reminding for taking medicines, allowing them for rest, providing food, and necessary support as and when required. Similarly, they defined care as something such as speaking of sweet words of encouragement, motivation to fight the disease, and discouragement for negative thoughts such as attempts for suicide and abandoning home.[ 9 ] In this particular case, despite initial resistance, the family members came forward, especially the husband and the mother-in-law, as narrated by the patient to help her fight for the disease. As she says.

"Yes my husband and as I mentioned he has a great role and supported a lot in my regular treatment and my sputum examination. He has always tried to accompany me to the hospital if he could find time to do so. He has arranged me proper food and taken utmost care so that I can take the medicines regularly without missing a single dose. Now my mother-in-law also supports a lot in this and in the absence of my husband she accompanies me to the hospital."

Limitations of family based care for TB

In this particular case study, limitation of the role of the families in providing care and support could not be elicited except a short span of resistance at the initial phase of detection of TB to the patient. This is primarily because the treatment was being rendered by a community health worker, known as Accredited Social Health Activist (ASHA). Under the RNTCP, anti-TB drugs are being provided under directly observed treatment short-course (DOTS) strategy. The community health worker who provides medication under this scheme is called a DOTS provider. Hence, the role of family member(s) in providing medication to the TB patients in India is limited and is being provided by the DOTS providers only. There could be several of limitations to this if the same is being provided by the family members. In a qualitative enquiry at Botswana, having 20 in-depth interviews, the respondents had different opinions about home-based TB care. The study revealed that patients feared that being treated for TB under Home-based Directly Observed Treatment (HB-DOT) could affect their adherence to medication. The concerns of the respondents were primarily related to the level of knowledge and skills of the home-based volunteers. The level of knowledge and skills that the trained health workers have is definitely higher than that of the home-based volunteers, which is the main reason of concern and fuels the perceived fear among the TB patients to receive care from them. Another concern of the TB patients in the same study was that the home-based volunteers may not be strict enough as the health workers. This is true as well in the sense that being a member of the family they may not be as strict as the trained health workers from the public system, as they do not belong to that family and will perform their duties without any compromise. The third concern pointed by these respondents was that in the absence of home-based volunteers there would be nobody to take care of them. Furthermore, the respondents suggested that in addition to home-based volunteers, the health workers should visit regularly to monitor the progress of treatment.[ 10 ]

Way ahead and further research

Albeit, research on health-seeking behavior and the influence of gender, culture, and families have begun in India, especially in the domain of TB care, however, studies on the role of families in care and support and the strength of families and culture is limited.[ 9 ] At the dearth of community health workers in India, research exploring the role of families in care and support is the need of the hour. This could be done by juxtaposing the role of health workers from public sectors and volunteers from families and communities so that the lacunas could be found out which would otherwise help in designing training programs for the home-based volunteers. Moreover, TB is a global public health crisis and around 25% of TB cases are found in India; hence, research on the role of families in rendering care and support and exploring the possibilities of utilizing home-based volunteers seems rational in the Indian context.

Financial Support and Sponsorship

Conflicts of interest.

There are no conflicts of interest.

Cited Here |
View Full Text | PubMed | CrossRef
PubMed | CrossRef
+ Favorites
View in Gallery

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Publications
Account settings
My Bibliography
Collections
Citation manager

Save citation to file

Email citation, add to collections.

Create a new collection
Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

Search in PubMed
Search in NLM Catalog
Add to Search

A case study of a patient with multidrug-resistant tuberculosis

Affiliation.

1 Community nurse working in South West England.
PMID: 30048191
DOI: 10.12968/bjon.2018.27.14.806

In this case study, a nurse presents her reflections on the challenges of supporting a patient through his treatment journey for multidrug-resistant tuberculosis. The patient has significant comorbidities and social issues, such as diabetes and homelessness. There was also a language barrier. All these aspects made the management of his treatment challenging. The medication side effects and his lifestyle were also a barrier to full engagement. The same multidisciplinary team was involved with the patient and, despite the obstacles, he seemed willing to engage with treatment and the team.

Keywords: Comorbidities; Language barrier; Multidisciplinary team; Multidrug-resistant tuberculosis; Pulmonary TB; Under-served population.

PubMed Disclaimer

Related information

Linkout - more resources, full text sources, other literature sources.

scite Smart Citations
Genetic Alliance

Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Open access
Published: 21 February 2023

Tuberculosis in older adults: case studies from four countries with rapidly ageing populations in the western pacific region

Alvin Kuo Jing Teo 1 , 2 , 11 ,
Kalpeshsinh Rahevar 3 ,
Fukushi Morishita 3 ,
Alicia Ang 15 ,
Takashi Yoshiyama 5 ,
Akihiro Ohkado 5 ,
Lisa Kawatsu 5 ,
Norio Yamada 5 ,
Kazuhiro Uchimura 5 ,
Youngeun Choi 6 ,
Zi Chen 7 ,
Siyan Yi 1 , 8 , 9 ,
Manami Yanagawa 3 ,
Kyung Hyun Oh 3 ,
Kerri Viney 10 na1 ,
Ben Marais 2 , 11 na1 ,
Heejin Kim 6 na1 ,
Seiya Kato 5 na1 ,
Yuhong Liu 12 na1 ,
Catherine W.M. Ong 4 , 13 , 14 na1 &
Tauhid Islam 3 na1

BMC Public Health volume 23 , Article number: 370 ( 2023 ) Cite this article

4955 Accesses

10 Citations

1 Altmetric

Metrics details

A Correction to this article was published on 07 June 2023

This article has been updated

The Western Pacific Region has one of the fastest-growing populations of older adults (≥ 65 years) globally, among whom tuberculosis (TB) poses a particular concern. This study reports country case studies from China, Japan, the Republic of Korea, and Singapore reflecting on their experiences in managing TB among older adults.

Across all four countries, TB case notification and incidence rates were highest among older adults, but clinical and public health guidance focused on this population was limited. Individual country reports illustrated a range of practices and challenges. Passive case finding remains the norm, with limited active case finding (ACF) programs implemented in China, Japan, and the Republic of Korea. Different approaches have been trialled to assist older adults in securing an early diagnosis, as well as adhering to their TB treatment. All countries emphasised the need for person-centred approaches that include the creative application of new technology and tailored incentive programs, as well as reconceptualisation of how we provide treatment support. The use of traditional medicines was found to be culturally entrenched among older adults, with a need for careful consideration of their complementary use. TB infection testing and the provision of TB preventive treatment (TPT) were underutilised with highly variable practice.

Older adults require specific consideration in TB response policies, given the burgeoning aging population and their high TB risk. Policymakers, TB programs and funders must invest in and develop locally contextualised practice guidelines to inform evidence-based TB prevention and care practices for older adults.

Peer Review reports

Globally the number of older adults (aged ≥ 65 years) is expected to triple by 2100 [ 1 ]. Life expectancy at birth has gradually increased since the 1950s, [ 1 ] with the number of older adults projected to exceed children aged < 5, across all economies by 2020 [ 2 ]. The Western Pacific Region has one of the fastest-growing populations of older adults globally. In 2019, life expectancy at birth of the population was 4 years above the global estimate of 73.3 years [ 3 ]. Japan, for instance, is the most aged country in the world (average life expectancy 85 years) [ 4 ]. Several other countries in the region, particularly the People’s Republic of China (hereinafter referred to as China), the Republic of Korea, and the Republic of Singapore (hereinafter referred to as Singapore), were estimated to have an increase in the proportion of older adults globally between 2019 and 2050.

Japan has the highest proportion of older adults in the world, at 28% of the total population in 2020, [ 5 ] and the proportion is expected to rise to 38% by 2065 [ 6 ]. China is also undergoing a profound demographic transition. In 2021, the number of older adults reached 191 million, amounting to 13.5% of the total population. The number rose by 4.6% compared to the previous year, [ 7 ] and the increasing trend is likely to persist with the number of older adults expected to exceed 480 million (34.6% of the total population) by 2050 [ 8 ]. Similarly, the demographic transition in the Republic of Korea has seen rapid growth in the older adult population. In 2020, 16% of the total population was ≥ 65 years; this is expected to reach 37% by 2050 [ 9 ]. The older adult population made up 13% of the total population in Singapore in 2020 [ 5 ]. The proportion is projected to rise on the same trajectory as China, Japan, and the Republic of Korea in the next few decades [ 1 ].

Improvements in life expectancy have been attributed to better nutrition, political stability, risk factor reduction and improved healthcare access [ 10 ]. Globally, older adults contributed 26% of the total disease burden (measured in disability-adjusted life years [DALYs]); 38% of the burden in high and upper-middle-income, and 16% in low-and lower-middle-income regions in 2019 [ 11 ]. Non-communicable diseases such as cancer, cardiovascular disease and neurological disorders accounted for most of this burden, but respiratory infections, including tuberculosis (TB), are a major contributor as well [ 11 ]. In fact, TB case notifications and estimated disease incidence rates were highest among older adults in China, Japan, the Republic of Korea, and Singapore (Fig. 1 ) [ 12 ].

The proportion of notified TB cases aged ≥ 65 in Japan, the Republic of Korea, Singapore, China, the Western Pacific region, and globally 2013−2020

Data source: World Health Organization Global TB Programme 2021

With increasing age, progressive immune dysfunction (immunosenescence) increases the risk for TB disease development [ 13 , 14 ]. The convergence of co-morbidities such as diabetes, chronic respiratory disease and undernutrition, and lifestyle behaviour like tobacco smoking also increase TB risk [ 15 , 16 , 17 , 18 ]. Considering the burgeoning aging population in the Western Pacific Region and the limited guidance specific to this population, it is clear that the needs of older adults require greater recognition in TB response policies. This study reports country case studies from China, Japan, the Republic of Korea, and Singapore, reflecting on country-specific experiences in TB diagnosis and management among older adults. The case studies are a part of a broader endeavour, including a narrative review and analysis of epidemiological trends, to understand and document TB management among older adults in the region. The findings from the narrative review (see supplementary materials for methods) were used to support the country-specific experiences reported in the case studies.

TB epidemiology

China has the second-highest TB burden globally. In 2020, China had an estimated 842,000 people with TB (59 per 100,000 population per year), with older adults accounting for a quarter of all cases [ 12 ]. China notified 159,100 cases (19% of all notified cases) among older adults, [ 12 ] and a consistently increasing trend has been observed over the last two decades [ 19 ]. The fifth national TB epidemiological survey performed in 2010 recorded prevalence rates for clinical and bacteriologically confirmed TB as 482 and 138/100,000 population, respectively, among older adults [ 20 ]. The TB prevalence rate in the survey increased with age and peaked in the 75−79 year age group at 866/100,000 population (overall prevalence rate: 442/100,000 population [95% CI 417−469]) [ 20 ]. A subsequent prospective cohort study reported high TB incidence rates, [ 21 ] highlighting the need to prioritise older adults in the TB response. TB-related mortality was also more prevalent among older adults [ 19 ] concentrated in China’s Central and South-Eastern regions, [ 22 ] among lower-income residents from rural areas and in those with comorbidities [ 23 ].

Health system structure and TB services

The Government of China introduced directly observed treatment (DOT) (now known as treatment support), [ 24 ] as a strategy, in 13 provinces in 1991, with a nationwide scale-up by 2001 [ 25 , 26 ]. The National Center for Disease Control (CDC) is responsible for coordination, policymaking, standards and technical guidance, while provincial CDCs conduct TB programs and monitor disease trends at the prefecture and county level.

TB diagnosis and treatment are provided under the free-TB service policy in China [ 27 , 28 ]. The policy is implemented through an integrated model where designated hospitals (specialised centres for infectious diseases and general hospitals with TB departments) provide TB diagnosis and treatment services [ 29 ]. Persons with drug-resistant TB are treated at designated prefectural or provincial-level hospitals [ 29 ]. Primary healthcare providers are responsible for referring older adults presenting with TB symptoms to designated county-level hospitals for diagnostic investigation. They also conduct follow-up, case management and support, and health education activities at the community level [ 29 ]. The private sector is not involved in TB management. This free-TB service policy is supported by funds from the central and local governments [ 29 ]. Expenditures incurred beyond the free-TB service policy are generally covered by other government-linked insurance schemes [ 27 ]. However, the insurance and financial support schemes vary between provinces, resulting in different levels of financial reimbursement.

Despite this policy, catastrophic costs due to TB remain substantial. Studies report that more than 50% of TB patients incurred out-of-pocket payments exceeding 40% of the household’s ‘capacity to pay, [ 30 , 31 ] indicating a need for better financial support plans and packages to be considered. In addition to financial support, social support through better health education, supportive services, and community engagement is important to improve TB treatment adherence and outcomes [ 32 ].

TB case finding

Active case finding (ACF), a provider-initiated systematic screening and testing for TB disease, [ 33 ] has been deployed in conjunction with the annual physical assessments program for older adults since 2015. Such an integrated approach was feasible and effective for TB case detection among older adults, especially those with comorbidities such as diabetes [ 34 , 35 , 36 ]. TB management is carried out by community health service centres under the technical guidance of the National Tuberculosis Plan [ 37 , 38 ].

Institutionalised aged care is a booming sector in China. As of September 2021, more than 2.6 million older adults were living in retirement facilities [ 39 ]. The quality of services provided in these facilities provides is inconsistent, often resulting in over-crowding in poorly ventilated conditions. Consequently, disease outbreaks are common and in one TB outbreak, 40% of the residents were diagnosed with TB disease [ 40 ]. Therefore, amendments to existing legislation regarding infection prevention and control in such facilities should be considered, with relevant TB information provided to residents and staff. ACF should be prioritised with the expansion of the annual physical assessments program.

TB treatment, care and support have been decentralised using various enablers such as the community, family members, local primary healthcare providers, and technology (electronic medication monitor) for people with TB who were clinically stable and non-infectious [ 26 , 41 ]. However, given their advancing age, existing chronic health conditions, and weakened immunity, older adults are prone to stopping treatment (loss to follow-up) and death [ 42 ]. Currently, limited guidance is provided at the country and global levels to address issues encountered by older adults, such as adverse event monitoring and management, drug-drug interactions and treatment adherence.

Traditional Chinese herbal medicines have been used to treat TB and as adjuvants to improve the tolerability of TB treatment. Although not endorsed by the WHO, herbal medicines are widely used in China [ 43 ]. In 2020, a meta-analysis reported some potential benefits, [ 44 ] but unfortunately, the evaluated studies lacked methodological rigour. Given how widely herbal medicines are used in China, especially in older adults, practice guidelines should consider how best to advise and regulate health practitioners.

TB prevention

The ramp-up of TB response through nationwide expansion of treatment support (historically known as directly observed treatment [DOT]) and improvement of disease reporting and referral system saw a 28% and 65% reduction in the prevalence of pulmonary TB and smear-positive TB, respectively, between 1990 and 2010 in China [ 45 ]. As the risk of infection and transmission decreases, it is expected that reactivation will be the main driver of TB disease occurrence as the population ages in the coming decades [ 46 , 47 ]. China has around 350 million people with TB infection, [ 48 ] however, screening for TB infection and the uptake of TB preventive treatment (TPT) among at-risk groups is limited. Incorporation of TB infection screening into the annual health assessment could improve awareness regarding TB infection and uptake of TPT among at-risk groups but requires further evaluation for feasibility and safety [ 49 ].

The estimated TB incidence in Japan has declined from 36 to 12/100,000 population per year from 2000 to 2020 [ 12 ]. The treatment coverage (notification rate as a proportion of the estimated incidence) was approximately 85%, with 69% reported among older adults (43% among those ≥ 80 years) [ 12 , 50 ]. The median age of a person notified with TB in 2020 was 77 years [ 50 ]. While the number and rate of TB cases have decreased in the last decade, the proportion of older adults among new cases has increased from 48% to 2000 to 59% in 2010 and 69% in 2020 [ 50 ]. Although this is mainly driven by reactivation, multiple outbreaks with transmission in health and aged-care facilities have been reported [ 51 , 52 ]. Despite increased rates of disease, TB mortality among older adults has declined from > 10 to < 5 deaths per 100,000 population in recent years [ 53 ]. Nevertheless, the highest number of TB deaths occurred among older adults in Japan [ 53 ].

The Ministry of Health, Labour, and Welfare of Japan comprises 18 departments, bureaus, and offices that oversee public health, workplace safety and sanitation, medical services, and health insurance [ 54 ]. Access to healthcare in Japan is facilitated by a universal health insurance scheme with population-wide coverage. While approximately 80% of the health facilities are privately owned, they are regulated by the government, and payment for services is controlled by the ministry under the insurance scheme [ 54 ]. These facilities provide direct medical care for people with TB, including in-patient care during TB treatment. Meanwhile, the public health centres provide other aspects of TB response such as screening for TB disease and infection, contact investigation, surveillance, and community-based treatment support [ 55 ]. National TB surveillance data are managed by the Research Institute of Tuberculosis [ 55 ].

TB care is covered by national health insurance and public subsidies at about 70% and 25%, respectively. The remaining 5% is paid out-of-pocket, but a public assistance fund can be used to offset costs for those who cannot afford it [ 55 ]. Medical expenses incurred during hospitalisation are wholly covered by health insurance and public subsidies. Given the rising need for dedicated and specialised care for older adults, a long-term care insurance scheme was introduced in 2020 to support those in need, as well as their family members [ 54 ].

In Japan, TB among older adults has historically been detected passively at outpatient and inpatient health facilities [ 56 , 57 ]. TB case finding among older adults in the health facilities were facilitated by the presence of comorbidities and routine follow-up for other health conditions, thus increasing the likelihood of care-seeking at the onset of TB symptoms and, subsequently, screening for TB [ 58 ]. However, case finding in such settings hinge on a clinician’s awareness of the need to screen for TB. On presentation to health facilities, TB diagnosis among older adults has historically been difficult due to atypical presentation in older adults. For example, one study reported that > 40% of older adults with TB presented with atypical features of TB in Japan [ 59 ]. The non-specific symptoms and low awareness of TB [ 60 ], especially in intermediate TB burden settings such as Japan, could lead to delayed diagnosis.

In addition, community-based ACF programs targeting older adults have been recommended and implemented [ 61 ]. Currently, these programs prioritise older adults ≥ 80 years for TB screening using mobile chest x-rays in the community. For institutionalised older adults, those aged ≥ 65 are offered annual TB disease screening to prevent outbreaks in aged-care facilities. Before admission to an aged-care facility, TB screening using chest radiography has also been implemented to facilitate the early detection of TB disease [ 62 ]. This obligation is also extended to frontline workers in health and aged-care facilities and welfare facilities.

Treatment of drug-susceptible TB in Japan generally follows the standard 6-month regimen [ 55 ] . However, a 9-month regimen without pyrazinamide (2 months of isoniazid, rifampicin, and ethambutol, followed by isoniazid and rifampicin for 7 months) is widely prescribed for older adults, particularly those aged ≥ 80, and the national guidelines did not recommend pyrazinamide for older adults with TB until 2018 [ 63 ]. While recent studies have reported that the pyrazinamide-based regimens do not lead to significantly higher rates of treatment discontinuation, liver injuries, and death than regimens without pyrazinamide, [ 63 , 64 ] the proportion of those receiving the standard 6-month regimen with pyrazinamide declines with age, particularly for those aged ≥ 75 years [ 50 ].

The overall treatment success rate of drug-susceptible TB was 66% in Japan in 2019 [ 50 ] . The treatment success rate among people with TB < 65 years was 82%, and the rate decreased among the older age groups (65−74 years; 76%, 75−84 years; 65%, ≥85 years; 46%) [ 50 ]. The low treatment success rate among older adults has been ascribed to a high death rate during TB treatment. In 2019, 33% of older adults with TB died during treatment, and > 60% died of non-TB-related causes [ 50 ]. Japan has a comprehensive treatment support strategy for people with TB disease or TB infection (TPT). For those who require in-patient care, treatment support that comprises patient education is provided in the hospital. Upon discharge, an individual support plan for community-based treatment support is developed based on the person’s risk of non-adherence, and the frequency and means of medication support are determined. A treatment support conference is held to evaluate and review the plan. Through a patient-centred care approach, treatment support may be provided by health professionals in the local community, community health workers and volunteers, or family members.

Tuberculin skin tests (TST) and interferon-gamma release assays (IGRA) are used to diagnose TB infection in Japan; both tests are covered by national health insurance. TB infection has been a notifiable condition since 2006 [ 50 ]. While there are no policies to actively screen older adults as a priority group for TB infection, notification of TB infection among older adults ≥ 65 has gradually increased since 2010 [ 50 ]. In 2020, approximately 49% of older adults with TB infection were detected via contact investigation, [ 50 ] a national policy to facilitate early detection of TB disease and infection and prevent onward transmission. There was also an increasing number of older adults diagnosed with TB infection in the hospital due to TB infection testing carried out before treating other medical conditions such as rheumatoid arthritis [ 65 ]. Nevertheless, the lower sensitivity of IGRA among older adults remains a concern. A 2017 study in Japan reported discrepancies between IGRA positivity rates and the corresponding estimated prevalence of TB infection among older adults, highlighting the utility and applicability of IGRA in this population [ 66 ]. Furthermore, providing TPT for older adults remains controversial among clinicians due to the risk of adverse events.

Republic of Korea

In 2020, in the Republic of Korea, there was an estimated TB incidence was 25,000 people with incident TB (49 per 100,000 population per year); 94% of these were notified to the national authorities, of which 49% were aged ≥ 65 [ 12 ]. [ 12 ] While the number of people with TB has steadily declined in the last decade, the proportion of older adults among new TB patients has risen (19.2% in 2001 and 49.1% in 2020) [ 67 ] . In 2020, the TB notification rate among older adults aged 65−69 was 58 per 100,000 population; the rate increased with age and peaked among those aged ≥ 80 years at 235 per 100,000 population [ 67 ]. TB mortality among older adults ≥ 65 was 13.8 per 100,000 population in 2020, the lowest rate recorded since 2001. However, of all TB-related deaths, 82.5% involved those aged ≥ 65, and the proportion has been above 80% since 2016 (2001; 58.0%, 2010; 72.1%). Despite the decline in overall TB burden and mortality rates, the course is predicted to reverse after 2032 (TB incidence) and 2026 (TB deaths) due to increasing trends among older adults, particularly those aged ≥ 80 69 .

The Ministry of Health and Welfare is the national policymaking and governing body for public health and medical services in the Republic of Korea [ 70 ]. The ministry also oversees the national health insurance scheme, a compulsory scheme that confers health care coverage and benefits for the entire population [ 70 ]. The Korean Disease Control and Prevention Agency (KDCA) is responsible for disease surveillance, public health response, disease prevention, and research and oversees the National TB Elimination Project [ 70 ]. While policies regarding TB prevention and care are established by the government, the private sector’s involvement in TB care began in the 1990s and was formalised in 2011 through the public-private mix model [ 71 ].

TB care in the public and private sectors is covered by national health insurance. A 10% co-payment scheme was in place until 2016 [ 71 ]. Since 2017, all expenses incurred during TB care, including hospitalisation, isolation orders (movement restriction to prevent further transmission), and an allowance for dependents, are included in the policy, thereby minimising out-of-pocket payments for people affected by TB. Interventions such as contact investigation and the screening of close contacts for TB disease and infection are also covered by health insurance. While there is specific funding for TB interventions targeted at older adults in the Republic of Korea, the budget only amounted to 1.8% of the total budget for the national TB control program in 2018 [ 71 ].

Since the inception of the national TB control program in the 1960s, early detection and treatment of TB has been a mainstay of TB policies, including for older adults [ 71 ]. In 2020, the TB incidence rate detected through TB screening among older adults ≥ 65 using mobile chest x-ray machines in 17 regions and provinces was 75 per 100,000, 1.9 times higher than the general population’s rate [ 72 ]. This high detection rate prompted KDCA to implement nationwide TB screening among older adults in the community and aged-care facilities as a key strategy to detect TB early in this group.

With an increasingly aging population and the growth in the proportion of women in the labour market, the number of older adults living in institutionalised settings has also increased proportionally, partly due to the introduction of long-term care insurance, which covers home and institutional care services, assistive equipment such as walker and wheelchair, and cash benefits [ 73 ]. In fact, the number of recipients of long-term care insurance increased three-fold between 2008 and 2019 [ 74 ]. Using mobile chest x-ray, TB screening among older adults ≥ 65 in long-term care facilities in 17 regions and provinces resulted in an incidence rate of 66 per 100,000 in 2020 [ 72 ].

Drug-susceptible TB in the Republic of Korea is treated using the standard 6-month regimen [ 75 ]. Partly due to concerns about potential pyrazinamide hepatotoxicity, there is ongoing work to optimise TB treatment among older adults through establishing an adverse events monitoring system, managing adverse events, and developing biomarkers that may predict diagnostic and therapeutic responses [ 76 ].

Overall, the TB treatment success rate has been sitting at approximately > 80% over the last two decades [ 12 ]. However, the treatment success rate among older adults was lower than the general population at approximately 70% between 2012 and 2015 [ 77 ]. Treatment support remains the main treatment administration option. Video-observed therapy and support through a comprehensive support centre for older adults living alone had been trialled with some success. Preliminary data showed higher treatment success rate among older adults enrolled in the programme [ 78 ]. Other incentives have been implemented, such as providing food, daily necessities, and medical accompaniment service for older adults who have trouble remaining on treatment. Furthermore, an approach to assess the vulnerability of people with TB, particularly older adults, with customised case management and linkage with social welfare services have also been implemented.

A comprehensive epidemiologic and contact investigations protocol is in place to screen at-risk populations for both TB disease and infection. TB disease and infection screening is mandatory for healthcare and nursery workers and teachers. Other at-risk groups, such as people living with HIV, those with silicosis, organ transplant recipients, and patients with kidney disease on dialysis, are recommended to be screened for TB infection. For household contacts of persons with TB, chest x-ray investigations and TB infection workups (TST/IGRA, but not mandatory) will be offered (covered by health insurance). TPT (3HP [isoniazid-rifapentine once-weekly for 3 months], 3RH [rifampicin-isoniazid daily for 3 months], or 4R [rifampicin daily for 4 months) is offered to those eligible for it. However, there is no policy to screen older adults for TB infection. In a 2016 study conducted across 11 regions, 40% of those aged ≥60 tested positive for TB infection using IGRA [ 79 ]. Considering the potentially high prevalence of TB infection and risk for reactivation among older adults, [ 80 ] a systematic approach to TB infection screening and TPT administration in this population is warranted.

An estimated 2700 people were affected by TB in 2020 (46 per 100,000 population per year); 89% of these (2400) were notified to the national authorities in 2020. [ 12 ] Of the 2400 people with TB, approximately 57% were Singapore citizens and permanent residents [ 81 ]. The TB incidence rate has been approximately 40 per 100,000 since the late 1990s and may be attributed to migration (about 30−50% of the cases were detected among short and long-term visitors, including work pass holders and students) and an aging population [ 82 , 83 ]. In 2020, older age groups (≥60) made up a significant proportion of the TB notifications both among Singaporean-born (58%) and foreign-born (46%) individuals [ 81 ]. While reactivation of past infections could sustain the TB epidemic among older residents, outbreaks in aged-care facilities have also been reported. Overall, TB mortality rates have remained < 1 per 100,000 population in the past few years [ 12 ]. However, the TB mortality rate among older adults ≥70 was higher at 5.1 per 100,000 population in 2018 [ 84 ].

The Ministry of Health (MOH) Singapore initiated the national TB program, named Singapore TB Elimination Program (STEP), to strengthen TB response efforts by detecting and treating TB disease and infection and preventing drug-resistance TB in 1997 [ 82 ]. STEP manages a notification registry and a treatment surveillance system to monitor TB notification and treatment outcomes. The TB Control Unit (TBCU) is the referral centre for TB management in Singapore, responsible for treatment support, contact investigation, and the administration of TPT. Both public and private hospitals can offer TB diagnosis and treatment. Referrals for TB diagnosis and treatment can be made by primary care physicians in public, private, and community-based organisations (such as the Singapore Anti-Tuberculosis Association).

Singapore’s health system is financed through subsidies, a national health savings account (MediSave), a basic national health insurance scheme (MediShield) for hospitalisation and treatment bills, and an endowment scheme for individuals who have exhausted other means of payment [ 85 ]. For citizens and permanent residents, TB diagnosis and treatment costs are covered by the health financing system. Non-residents, such as those holding long-term work visas, are not covered by these schemes and must rely on employer coverage, private health insurance, or out-of-pocket payment for TB screening and diagnosis costs. TB medications are provided to all at no cost.

TB is diagnosed by passive case finding when a person presents with symptoms or a chest x-ray for other medical conditions (yet suggestive of TB) [ 86 ]. TB screening is conducted for foreigners applying for work or long-term visas and during the visa renewal process using chest radiography [ 86 ]. The same approach applies to everyone regardless of age. There are currently no specific interventions targeting TB screening among older adults. Some aged-care facilities may enquire about the date and information of the last chest x-ray or require a chest x-ray to be conducted before admission. However, this policy is inconsistent.

Treatment of drug-susceptible TB in Singapore follows the standard 6-month regimen (2 months of isoniazid, rifampicin, ethambutol, and pyrazinamide followed by isoniazid and rifampicin for 4 months) [ 86 ]. However, for older adults who may be unlikely to tolerate pyrazinamide, a 9-month regimen comprising ethambutol, rifampicin, and isoniazid for 2 months (intensive phase), followed by rifampicin and isoniazid for 7 months (continuation phase) is used [ 86 ]. It is required by law under the Infectious Diseases Act for all clinicians to report treatment progress and the outcomes of people with TB to the MOH [ 86 ].

The treatment success rate was 79% in 2019, [ 12 ] and the treatment success rate among older adults was not known at the time of writing. Different forms of treatment support have been implemented to support treatment. Outpatient treatment support is implemented at all the 18 public polyclinics and the TBCU. People with TB disease can choose to have their treatment administered at the nearest public health facility. In a 2016 study among adults with TB in Singapore, 72% of respondents reported that they could accept the schedule of facility-based treatment support [ 87 ]. However, 45% perceived the arrangement to be disruptive to their work, school, and social activities [ 87 ]. The results were not age disaggregated. While not specific to older adults, incentive-based intervention to encourage treatment adherence among people with TB from the lower-income bracket has shown promise compared to the non-intervention group in Singapore [ 88 ]. Other forms of treatment support, such as administration in aged-care facilities and through outreach for recipients who have mobility issues or are frail, have been implemented [ 86 ]. Self-administered treatment is also used in Singapore, especially for those treated in the private sector. However, there are no data on self-administered treatment among older adults.

The STEP strategy includes contact investigation to identify close contacts of persons with bacteriologically confirmed TB [ 86 ]. Testing for TB infection is recommended for high-risk groups if there is an intention to treat with TPT [ 86 ]. IGRA is the preferred test for TB infection; variable TPT options are available, including 6 H (isoniazid daily for 6 months), or 4R (daily rifampicin for 4 months) [ 86 ]. Although TB infection has been estimated to affect up to 30% of older adults in Singapore, [ 81 ] there is no policy or specific guideline regarding TPT use for older adults. The risk of adverse events associated with TPT remains a concern, and further studies are needed.

Summary and conclusion

The four case studies presented exemplify the range of practices and challenges in managing TB among older adults in the Western Pacific Region. Here we summarise the key findings and action points based on country experiences:

While passive case finding remains the mainstay intervention contributing to TB case detection in older adults, routine ACF has been implemented in China, Japan, and the Republic of Korea. The optimal use of ACF requires further consideration and assessment in different contexts.

A person-centred approach to TB care was identified as a key theme in all four countries. Nevertheless, limited guidance is available at the country and global levels to address issues older adults encounter (e.g., adverse event monitoring and management, consideration of drug-drug interactions, and treatment adherence). Existing evidence (and evidence gaps) related to the optimal management of older adults with TB require systematic review and guideline development.

There is a lack of standardisation in the approach to TB infection testing and TPT provision. It is important to generate the necessary evidence to inform benefit-risk estimates for managing TB infection among older adults.

The use of traditional medicines is culturally deeply rooted among older adults in this region. Their complementary use in TB care should be explored and carefully considered.

A summary of the approaches implemented by the four countries is outlined in Fig. 2 .

TB management among older adults in China, Japan, the Republic of Korea, and Singapore

TB management experiences were mapped based on three main domains: case finding and detection, treatment, and prevention. The relevant policies and interventions are tagged to the respective implementing countries. Areas shaded in grey represent cross-cutting interventions that extend beyond the older adult population. Treatment support was historically known as directly observed treatment (DOT)

Abbreviations: TB; tuberculosis, TPT; TB preventive treatment, CN; China, JP; Japan, KR; Republic of Korea, SG; Singapore

Cross-cutting challenges

The COVID-19 pandemic has severely disrupted access to TB care. with an 18% decrease in TB disease notifications globally [ 89 ]. Health services for older people were often more severely.

affected and TB services in South-East Asia and the Western Pacific Regions were highly impacted [ 12 ]. Of the four countries included in this study, TB notification data for 2020 and 2021 were lower than 2019 [ 12 , 90 ].

In addition to key risk factors for TB such as living in crowded and poorly ventilated conditions, indoor air pollution and tobacco smoking, [ 91 ] older adults also have age-associated immune dysfunction, financial resources constraints, and frequent comorbidities [ 92 ]. Caring for older adults with TB can be difficult and complex cases may require referral to appropriate specialty care. Service integration between relevant specialties should facilitate better care for older adults with TB. TB diagnosis in older adults is further complicated by challenges related to cognitive impairment, communication challenges and difficulties in obtaining quality sputum samples [ 93 , 94 ]. Dementia, which primarily affects older adults, has been associated with difficult TB diagnosis and poor adherence to TB treatment in the absence of tailored treatment support [ 95 ].

TB-related stigma (perceived and experienced) remains pervasive in many communities [ 96 ]. While systematic assessment of TB-related stigma among older adults has not been conducted, older adults in China have reported unwillingness to seek TB care due to fear of discrimination and self-isolation because of stigma [ 32 , 97 ]. Anecdotally, aged care facilities in Japan were reluctant to accept returning residents after hospital admission for TB care due to fear and stigma, while unwillingness to be screened for TB has also been ascribed to stigma.

The high standards of living and well-being may have shaped the notion that TB is not a significant concern, rendering a lower index of TB suspicion, including among healthcare workers [ 98 ]. A recent cross-sectional study conducted among older adults in Shenzhen, China, reported that > 70% of the respondents were aware of TB as a contagious disease, its symptoms, and preventive measures. Yet, less than half knew that TB was curable, and only one-third were willing to be screened for TB if they were to develop suspicious symptoms [ 99 ].

Considering that aging populations is a global phenomenon that implies a heightened TB risk, particularly in areas with historically high rates of TB, policymakers and funders must invest more to deliver and generate the evidence required to inform optimal TB prevention and care initiatives in older adults.

Data Availability

Data used for this case study comprised published literature, reports, and co-authors’ experience in managing TB among older adults in their respective settings.

Change history

07 june 2023.

A Correction to this paper has been published: https://doi.org/10.1186/s12889-023-15576-0

United Nations Population Division. World Population Prospects. 2019. New York: United Nations Department of Economic and Social Affairs, 2019 (accessed Sept 8, 2021). https://population.un.org/wpp/Graphs/DemographicProfiles/Line/900

World Bank. Population ages 65 and above, total. 2020. (accessed Sept 13, 2021). https://data.worldbank.org/indicator/SP.POP.65UP.TO

The Global Health Observatory. Life expectancy at birth (years). 2020. (accessed Sept 10, 2021). https://www.who.int/data/gho/data/indicators/indicator-details/GHO/life-expectancy-at-birth-(years)

World Bank. Life expectancy at birth, total (years) - Japan. 2020. (accessed Oct 27, 2022). https://data.worldbank.org/indicator/SP.DYN.LE00.IN?locations=JP

World Bank. Population ages 65 and above (% of total population). 2020. (accessed Sept 13, 2021). https://data.worldbank.org/indicator/SP.POP.65UP.TO.ZS?most_recent_value_desc=true

National Institute of Population and Social Security Research. Population projections for Japan (2016–2065). 2017. https://www.ipss.go.jp/pp-zenkoku/e/zenkoku_e2017/pp_zenkoku2017e_gaiyou.html

National Bureau of Statistics of China. The 7th national population census. 2021. (accessed April 1, 2022). http://www.stats.gov.cn/tjsj/tjgb/rkpcgb/qgrkpcgb/202106/t20210628_1818824.html

Department of Economic and Social Affairs, United Nations. World population prospects: population by age and sex (China). 2019. (accessed April 1, 2022). https://population.un.org/wpp/DataQuery/

Korean Statistical Information Service. Projected population by age (Korea). 2021. &tblId=DT_1BPA001&vw_cd=MT_ETITLE&list_id=A41_10&scrId=&language=en&seqNo=&lang_mode=en&obj_var_id=&itm_id=&conn_path=MT_ETITLE&path=%252Feng%252FstatisticsList%252FstatisticsListIndex.do.. https://kosis.kr/statHtml/statHtml.do?orgId=101

Mathers CD, Stevens GA, Boerma T, White RA, Tobias MI. Causes of international increases in older age life expectancy. The Lancet. 2015;385:540–8.

Article Google Scholar

Institute for Health Metrics and Evaluation (IHME). GBD compare data visualizations, Seattle. Washington: IHME, University of Washington, 2021 (accessed Sept 14, 2021). http://vizhub.healthdata.org/gbd-compare

World Health Organization. Global tuberculosis report 2021.Geneva, 2021 https://www.who.int/publications/i/item/9789240037021

Mori T, Leung CC. Tuberculosis in the Global Aging Population. Infect Dis Clin N Am. 2010;24:751–68.

Shaw AC, Joshi S, Greenwood H, Panda A, Lord JM. Aging of the Innate Immune System. Curr Opin Immunol. 2010;22:507–13.

Article CAS PubMed PubMed Central Google Scholar

Uchimura K, Ngamvithayapong-Yanai J, Kawatsu L, et al. Characteristics and treatment outcomes of tuberculosis cases by risk groups, Japan, 2007–2010. Western Pac Surveillance Response Journal: WPSAR. 2013;4:11–8.

Article PubMed PubMed Central Google Scholar

Byrne AL, Marais BJ, Mitnick CD, Lecca L, Marks GB. Tuberculosis and chronic respiratory disease: a systematic review. Int J Infect Dis. 2015;32:138–46.

Article PubMed Google Scholar

Donini LM, ScarDeLLa P, PioMbo L. Malnutrition in elderly: social and economic determinants. J Nutr Health Aging. 2013;17:7.

Bates MN, Khalakdina A, Pai M, Chang L, Lessa F, Smith KR. Risk of Tuberculosis from exposure to Tobacco smoke: a systematic review and Meta-analysis. Arch Intern Med. 2007;167:335–42.

Wang L, Wang W. Temporal Trends in notification and mortality of tuberculosis in China, 2004–2019: a joinpoint and age-period-cohort analysis. Int J Environ Res Public Health. 2021;18:5607.

The Office of the Fifth National TB Epidemiological Survey. The fifth national tuberculosis epidemiological survey 2010. China J Antituberc. 2012;34:485–508.

Google Scholar

Cheng J, Sun Y-N, Zhang C-Y, et al. Incidence and risk factors of tuberculosis among the elderly population in China: a prospective cohort study. Infect Dis Poverty. 2020;9:13.

Amsalu E, Liu M, Li Q, et al. Spatial-temporal analysis of tuberculosis in the geriatric population of China: an analysis based on the bayesian conditional autoregressive model. Arch Gerontol Geriatr. 2019;83:328–37.

Wang L, Xu C, Hu M, et al. Spatio-temporal variation in tuberculosis incidence and risk factors for the disease in a region of unbalanced socio-economic development. BMC Public Health. 2021;21:1817.

World Health Organization. WHO operational handbook on tuberculosis: module 4: treatment: drug-susceptible tuberculosis treatment. Geneva: World Health Organization., 2022 (accessed Jan 25, 2022). https://www.who.int/publications/i/item/9789240048126

Xianyi C, Fengzeng Z, Hongjin D, et al. The DOTS strategy in China: results and lessons after 10 years. Bull World Health Organ. 2002;80:430–6.

PubMed PubMed Central Google Scholar

Hou W-L, Song F-J, Zhang N-X, et al. Implementation and community involvement in DOTS strategy: a systematic review of studies in China [Review article]. Int J Tuberc Lung Dis. 2012;16:1433–40.

Chen S, Zhang H, Pan Y, et al. Are free anti-tuberculosis drugs enough? An empirical study from three cities in China. Infect Dis Poverty. 2015;4:47.

Wei X, Zou G, Walley J, et al. China Tuberculosis Policy at Crucial Crossroads: comparing the practice of different hospital and tuberculosis control collaboration models using Survey Data. PLoS ONE. 2014;9:e90596.

Long Q, Guo L, Jiang W, Huan S, Tang S. Ending tuberculosis in China: health system challenges. The Lancet Public Health. 2021;6:e948–53.

Zhou C, Long Q, Chen J, et al. Factors that determine catastrophic expenditure for tuberculosis care: a patient survey in China. Infect Dis Poverty. 2016;5:6.

Duan W, Zhang W, Wu C, et al. Extent and determinants of catastrophic health expenditure for tuberculosis care in Chongqing municipality, China: a cross-sectional study. BMJ Open. 2019;9:e026638.

Li X, Wang B, Tan D, et al. Effectiveness of comprehensive social support interventions among elderly patients with tuberculosis in communities in China: a community-based trial. J Epidemiol Commun Health. 2018;72:369–75.

Article CAS Google Scholar

World Health Organization. WHO consolidated guidelines on tuberculosis. Module 2: screrening - systematic screening for tuberculosis disease. Geneva: World Health Organization., 2021 (accessed March 9, 2022). https://www.who.int/publications/i/item/9789240001503

Zhang C, Xia L, Rainey JJ, et al. Findings from a pilot project to assess the feasibility of active tuberculosis case finding among seniors in rural Sichuan Province, China, 2017. PLoS ONE. 2019;14:e0214761.

Zhang X-L, Li S-G, Li H-T, et al. Integrating tuberculosis screening into annual health examinations for the rural elderly improves case detection. Int J Tuberculosis Lung Disease: Official J Int Union Against Tuberculosis Lung Disease. 2015;19:787–91.

Li J, Liu X-Q, Jiang S-W, et al. Improving tuberculosis case detection in underdeveloped multi-ethnic regions with high disease burden: a case study of integrated control program in China. Infect Dis Poverty. 2017;6:151.

Office of the State Council. People’s Republic of China. National tuberculosis plan for the 13th five-year plan period. Beijing: Office of the State Council, People’s Republic of China; 2017.

National Health Commission of China. Chinese tuberculosis control and prevention: technical work regulations. Beijing: National Health Commission of China; 2020. http://tb.chinacdc.cn/ggl/202004/P020200414515703939844.pdf

National Ministry of Civil Affairs. Third quarterly report of social welfare institutions 2021. Beijing: National Ministry of Civil Affairs; 2021. http://www.mca.gov.cn/article/sj/tjjb/2021/202103jgqk.html

Hu Y, Hu Y, Ye X. Epidemiological investigation of an outbreak of tuberculosis. Mod Prev Med. 2015;42:982–3.

Wang N, Shewade HD, Thekkur P, et al. Electronic medication monitor for people with tuberculosis: implementation experience from thirty counties in China. PLoS ONE. 2020;15:e0232337.

Abuaku B, Tan H, Li X, Chen M, Huang X. Treatment default and death among tuberculosis patients in Hunan, China. Scand J Infect Dis. 2010;42:281–7.

Tomioka H, Tatano Y, Shimizu T, Sano C. Clinical and basic studies on therapeutic efficacy of Herbal Medicines against Mycobacterial Infections. Med (Basel). 2019;6:E67.

Li X, Li X, Liu Q, et al. Traditional chinese medicine combined with western medicine for the treatment of secondary pulmonary tuberculosis: a PRISMA-compliant meta-analysis. Med (Baltim). 2020;99:e19567.

Wang L, Zhang H, Ruan Y, et al. Tuberculosis prevalence in China, 1990–2010; a longitudinal analysis of national survey data. The Lancet. 2014;383:2057–64.

Huynh GH, Klein DJ, Chin DP, et al. Tuberculosis control strategies to reach the 2035 global targets in China: the role of changing demographics and reactivation disease. BMC Med. 2015;13:88.

Cui X, Gao L, Cao B. Management of latent tuberculosis infection in China: exploring solutions suitable for high-burden countries. Int J Infect Dis. 2020;92:37–40.

Gao L, Bai L, Liu J, et al. Annual risk of tuberculosis infection in rural China: a population-based prospective study. Eur Respir J. 2016;48:168–78.

Article CAS PubMed Google Scholar

Gao L, Zhang H, Xin H, et al. Short-course regimens of rifapentine plus isoniazid to treat latent tuberculosis infection in older chinese patients: a randomised controlled study. Eur Respir J. 2018;52:1801470.

Tuberculosis Surveillance Center. Tuberculosis in Japan: annual report 2021. Tokyo: Department of Epidemiology and Clinical Research, Research Institute of Tuberculosis Japan., 2021 https://jata.or.jp/english/dl/pdf/TB_in_Japan_2021.pdf

Toyota M, Sasaki Y. [The issue of tuberculosis in the elderly in Japan]. Kekkaku. 2010;85:881–94.

PubMed Google Scholar

Seto J, Wada T, Suzuki Y, et al. Mycobacterium tuberculosis Transmission among Elderly Persons, Yamagata Prefecture, Japan, 2009–2015. Emerg Infect Dis. 2017;23:448–55.

Hagiya H, Koyama T, Zamami Y, et al. Trends in incidence and mortality of tuberculosis in Japan: a population-based study, 1997–2016. Epidemiol Infect. 2018;147:e38.

Sakamoto H, Rahman MdM, Nomura S et al. Japan health system review. New Delhi:World Health Organization, Regional Office forSouth-East Asia, 2018 https://apps.who.int/iris/handle/10665/259941

Katsuda N, Hirosawa T, Reyer JA, Hamajima N. Roles of public health centers (hokenjo) in tuberculosis control in Japan. Nagoya J Med Sci. 2015;77:19–28.

Hoshino H, Ohmori M, Uchimura K, Yamauchi Y. [TB epidemiology in an elderly group in Japan]. Kekkaku. 2008;83:423–9.

Kawatsu L, Yoshiyama T, Kato S. Tuberculosis in the Elderly. In: Migliori GB, Raviglione MC, editors. Essential tuberculosis. Cham: Springer International Publishing; 2021. pp. 253–8.

Chapter Google Scholar

Yoshikawa R, Kawatsu L, Uchimura K, Ohkado A. Delay in health-care-seeking treatment among tuberculosis patients in Japan: what are the implications for control in the era of universal health coverage? Western Pac Surveill Response J. 2020;11:37–47.

Hikone M, Ainoda Y, Sakamoto N, Ohnishi K. Clinical characteristics of elderly pulmonary tuberculosis in an acute-care general hospital in Tokyo, Japan: a 12-year retrospective study. J Infect Chemother. 2020;26:245–50.

Nakao M, Sone K, Kagawa Y, et al. Diagnostic delay of pulmonary tuberculosis in patients with acute respiratory distress syndrome associated with aspiration pneumonia: two case reports and a mini-review from Japan. Exp Ther Med. 2016;12:835–9.

Ghotbi N, Nishimura S, Takatsuka N. Japan’s national tuberculosis control strategies with economic considerations. Environ Health Prev Med. 2005;10:213–8.

Ohmori M, Wada M, Mitarai S, et al. [Tuberculosis control in health care facilities for the elderly, from the viewpoint of risk management]. Kekkaku: [Tuberculosis]. 2006;81:71–7.

Hagiwara E, Suido Y, Asaoka M, et al. Safety of pyrazinamide-including regimen in late elderly patients with pulmonary tuberculosis: a prospective randomized open-label study. J Infect Chemother. 2019;25:1026–30.

Takaku T, Saito T, Nemoto K, Oisihi S, Hayashihara K. Comparison of adverse effects in tuberculosis patients over 80 years of age with and without pyrazinamide treatment. Eur Respir J. 2017;50. https://doi.org/10.1183/1393003.congress-2017.PA3045

Kawatsu L, Uchimura K, Ohkado A. Trend and treatment outcomes of latent tuberculosis infection among migrant persons in Japan: retrospective analysis of Japan tuberculosis surveillance data. BMC Infect Dis. 2021;21:42.

Kato S, Ota M, Suenaga M, Hirayama T, Yoshiyama T. Estimation of positive rates of interferon-gamma release assay by age group in Japan. Kekkaku. 2017;92:365–70.

Korea Disease Control and Prevention Agency. Annual report on notified tuberculosis in Korea 2020. Seoul: Korea Disease Control and Prevention Agency, 2021.

Hyewon L, Jinsun K, Hye Kyung I, Youmi K. Characteristics and trends in deaths from tuberculosis in the Republic of Korea 2001–2020. Seoul: Division of Tuberculosis Prevention and Control, Korea Disease Control and Prevention Agency, 2021 &bid=0031&list_no=717657&act=view. https://kdca.go.kr/board/board.es?mid=a30501000000

Jang SY, Kim MJ, Cheong H-K, Oh I-H. Estimating disability-adjusted life years due to tuberculosis in Korea through to the Year 2040. Int J Environ Res Public Health. 2020;17:5960.

Sooman K, Tae-jin L, Chang-yup K. Republic of Korea health system review. Manila: World Health Organization Regional Office for the Western Pacific, 2015 (accessed April 9, 2022). https://apps.who.int/iris/handle/10665/208215

Cho KS. Tuberculosis control in the Republic of Korea. Epidemiol Health. 2018;40:e2018036.

HeeAe K, JeeYeon S, HyeKyung I, EunHye S, WeonJung J, YunSoo K. Results of the 2020 elderly tuberculosis screening in Republic of Korea. Seoul: Korea Disease Control and Prevention Agency; 2021. &bid=0031&list_no=713470&act=view. https://www.kdca.go.kr/board/board.es?mid=a30501000000

Chung Yu L, Euisook K. Long-term care in developing countries: ten case studies (Korea). Geneva: World Health Organization and WHO Collaborating Centre for Health and Long-Term Care Policy and Research; 2003. https://apps.who.int/iris/handle/10665/42769

Ga H. Long-term Care System in Korea. Ann Geriatr Med Res. 2020;24:181–6.

Joint Committee for the Revision of Korean Guidelines for Tuberculosis. Korean guidelines for tuberculosis. 4th ed. Seoul: Korea Disease Control and Prevention Agency; 2020. https://tbzero.kdca.go.kr/tbzero/contents.do

Min J. Study on the optimization of pulmonary TB treatment in elderly 2020–2022. Seoul: Catholic University and the Korean National Institute of Health, 2022 https://cris.nih.go.kr/cris/search/detailSearch.do/17218#step8

Son H, Mok J, Lee M, et al. Status and determinants of treatment outcomes among New Tuberculosis Patients in South Korea: a retrospective cohort study. Asia Pac J Public Health. 2021;33:907–13.

Kim HJ. Enhanced video observed therapy (VOT) to the elderly TB patients: case study of Korea [PowerPoint presentation]. 2021.

Hwang H-J, Park J-H, Nam S, et al. Characteristics of a large latent tuberculous infection screening programme using QuantiFERON®-TB gold in Korea. Int J Tuberc Lung Dis. 2018;22:504–9.

Rajagopalan S. Tuberculosis and aging: A Global Health Problem. Clin Infect Dis. 2001;33:1034–9.

Ministry of Health Singapore. Update on tuberculosis situation in Singapore. Singapore: Ministry of Health Singapore., 2021 https://www.moh.gov.sg/news-highlights/details/update-on-tuberculosis-situation-in-singapore-2022

Loh KS, Hsu LY. Tuberculosis in Singapore: past and future. Ann Acad Med Singap. 2019;48:72–4.

Tam G, Lai SW. Is Singapore on track to eliminate tuberculosis by 2030? A policy case study. SAGE Open Medicine. 2019;7. https://doi.org/10.1177/2050312119851331

Communicable Diseases Division, Ministry of Health Singapore. Communicable diseases surveillance Singapore 2018. Singapore: Ministry of Health Singapore, 2019 https://www.moh.gov.sg/docs/librariesprovider5/diseases-updates/communicable-diseases-surveillance-in-singapore-2018210c9a3beaa94db49299c2da53322dce.pdf

Lim J. Sustainable Health Care Financing: the Singapore experience. Global Policy. 2017;8:103–9.

Ministry of Health Singapore. Prevention, diagnosis, and management of tuberculosis: MOH clinical practice guidelines. Singapore: Ministry of Health Singapore., 2016 https://www.moh.gov.sg/hpp/doctors/guidelines/GuidelineDetails/cpgmed_prevention_diagnosis_management_tuberculosis

Lee HK, Teo SSH, Barbier S, Tang SC, Yeo GH, Tan NC. The impact of direct observed therapy on daily living activities, quality of life and socioeconomic burden on patients with tuberculosis in primary care in Singapore. Proceedings of Singapore Healthcare 2016; 25: 235–42.

Chua AP-G, Lim LK-Y, Ng H, Chee CB-E, Wang YT. Outcome of a grocery voucher incentive scheme for low-income tuberculosis patients on directly observed therapy in Singapore. Singap Med J. 2015;56:274–9.

Migliori GB, Thong PM, Alffenaar J-W, et al. Gauging the impact of the COVID-19 pandemic on tuberculosis services: a global study. Eur Respir J. 2021. 58. https://doi.org/10.1183/13993003.01786-2021

Morishita F. COVID-19 impact on TB notification in the Western Pacific Region. 2022.

Lönnroth K, Jaramillo E, Williams BG, Dye C, Raviglione M. Drivers of tuberculosis epidemics: the role of risk factors and social determinants. Soc Sci Med. 2009;68:2240–6.

Negin J, Abimbola S, Marais BJ. Tuberculosis among older adults–time to take notice. Int J Infect diseases: IJID: official publication Int Soc Infect Dis. 2015;32:135–7.

Sharma M, Onozaki I, Nunn P. TB in older people in Asia: why it is important. Int J Tuberc Lung Dis. 2021;25:521–4.

Hong JM, Lee H, Menon NV, Lim CT, Lee LP, Ong CWM. Point-of-care diagnostic tests for tuberculosis disease. Sci Transl Med. 2022. published online April 6. https://doi.org/10.1126/scitranslmed.abj4124

Bea S, Lee H, Kim JH, et al. Adherence and Associated factors of treatment regimen in drug-susceptible tuberculosis patients. Front Pharmacol. 2021;12:625078.

Courtwright A, Turner AN. Tuberculosis and stigmatization: pathways and interventions. Public Health Rep. 2010;125:34–42.

Wang Y, Feng J, Zhang J, et al. Willingness to seek medical care for tuberculosis and associated factors among the elderly population in Shenzhen: a cross-sectional study. BMJ Open. 2021;11:e051291.

Watts K. Is Australia neglecting the local Topography when it comes to Catastrophic costs and ending tuberculosis? Trop Med Infect Disease. 2018;3:126.

Wang Y, Gan Y, Zhang J, et al. Analysis of the current status and associated factors of tuberculosis knowledge, attitudes, and practices among elderly people in Shenzhen: a cross-sectional study. BMC Public Health. 2021;21:1163.

Download references

Acknowledgements

We would like to thank Jeffery Cutter, Jun Yang Tay, and Shera Tan of the Singapore TB Program for verifying the section on Singapore’s experience.

The author(s) received no specific funding for this work.

Author information

Kerri Viney, Ben Marais, Heejin Kim, Seiya Kato, Yuhong Liu, Catherine W.M. Ong and Tauhid Islam are joint senior authors.

Authors and Affiliations

Saw Swee Hock School of Public Health, National University of Singapore and National University Health System, Singapore, Singapore

Alvin Kuo Jing Teo & Siyan Yi

Faculty of Medicine and Health, University of Sydney, Sydney, NSW, Australia

Alvin Kuo Jing Teo & Ben Marais

World Health Organization, Regional Office for the Western Pacific, Manila, Philippines

Kalpeshsinh Rahevar, Fukushi Morishita, Manami Yanagawa, Kyung Hyun Oh & Tauhid Islam

Division of Infectious Diseases, Department of Medicine, National University Hospital, Singapore, Singapore

Catherine W.M. Ong

Research Institute of Tuberculosis, Anti-Tuberculosis Association, Tokyo, Japan

Takashi Yoshiyama, Akihiro Ohkado, Lisa Kawatsu, Norio Yamada, Kazuhiro Uchimura & Seiya Kato

Korean National Tuberculosis Association, Seoul, Republic of Korea

Youngeun Choi & Heejin Kim

Office of International Cooperation, Innovation Alliance on Tuberculosis Diagnosis and Treatment, Beijing, China

KHANA Center for Population Health Research, Phnom Penh, Cambodia

Center for Global Health Research, Public Health Program, Touro University California, Vallejo, CA, USA

Global Tuberculosis Programme, World Health Organization, Geneva, Switzerland

Kerri Viney

The University of Sydney Institute for Infectious Diseases (Sydney ID) and the Centre of Research Excellence in Tuberculosis (TB-CRE), Sydney, NSW, Australia

Beijing Chest Hospital, Capital Medical University, Beijing, China

Infectious Diseases Translational Research Programme, Department of Medicine, National University of Singapore, Singapore, Singapore

Institute of Health Innovation and Technology (iHealthtech), National University of Singapore, Singapore, Singapore

Division of Infectious Diseases, Department of Medicine, Woodlands Health, Singapore, Singapore

You can also search for this author in PubMed Google Scholar

Contributions

KR, FM, TI, and SY conceptualised the study. AA, TY, AO, LK, NY, KU, YC, ZC, SK, YL, and CWMO contributed to the country experiences reported in this study. AKJT collated the information. AKJT, BM, KV, KR, FM, MY, and KHO wrote the initial draft. All authors critically reviewed the manuscript and approved the final version.

Corresponding author

Correspondence to Kalpeshsinh Rahevar .

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Competing interests.

The authors declare that they have no competing interests. The authors alone are responsible for the views expressed in this article, and they do not necessarily represent the views, decisions, or policies of the institutions with which they are affiliated.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This article has been updated to correct a typo and a missing disclaimer.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Teo, A., Rahevar, K., Morishita, F. et al. Tuberculosis in older adults: case studies from four countries with rapidly ageing populations in the western pacific region. BMC Public Health 23 , 370 (2023). https://doi.org/10.1186/s12889-023-15197-7

Download citation

Received : 28 October 2022

Accepted : 02 February 2023

Published : 21 February 2023

DOI : https://doi.org/10.1186/s12889-023-15197-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Best practices

BMC Public Health

ISSN: 1471-2458

General enquiries: [email protected]

Research Article
Open access
Published: 19 August 2024

Global, regional, and national burden of HIV-negative tuberculosis, 1990–2021: findings from the Global Burden of Disease Study 2021

Shun-Xian Zhang 1 , 2 na1 ,
Feng-Yu Miao 3 na1 ,
Jian Yang 4 ,
Wen-Ting Zhou 5 ,
Shan Lv 2 , 6 ,
Fan-Na Wei 2 , 6 ,
Yu Wang 1 ,
Xiao-Jie Hu 1 ,
Ping Yin 1 ,
Pei-Yong Zheng 1 ,
Ming Yang 1 ,
Mei-Ti Wang 7 ,
Xin-Yu Feng 2 , 6 ,
Lei Duan 2 , 6 ,
Guo-Bing Yang 8 ,
Ji-Chun Wang ORCID: orcid.org/0009-0002-0862-2704 4 &
Zhen-Hui Lu 1

Infectious Diseases of Poverty volume 13 , Article number: 60 ( 2024 ) Cite this article

472 Accesses

Metrics details

Tuberculosis (TB) is a major infectious disease with significant public health implications. Its widespread transmission, prolonged treatment duration, notable side effects, and high mortality rate pose severe challenges. This study examines the epidemiological characteristics of TB globally and across major regions, providing a scientific basis for enhancing TB prevention and control measures worldwide.

The ecological study used data from the Global Burden of Disease (GBD) Study 2021. It assessed new incidence cases, deaths, disability-adjusted life years (DALYs), and trends in age-standardized incidence rates (ASIRs), mortality rates (ASMRs), and DALY rates for drug-susceptible tuberculosis (DS-TB), multidrug-resistant tuberculosis (MDR-TB), and extensively drug-resistant tuberculosis (XDR-TB) from 1990 to 2021. A Bayesian age-period-cohort model was applied to project ASIR and ASMR.

In 2021, the global ASIR for all HIV-negative TB was 103.00 per 100,000 population [95% uncertainty interval (UI): 92.21, 114.91 per 100,000 population], declining by 0.40% (95% UI: − 0.43, − 0.38%) compared to 1990. The global ASMR was 13.96 per 100,000 population (95% UI: 12.61, 15.72 per 100,000 population), with a decline of 0.44% (95% UI: − 0.61, − 0.23%) since 1990. The global age-standardized DALY rate for HIV-negative TB was 580.26 per 100,000 population (95% UI: 522.37, 649.82 per 100,000 population), showing a decrease of 0.65% (95% UI: − 0.69, − 0.57 per 100,000 population) from 1990. The global ASIR of MDR-TB has not decreased since 2015, instead, it has shown a slow upward trend in recent years. The ASIR of XDR-TB has exhibited significant increase in the past 30 years. The projections indicate MDR-TB and XDR-TB are expected to see significant increases in both ASIR and ASMR from 2022 to 2035, highlighting the growing challenge of drug-resistant TB.

Conclusions

This study found that the ASIR of MDR-TB and XDR-TB has shown an upward trend in recent years. To reduce the TB burden, it is essential to enhance health infrastructure and increase funding in low-SDI regions. Developing highly efficient, accurate, and convenient diagnostic reagents, along with more effective therapeutic drugs, and improving public health education and community engagement, are crucial for curbing TB transmission.

Graphical Abstract

Tuberculosis (TB) is an infectious disease caused by Mycobacterium tuberculosis ( Mtb ), primarily affecting the lungs and potentially leading to a chronic, systemic wasting disease. It spreads via airborne droplets through the respiratory tract and remains a significant global public health issue, causing considerable morbidity and mortality [ 1 , 2 , 3 ]. According to the World Health Organization (WHO)'s Global Tuberculosis Report 2023, an estimated 10.6 million people [95% uncertainty intervals (UIs): 9.9, 11.4 million] developed TB in 2022, with an incidence rate of 133.0 per 100,000 population. TB caused 1.30 million deaths worldwide in 2022, including 1.13 million deaths among human immunodeficiency virus (HIV)-negative individuals (95% UI: 1.02, 1.26 million) and 0.17 million deaths among people living with HIV (PLWH) (95% UI: 0.14, 0.19 million) [ 2 ].

TB can cause long-term damage to the lungs and other organs, leading to various sequelae, complications, and comorbidities [ 4 , 5 ]. TB can result in structural lung diseases such as obstructive lung disease, bronchiectasis, lung destruction, and atelectasis, as well as permanent lung function impairment, including restrictive, obstructive, and mixed pulmonary dysfunction [ 4 , 5 ]. Lung function impairment may occur even in asymptomatic patients, increasing all-cause mortality and reducing life expectancy. Additionally, TB has significant physical and psychological impacts on patients [ 5 ].

TB remains a significant contributor to global morbidity and mortality despite being preventable and curable. It is the leading cause of death from a single infectious agent [ 1 , 3 ]. Effective anti-TB drugs, such as rifampicin and isoniazid, have reduced TB incidence globally by 1.9% annually [ 6 ]. However, the rise of HIV and drug-resistant TB strains, along with socioeconomic challenges like poverty, conflict, and natural disasters, have hindered progress, falling short of the WHO’s target reduction rate of 4–5% [ 2 , 7 , 8 ].

In low- and middle-income countries (LMICs), TB diagnosis primarily relies on sputum smear microscopy, detecting only 50–60% of cases [ 9 ]. Although more sensitive diagnostic methods and drug resistance tests, such as next-generation sequencing, are available, their high costs limit widespread application in high-burden countries [ 10 ]. Delays in diagnosis and treatment initiation significantly contribute to TB transmission [ 11 ].

TB treatment regimens, requiring multiple drugs over several months, pose challenges for patients and healthcare systems, particularly in LMICs. The rising incidence of drug-resistant TB necessitates longer, more expensive, and less tolerable treatment courses, exacerbating transmission [ 12 ]. Additionally, the spread of HIV/acquired immune deficiency syndrome (AIDS) complicates TB control, with 6.3% of new TB cases being HIV-positive in 2022, and co-infection rates reaching 50.0% in some sub-Saharan African regions [ 2 ]. If current trends continue, achieving the 2030 Sustainable Development Goals (SDGs) and the WHO End TB Strategy—aiming for a 90% reduction in TB mortality and an 80% reduction in incidence from 2015 levels, while eliminating catastrophic costs for TB-affected households—will be challenging [ 13 ].

A comprehensive understanding of the burden and epidemiological trends of TB is crucial for assessing progress toward TB elimination and guiding the formulation of control policies and strategies. The Global Burden of Disease (GBD) Study 2021, one of the most comprehensive observational epidemiologic studies globally, provides essential data to explore and understand the epidemiological characteristics of HIV-negative TB [ 1 , 14 , 15 ]. The study aims to detail the epidemiologic features of TB, drug-susceptible TB (DS-TB), multidrug-resistant TB (MDR-TB), and extensively drug-resistant TB (XDR-TB) on a global scale and across major geographic regions. The findings underscore the urgency of TB control within the global health framework and provide a scientific basis for developing more effective public health strategies and programs to curb TB transmission.

Date source

The GBD 2021 comprehensively evaluated the burden of diseases, injuries, and risk factors across different age and gender groups globally. It provided data on 371 diseases or injuries and 88 risk factors from 204 countries and territories spanning 1990 to 2021 [ 15 ]. The GBD 2021 project estimated the rates, numbers and percentages change of incidences, prevalences, deaths and disability adjusted life years (DALYs) for diseases, details of these estimated indices were presented in the appendix of the GBD 2021 capstone paper [ 1 ].

The GBD 2021 estimates for the global TB burden have been updated with improved data sources, estimation models, and statistical analysis methods, adhering to the Guidelines for Accurate and Transparent Health Estimates Reporting [ 15 , 16 ]. The Disease-Model Bayesian Meta-Regression (DisMoD-MR) tool (version 2.1), based on the Bayesian Priors, Regularisation, and Trimming (MR-BRT) model, utilizes all available morbidity and mortality data, epidemiological relationships, and spatial relationships to provide consistent disease burden estimates. Detailed information on the design, data collection, and estimation methods for TB and its subtypes in the GBD study 2021 is available elsewhere [ 1 , 15 , 16 ].

For HIV-negative TB, DS-TB, MDR-TB, and XDR-TB, data on annual incident cases, incidence rates, number of deaths, mortality rates, DALY numbers, and DALY rates from 1990 to 2021 were obtained from the Global Health Data Exchange tool, categorized by year, age, gender, region, and country ( https://ghdx.healthdata.org ). Notably, data on the age-standardized mortality rates (ASMRs) and death cases for XDR-TB were available from 1993 [ 1 ]. Therefore, data from 2010 to 2021 were used to analyze the annual changes in age-standardized incidence rates (ASIRs), ASMRs, and age-standardized DALY rates for XDR-TB. For TB, DS-TB, and MDR-TB, data from 1990 to 2021 were used to assess annual average rate changes [ 1 ].

The Socio-demographic Index (SDI) represents the combined level of health-related social and economic conditions in each region. SDI is a composite measure derived from the total fertility rate in women under 25 years, the mean education level in individuals aged 15 years and older, and per capita income. SDI values range from 0.00 to 1.00 and are multiplied by 100. Countries and territories in the GBD 2021 are grouped into five development levels: low (< 0.46), low-middle (0.46–0.60), middle (0.61–0.69), high-middle (0.70–0.81), and high (> 0.81) [ 15 , 16 ].

The GBD 2021 estimated mortality and DALYs for various risk factors and their combinations across different geographical units, providing a standardized and comprehensive assessment of risk exposure and attributable burden [ 15 , 16 ]. Data on ASMR and age-standardized DALY rates due to risk factors were categorized under level 2, including air pollution, alcohol use, child and maternal malnutrition, childhood sexual abuse and bullying, dietary risks, drug use, high body-mass index, high fasting plasma glucose, high low density lipoprotein cholesterol, high systolic blood pressure, intimate partner violence, kidney dysfunction, low bone mineral density, low physical activity, non-optimal temperature, occupational risks, other environmental risks, tobacco use, unsafe sex, unsafe water, sanitation and hand washing [ 15 , 16 ].

Case definition

The classification of TB follows the International Statistical Classification of Diseases and Related Health Problems (ICD). This includes all forms of TB, both pulmonary and extrapulmonary, whether bacteriologically confirmed or clinically diagnosed. The relevant ICD-10 codes for TB are A10–A14, A15–A19.9, B90–B90.9, K67.3, K93.0, M49.0, N74.1, P37.0, and U84.3. The corresponding ICD-9 codes are 010–019.9, 137–137.9, 138.0, 138.9, 320.4, and 730.4–730.6. For HIV-associated TB, the ICD-10 code is B20.0 [ 1 , 15 ].

DS-TB is defined as TB that is susceptible to isoniazid and rifampicin. MDR-TB without extensive drug resistance is defined as TB that is resistant to isoniazid and rifampicin but not resistant to any fluoroquinolone or second-line injectable drug. XDR-TB is defined as TB that is resistant to isoniazid and rifampicin, any fluoroquinolone, and at least one second-line injectable drug [ 1 , 2 , 16 ].

Statistical analysis

The disease burden of TB, DS-TB, MDR-TB, and XDR-TB among HIV-negative individuals was quantified by ASIRs, ASMRs, and age-standardized DALYs, along with the numbers of incidence, death, and DALY. Age-standardized rates (ASRs), specific rates for various age groups, and corresponding numbers were extracted from the GBD 2021 database, represented as estimated values with 95% UIs [ 17 ]. The formula for calculating the ASR is:

where \(a_{i}\) the age-specific rate in the \(i\) th age group and \(w_{{\text{i}}}\) represents the number of persons (or the weight) in the same age group among the GBD 2021 standard population. \(N\) is the number of age groups. 95% UIs were defined as the 2.5th and 97.5th percentile values of the ordered 1000 draws.

The percentage change in incidence, death, and DALY numbers and rates from 1990 to 2021 was calculated using the equation [ 18 ]:

Percentage changes = (value behind −value before )/value before × 100%. The GBD 2021 database used UIs instead of precise statistical values. Consequently, when comparing two numerical values (numbers, rates, or percentages), statistical significance could not be directly calculated. If the UIs overlapped, it indicated no significant difference ( P > 0.05). Conversely, if the UIs did not overlap, a statistical difference existed ( P < 0.05).

Smoothing spline models were used to evaluate the relationship between the ASRs (ASIRs, ASMRs, and age-standardized DALY rates) of HIV-negative TB, DS-TB, MDR-TB, and XDR-TB and the SDI across 21 geographical regions and 204 countries and territories. Locally Weighted Scatterplot Smoothing was applied to fit the splines, automatically determining the degree, number, and location of knots based on the data and the span parameter. Spearman's rank correlation coefficient was used to verify the correlations between ASRs and SDI. A P -value of less than 0.05 was considered statistically significant.

For the ASIRs, ASMRs, and age-standardized DALY rates from 1990 to 2021, the estimated annual percentage changes (EAPCs) were calculated to depict trends in HIV-negative TB, DS-TB, MDR-TB, and XDR-TB using a linear regression model \({\text{ln}}(ASR) = \alpha + \beta x + \varepsilon\) \(x\) signifies the calendar year, and \(\varepsilon\) denotes an independent, normally distributed error term [ 19 ]. Then, the EAPC is equal to \(100 \times (e^{\beta } - 1)\) , the EAPCs and their 95% confidence intervals ( CI s) are utilized to describe trends over specified time intervals [τ j—1 , τ j ]. If the upper limit of the EAPC (95% CI ) is less than zero, the rate exhibits a statistically significant decline over the observed period. Conversely, if the lower limit of the EAPC (95% CI ) is greater than zero, the rate shows a statistically significant increase. When the 95% CI of EAPC includes zero, the change of the ASRs is considered statistically non-significant, indicating no meaningful trend. Two-tailed tests were used for all statistical assessments, and the significance level was set at P < 0.05.

The Bayesian age-period-cohort model (BAPC, with the default parameters) examined the multiplicative effects of age, period, and cohort [ 20 , 21 ]:

\(n_{ij} = \mu + \alpha_{i} + \beta_{j} + \gamma_{k}\) In the model, \(n_{ij}\) stand for the ASR, \(\mu\) denotes the intercept, and \(\alpha_{i}\) and \(\gamma_{k}\) were age, period, and cohort effects, respectively. BAPC model was implemented with the integrated nested Laplace approximation (INLA) and BAPC packages in R software [ 20 , 21 ].

All statistical analyses were conducted using R software (version 4.4.1. R Foundation for Statistical Computing, Vienna, Austria, https://cran.r-project.org ).

Incidence and temporal trend

In 2021, the global ASIR for all HIV-negative TB was 103.00 per 100,000 population (95% UI: 92.21, 114.91 per 100,000 population), reflecting a percentage change of − 0.40% (95% UI: − 0.43, − 0.38%) compared to 1990. The ASIR for DS-TB was 97.29 per 100,000 population (95% UI: 85.79, 110.48 per 100,000 population), with a decline of − 0.43% (95% UI: − 0.47, − 0.41%) compared to 1990. For MDR-TB, the ASIR was 5.42 per 100,000 population (95% UI: 3.17, 9.34 per 100,000 population), showing an increase of 4.09% (95% UI: 0.99, 12.15%) compared to 1990. XDR-TB had an ASIR of 0.29 per 100,000 population (95% UI: 0.21, 0.42 per 100,000 population), with a percentage change of 0.03% (95% UI: − 0.25, 0.47%) compared to 1990 (Table 1 ). In addition, the EAPC for ASIR from 1990 to 2021 were − 1.91 (95% CI: − 2.01, − 1.82) for TB, − 2.40 (95% CI: − 2.21, − 1.97) for DS-TB, and 2.05 (95% CI: 0.58, 3.54) for MDR-TB, respectively, from 1990 to 2021.

In 2021, the global incidence of TB was 8.41 million cases (95% UI: 7.52, 9.39 million), with DS-TB accounting for 7.94 million cases (95% UI: 7.01, 9.02 million). MDR-TB accounted for 0.44 million cases (95% UI: 0.26, 0.77 million), and XDR-TB accounted for 24,036 cases (95% UI: 17,144, 34,587 persons. Additional file 1 : Table S1).

In 2021, the ASIR for all HIV-negative TB was 115.34 per 100,000 population (95% UI: 103.71, 128.58 per 100,000 population) in males, declining by − 0.39% (95% UI: − 0.42, − 0.36%) compared to 1990. For females, the ASIR was 91.96 per 100,000 population (95% UI: 81.49, 102.62 per 100,000 population) in female, declining by − 0.42% (95% UI: − 0.45, − 0.40%) compared to 1990 (Table 1 ). The ASIR for TB, DS-TB, MDR-TB, and XDR-TB showed no significant differences between males and females ( P > 0.05). Compared to 1990, the ASIR of TB and DS-TB declined in both genders by 2021. However, the ASIR of MDR-TB increased in both genders in 2021. For XDR-TB, the ASIR did not show a significant trend in either direction in males or females in 2021 compared to 2010 (all P > 0.05. Table 1 ).

In 2021, the ASIR of TB, DS-TB, and MDR-TB were highest in low SDI regions and lowest in high SDI regions. Conversely, the ASIR of XDR-TB was highest in high-middle SDI countries and lowest in high SDI regions (Table 1 ). Over the past 30 years, the ASIR of TB and DS-TB declined across all SDI categories (Additional file 1 : Table S2). For MDR-TB, the ASIR declined in high SDI regions (EAPC = − 3.33, 95% CI: − 4.30, − 2.26). Conversely, ASIR increased in low-middle SDI (EAPC = 6.02, 95% CI: 3.61, 6.49) and low SDI (EAPC = 4.39, 95% CI: 2.42, 6.40) regions. XDR-TB demonstrated an increase in ASIR across all SDI categories, regions with lower SDI values exhibited greater increases in ASIR. For example, low SDI regions had the highest increase in ASIR (EAPC = 15.30, 95% CI: 10.90, 19.67. Additional file 1 : Table S2).

In 2021, the ASIR of TB and DS-TB was highest in Southern sub-Saharan Africa and lowest in high-income North America. The highest ASIR of MDR-TB was observed in Eastern Europe, while the lowest was in high-income North America. Similarly, the ASIR of XDR-TB was highest in Eastern Europe (Table 1 ). From 1990 to 2021. The ASIR of TB and DS-TB showed a global decline across all 21 regions, with the most significant reductions observed in Andean Latin America ( P < 0.05). However, the ASIR of MDR-TB increased in several regions, including Southeast Asia, Oceania, Central Asia, Eastern Europe, Central Latin America, Tropical Latin America, North Africa and the Middle East, South Asia, and various sub-Saharan African regions (all P < 0.05). The most substantial increase was in Central Asia ( P < 0.05), while high-income North America experienced the largest decrease in ASIR of MDR-TB ( P < 0.05. Table 1 ).

In 2021, Somalia had the highest ASIR of TB, DS-TB, and MDR-TB, and the Republic of Moldova had the highest ASIR of XDR-TB (Additional file 1 : Table S3). Compared to 1990, the ASIR of TB and DS-TB decreased in all countries and regions except for the Philippines (all P < 0.05. Additional file 1 : Table S3). Significant increases in ASIR of MDR-TB were observed in Kyrgyzstan. In addition, the most substantial increase in ASIR of XDR-TB was recorded in Papua New Guinea in 2021 compared to 2010 ( P < 0.05. Additional file 1 : Table S3).

Death and temporal trend

In 2021, the global ASMR of TB was 13.96 per 100,000 population (95% UI: 12.61, 15.72 per 100,000 population), reflecting a decline of – 0.44% (95% UI: − 0.61, − 0.23%) compared to 1990. The ASMR for DS-TB, MDR-TB, and XDR-TB in 2021 were 12.58 per 100,000 population (95% UI: 10.91, 14.42 per 100,000 population), 1.28 per 100,000 population (95% UI: 0.50, 2.53 per 100,000 population), and 0.09 per 100,000 population (95% UI: 0.04, 0.18 per 100,000 population), respectively. DS-TB showed a decline in ASMR of 0.49% (95% UI: − 0.63, − 0.28%) compared to 1990, while the ASMR of MDR-TB increased substantially by 36.08% (95% UI: 5.11, 208.98%) compared to 1990, and the ASMR of XDR-TB exhibited a percentage change of 2.24% (95% UI: − 0.26, 9.60%) compared to 2010 (Table 2 ). The EAPC for ASMR from 1990 to 2021 were − 3.53 (95% CI: − 3.71, – 3.36%) for TB, − 3.79% (95% CI: − 3.92, − 3.67) for DS-TB, and 1.18 (95% CI: − 0.43, 2.82) for MDR-TB, respectively.

In 2021, TB resulted in 1.16 million deaths (95% UI: 1.05, 1.31 million), DS-TB accounted for 1.05 million deaths (95% UI: 0.91, 1.20 million), MDR-TB caused 0.11 million deaths (95% UI: 0.04, 0.21 million), and XDR-TB led to 7946 deaths (95% UI: 3326, 14,859 persons. Additional file 1 : Table S4).

In 2021, the ASMR for TB in males was 18.19 per 100,000 population (95% UI: 16.16, 21.80 per 100,000 population), reflecting a decline of − 0.65% (95% UI: − 0.72, − 0.47%) compared to 2021. In females, the ASMR was 10.22 per 100,000 population (95% UI: 9.28, 11.33 per 100,000 population), with a decline of – 0.66% (95% UI: − 0.71, − 0.62%) compared to 2021. ASMR for TB, DS-TB were higher in males than in females (all P < 0.05), whereas the ASMR for MDR-TB and XDR-TB showed no significant differences between genders (all P > 0.05). Compared to 1990, the ASMR for TB and DS-TB generally declined in both males and females by 2021. However, the ASMR for MDR-TB increased in both genders in 2021 (Table 2 ).

In 2021, the ASMR for TB, DS-TB, and MDR-TB were highest in low SDI regions, with TB and DS-TB rates significantly exceeding those in other regions. The ASMR for XDR-TB was highest in low-middle SDI countries. Over the past 30 years, the ASMR for TB and DS-TB declined across varying SDI categories (Additional file 1 : Table S2). For MDR-TB, declines in ASMR were also noted in high SDI regions (EAPC = − 6.13, 95% CI: − 6.71, − 5.08), high-middle SDI regions (EAPC = − 3.67, 95% CI: − 5.42, − 1.88), and middle SDI regions (EAPC = − 2.51, 95% CI: − 3.70, − 1.31). In contrast, increases in ASMR were observed in low-middle SDI regions (EAPC = 4.80, 95% CI: 2.19, 7.48) and low SDI regions (EAPC = 3.61, 95% CI: 1.49, 5.77). For XDR-TB, the ASMR exhibited significant increase in high-middle SDI regions (EAPC = 4.53, 95% CI: 1.36, 7.60), middle SDI regions (EAPC = 3.50, 95% CI: 1.37, 5.68), low-middle SDI regions (EAPC = 9.78, 95% CI: 6.57, 13.09), and low SDI regions (EAPC = 10.88, 95% CI: 7.65, 14.19. Additional file 1 : Table S2).

In 2021, the overall ASMR due to TB was highest in Central sub-Saharan Africa and lowest in high-income North America. The ASMR for DS-TB was also highest in Central sub-Saharan Africa and lowest in Western Europe. The ASMR for MDR-TB peaked in South Asia and was lowest in high-income North America, while the highest ASMR for XDR-TB was observed in Central Asia.

Comparing 2021 to 1990, the ASMR of TB and DS-TB decreased across all 21 global geographical regions, with the most rapid declines observed in East Asia and the slowest in Southern sub-Saharan Africa. However, the ASMR MDR-TB increased in Oceania, Central Asia, South Asia, Eastern Europe, Tropical Latin America, Eastern sub-Saharan Africa, Southern sub-Saharan Africa, and Western sub-Saharan Africa (all P < 0.05), while it decreased in Western Europe and high-income North America (all P < 0.05) (Table 2 ).

In 2021, the highest ASMR for TB was recorded in the Central African Republic, which also had the highest ASMR for DS-TB. The highest ASMR for MDR-TB was observed in Somalia, while Mongolia had the highest ASMR for XDR-TB (Additional file 1 : Table S3). Compared to 1990, the ASMR for TB in 2021 did not significantly increase in most countries and regions, with many showing a declining trend (all P < 0.05. Additional file 1 : Table S3). However, the decline was slowest in Lesotho. The ASMR for DS-TB decreased in many countries but increased in several countries and territories, with notable rises in Zimbabwe. The highest increases in the ASMR for MDR-TB were observed in Somalia ( P < 0.05. Additional file 1 : Table S3).

DALY and temporal trend

In 2021, the global age-standardized DALY rate for all TB was estimated at 580.26 per 100,000 population (95% UI: 522.37, 649.82 per 100,000 population), with a declining by − 0.65% (95% UI: − 0.69, − 0.57%) compared to 1990. For DS-TB, the age-standardized DALY rate was 526.03 per 100,000 population (95% UI: 457.25, 596.46 per 100,000 population), declining by – 0.68% (95% UI: − 0.72, − 0.60%) compared to 1990. MDR-TB had an age-standardized DALY rate of 50.76 per 100,000 population (95% UI: 21.28, 99.37 per 100,000 population), with an increase of 2.68% (95% UI: 0.53, 7.70%) compared to 1990. XDR-TB showed an age-standardized DALY rate of 0.58 per 100,000 population (95% UI: 0.19, 1.39 per 100,000 population), with a percentage change of − 0.25% (95% UI: − 0.47, 0.05%) compared to 1990. The EAPC of the age-standardized DALY rate of TB, DS-TB, and MDR-TB were -3.50 (95% CI: − 3.67, − 3.32), − 3.75 (95% CI: − 3.87, − 3.62), and 1.36 (95% CI: − 0.30, 3.04) from 1990 to 2021, respectively.

In 2021, TB caused a total of 46.98 million DALYs (95% UI: 42.48, 52.46 million), with DS-TB contributing 42.56 million DALYs (95% UI: 36.84, 48.24 million), MDR-TB contributing 4.13 million DALYs (95% UI: 1.71, 8.06 million), and XDR-TB contributing 0.29 million DALYs (95% UI: 0.12, 0.53 million. Additional file 1 : Table S5).

In 2021, the age-standardized DALY rate of all HIV-negative TB in males was 705.23 per 100,000 population (95% UI: 623.40, 834.06 per 100,000 population), with a decline of -0.63% (95% UI: − 0.69, − 0.47%) compared to 1990 (Table 3 ). For females, the age-standardized DALY rate for all HIV-negative TB was 463.18 per 100,000 population (95% UI: 417.49, 512.9 per 100,000 population), with a decline of − 0.67% (95% UI : − 0.71, − 0.63%) compared to 1990 (Table 3 ). In 2021, the age-standardized DALY rate caused by TB was higher in males compared to females ( P < 0.05). Specifically, the age-standardized DALY rate for DS-TB was higher in males than that of in females ( P < 0.05), while the rates for MDR-TB and XDR-TB showed no significant differences between genders ( all P > 0.05). Compared to 1990, the age-standardized DALY rate for TB and DS-TB in both males and females showed a declining trend in 2021 (all P < 0.05), with no difference in the rate of decline between genders ( P > 0.05). However, the age-standardized DALY rates for MDR-TB increased in both males and females (all P < 0.05), with no significant difference in the rate of increase between genders ( P > 0.05. Table 3 ).

In 2021, the age-standardized DALY rate of TB, DS-TB, and MDR-TB was highest in regions with low SDI, followed by low-middle SDI regions. The age-standardized DALY rates of TB and DS-TB in Low SDI areas were significantly higher than in other regions. The age-standardized DALY rate of XDR-TB was highest in low-middle SDI region, followed by low SDI region. Over the past 30 years, the age-standardized DALY rate for TB and DS-TB showed a significant decline across five SDI regions (all P < 0.05. Additional file 1 : Table S2). For MDR-TB, the age-standardized DALY rate declined in high SDI regions (EAPC = – 5.98, 95% CI: − 6.99, − 4.96), high-middle SDI regions (EAPC = − 3.30, 95% CI: − 5.15, − 1.40), and middle SDI regions (EAPC = − 2.34, 95% CI: − 5.15, − 1.40). However, increases were observed in low-middle SDI regions (EAPC = 4.60, 95% CI: 2.02, 7.24) and low SDI regions (EAPC = 3.25, 95% CI: 1.13, 5.42). XDR-TB experienced an increase in the age-standardized DALY rate across five SDI regions, regions with lower SDI values showed the greatest increases in age-standardized DALY rates. For instance, low SDI regions had the highest increase (EAPC = 15.30, 95% CI: 10.90, 19.67) in in age-standardized DALY rates (Additional file 1 : Table S2).

In 2021, the age-standardized DALY rates for TB, DS-TB, and MDR-TB were highest in Central sub-Saharan Africa. The lowest age-standardized DALY rates for TB and DS-TB were observed in Australasia, while the lowest rates for MDR-TB were in high-income North America. The highest age-standardized DALY rates for XDR-TB were found in Southeast Asia. Compared to 1990, the age-standardized DALY rates for TB and DS-TB decreased across all 21 global regions by 2021, with the largest declines observed in the high-income Asia Pacific (all P < 0.05. Table 3 ). Trends for MDR-TB varied, with the highest increases in age-standardized DALY rates observed in Oceania ( P < 0.05. Table 3 ), and most significant decreases in high-income North America ( P < 0.05. Table 3 ).

In 2021, the Central African Republic had the highest age-standardized DALY rates for TB and DS-TB. Somalia recorded the highest age-standardized DALY rates for MDR-TB, while Mongolia had the highest rates for XDR-TB. Compared to 1990, the age-standardized DALY rates for TB in 2021 did not significantly increase globally ( P > 0.05), with many countries showing a declining trend. However, the decrease in Lesotho is the smallest. The age-standardized DALY rate for DS-TB increased in Zimbabwe but decreased in other regions. The fastest increases in age-standardized DALY rates for MDR-TB were observed in Somalia (Additional file 1 : Table S3).

Age-gender characteristics

In the 80–84 age group and above, the specific incidence rates of TB and DS-TB are higher in males than in females ( P < 0.05), with no significant differences observed in other age groups (all P > 0.05). The specific incidence rates for MDR-TB and XDR-TB show no differences between males and females across all age groups (all P > 0.05. Fig. 1 A-D).

The specific incidence of TB, DS-TB, MDR-TB, and XDR-TB showed notable differences across age and gender distribution in 2021 year ( A Incidence of TB, B Incidence of DS-TB. C Incidence of MDR-TB. D Incidence of XDR-TB. DS-TB drug-susceptible tuberculosis, MDR-TB multidrug-resistant tuberculosis without extensive drug resistance, TB Tuberculosis, XDR-TB extensively drug-resistant tuberculosis)

In the 30–34 age group and above, the specific mortality rate of TB is higher in males than in females ( P < 0.05), with no significant differences in other age groups (all P > 0.05). In the 25–29 age group and those aged 30 and above, the specific mortality rate of DS-TB is higher in males than in females (all P < 0.05). The specific mortality rates for MDR-TB and XDR-TB show no differences between males and females across all age groups (all P > 0.05. Additional file 1 : Fig.S1 A-D).

In the 25–29 age group and above, the specific DALY rates for TB and DS-TB are higher in males than in females (all P < 0.05), with no significant differences in other age groups (all P > 0.05). The specific DALY rates for MDR-TB and XDR-TB show no differences between males and females across all age groups (all P > 0.05. Additional file 1 : Fig.S2 A-D).

Association between ASRs and SDI

In 2021, across 204 countries and territories, the ASIR, ASMR, and age-standardised DALY rate for TB exhibited a significant negative correlation with the SDI, demonstrating a rapid decline as SDI increased (ASIR: r = − 0.799, P < 0.001; ASMR: r = − 0.857, P < 0.001; DALYs: r = − 0.862, P < 0.001. Respectively. Additional file 1 : Fig. S3 A–C). Similarly, for DS-TB, a comparable trend was observed, with substantial decreases in ASIR ( r = − 0.804, P < 0.001), ASMR ( r = − 0.859, P < 0.001), and age-standardized DALY rates ( r = − 0.866, P < 0.001) accompanying increases in SDI (Additional file 1 : Fig. S4A–C). In addition, a pronounced negative correlation was observed between the ASIR, ASMR, and age-standardised DALY rate for MDR-TB and SDI (ASIR: r = − 0.642, P < 0.001; ASMR: r = − 0.782, P < 0.001; DALYs: r = − 0.788, P < 0.001. Respectively. Additional file 1 : Fig. S5 A-C). For XDR-TB, significant inverse associations were also noted with SDI, as evidenced by the trends in ASIR ( r = − 0.485, P < 0.001), ASMR ( r = − 0.611, P < 0.001), and age-standardized DALY rates ( r = − 0.625, P < 0.001. Additional file 1 : Fig. S6 A–C).

From 1990 to 2021, the ASIR ( r = − 0.865, P < 0.001), ASMR ( r = − 0.872, P < 0.001), and age-standardized DALY rate ( r = − 0.865, P < 0.001) for TB declined rapidly with increases in the SDI globally. However, these trends varied across different GBD regions. TB incidence rates modestly increased with rising SDI in Central and Southern sub-Saharan Africa, as well as Southeast Asia. Conversely, TB mortality and age-standardized DALY rate showed rapid declines in Central sub-Saharan Africa and South Asia as SDI increased (Additional file 1 : Fig.S7 A–C). The trends in ASIR ( r = − 0.813, P < 0.001), ASMR ( r = − 0.886, P < 0.001), and age-standardized DALY rate ( r = − 0.879, P < 0.001) for DS-TB mirrored those of TB (Additional file 1 : Fig. S8 A–C). For MDR-TB, the ASIR ( r = − 0.504, P < 0.001), ASMR ( r = − 0.686, P < 0.001), and age-standardized DALY rate ( r = − 0.674, P < 0.001) decreased slowly with rising SDI. However, the ASIR of MDR-TB in South Asia showed a continuous increase with rising SDI. The ASIR, ASMR, and DALY rates for MDR-TB initially rose sharply with increasing SDI, peaked, and then declined rapidly as SDI continued to rise in East Asia, Eastern Europe, and Central sub-Saharan Africa (Additional file 1 : Fig. S9 A–C). The ASIR and ASMR of XDR-TB showed a rapid upward trend with increasing SDI in Central sub-Saharan Africa and South Asia. In Eastern Europe, Central Asia, and East Asia, the ASMR and age-standardized DALY rate for XDR-TB rose swiftly with increasing SDI, reached a peak, and then declined sharply with further increases in SDI (Additional file 1 : Fig. S10 A–C).

Risk factors for ASMRs and age-standardized DALY rates

From 1990 to 2021, the major risk factors contributing to the ASMR and age-standardized DALY rate for TB were ranked as follows: tobacco use, high fasting plasma glucose, high body mass index, dietary risks, and low physical activity. Notably, the proportion attributed to tobacco use has been steadily declining, whereas the proportions due to high fasting plasma glucose and high body mass index have been increasing (Fig. 2 A, B). The risk factors and trends for DS-TB and MDR-TB mirrored those of TB (Fig. 2 C–F). For XDR-TB over the same period, the risk factors for ASMR and age-standardized DALY rate were similarly ranked (Additional file 1 : Fig. S11 A, B). However, the contribution of tobacco to the ASMR and DALY rates for XDR-TB initially rose gradually from 1990 to 2005 and then declined rapidly thereafter. Meanwhile, the proportions due to high fasting plasma glucose and high body mass index have continued to rise (Additional file 1 : Fig. S11 A, B).

The association between risk factors and the age-standardized mortality rate and DALY rate of TB, DS-TB, and MDR-TB in 21 GBD regions from 1990 to 2021 ( A mortality rate of TB. B DALY rate of TB. C mortality rate of DS-TB. D DALY rate of DS-TB. E mortality rate of MDR-TB. F DALY rate of MDR-TB. DALYs disability-adjusted life years, DS-TB drug-susceptible tuberculosis, GBD Global Burden of Disease, MDR-TB multidrug-resistant tuberculosis without extensive drug resistance, TB Tuberculosis)

The contributions of dietary risks and low physical activity to the ASMR and age-standardized DALY rates for TB, DS-TB, MDR-TB, and XDR-TB have remained relatively unchanged and consistently low (Fig. 2 A–F. Additional file 1 : Fig. S11 A, B).

Projecting disease burden

By 2035, the projected ASIR for TB is 76.76 per 100,000 population (95% CI: 69.61, 83.99 per 100,000 population), and the ASMR is 8.70 per 100,000 population (95% CI: 7.69, 9.70 per 100,000 population). Between 2022 and 2035, its projections indicate declining trends for DS-TB, while MDR-TB and XDR-TB are expected to see significant increases in both ASIR and ASMR, highlighting the growing challenge of drug-resistant TB (Table 4 . Additional file 1 : Table S6).

The study is the first to use the GBD 2021 database to assess the burden and trends of HIV-negative TB over the past 30 years. Significant differences in ASIR, ASMR, and age-standardized DALY rates for HIV-negative TB, DS-TB, MDR-TB, and XDR-TB were observed across different countries and regions. The ASIR, ASMR, and age-standardized DALY rate for HIV-negative TB have all decreased from 1990 to 2021, indicating a global decline in the overall burden. However, TB remains a persistent threat in sub-Saharan Africa, Southeast Asia, and Eastern Europe, particularly in low SDI regions. In addition, the ASIR and ASMR of MDR-TB and XDR-TB have increased in recent years, highlighting drug-resistant TB as a severe global public health issue. These findings provide essential technical support and decision-making evidence for governments worldwide to formulate key TB control measures and to plan and allocate health resources effectively. These findings supports the development of national health plans, rational allocation of medical institutions, human resources, equipment, and funding, and the concentration of resources on priority health issues to achieve greater cost-effectiveness and social impact.

Efficient and innovative diagnostic technologies and strategies are needed to control the spread of TB

Significant progress has been made in global TB control over the past decades, aligned with the SDGs to end the TB epidemic. However, the decline in TB incidence remains disappointing, with one in three TB patients undiagnosed and many not receiving timely diagnosis and appropriate treatment [ 22 ]. In high-burden countries, case detection and treatment success rates are still alarmingly low [ 1 , 2 ]. Of greater concern is the rising incidence and mortality of MDR-TB and XDR-TB globally, driven primarily by transmission between individuals rather than by the mutation of DS-TB strains due to inadequate treatment [ 23 ]. Early detection and standardized treatment of new TB cases, particularly MDR-TB and XDR-TB, are crucial for accelerating recovery and curbing the community transmission of drug-resistant TB [ 24 ].

Effective TB control requires significant breakthroughs across various fronts. Key areas include developing highly sensitive and specific rapid diagnostic tools, creating more effective drugs for both drug-susceptible and drug-resistant TB strains, and innovating more effective vaccines [ 2 ]. Governments must provide substantial funding, technology, and healthcare services to transition from traditional Directly Observed Treatment, Short-Course (DOTS) passive case detection to proactive case identification in high-burden areas [ 2 ]. Comprehensive health education, care, and medication should be provided to the most vulnerable populations, integrated with other health services, particularly HIV/AIDS services, to deliver efficient and holistic healthcare. Ensuring adequate health surveillance at primary healthcare and community health service institutions is also crucial [ 25 , 26 ].

To effectively control TB, early and accurate case detection, prompt initiation, and adherence to effective treatment are crucial for breaking the transmission chain [ 27 ]. In high-burden countries, many TB patients are asymptomatic, and treating infections only after symptoms appear is insufficient to significantly reduce community transmission and incidence rates [ 2 ]. Proactive strategies are needed to address health system barriers to TB control, including routine screening of household contacts of TB patients and shifting from empirical detection based on clinical symptoms to active case detection through sputum smear and culture [ 28 ].

Advances in diagnostic technology have made early TB diagnosis more feasible. Tools like the molecular diagnostic GeneXpert MTB /RIF reduce TB diagnosis time from 1–2 weeks to a few hours [ 29 ]. Second-generation sequencing technology is also time-efficient, providing diagnostic directions in cases of co-infection with rare or multiple pathogens [ 30 ]. The emergence of novel biomarkers, specifically TB-specific host biomarkers and Mtb biomarkers, allows the development of evaluation models that enable rapid, accurate, and effective monitoring of Mtb infection and TB treatment efficacy [ 31 , 32 , 33 ]. Future efforts must focus on developing affordable, accessible, highly sensitive, and specific screening methods and indicators. Continuously optimizing screening strategies and proactively identifying TB-infected patients are essential to effectively control the spread of the TB epidemic.

Effective TB control requires low-toxicity anti-TB drugs, effective vaccines, and an efficient primary healthcare system

The study confirms previous findings [ 13 ], indicating that the TB burden is more severe among males than females, with TB incidence rates being 50% higher and mortality rates 100% higher in males in many countries. Across all levels of sociodemographic development, the TB mortality rate for HIV-negative males over 30 years of age consistently surpasses that for HIV-negative females. This underscores the importance of considering gender factors in TB epidemiology and calls for gender-sensitive public health interventions.

The study found that all ASRs of HIV-negative TB were inversely related to the SDI, with high-SDI regions exhibiting low ASIR, ASMR, and age-standardized DALY rates for DS-TB, MDR-TB, and XDR-TB. Conversely, the ASIR, ASMR, and age-standardized DALY rates for MDR-TB and XDR-TB remain high in low-SDI regions, particularly in Central sub-Saharan Africa and South Asia, where these rates have been persistently rising. This highlights the ongoing public health challenge in these areas and the need for targeted interventions.

The WHO-recommended strategies for TB control have evolved significantly over time. Initially, the TB control strategy was clinical and programmatic, focusing mainly on providing standardized regimens and medications [ 34 ]. The underlying assumption was that existing biomedical tools could primarily solve TB transmission, with the premise that curing patients with active disease would reduce mortality, lower disease prevalence, decrease transmission, and subsequently reduce incidence [ 35 ]. However, the actual situation in many countries is more complex, health systems are inefficiently managed, financially under-resourced, and severely understaffed in many LMICs. Additionally, there is a lack of adequate drug production capacity and low levels of health system informatization [ 36 ]. These issues significantly hinder effective TB control and have not been adequately addressed in TB control efforts.

There is an urgent need for implementation research to evaluate the behavioral factors and conditions affecting the efficacy of drug treatments for TB. Such studies are essential to modify and improve TB control strategies, enhancing the real-world effectiveness of anti-TB medications, improving patient outcomes, reducing mortality rates, shortening the duration of bacterial shedding, and decreasing the risk of community transmission and overall incidence [ 2 , 37 ]. Chemotherapy remains the most crucial treatment for drug-resistant TB [ 37 ]. However, challenges such as prolonged treatment duration, poor clinical efficacy, numerous adverse effects, and high mortality rates persist. Therefore, developing new drugs and optimizing chemotherapy regimens are vital to increasing cure and survival rates [ 38 ]. These advancements are crucial for the clinical treatment and control of TB.

Currently, the only licensed vaccine for TB prevention is the bacille Calmette-Guérin (BCG) vaccine. Developed nearly a century ago, BCG is effective in preventing severe forms of TB in children, essentially mitigating severe disease and reducing the severity of clinical symptoms in pediatric TB cases. This vaccine has been widely administered to children globally for many years [ 2 ]. However, its effectiveness has shown significant geographical variability. Moreover, there is no licensed vaccine that effectively prevents TB in adults, either before or after exposure to the infection. The development and widespread administration of more effective preventive vaccines in high TB burden settings are crucial for advancing TB elimination efforts. Encouragingly, the M72/AS01E vaccine has shown promise in inducing an immune response and providing protection against the progression to pulmonary tuberculosis for at least three years [ 39 ].

Low-SDI regions require increased focus on MDR-TB and XDR-TB

The study found that the ASIR of MDR-TB is increasing rapidly in low-SDI regions, while it is declining in high-SDI regions, consistent with previous studies [ 2 , 40 ]. This disparity can be attributed to two main factors: (1) Detection Coverage. High-SDI regions have a high coverage rate for MDR-TB detection. Detected patients are promptly isolated and treated, reducing disease spread and decreasing incidence rates. In contrast, low-SDI regions have poor accessibility to drug resistance testing services and low population coverage in the early stages. As detection coverage increases in low-SDI regions, more potential patients are identified, leading to an apparent increase in incidence [ 41 ]. (2) Insufficient Community Control. Low-SDI regions lack mature community management plans and have limited services for patient care and treatment. Approximately one-third of MDR-TB patients remain smear-positive at discharge, posing a risk of community transmission. These findings highlight the importance of enhancing detection coverage and improving community-based control measures, particularly in low-SDI regions, to curb the spread of MDR-TB [ 2 ].

The detection of Mtb strain drug resistance has traditionally relied on bacterial culture methods. However, rapid molecular diagnostic tests and sequencing technologies are now being introduced for the diagnosis of drug-resistant TB. Despite the emergence of new drugs such as bedaquiline, pretomanid, and linezolid [ 42 ], which have significantly improved the cure rates of refractory TB, the proportion of drug-resistant TB patients receiving and completing standardized treatment remains low [ 43 ].

Currently, the global registration level for rifampicin-resistant (RR)/MDR-TB treatment corresponds to only 43% of newly diagnosed RR/MDR-TB patients annually [ 2 ]. From 2018 to 2022, only 55% of the targeted number of MDR/RR-TB patients received treatment [ 2 ]. By the end of 2022, 40 countries had adopted the new six-month BPaLM/BPaL treatment regimen for MDR-TB/RR-TB or pre-extensively drug-resistant TB (pre-XDR-TB) [ 2 ], and 92 countries had implemented the shorter nine-month oral treatment regimen for MDR/RR-TB. The treatment success rate for RR/MDR-TB patients has shown steady improvement, increasing to 63% in 2020 from 60% in 2021, though there is still significant room for progress [ 2 ].

Clinicians should adhere to the latest TB treatment guidelines and consider early patient characteristics to predict clinical outcomes. For patients with poor prognostic features, such as thick-walled cavities or persistent positive sputum cultures at 3 months, targeted individualized measures should be implemented. Enhancing the treatment success rate for drug-resistant TB patients and reducing the risk of transmission within families and communities is a crucial control strategy [ 44 , 45 ].

Preventing latent TB infection (LTBI) can curb TB transmission

Close contacts of TB patients are highly susceptible to LTBI. Early screening and preventive treatment for these individuals are crucial measures to prevent TB transmission [ 46 ]. Individuals with LTBI do not experience adverse health effects and do not transmit Mtb . However, they face a continuous risk of progressing to active TB through reactivation. For those with long-term infection, the annual risk of developing active TB is relatively low, with empirical estimates ranging from 10 to 20 per 100,000 population [ 47 ].

LTBI remains widespread in regions with high TB prevalence, and reactivation can account for a significant proportion of incident TB cases. This phenomenon is observed even in countries where TB transmission has been steadily declining. It is estimated that one-quarter of the world's population is infected with LTBI, highlighting its potential as a substantial reservoir for future active TB cases. Interventions to prevent the progression of LTBI to active TB disease are critical for TB control [ 48 ].

In countries with low TB incidence, close contacts of pulmonary TB cases, such as family members, are systematically screened and treated for LTBI. The recommended treatment rate for individuals with LTBI should be at least 85%, with a completion rate of 75% [ 49 ]. However, the number of household contacts diagnosed with LTBI and prescribed preventive treatment remains very low [ 50 ]. Identifying and treating individuals with LTBI is crucial for preventing the development of active TB and controlling its spread.

Some studies suggest that in countries with severe TB epidemics or in LMICs, the cost-effectiveness of detecting and treating latent TB infections is lower than treating active TB cases. This is due to the large population with latent infections, high costs, poor acceptability, and difficulties in managing treatment. Nonetheless, addressing LTBI remains essential for comprehensive TB control efforts [ 2 , 51 ].

Control strategies developed under the guidance of the One Health approach will better curb TB transmission

In response to the persistent threat of MDR-TB and XDR-TB amid a slow decline in the TB burden, a One Health-based strategy is essential for effective and comprehensive TB control. This approach integrates interdisciplinary collaboration, environmental management, animal health monitoring, community engagement, strengthened research, robust policy support, and international cooperation [ 52 , 53 , 54 , 55 , 56 , 57 , 58 ].

The study has several limitations. First, it cannot avoid the inherent limitations of the GBD methodology. Data from some countries and regions, particularly for MDR-TB and XDR-TB incidence, DALYs, and mortality rates between 1990 and 2021, are missing, significantly affecting the accuracy and completeness of model estimates. Even when data are available, variations in quality, accuracy, and comparability can introduce biases [ 1 ]. For instance, the GBD 2021 dataset was released internally at the end of 2023, but comprehensive adjustments to the model parameters were made in early 2024, resulting in discrepancies between the updated data (used in this study) and data from other papers concerning TB burden [ 1 ]. Second, GBD 2021 data are model based rather than real-world data, which may result in overestimation or underestimation. Third, TB and subtype indicators for 204 countries and regions were calculated using global population standardization to ensure comparability. However, these indicators may not accurately reflect the true epidemiological situation of TB in each country. Fourth, the 95% CI of the EAPC may be underestimated because it estimates the average trend over the past thirty years without considering the uncertainty of these rates. Moreover, EAPC is accurate under linear trends, but when the rates show non-linear trends such as U-shaped, V-shaped, or L-shaped patterns, EAPC results can be erroneous. Fifth, a comprehensive assessment of disease burden requires broader consideration of economic, familial, and social factors. Sixth, the study did not include LTBI, HIV-positive TB cases, or TB cases resistant solely to rifampicin. Future studies should adopt multidimensional analyses to improve the accuracy and robustness of results.

The study highlights that the incidence of MDR-TB and XDR-TB remained steady from 2015 but began to rise slowly between 2019 and 2021. In addition, TB incidence is notably high in women, while men exhibit higher mortality rates. To address these issues, gender-specific TB screening and treatment programs should be implemented to improve early detection and treatment outcomes. Enhancing health infrastructure and increasing funding in low SDI regions are crucial for accelerating TB elimination. Developing rapid, accurate diagnostic tools and shorter, more effective, and less toxic treatment regimens are essential for combating MDR-TB and XDR-TB. Moreover, improving public health education and community engagement can raise awareness and ensure better prevention and treatment adherence. These measures are vital for reducing the global TB burden, particularly in protecting vulnerable populations.

Availability of data and materials

The datasets analysed during the current study are available at http://ghdx.healthdata.org/gbd-results-tool .

Abbreviations

Acquired immune deficiency syndrome

Age-standardized incidence rate

Age-standardized mortality rate

Age-standardized rate

Bayesian age-period-cohort

Bacille Calmette-Guérin

Confidence interval

Corona Virus Disease 2019

Disability-adjusted life years

Disease-model-Bayesian meta-regression

Directly observed treatment, short-course

Drug-susceptible tuberculosis

Estimated annual percentage change

Global Burden of Disease

Human immunodeficiency virus

Integrated nested Laplace approximation

Low- and middle-income countries

Latent tuberculosis infection

Multidrug-resistant tuberculosis without extensive drug resistance

Bayesian priors, regularisation, and trimming

Mycobacterium tuberculosis

Pre-extensively drug-resistant tuberculosis

Rifampicin resistance tuberculosis

Sociodemographic Index

Sustainable Development Goals

Tuberculosis

Uncertainty interval

World Health Organization

Extensively drug-resistant tuberculosis

GBD 2021 Tuberculosis Collaborators. Global, regional, and national age-specific progress towards the 2020 milestones of the WHO End TB Strategy: a systematic analysis for the Global Burden of Disease Study 2021. Lancet Infect Dis. 2024;24:698–725.

Article Google Scholar

World Health Organization. Global tuberculosis report 2023. https://www.who.int/publications/i/item/9789240083851 . Accessed 30 May 2024.

Deng LL, Zhao F, Li ZW, Zhang WW, He GX, Ren X. Epidemiological characteristics of tuberculosis incidence and its macro-influence factors in Chinese mainland during 2014–2021. Infect Dis Poverty. 2024;13:34.

Article PubMed PubMed Central Google Scholar

Menzies NA, Quaife M, Allwood BW, Byrne AL, Coussens AK, Harries AD, et al. Lifetime burden of disease due to incident tuberculosis: a global reappraisal including post-tuberculosis sequelae. Lancet Glob Health. 2021;9:e1679–87.

Article CAS PubMed PubMed Central Google Scholar

Ivanova O, Hoffmann VS, Lange C, Hoelscher M, Rachow A. Post-tuberculosis lung impairment: systematic review and meta-analysis of spirometry data from 14 621 people. Eur Respir Rev. 2023;32:220221.

World Health Organization. Global tuberculosis report 2021. https://www.who.int/publications/i/item/9789240037021 . Accessed 30 May 2024.

Chen Y, Liu J, Zhang Q, Wang Q, Chai L, Chen H, et al. Epidemiological features and temporal trends of HIV-negative tuberculosis burden from 1990 to 2019: a retrospective analysis based on the Global Burden of Disease Study 2019. BMJ Open. 2023;13: e074134.

Zhang S, Qiu L, Wu D, Zhang S, Pan C, Li C, et al. Predictors for treatment outcomes in patients with multi-drug resistant tuberculosis—China, 2018–2020. China CDC Wkly. 2022;4:907–11.

PubMed PubMed Central Google Scholar

Kwak SH, Choi JS, Lee EH, Lee SH, Leem AY, Lee SH, et al. Characteristics and risk factors associated with missed diagnosis in patients with smear-negative pulmonary tuberculosis. Korean J Intern Med. 2021;36:S151–9.

Article PubMed Google Scholar

Li J, Ouyang J, Yuan J, Li T, Luo M, Wang J, et al. Establishment and evaluation of an overlap extension polymerase chain reaction technique for rapid and efficient detection of drug-resistance in Mycobacterium tuberculosis . Infect Dis Poverty. 2022;11:31.

Diallo A, Combary A, Veronese V, Dahourou DL, Ouédraogo S, Traoré IT, et al. Delays in TB diagnosis and treatment initiation in Burkina Faso during the COVID-19 Pandemic. Trop Med Infect Dis. 2022;7:237.

Wang D, Cai XL, Lin X, Zheng J, Wu YL, Peng XE. Hepatoprotective drugs for prevention of liver injury resulting from anti-tuberculosis treatment: a meta-analysis of cohort studies. Infect Med. 2022;1:154–62.

GBD 2019 Tuberculosis Collaborators. Global, regional, and national sex differences in the global burden of tuberculosis by HIV status, 1990–2019: results from the Global Burden of Disease Study 2019. Lancet Infect Dis. 2022;22:222–41.

Ezzati M, Hoorn SV, Rodgers A, Lopez AD, Mathers CD, Murray CJ, et al. Estimates of global and regional potential health gains from reducing multiple major risk factors. Lancet. 2003;362:271–80.

GBD 2021 Diseases and Injuries Collaborators. Global incidence, prevalence, years lived with disability (YLDs), disability-adjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2024;403:2133–61.

GBD 2021 Risk Factors Collaborators. Global burden and strength of evidence for 88 risk factors in 204 countries and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2024;403:2162–203.

Fu L, Tian T, Wang B, Lu Z, Bian J, Zhang W, et al. Global, regional, and national burden of HIV and other sexually transmitted infections in older adults aged 60–89 years from 1990 to 2019: results from the Global Burden of Disease Study 2019. Lancet Healthy Longev. 2024;5:e17-30.

Cao G, Liu J, Liu M. Global, regional, and national trends in incidence and mortality of primary liver cancer and its underlying etiologies from 1990 to 2019: results from the Global Burden of Disease Study 2019. J Epidemiol Glob Health. 2023;13:344–60.

Bu Q, Qiang R, Fang L, Peng X, Zhang H, Cheng H. Global trends in the incidence rates of MDR and XDR tuberculosis: findings from the global burden of disease study 2019. Front Pharmacol. 2023;14:1156249.

Wang W, Wang Y, Wang F, Chen H, Qin X, Yang L, et al. Notable dysthymia: evolving trends of major depressive disorders and dysthymia in China from 1990 to 2019, and projections until 2030. BMC Public Health. 2024;24:1585.

Zhang T, Sun L, Yin X, Chen H, Yang L, Yang X. Burden of drug use disorders in the United States from 1990 to 2021 and its projection until 2035: results from the GBD study. BMC Public Health. 2024;24:1639.

Guo J, Feng YP, Liu ZD, Luo SR, Wu QY. Analysis of factors influencing patient delay by patients with pulmonary tuberculosis in Lishui City Zhejiang Province. BMC Pulm Med. 2023;23:264.

Morey-León G, Mejía-Ponce PM, Granda Pardo JC, Muñoz-Mawyin K, Fernández-Cadena JC, García-Moreira E, et al. A precision overview of genomic resistance screening in Ecuadorian isolates of Mycobacterium tuberculosis using web-based bioinformatics tools. PLoS ONE. 2023;18: e0294670.

Jhaveri TA, Jhaveri D, Galivanche A, Lubeck-Schricker M, Voehler D, Chung M, et al. Barriers to engagement in the care cascade for tuberculosis disease in India: a systematic review of quantitative studies. PLoS Med. 2024;21: e1004409.

Fang G, Yang D, Wang L, Wang Z, Liang Y, Yang J. Experiences and challenges of implementing universal health coverage With China’s national basic public health service program: literature review, regression analysis, and insider interviews. JMIR Public Health Surveill. 2022;8: e31289.

Hedayatipour M, Etemadi S, Hekmat SN, Moosavi A. Challenges of using evidence in managerial decision-making of the primary health care system. BMC Health Serv Res. 2024;24:38.

Visca D, Tiberi S, Pontali E, Spanevello A, Migliori GB. Tuberculosis in the time of COVID-19: quality of life and digital innovation. Eur Respir J. 2020;56:2001998.

Morishita F, Eang MT, Nishikiori N, Yadav RP. Increased case notification through active case finding of tuberculosis among household and neighbourhood contacts in Cambodia. PLoS ONE. 2016;11: e0150405.

Hu X, Wang J, Ju Y, Zhang X, Qimanguli W, Li C, et al. Combining metabolome and clinical indicators with machine learning provides some promising diagnostic markers to precisely detect smear-positive/negative pulmonary tuberculosis. BMC Infect Dis. 2022;22:707.

Konstantynovska O, Rekrotchuk M, Hrek I, Rohozhyn A, Rudova N, Poteiko P, et al. Severe clinical outcomes of tuberculosis in kharkiv region, ukraine, are associated with Beijing strains of Mycobacterium tuberculosis. Pathogens. 2019;8:75.

Zhang F, Zhang F, Dong Y, Li L, Pang Y. New insights into biomarkers for evaluating therapy efficacy in pulmonary tuberculosis: a narrative review. Infect Dis Ther. 2023;12:2665–89.

Zhao J, Han X, Li H, Luo Y, Fang Y, Wang Y, et al. Analysis of the immune response by standardized whole-blood stimulation with metabolism modulation. Phenomics. 2023;4:81–9.

Hou X, Guo Q, Lin Q, Ye T, Bi J, Liang J, et al. Determination of the predictive factors for diagnostic positivity of nucleic acid amplification tests for diagnosing pulmonary tuberculosis. Infect Med. 2022;1:17–22.

Alsdurf H, Benedetti A, Buu TN, Adjobimey M, Cook VJ, Fisher D, et al. Human resource implications of expanding latent tuberculosis patient care activities. Front Med. 2024;10:1265476.

Coleman M, Hill J, Timeon E, Tonganibeia A, Eromanga B, Islam T, et al. Population-wide active case finding and prevention for tuberculosis and leprosy elimination in Kiribati: the PEARL study protocol. BMJ Open. 2022;12: e055295.

Stosic M, Grujicic SS, Grgurevic A, Kuruc V, Ristic L, Antonijevic G, et al. Trends in tuberculosis notification and mortality and factors associated with treatment outcomes in Serbia, 2005 to 2015. Euro Surveill. 2020;25:1900322.

Gu P, Lu P, Ding H, Liu Q, Ding X, Chen Y, et al. Effectiveness, cost, and safety of four regimens recommended by WHO for RR/MDR-TB treatment: a cohort study in Eastern China. Ann Med. 2024;56:2344821.

Wahid A, Ghafoor A, Khan AW, Al-Worafi YM, Latif A, Shahwani NA, et al. Comparative effectiveness of individualized longer and standardized shorter regimens in the treatment of multidrug resistant tuberculosis in a high burden country. Front Pharmacol. 2022;13:973713.

Tait DR, Hatherill M, Van Der Meeren O, Ginsberg AM, Van Brakel E, Salaun B, et al. Final analysis of a trial of M72/AS01E vaccine to prevent tuberculosis. N Engl J Med. 2019;381:2429–39.

Article CAS PubMed Google Scholar

Chandra Deb L, Ahmed SSU, Baidhya CC, Deb Nath N, Ghosh S, Paul S. Prevalence of Eimeria spp. with associated risk factors in dairy calves in Sylhet Bangladesh. Vet Med Sci. 2022;8:1250–7.

Li ZP, Mao WH, Huang F, Wang N, Ma LP, Zhang LQ, et al. Access to quality diagnosis and rational treatment for tuberculosis: real-world evidence from China-gates tuberculosis control project phase III. Infect Dis Poverty. 2021;10:92.

Guinn KM, Rubin EJ. Implementing new approaches to tuberculosis control. China CDC Wkly. 2021;3:256–9.

Liu P, Wang X, Liang J, Dong Q, Zhang J, Liu D, et al. A recombinase polymerase amplification-coupled Cas12a mutant-based module for efficient detection of streptomycin-resistant mutations in Mycobacterium tuberculosis . Front Microbiol. 2022;12:796916.

Lu ZH, Yang M, Pan CH, Zheng PY, Zhang SX. Multi-modal deep learning based on multi-dimensional and multi-level temporal data can enhance the prognostic prediction for multi-drug resistant pulmonary tuberculosis patients. SOH. 2022;1:100004.

Google Scholar

Ryckman TS, Schumacher SG, Lienhardt C, Sweeney S, Dowdy DW, Mirzayev F, et al. Economic implications of novel regimens for tuberculosis treatment in three high-burden countries: a modelling analysis. Lancet Glob Health. 2024;12:e995–1004.

Kim M, Johnson CE, Schmalstig AA, Annis A, Wessel SE, Van Horn B, et al. A long-acting formulation of rifabutin is effective for prevention and treatment of Mycobacterium tuberculosis . Nat Commun. 2022;13:4455.

Menzies NA, Wolf E, Connors D, Bellerose M, Sbarra AN, Cohen T, et al. Progression from latent infection to active disease in dynamic tuberculosis transmission models: a systematic review of the validity of modelling assumptions. Lancet Infect Dis. 2018;18:e228–38.

Dirix V, Dauby N, Hites M, Watelet E, Van Praet A, Godefroid A, et al. Optimal detection of latent Mycobacterium tuberculosis infection by combined heparin-binding hemagglutinin (HBHA) and early secreted antigenic target 6 (ESAT-6) whole-blood interferon gamma release assays. J Clin Microbiol. 2022;60:e0244321.

Cohen A, Mathiasen VD, Schön T, Wejse C. The global prevalence of latent tuberculosis: a systematic review and meta-analysis. Eur Respir J. 2019;54:1900655.

Domínguez Á, Soldevila N, Toledo D, Parrón I, Millet JP, Barrabeig I, et al. Factors associated with treatment prescription to pulmonary tuberculosis contacts in catalonia (2019–2021): a population-based epidemiological study. Vaccines. 2023;11:1800.

Zignol M, Dean AS, Alikhanova N, Andres S, Cabibbe AM, Cirillo DM, et al. Population-based resistance of Mycobacterium tuberculosis isolates to pyrazinamide and fluoroquinolones: results from a multicountry surveillance project. Lancet Infect Dis. 2016;16:1185–92.

Huang L, He J, Zhang C, Liu J, Guo Z, Lv S, Zhang X, Li S. China’s One Health governance system: the framework and its application. Sci One Health. 2023;2:100039.

Python Ndekou P, Drake A, Lomax J, Dione M, Faye A, Daly Njiemessa Nsangou M, et al. An agent-based model for collaborative learning to combat antimicrobial resistance: proof of concept based on broiler production in Senegal. Sci One Health. 2023;2:100051.

Sharan M, Vijay D, Prakash Yadav J, Singh Bedi J, Dhaka P. Surveillance and response strategies for zoonotic diseases: a comprehensive review. Sci One Health. 2023;2:100050.

Zhang L, Guo W, Lv C. Modern technologies and solutions to enhance surveillance and response systems for emerging zoonotic diseases. Sci One Health. 2024;3:100061.

Biswas R, Debnath C, Bandyopadhyay S, Samanta I. One Health approaches adapted in low resource settings to address antimicrobial resistance. Sci One Health. 2022;1:100011.

Qian J, Wu Z, Zhu Y, Liu C. One Health: a holistic approach for food safety in livestock. Sci One Health. 2022;1:100015.

Chen Y, Chen W, Cheng Z, Chen Y, Li M, Ma L, et al. Global burden of HIV-negative multidrug- and extensively drug-resistant tuberculosis based on Global Burden of Disease Study 2021. Sci One Health. 2024;3:100072.

Download references

Acknowledgements

The authors appreciate the works by the GBD Study 2021 collaborators.

The study was supported by the fund of Shanghai 2021 “Science and Technology Innovation Action Plan” (21Y11922500, 21Y11922400), the Medical Innovation Research Special Project of the Shanghai Natural Science Foundation (23ZR1464000, 23ZR1463900), the International Joint Laboratory on Tropical Diseases Control in Greater Mekong Subregion from Shanghai Municipality Government (21410750200), the Three-year Action Plan for Promoting Clinical Skills and Innovation Ability of Municipal Hospitals (SHDC2022CRS039), the Talent Fund of Longhua Hospital affiliated to Shanghai University of Traditional Chinese Medicine (LH001.007), and the Bill & Melinda Gates foundation. The Funders had no role in the study design or in the collection, analysis, and interpretation of the data, writing of the report, or decision to submit the article for publication.

Author information

Shun-Xian Zhang and Feng-Yu Miao have contributed equally to this work.

Authors and Affiliations

Longhua Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai, 200032, China

Shun-Xian Zhang, Yu Wang, Xiao-Jie Hu, Ping Yin, Pei-Yong Zheng, Ming Yang & Zhen-Hui Lu

National Institute of Parasitic Diseases at Chinese Center for Disease Control and Prevention (Chinese Center for Tropical Diseases Research), NHC Key Laboratory of Parasite and Vector Biology, WHO Collaborating Centre for Tropical Diseases; National Center for International Research On Tropical Diseases; National Key Laboratory of Intelligent Tracking and Forecasting for Infectious Diseases, Shanghai, 200025, China

Shun-Xian Zhang, Shan Lv, Fan-Na Wei, Xin-Yu Feng & Lei Duan

Beijing Municipal Health Big Data and Policy Research Center, Beijing Institute of Hospital Management, Beijing, 101100, China

Feng-Yu Miao

Department of Science and Technology, Chinese Center for Disease Control and Prevention;, National Key Laboratory of Intelligent Tracking and Forecasting for Infectious Diseases, Beijing, 102206, China

Jian Yang & Ji-Chun Wang

National Health Commission (NHC) Key Laboratory of Biosafety, National Institute for Viral Disease Control and Prevention, Chinese Center for Disease Control and Prevention; National Key Laboratory of Intelligent Tracking and Forecasting for Infectious Diseases, Beijing, 102206, China

Wen-Ting Zhou

School of Global Health, Chinese Center for Tropical Diseases Research-Shanghai Jiao Tong University School of Medicine, Shanghai, 200025, China

Shan Lv, Fan-Na Wei, Xin-Yu Feng & Lei Duan

Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine, Shanghai, 200032, China

Mei-Ti Wang

Gansu Provincial Center for Disease Control and Prevention, Lanzhou, 730000, China

Guo-Bing Yang

You can also search for this author in PubMed Google Scholar

Contributions

SX-Z, ZH-L and PY-Z conceived, designed the manuscript. JC-W, FN-W, GB-Y, FY-M, JY, XJ-H and YW did a literature search and download the data. SL and DL conducted the analysis and interpretation of the data. YM compiled the tables and figures. SX-Z and PY drafted the manuscript. MT-W, XY-F and WT-Z proofed and interpreted the report. SX-Z and FY-M contributed equally to the paper. JC-W and ZH-L are the corresponding authors. All authors participated in data analysis, interpretation, discussion and writing of the manuscript, and all authors read and approved the final version of the paper.

Corresponding authors

Correspondence to Ji-Chun Wang or Zhen-Hui Lu .

Ethics declarations

Ethical approval and consent to participate.

The protocol of the GBD 2021 has been approved by the research ethics board at the University of Washington. The GBD 2021 shall be conducted in full compliance with University of Washington policies and procedures, as well as applicable federal, state, and local laws.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Supplementary Information

40249_2024_1227_moesm1_esm.docx.

Additional file 1: Table S1 The number of incidence cases of TB, DS-TB, MDR-TB, and XDR-TB in HIV-negative individuals in 2021, and percentage change of the number of incidence cases were analyzed across GBD regions. Table S2 The EAPC of ASRs for TB, DS-TB, MDR-TB, and XDR-TB in HIV-negative individuals were analyzed across five SDI regions. Table S3 ASRs of TB, DS-TB, MDR-TB, and XDR-TB in HIV-negative individuals in 2021, and percentage change of ASRs for 204 countries and territories. Table S4 The number of death cases of TB, DS-TB, MDR-TB, and XDR-TB in HIV-negative individuals in 2021, and percentage of change rates of death number for GBD regions. Table S5 The number of DALY cases of TB, DS-TB, MDR-TB, and XDR-TB in HIV-negative individuals in 2021, and percentage change of number of DALY cases were analyzed across GBD regions. Table S6 Predicted ASRs for HIV-DS-TB, HIV-MDR-TB, and HIV-XDR-TB from spanning 2022–2035, based on the Bayesian Age-Period-Cohort Model. Fig. S1 The specific mortality of TB, DS-TB, MDR-TB, and XDR-TB showed notable differences across age and gender distributions in 2021 year. Fig. S2 The specific DALY of TB, DS-TB, MDR-TB, and XDR-TB showed notable differences across age and gender distributions in 2021 year. Fig. S3 The association between the SDI and the ASIR, ASMR, and age-standardized DALY rate of TB across 204 countries and territories in 2021 year. Fig. S4 The association between the SDI and the ASIR, ASMR, and age-standardized DALY rate of DS-TB across 204 countries and territories in 2021 year. Fig. S5 The association between the SDI and the ASIR, ASMR, and age-standardized DALY rate of MDR-TB across 204 countries and territories in 2021 year. Fig. S6 The association between the SDI and the ASIR, ASMR, and age-standardized DALY rate of XDR-TB across 204 countries and territories in 2021. Fig. S7 The association between the SDI and the ASIR, ASMR, and age-standardized DALY rate from 1990 to 2021 year. Fig. S8 The association between the SDI and the ASIR, ASMR, and age-standardized DALY rate of DS-TB from 1990 to 2021 year. Fig. S9 The association between the SDI and the ASIR, ASMR, and age-standardized DALY rate of MDR-TB from 1990 to 2021 year. Fig. S10 The association between the SDI and the ASIR, ASMR, and age-standardized DALY rate of XDR-TB from 1990 to 2021 year. Fig. S11 The association between risk factors and the ASMR and age-standardized DALY rate of XDR-TB in 21 GBD regions from 1990 to 2021.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Zhang, SX., Miao, FY., Yang, J. et al. Global, regional, and national burden of HIV-negative tuberculosis, 1990–2021: findings from the Global Burden of Disease Study 2021. Infect Dis Poverty 13 , 60 (2024). https://doi.org/10.1186/s40249-024-01227-y

Download citation

Received : 03 June 2024

Accepted : 01 August 2024

Published : 19 August 2024

DOI : https://doi.org/10.1186/s40249-024-01227-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Epidemiology
Global burden of disease

Infectious Diseases of Poverty

ISSN: 2049-9957

Submission enquiries: Access here and click Contact Us
General enquiries: [email protected]

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

View all journals
Explore content
About the journal
Publish with us
Sign up for alerts
Open access
Published: 23 August 2024

Nationwide surveys of awareness of tuberculosis in India uncover a gender gap in tuberculosis awareness

Ranganath Thimmanahalli Sobagaiah 1 ,
Nitu Kumari 2 ,
Divya Bharathi Gattam 3 &
Mohammed Shoyaib Khazi ORCID: orcid.org/0000-0003-4682-0306 4

Communications Medicine volume 4 , Article number: 168 ( 2024 ) Cite this article

230 Accesses

1 Altmetric

Metrics details

Epidemiology
Tuberculosis

Tuberculosis remains a major challenge in India, with an estimated 2.69 million cases each year. Although men are more affected than women, gender differences and related factors affect awareness of tuberculosis and thus impact tuberculosis diagnosis and access to treatment. Understanding the gender-specific needs and complexities when diagnosing and treating tuberculosis is essential to manage cases in India.

We undertook a comparative study using data from three National Family and Health Surveys (NFHS), specifically NFHS-3, NFHS-4 and NFHS-5. We investigated the prevalence and gender disparity in awareness about tuberculosis, and associated factors, using regression analysis.

Most men and women surveyed are between the ages of 15 and 19. Across the surveys, the proportion of men and women who are unaware of spreading of tuberculosis decreases from 44.9% during NFHS 3 to 29.6% during NFHS 5. However, the prevalence ratio of men to women with no knowledge about modes of transmission of Tuberculosis increases from 0.92 during NFHS 3 to 0.98 during NFHS 5. Higher odds with younger age (NFHS 5, aOR: 1.07 (1.01–1.13)) and rural residency (NFHS 5, aOR: 1.12 (1.06–1.18)), and lower odds with unmarried marital status (NFHS 5, aOR: 0.92 (0.86–0.98)) are noteworthy associations. Women and men have differences in knowledge.

Conclusions

Gender disparity associated with awareness about tuberculosis in India is observed across all three nationwide surveys. Being aged fifteen to nineteen years and residing in rural area are risk factors. Being unmarried is a protective factor for women, but not for men.

Plain Language Summary

Lack of awareness of the spread of tuberculosis may be an important factor contributing to the current burden of disease. We used datasets from three rounds of the National Family Health Survey conducted in India to determine the proportion of men and women who knew how tuberculosis spreads. Using a predictive model, we showed that misconceptions are more common among both men and women. For women, younger age and living in rural areas were risk factors for lack of awareness, which was not the case for men. Such differences may represent a barrier to reducing the burden of disease. These findings can be used to develop gender-specific, comprehensive people awareness programs to raise awareness about tuberculosis.

Drivers of sex differences in the South African adult tuberculosis incidence and mortality trends, 1990–2019

Identification and Prediction of Tuberculosis in Eastern China: Analyses from 10-year Population-based Notification Data in Zhejiang Province, China

Long-term trends of tuberculosis incidence and mortality in four central African countries

Introduction.

Tuberculosis (TB) is a disease vastly influenced and prevented by the social factors in the community. Lack of knowledge regarding the disease can contribute to underuse of medical services, delay in diagnosis and poor treatment adherence in people living with tuberculosis. Enhancing the dissemination of information on tuberculosis to increase the public awareness and health promotion is crucial to achieve the global targets for reduction in disease burden of tuberculosis. Studies have revealed that irrespective of a general overview about the disease, there is a breach in knowledge regarding the transmission, diagnosis, management, and its prevention. Also, poor knowledge or comprehension of tuberculosis disease and its treatment frequently contributes to non-adherence to therapy 1 , 2 , 3 .

Currently in India, the National Tuberculosis Elimination Programme (NTEP) with the development of National Strategic Plan 2017–25 is an ambitious attempt by the Government to eliminate tuberculosis by 2025 4 . Despite being a preventable and curable disease, tuberculosis is the most infectious killer disease attributing to almost 10 million cases per year globally, out of which 1.9 million cases are from India 5 , 6 . Numerous guidelines and tools have been released and made accessible by the Ministry of Health and Family Welfare (MoHFW) to tackle the problem of tuberculosis. The policies have been constantly updated after gathering considerable implementation-related learnings and the expansion of programme activities is still happening 7 .

Lack of knowledge about TB is a continuing problem and pose a risk for its prevention and care in China 8 . Similar situation can also be expected in India. In addition, because of the lack of knowledge about the disease and fear of being ostracized, persons with TB often hide their symptoms and fail to receive appropriate treatment which is a stumbling block in the prevention and care of the disease 9 .

The trend analysis of National Family Health Survey (NFHS) 10 aids to give us key themes to improve the National Tuberculosis Elimination Programme’s (NTEP) coverage, quality, equity, efficiency, and effectiveness.

Internationally, in countries with high disease burden of tuberculosis, the routine diagnosis of tuberculosis, treatment compliance and health seeking habits are observed to be affected by gender and their knowledge and perception towards the disease 11 , 12 . The overall misconceptions about the transmission of TB ranges from 43–68 percent of women and 35–66 percent of men in all subgroups of background characteristics 13 , 14 .

With the disease burden of 1,933,381 cases from India in 2021 6 , out of which 6% were children aged 0 to 14 years, 58% were men and 36% were women, it becomes even more crucial to address the gap in awareness of transmission of TB among the two genders. Therefore, the Central TB Division formed the National Framework for Gender Responsive Approach to TB in India guidelines which reports that gender differences and inequalities play a crucial role in how people access and receive healthcare due to TB 15 . Gender is an important variable in the incidence, exposure, risks, health seeking behaviour and in treatment outcomes of tuberculosis.

Globally, studies have also shown that men may repress their illnesses knowingly or unknowingly in an effort to avoid being perceived as weak or feminine, or as a form of compensation. They achieve this, among other things, by believing that they are physically superior to women. They ignore disease as they work to fulfil their obligations to support and uplift their families, something many people are finding harder and harder to accomplish 7 . Men perceive control as a fundamental component of acceptable manhood and efforts to obtain it have also led men to put their health on the back burner, men were afraid of being perceived as being less than men 11 , 16 .

In countries like Malawi, role constructions as primary material providers for their immediate family along with the opportunity costs of acknowledging illness seem important barriers to care-seeking. Upon that, Men’s sense of adequacy as providers was influenced by limited employment opportunities and small incomes. It has been suggested that there is a need to address harmful masculinity and promote gender equality to support interventions for TB and chronic cough 16 .

In India, men are more affected with TB compared to women, but women are at a higher risk of manifesting the disease easily due to undernutrition mainly because of social norms which prevent prioritizing of their nutrition, health, and well-being. Whereas men are at the risk of developing TB due to their employment like mining and construction industries 15 .

Moreover, the factors affecting the health seeking due to gender remains the same as that found globally and access to services is greatly impacted by gender disparities that affect care-seeking, as well as health system variables such access restrictions, a lower index of suspicion of TB in women, and the provision of insufficient information to care-seekers 17 .

Across the globe, there has been a trend that the female participation in the surveys exceeded male participation in TB related surveys 18 . During NFHS less number of men were interviewed when compared to women. Hence, gender-specific factors in tuberculosis prevention and treatment can have a wide range such as differences in care-seeking behavior, diagnostic challenges, risk factors, disease burden of HIV and tuberculosis coinfection, and delayed treatment. Addressing these factors is crucial for achieving equity in tuberculosis care and reducing the burden of the disease among both men and women 19 . Another important factor that can be considered is the sex assortativity among the contacts of the existing patients that might have contributed to sex disparities in disease burden of tuberculosis among adults 20 . According to the NFHS-5 data, although, gender influence in knowledge and perception towards the disease affects the tuberculosis management and care, the extent of the influence is not explicitly explored in India.

This study observes the trends in gender influence in awareness of transmission of tuberculosis at national level to understand the factors that affect this. Noteworthy variation in awareness regarding transmission of tuberculosis is observed among men and women at the national level. On exploring the factors that influence TB awareness, interesting results are obtained which have major implications for TB prevention and care initiatives such as the NTEP in India. The most important factors among women are socioeconomic status, rural residence, age, and education. Our results suggest empowering women and promoting the education of mothers could improve TB awareness, a goal of the TB prevention and care program in India.

Study design

It is a cross-sectional study that compares three complex sample surveys of nationally representative population.

Data sources

Datasets of Demographic Health Survey (DHS) which is also known as National Family Health Survey (NFHS) in India. After permission we obtained the recoded datasets for all three NFHS from DHS. Individual Recode file that contains the data on all the women and Mens Recode file that contains data on all men interviewed during NFHS were used in data analysis. These files shall be referred as Womens dataset and Mens dataset in the article.

For NFHS 3 conducted during 2005 to 2006, the survey included participants from 29 states. For NFHS 4 conducted during 2015 to 2016 and NFHS 5 conducted during 2019 to 2021, the survey included participants from all states and union territories.

Sample characteristics

For NFHS surveys, the multistage cluster sampling is adopted along with population proportion to sampling technique.

Participants

NFHS-3 and NFHS-4 adopted different sample designs for data collection. NFHS-3 used a two-stage approach for rural areas and a three-stage approach for urban areas. In rural areas, the first stage involved selecting villages as Primary Sampling Units (PSUs) using probability proportional to population size (PPS), and in the second stage, households were systematically chosen within each village. In urban areas, three stages were used, with the selection of wards, Census Enumeration Blocks (CEBs), and households 21 . NFHS-4 employed a stratified two-stage sample design with the 2011 census serving as the sampling frame. In rural areas, PSUs (villages) were selected using PPS, and the strata were defined based on the number of households and the percentage of the population belonging to scheduled castes and tribes. In urban areas, CEBs were selected using PPS, considering the SC/ST population percentage. Complete household mapping was conducted in selected PSUs, which were segmented into clusters. Random sampling was used to choose clusters, and within each selected cluster, 22 households were randomly selected in the second stage of data collection. This design resulted in NFHS-4 clusters being either complete PSUs or segments of PSUs 22 . NFHS-5 used the same sample design as that of NFHS-4 21 . From each household one woman from the eligible age group was selected randomly for interview. However for men, during NFHS 3, only those men were interviewed who were usual residents of the sample household or visitors who stayed in the sample household the night before the survey 21 . However, during NFHS 4 and NFHS 5, only men who were selected only in the subsample of households selected for the state module 22 , 23 . In addition to the above, during NFHS 3, the union territories were not considered. Moreover, Telangana was formed in June 2014. Therefore, it is not available as a separate state in NFHS 3. Similarly, Ladakh as a union territory was formed in October 2019. Hence it is not available as a separate state or union territory in NFHS 4.

A total of 74369 cases in Mens dataset and 124385 cases in Womens dataset during NFHS 3, 112122 cases in Mens dataset and 169686 cases in Womens dataset during NFHS 4 and 101839 cases in Mens dataset and 724115 cases in Womens dataset during NFHS 5 were available. Inclusion criteria for analysis for our research was, first: age group of the respondent between 15 to 45 years of age, and second was “Yes” as response to the question: Ever heard about tuberculosis. Detailed inclusion criteria are given in Supplementary Figs. 1 and 2 .

We created the variable on awareness about tuberculosis based on the respondent’s response as “Yes” or “No” to the following questions that were asked during NFHS survey:

Q1: Tuberculosis spread by: Air when coughing or sneezing.

Q2: Tuberculosis spread by: Sharing utensils.

Q3: Tuberculosis spread by: Touching a person with tuberculosis.

Q4: Tuberculosis spread by: Food.

Q5: Tuberculosis spread by: Sexual contact.

Q6: Tuberculosis spread by: Mosquito bites.

Based on the responses, we derived four categories in the dependent variable which are as follows: Category 1: Knowledge without misconceptions: if the response was “Yes” to Q1 and “No” to all other questions. Category 2: Knowledge with misconceptions: if the response was “Yes” to Q1 and “Yes” to any other questions from Q2 to Q6. Category 3: No knowledge without misconceptions: if the response was “No” to all questions from Q1 to Q6. Category 4: No knowledge with misconceptions: if the response was “No” to Q1 and “Yes” to any other questions from Q2 to Q6. For data representation and analysis, Category 3 and Category 4 were added and was considered as single category. Category 1 was used as reference for regression analysis. The categorization in the dependent variable was made based on previous study 24 .

Independent

Based on the review of literature 15 , we selected the following variables for the regression model.

Age in five-year groups

The current age of the respondent was divided into groups of five years each. The participants from all surveys selected in the study were belonging to the age group of fifteen to forty-five years of age. Age group of 45 to 49 years was used as reference category.

Type of place of residence

It is where the respondent was interviewed as either urban or rural which was created based on whether the cluster or sample point number is defined as urban or rural and urban area was considered as a reference category.

Region in which the respondent was interviewed. During NFHS 3, only twenty-nine states were included. However, during NFHS 4 and NFHS 5, states along with Union Territories were also included in the survey. Kerala state was taken as a reference category.

Highest education level

This is a standardized variable providing level of education in the following categories: No education, Primary, Secondary, and Higher which was used as reference category.

Wealth Index

The wealth index is a composite measure of a household’s cumulative living standard. The wealth index is calculated using easy-to-collect data on a household’s ownership of selected assets, such as televisions and bicycles; materials used for housing construction; and types of water access and sanitation facilities. Richest category was used for reference in regression analysis.

Current marital status

It is the current marital status of the respondent. The original variable in the dataset was recoded to form three categories as the distribution of data among various categories in the original variable was skewed. The recoded variable had three categories: “Never married”, “Married” and “Others” which was used as reference category.

Response to the question

Tuberculosis can be cured: The response had three categories: “No”, “Yes” and “Don’t know”. The response “Yes” was taken as reference category.

Keep secret if family member gets tuberculosis: The response had three categories: “No”, “Yes, remain a secret” and “Don’t know/Not sure/It depends”. Response “No” was taken as reference category.

Frequency of reading newspaper or magazine

The response had four categories: “Not at all”, “Less than once a week”, “at least once a week” and “almost every day”.

Frequency of listening to radio

The response was had four categories like those of frequency of reading newspaper or magazine.

Frequency of watching television

The response was had four categories like those of frequency of reading newspaper or magazine. Reference category for frequency of reading newspaper or magazine, listening to radio and watching television was “at least once a week” for regression analysis.

During NFHS 3, the Men’s Questionnaire was employed to interview men aged 15–54 who were usual residents of the sample household or visitors who stayed in the sample household the night before the survey 21 . However, during NFHS 4 and NFHS 5, the Men’s Questionnaire was administered only in the subsample of households selected for the state module 14 , 21 . Hence, the number of cases in the Mens dataset are less in number when compared to those in Womens dataset. Moreover, those who were not interviewed may have contributed notably to the results of our study.

64,212 cases from Mens dataset and 109,032 cases from Womens dataset file for NFHS 3, 91,293 cases from Mens dataset and 61,8274 cases from Womens dataset for NFHS 4 and 85,751 cases from Mens dataset and 671,750 cases from Womens dataset for NFHS 5 were included in the study for further analysis.

Ethical considerations

Our study used secondary data for analysis from the datasets provided by the Demographic Health Surveys Program (DHS). We applied for access, and this was granted based on us providing information about our planned use. All the datasets provided were re coded and already anonymized to completely protect the privacy of the survey participants. Informed consent was obtained from the participant or guardian (for children) before the interview for all surveys by DHS 25 . We did not obtain approval from institutional review boards as the data we were using was deidentified and recoded, that has already been reviewed for privacy and ethical concerns before by DHS. Moreover, this data is available public domain in form of datasets and national and state level reports. The authors were not allowed to share the datasets with each other. Hence all authors have obtained authorization to use the datasets separately from DHS.

Statistics and reproducibility

The datasets were imported to STATA® MP 4 core v17, and declaration for survey design for each dataset was done for weights, primary sampling unit and strata as per instructions by DHS in order to accommodate for stratification by province and state, size group. Dependent variables were computed and required independent variables were recoded. Association between categorical variables was assessed using design adjusted Chi square test. Further, Multinominal Logistic Regression analysis was used to derive adjusted odds ratio with Category 1 as the reference category in the dependent variable. The regression models were derived separately for men and women. Subsequently Poissons Regression analysis was used to derive adjusted prevalence ratio for similar models as it is difficult to interpret an odds ratio for a cross-sectional study as there is confusion between risk or odds leading to incorrect quantitative interpretation 26 . Moreover, the prevalence of no knowledge was higher than 10% and the odds ratio would overestimate the prevalence ratio. However, due to the limitation of Poissons regression with svy commands, in the dependent variable was converted into binomial variable combining Category 1 and Category 2 into a single category as “Knowledge about the spread of Tuberculosis” and Category 3 and Category 4 into a single category as “No knowledge about the spread of Tuberculosis”. All the statistical analysis was carried out under the subset of svy commands that has inherent property for measures similar to robust measures for poisons regression 27 . Microsoft® Excel 365 was used to make line charts. QGIS® Desktop 3.30.1 was used to make maps for prevalence ratio of Men: Women of No knowledge about spreading tuberculosis among men and women. To limit the length of the manuscript, the details on odds ratio are given in the main manuscript and details on prevalence ratio are given in the Supplementary Table No. 7 to Supplementary Table No. 10 .

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

In NFHS 3, 64,212 cases from Mens dataset (containing data from interview of eligible men at household) and 109,032 cases from Womens dataset (containing data from interview of eligible women at household) were included as they fulfilled inclusion criteria. Similarly, from NFHS 4, 91,293 cases from Mens dataset and 618,274 cases from Womens dataset were included for analysis. In addition to the above, from NFHS 5, 85,751 cases from Mens dataset and 691,750 cases from Womens dataset were included for analysis.

Descriptive data

In NFHS 3, 18.26% men and 19.93% women were from the age group fifteen to nineteen years. 38.00% of the men were residing in urban area whereas 64.53% women were residing in rural areas. Most of the men and women had Secondary level of education and belonged to richest level of wealth index. More than half of them were married. While highest proportion of the men belonged from Central zone followed by South zone, most of the women were from Central zone followed by East and South zone in similar proportion. 36.41% men read newspaper or magazine, 25.23% listened to the radio, and 47.83% watched television almost every day. However, for most of the women watching television daily was the only mode of exposure to mass media on almost daily basis. More than 50% of the women never read newspaper or magazine or listened to radio. More than 75% believed that tuberculosis can be cured and would not keep a secret if family member gets tuberculosis.

In NFHS 4, the proportions for age group, level of education, wealth index, current marital status, belief that tuberculosis can be cured, belonging to zone and response for keeping secret if family member gets tuberculosis, were like those in NFHS 3 among men. However, most of the women belonged to the age group twenty to twenty-four years followed by fifteen to nineteen years. Among women similar proportions as that of NFHS 3 were seen with respect to reading newspaper or magazine, listening to radio, and watching television. However, more than 60% of both men and women resided in rural area. In addition to that, among men only 34.53% red newspaper or magazine, and only 6.98% listened to radio and 63.28% watched television almost every day. There was a major change in proportion among men for frequency of mode of exposure to mass media with respect to and listening to radio and watching television.

In NFHS 5, like that in NFHS 4, the proportion of majority of men and women remained unchanged in terms of, type of place of residence, education level, current marital status, belief that tuberculosis can be cured, and response to the question that will they keep secret if family member gets tuberculosis. Among women most of them belonged to the age group of fifteen to nineteen years. In addition to that, most of the men belonged from East zone followed by West zone. Also, there was an increase in proportion of men and women who would never read a newspaper or magazine and listen to radio. In addition to that, there were no respondents who would read newspaper or magazine, listen to radio, or watch television almost every day among both men and women who participated in NFHS 5.

The detailed distribution of eligible men and women during three NFHS surveys are given in Tables 1 and 2 respectively. Moreover, the state and union territory wise distribution for eligible men and women is given in Supplementary Table No. 1 and Supplementary Table No. 2 respectively.

Trend of Knowledge and Misconceptions about spreading to tuberculosis:

Across three surveys, there has been a decrease in the proportion of men who had “No knowledge” and consequently rise in proportion of those who had knowledge about spreading of tuberculosis. Moreover, during NFHS 4, the proportion of those with “No Knowledge” was less than that of those who “Had Knowledge”. In addition to that, from NFHS 4 to NFHS 5, there is an increase in proportion of those who “Had knowledge with misconceptions” but decrease in the proportion of those men who “Had knowledge without misconceptions” about the spread of tuberculosis. Hence there was an increase of misconceptions among men. (Fig. 1 ).

Proportion of knowledge and misconceptions about spreading of tuberculosis among men.

Across three surveys, there has been a decrease in the proportion of women who had “No knowledge” and consequently rise proportion of women who “Had knowledge” about the spread of tuberculosis. However, the proportion of women with “No Knowledge” has always been higher than that of those who “Had knowledge without misconception”. The difference between those who “Had knowledge without misconception” and those who “Had knowledge with misconceptions” about the spread of tuberculosis had been increasing. Hence, there was an increase of misconceptions among women. (Fig. 2 ).

Proportion of knowledge and misconceptions about spreading of tuberculosis among women.

For comparison of prevalence of “No knowledge” about spreading tuberculosis among men and women across the states, during NFHS 3, NFHS 4 and NFHS 5 are presented as Prevalence Ratio on geographical basis on map of India with political boundaries denoting state and union territories. (Figs. 3 , 4 and 5 respectively)

Prevalence ratio of men: women for no knowledge regarding spreading of TB during NFHS 3.

Prevalence ratio of men: women for no knowledge regarding spreading of TB during NFHS 4.

Prevalence ratio of men: women for no knowledge regarding spreading of TB during NFHS 5.

Outcome data

Adjusted odds ratio for men and women for various factors affecting the response as “No knowledge” about spreading of tuberculosis.

Main results

The Crude odds ratio for the independent variables for men and women during three rounds of NFHS is given in Supplementary Table No. 3 and Supplementary Table No. 4 respectively. While any age group was not a significant factor among men, among women the age group of fifteen to twenty four years had higher odds of having “No knowledge” about the spread of tuberculosis during NFHS 3 (aOR: 1.28 (1.12–1.44) for age group fifteen to nineteen years and 1.15 (1.03–1.28) for age group of twenty to twenty four years) and NFHS 5 (aOR: 1.07 (1.01–1.13) for both age groups) and it was statistically significant.

Like the age groups, residing in rural areas was not a significant factor among men. However, women had the higher odds ratio of “No knowledge” who were residing in rural areas and the odds had marginal change across three surveys (aOR: 1.18 (1.07–1.30) during NFHS 3, aOR: 1.09 (1.03–1.14) during NFHS 4 and aOR: 1.12 (1.06–1.18) during NFHS 5) and it was statistically significant. It shows that there has been a disparity between men and women with respect to residing in rural areas.

Education level

No education has constantly been associated significantly with higher odds ratio of having “No knowledge” about spread of tuberculosis among both men and women, however the odds ratio were higher among women (aOR: 2.72 (2.37–3.11) during NFHS 3, aOR: 2.08 (1.96–2.22) during NFHS 4 aOR: 1.66 (1.57–1.76) during NFHS 5) when compared to men through three surveys and the difference of odds ratio between men and women have been decreasing.

For both genders, all categories of wealth index were associated significantly with higher odds ratio of “No Knowledge” about spread of tuberculosis and the odds ratio were highest among the respondents belonging to the poorest category of wealth index during NFHS 3 and NFHS 4. However, during NFHS 5, among men only those belonging to the poorest category of wealth index were associated with higher odds ratio.

For men the marital status was not a significant factor. However, among women never married had lesser odds ratio of “No knowledge” about spreading tuberculosis during NFHS 3 and NFHS 5 which was statistically significant.

Tuberculosis can be cured

Both among men and women those who believed that tuberculosis cannot be cured, were associated with higher odds ratio of having “No knowledge” about spreading of tuberculosis. Among men the odds ratio had increased during NFHS 5 (aOR: 1.95 (1.62–2.34)) when compared to NFHS 3 (aOR: 1.74 (1.51–2.00)). However, among women the odds ratio had decreased during NFHS 5 (aOR: 1.62 (1.54–1.71)) when compared to NFHS 3 (aOR: 1.76 (1.57–1.98)) and these findings were statistically significant.

Would keep secret if family member gets tuberculosis

For men, only during NFHS 5 had higher odds ratio (aOR: 1.21 (1.05–1.39)) for “No knowledge” about spreading of tuberculosis if they wanted keep secret if family member gets tuberculosis. However, for women, the odds ratio of “No knowledge” about spreading of tuberculosis were higher during all three surveys. Moreover, the odds ratio among women have reduced over time marginally. ((aOR 1.27 (1.16–1.38) in NFSH 3, aOR: 1.23 (1.17–1.30) in NFHS 4 and aOR: 1.26 (1.16–1.38) in NFHS 5).

Not reading newspaper or magazine at all among men was associated with increased odds ratio of having “No knowledge” about spreading tuberculosis among men during NFHS 3 (aOR: 1.33 (1.16–1.50)) and NFHS 4 (aOR: 1.24 (1.12–1.36)) only. However, the odds ratio was insignificant during NFHS 5. Among women not reading newspaper or magazine at all was associated with increased odds ratio of having “No knowledge” about spreading of tuberculosis during three surveys. In addition to that the odds ratio had decreased from NFHS 3 (aOR: 1.23 (1.12–1.35)) to NFHS 5 (aOR: 1.16 (1.10–1.22)).

Among both genders, not listening to radio was associated with higher odds ratio of having “No knowledge” about spreading of tuberculosis during NFHS 3. How ever during NFHS 5, the odds ratio were insignificant in males and protective among females (aOR: 0.93 (0.86–0.99)).

For men not watching television at was associated with increased odds ratio of “No knowledge” about spread of tuberculosis during NFHS 4 (aOR: 1.14 (1.02–1.27)). However, among females, not watching television at all was associated with increased odds ratio during NFHS 3 (aOR: 1.12 (1.01–1.24)) and NFHS 5 (aOR: 1.08 (1.04–1.12)).

The detailed adjusted odds ratio for men and women are given in Tables 3 and 4 respectively.

Other analyses

In our regression model for all three surveys, we included the State or Union Territory of residence of respondents to derive aOR for residents of other states and union territories when compared to the residents of the state Kerala. During NFHS 3 for men, the highest odds ratio for “No knowledge” about spreading of tuberculosis was among those who were residing in Jharkhand (aOR: 11.02 (6.71–18.10)) followed by Madhya Pradesh (aOR: 4.96 (3.43–7.17)). Similarly, during NFHS 4 the highest odds ratio was among those who were residing in Uttarakhand (aOR: 7.33 (5.07–10.58)) followed by Himachal Pradesh (aOR: 6.04 (4.44–8.21)). However, during NFHS 5, Dadra & Nagar Haveli and Daman & Diu (aOR: 47.76 (24.25–94.07)) had the highest odds ratio followed by Bihar (aOR: 14.49 (10.02–20.97)). For women, during NFHS 3, the highest odds ratio for having “No knowledge” about spreading of tuberculosis was among those who were residing in Bihar (aOR: 15.00 (10.52–21.38)) followed by Assam (aOR: 10.07 (7.82–12.96)). Similarly, during NFHS 4, the highest odds ratio was among those who were residing in Jharkhand (aOR: 14.93 (13.18–16.90)) followed by Assam (aOR 8.49 (7.47–9.63)). However, during NFHS 5, the highest odds ratio was in those women who were residing in Bihar (aOR: 31.36 (27.65–35.57)) followed by Jharkhand (aOR: 25.46 (22.16–29.25)).

The detailed adjusted odds ratio for men and women for state and union territories is given in Supplementary Table 5 and Supplementary Table 6 respectively.

The details on unadjusted and adjusted Prevalence Ratio for Men and Women are given in Supplementary Table 7 to Supplementary Table 10 .

India being a signatory to the 2030 Agenda for Sustainable Development 28 , we are currently implementing National Strategic Plan (NSP – 2017–2025) 29 , 30 and envision tuberculosis free India by 2025. In order to achieve this goal, there is a need to adopt a comprehensive approach to gender specific and gender sensitive interventions 15 . This study was undertaken to find out gender disparity and its associated factors regarding awareness of tuberculosis in India by comparing data of three nationwide surveys viz. NFHS-3, NFHS-4 and NFHS-5.

Across the three surveys comparison, we found that there has been a decrease in the proportion of men with “no knowledge” about the spread of tuberculosis and consequently rise in proportion of those who had knowledge over the stretch of years in India. This depicts success of various strategies involved to increase public awareness viz. availability of health information sources in vernacular language and according to local needs; regular training of concerned human resources and promotion of e- learning modes. Moreover, the decrease in the proportion of women who had “no knowledge” was more as compared to men, may be due to improved access of women to electronic media via mobile and internet usage, which could not be assessed due to limitations of the study and may also be due to inclusion of females in health manpower. Moreover, the proportion of men and women with knowledge about the spread of tuberculosis was associated with misconception regarding awareness of tuberculosis transmission which can be attributed to easier access to electronic media via mobile and internet and also lack of awareness of trusted sources of correct information regarding health-related states, particularly TB. In addition, it points towards gender being an important social construct which influences the level of awareness of people about health and illness. As per social norms men have a greater public involvement and hence greater exposure to information, which leads to greater awareness about tuberculosis among men 31 . A previous study from Gujarat showed similar findings of higher proportion of men with better knowledge. It was seen that men were more aware about the mode of transmission and symptoms of tuberculosis 32 . Another study from Nanded, Maharashtra showed similar finding of higher knowledge (33.6%) and more positive attitude (53%) in men about tuberculosis compared to women 33 .

In present study, age was an important risk factor which is associated with gender disparity in awareness regarding tuberculosis transmission. Women in age group of fifteen to nineteen years age and twenty to twenty-four years of age were having “no knowledge” about the spread of tuberculosis when compared to men in same age group. Our analysis was concordant to previous similar studies which showed that women with higher age group are more aware and knowledgeable about TB 34 , 35 , 36 . Increase in age may add more health experience, hence, better aware about TB and identify the mode of infection. In addition to that, culturally higher aged women in India enjoy greater autonomy and freedom than younger one, thus find little or no hindrance in order to seek medical help for self thus more aware and knowledgeable than younger one 37 .

Overall, the analysis shows that the odds among those with “no education” having lesser awareness of tuberculosis transmission have reduced in both men and women but still the odds were more in women as compared to men over the years. As women with higher education have higher odds for awareness and correct knowledge regarding spread of TB, it was in the line with other studies 38 , 39 . It could be that educated people have greater access to various sources of information leading to more awareness about health, availability of healthcare services and use this awareness and information in accessing the health care services 40 , 41 .

Furthermore, women residing in rural area and belonging to low income households were acting as a risk factor for decreased awareness on transmission of tuberculosis, while, such was not the case among men, which is in line with similar other studies 42 , 43 , 44 , 45 . Women with better socio-economic status and those in urban areas are more likely to have better access to health, better media access to TB information, as well as good communication, transportation, and other necessities 35 , 46 . In addition, the rural-urban divide in knowledge and awareness can also attribute to awareness disparity depending upon the place of residence. Therefore, urbanized women and from higher socioeconomic backgrounds have a much better chance than women from rural areas and lower socioeconomic backgrounds of meeting their needs and demands thus knowledge and awareness regarding TB 47 .

It is also found that women who would like to keep it a secret if any member suffers from tuberculosis have higher odds of having no knowledge about spread of tuberculosis than men during NFHS 3 and NFHS 4 which was not so during NFHS 5. Its probable explanation could be that women with more hindrance feel lesser autonomy in terms of medical seeking behaviour thus do not easily disclose a family member’s tuberculosis 47 . During NFHS 5 there was an overlap in the odds for men and women thus eliminating the gender disparity. Usually, NFHS is completed within a year. However, NFHS 5 was completed in three years that is from 2019 to 2021. The duration of completion of survey was increased due to lockdown. However, we are of the opinion that the pandemic has not affected awareness about the methods of spreading tuberculosis in a notable way. The results can be generalised to whole population of the country as the NFHS was conducted among the nationally representative population in the country.

Various strategies to improve access to knowledge regarding tuberculosis and its transmission include creating a culture of evidence-based decision-making by the use of ICT based applications from grass root level upwards, supporting integration and improvement in TB information systems, including NIKSHAY for achievement of TB elimination goals and establishing a TB Knowledge Network (TBKN), inter-connecting all knowledge and research institutions in the country through a virtual network. The overarching role will be to establish a backbone connectivity which will enable knowledge and information sharing amongst TBKN connected institutes, enabling collaborative research, development and innovation amongst TBKN connected institutes, facilitating advanced distance education in specialized sub-areas of TB, facilitating connection between different sectoral networks in the field of research.

The key strategy is to move towards an e-learning mode utilizing the web based and mobile based learning experiences and translating the content to vernacular language and adding relevant content as per local needs at the State level. There has been also high visibility media campaign involving Amitabh Bacchan, India’s biggest film star and an ex-TB patient, as the TB brand ambassador, a big impact on conveying the threat of TB to the public at large. Moreover, TB Champions from amongst patients, technical experts, political representatives, public figures, sportsperson, and celebrities added their voice to increase visibility and action on TB. Substantial efforts have been made towards capacity building of programme managers, state IEC officers and communication facilitators with dedicated national, regional and state level trainings and workshops, to increase awareness about TB 29 .

This study has few limitations such as since it is secondary data analysis, all aspects about knowledge and awareness of tuberculosis could not be explored. Secondly, NFHS which produced the data for this study, was based on respondents’ self-reported information, with no objective validation of the information provided. Furthermore, the dataset of women used in the analysis is limited to the reproductive age group women (fifteen years to forty-nine years of age), which is insufficient to generalize the result for all the women. Similarly, elderly men dataset is not available for analysis. The proportion of men included in the survey was not comparable to that of women. Lastly, since the data for this study came from a cross-sectional survey, we were only able to look at the association between independent and dependent variables and hence any conclusions about causality could not be drawn. In the models derived using Poisson’s Regression providing Prevalence Ratio, multiple independent variables can be seen having a varied level of significance, when compared with Odds Ratio Logistic Regression providing Relative Risk Ratio. This can be attributed to the fact that Poisson’s Regression was derived after converting the dependent variable into a binomial variable for the feasibility of the statistical analysis based on the available expertise of the authors.

Based upon our study, we recommend increased usage of mass media and social media platforms for disseminating health education, since television and radio as media of communication does not hold much value in today’s era. Future research should investigate the reasons that could explain the unexplained differences in tuberculosis awareness, knowledge, and attitude amongst men and women. Moreover, frequent community health contact activities considering gender- specific needs in tuberculosis prevention and care initiatives should be promoted. Furthermore, the fear of stigma and discrimination in different ways at their homes, workplaces, healthcare settings and in communities may prevent people, women, and transgenders, from seeking healthcare. This can be tackled by adopting social behaviour change communication (SBCC) campaigns, especially targeting women, may yield greater results in tuberculosis awareness and knowledge, leading to better tuberculosis notification rates, hence, achieve the goal of ending tuberculosis in India by 2025.

Data availability

The datasets that support the findings of this study are available from DHS at https://dhsprogram.com/ but restrictions apply to the availability of these data, which were used under license for the current study. Though the datasets are available in the public domain, a formal request is required to be placed with DHS which should mention the project details such as Title, Objectives and description of tentative analysis that will be carried out. The numerical data for the Figs. 1 and 2 can be found in file named Supplementary Table 11 . Further, the numerical data for Figs. 3 , 4 and 5 is provided in Supplementary Tables 12 , 13 and 14 respectively.

Mbuthia, G. W., Olungah, C. O. & Ondicho, T. G. Knowledge and perceptions of tuberculosis among patients in a pastoralist community in Kenya: A qualitative study. Pan Afr Med J 30 , 287 (2018).

Article PubMed PubMed Central Google Scholar

Matakanye H., Tshitangano T. G., Mabunda J. T. & Maluleke T. X. Knowledge, beliefs, and perceptions of tb and its treatment amongst tb patients in the limpopo province, South Africa. Int J Environ Res Public Health . 18 , 10404, (2021). Available from: /pmc/articles/PMC8508321/

Mondal, M. N., Nazrul, H. M., Chowdhury, M. R. K. & Howard, J. Socio-demographic factors affecting knowledge level of Tuberculosis patients in Rajshahi City, Bangladesh. Afr Health Sci [Internet] 14 , 855 (2014).

Article CAS PubMed Google Scholar

Family N. & Survey H. Government of India Ministry of Health and Family Welfare COMPENDIUM OF FACT SHEETS INDIA AND 14 STATES/UTs (Phase-11).

Khanna, A., Saha, R. & Ahmad, N. National TB elimination programme - What has changed. Indian J Med Microbiol [Internet] 42 , 103 (2023).

Tuberculosis. [cited 2023 Apr 25]. https://www.who.int/health-topics/tuberculosis#tab=tab_1

Mavhu, W. et al. Chronic cough and its association with TB–HIV co‐infection: factors affecting help‐seeking behaviour in Harare, Zimbabwe. Tropical Medicine & International Health 15 , 574–579 (2010).

Google Scholar

Chen, W. et al. Is tuberculosis health education reaching the public in China? A cross-sectional survey in Guizhou Province. Bmj Open 6 , e013534 (2016).

Mathew, A. S. & Takalkar, A. M. Living with tuberculosis: the myths and the stigma from the Indian perspective. Clinical infectious diseases 45 , 1247 (2007).

Article PubMed Google Scholar

TB statistics - India.: https://tbfacts.org/tb-statistics-india/

Chikovore, J. et al. Missing men with tuberculosis: the need to address structural influences and implement targeted and multidimensional interventions. BMJ Global Health 5 , e002255 (2020).

Chikovore, J., Hart, G., Kumwenda, M., Chipungu, G. A. & Corbett, L. For a mere cough, men must just chew Conjex, gain strength, and continue working’: the provider construction and tuberculosis care-seeking implications in Blantyre, Malawi. Global Health Action 8 , 26292 (2015).

India announces plan to end tuberculosis by 2025 [Internet]. Bill & Melinda Gates Foundation. Available from: https://www.gatesfoundation.org/ideas/articles/india-announces-plan-to-end-tuberculosis-by-2025

International Institute for Population Sciences. National Family Health Survey(NFHS-5) India 2019-21. Minist Heal Fam Welf Natl . 1 , 1–714 (2022).

National Framework For a Gender-responsive approach to TB in India [Internet]. [cited 2023 Aug 23]. Available from: https://tbcindia.mohfw.gov.in/wp-content/uploads/2023/05/388838054811-NTEP-Gender-Responsive-Framework_311219.pdf

Chikovore, J. et al. Control, struggle, and emergent masculinities: a qualitative study of men’s care-seeking determinants for chronic cough and tuberculosis symptoms in Blantyre, Malawi. BMC Public Health 14 , 1–2 (2014).

Article Google Scholar

National Family Health Survey. http://rchiips.org/nfhs/

Horton, K. C., MacPherson, P., Houben, R. M., White, R. G. & Corbett, E. L. Sex differences in tuberculosis burden and notifications in low-and middle-income countries: a systematic review and meta-analysis. PLoS Medicine 13 , e1002119 (2016).

Ledesma, J. R. et al. Global, regional, and national sex differences in the global burden of tuberculosis by HIV status, 1990–2019: results from the Global Burden of Disease Study 2019. The Lancet Infectious Diseases 22 , 222–241 (2022).

Horton, K. C., Hoey, A. L., Béraud, G., Corbett, E. L. & White, R. G. Systematic review and meta-analysis of sex differences in social contact patterns and implications for tuberculosis transmission and control. Emerging Infectious Diseases 26 , 910 (2020).

Danarastri, S., Perry, K. E., Hastomo, Y. E., Kurniawati & Priyonugroho, K. Gender differences in health-seeking behaviour, diagnosis and treatment for TB. Int J Tuberc Lung Dis 26 , 568–570 (2022).

Article CAS PubMed PubMed Central Google Scholar

International Institute for Population Sciences. 1, National Family Health Survey (NFHS-3) 2005-06 India . 12–13. (Mumbai, International Institute for Population Sciences, 2007).

International Institute for Population Sciences. 1, National Family Health Survey (NFHS-4) 2015-2016 India . 1–2. (Mumbai, International Institute for Population Sciences, 2017).

Khandoker, A., Khan, M. M., Krämer, A. & Mori, M. Knowledge about tuberculosis transmission among ever-married women in Bangladesh. The International Journal of Tuberculosis and Lung Disease 15 , 379–384 (2011).

CAS PubMed Google Scholar

International Institute for Population Sciences. 1, National Family Health Survey (NFHS-5) 2019-2021 India . 2–3. (Mumbai, International Institute for Population Sciences, 2022).

Krishnamoorthy, Y. & Ganesh, K. Prevalence and determinants of physical violence and its impact on birth outcomes during pregnancy in India: evidence from a nationally representative survey. Journal of Interpersonal Violence 37 , 2615–2632 (2022).

StataCorp. [stata.com]. Stata Statistical Software: Release 17. College Station, TX: StataCorp LLC, (2021). https://www.stata.com/manuals/svy.pdf

The DHS program [Internet]. The DHS Program - Protecting the Privacy of DHS Survey Respondents. The Demographic and Health Surveys (DHS) Program. https://dhsprogram.com/Methodology/Protecting-the-Privacy-of-DHS-Survey-Respondents.cfm

Sustainable Development Goals. United Nations India Annual Report 2021.

National Strategic Plan for Tuberculosis 2017 - 25: Elimination by 2025 [Internet]. Central Tuberculosis Division National Tuberculosis Elimination Programme. Ministry of Health and Family Welfare; 2017 [cited 2023 Aug 24]. Available from: https://tbcindia-wp.azurewebsites.net/wp-content/uploads/2023/05/National-Strategic-Plan-2017-25.pdf

World Health Organization. Gender and tuberculosis . (World Health Organization, 2002)

Bhalla, S., Chndwani, H., Chudasama, R., Somasundaram, C. & Pathi, S. Awareness regarding pulmonary Tuberculosis In a rural population of Gujrat. Health and Population 27 , 29–39 (2004).

Pichamuthu, B., Putthawar, P. & Akinola, A. Gender differences in knowledge and attitude about tuberculosis among rural population in Nanded district Maharashtra State India. IOSR Journal of Humanities and Social Science 19 , 81–84 (2014).

Agho, K. E., Hall, J. & Ewald, B. Determinants of the knowledge of and attitude towards tuberculosis in Nigeria. J Heal Popul Nutr 32 , 520–538 (2014).

CAS Google Scholar

Pengpid, S. & Peltzer, K. Knowledge, attitudes, and practices regarding tuberculosis in Timor-Leste: Results from the demographic and health Survey 2016. J Prev Med Public Heal 52 , 115–122 (2019).

Navio, J. L. P., Yuste, M. R. & Pasicatan, M. A. Socio-economic determinants of knowledge and attitudes about tuberculosis among the general population of Metro Manila, Philippines. Int J Tuberc Lung Dis 6 , 301–306 (2002).

All, U. T. C. Evidence of women’s empowerment in India: a study of socio-spatial disparities. GeoJournal 65 , 365–380 (2016).

Tolossa, D., Medhin, G. & Legesse, M. Community knowledge, attitude, and practices towards tuberculosis in Shinile town, Somali regional state, eastern Ethiopia: A cross-sectional study. BMC Public Health 14 , 1–13 (2014).

Datiko, D. G., Habte, D., Jerene, D. & Suarez, P. Knowledge, attitudes, and practices related to TB among the general population of Ethiopia: Findings from a national cross-sectional survey. PLoS One 14 , 1–16 (2019).

Mazumdar, S., Satyanarayana, S. & Pai, M. Self-reported tuberculosis in India: Evidence from NFHS-4. BMJ Glob Heal 4 , 1–8 (2019).

Feuser T. Investigation of Tuberculosis Stigma in 2008 Ghana. Thesis https://scholarworks.gsu.edu/cgi/viewcontent.cgi?article=1351&context=iph_theses (2014).

Pengpid, S. et al. Knowledge, attitudes, and practices about tuberculosis and choice of communication channels in Thailand. J Infect Dev Ctries 10 , 687–693 (2016).

Hoa, N. P., Chuc, N. T. K. & Thorson, A. Knowledge, attitudes, and practices about tuberculosis and choice of communication channels in a rural community in Vietnam. Health Policy (New York) 90 , 8–12 (2009).

de Freitas, I. M. et al. Fatores associados ao conhecimento sobre tuberculose e atitudes das famílias de pacientes com a doença em Ribeirão Preto, São Paulo. Rev Bras. Epidemiol. 18 , 326–340 (2015).

Turk, T., Newton, F. J., Netwon, J. D., Naureen, F. & Bokhari, J. Evaluating the efficacy of tuberculosis Advocacy, Communication and Social Mobilization (ACSM) activities in Pakistan: A cross-sectional study. BMC Public Health 13 , 887 (2013).

Gautam, N., Karki, R. R. & Khanam, R. Knowledge on tuberculosis and utilization of DOTS service by tuberculosis patients in Lalitpur District, Nepal. PLoS One 16 , 1–14 (2021).

Mondal, D., Karmakar, S. & Banerjee, A. Women’s autonomy and utilization of maternal healthcare in India: Evidence from a recent national survey. PLoS One 15 , 1 (2020).

Download references

Acknowledgements

We would like to thank DHS for approving the authorization to use datasets and providing datasets free of cost upon request. The authors would like to state that we did not receive any external funding for our work. This research project was entirely self-funded, and we did not have the support of any grants or other funding sources.

Author information

Authors and affiliations.

Bangalore Medical College and Research Institute, Bengaluru, India

Ranganath Thimmanahalli Sobagaiah

World College of Medical Sciences and Research, Jhajjar, India

Nitu Kumari

Kempegowda Institute Medical Sciences, Bengaluru, India

Divya Bharathi Gattam

All India Institute of Medical Sciences, Mangalagiri, India

Mohammed Shoyaib Khazi

You can also search for this author in PubMed Google Scholar

Contributions

Ranganath T S (R.T.S.) conceptualized the study. In addition to that, he performed a review of literature, and supervised the research team. Nitu Kumari (N.K.) and Divya Bharathi G (D.B.G.) curated the data, conducted the investigation for filtering the variables, developed, and validated the methodology, and created the visualizations. K Md Shoyaib (K.M.D.S.) conducted the final statistical analysis on the final datasets. N.K., D.B.G. and K.M.D.S. drafted the manuscript and all authors reviewed, edited, and approved the final manuscript. All authors had complete access to the D.H.S. datasets, which were accessed upon approval of individual requests, by D.H.S. All authors have verified the data.

Corresponding author

Correspondence to Mohammed Shoyaib Khazi .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

: Communications Medicine thanks the anonymous reviewers for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplement file, reporting summary, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Thimmanahalli Sobagaiah, R., Kumari, N., Bharathi Gattam, D. et al. Nationwide surveys of awareness of tuberculosis in India uncover a gender gap in tuberculosis awareness. Commun Med 4 , 168 (2024). https://doi.org/10.1038/s43856-024-00592-x

Download citation

Received : 08 June 2023

Accepted : 14 August 2024

Published : 23 August 2024

DOI : https://doi.org/10.1038/s43856-024-00592-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

Explore articles by subject
Guide to authors
Editorial policies

Subscriptions
Advanced search

Advanced Search

The incidence of tuberculosis among adolescents and young adults: a global estimate

Find this author on Google Scholar
Find this author on PubMed
Search for this author on this site
ORCID record for Kathryn J. Snow
For correspondence: [email protected]
Figures & Data
Info & Metrics

Historical data show that the risk of tuberculosis increases dramatically during adolescence, and young people face unique challenges in terms of case detection and effective treatment. However, little is known about the burden of tuberculosis among young people in the modern era. This study aimed to provide the first estimates of the global and regional incidence of tuberculosis among young people aged 10–24 years.

Using the World Health Organization (WHO) database of tuberculosis notifications for 2012, we estimated the burden of tuberculosis among young people by WHO region. Adjustments were made for incomplete age disaggregation and underreporting, using supplementary data from several countries representing diverse tuberculosis epidemics.

We estimate that 1.78 million (uncertainty interval (UI) 1.23–3.00 million) young people developed tuberculosis in 2012, accounting for 17% of all new tuberculosis cases globally. Young people in the WHO South East Asian Region (721 000, UI 473 000–1.35 million) and the WHO African Region (534 000, UI 359 000–912 000) experienced the greatest number of tuberculosis episodes.

Young people suffer a considerable burden of tuberculosis. Age-specific burden of disease estimation for this age group is complicated by incomplete age disaggregation of tuberculosis data, highlighting the importance of continued surveillance system strengthening.

An estimated 1.78 million young people (aged 10–24 years) around the world developed tuberculosis in 2012 http://ow.ly/o9vq30hoepv

Introduction

Adolescence is increasingly recognised as a period of development that underpins many aspects of future health and well-being [ 1 , 2 ]. The success of the child survival agenda within the Millennium Development Goals (2000–2015), among other factors, has resulted in many countries now having more adolescents (10–19 years) and young adults (20–24 year olds) than younger children; young people aged 10–24 years now comprise a quarter of the world's population [ 3 ].

Adolescence and early adulthood is increasingly recognised as a key risk period for tuberculosis infection, disease and adverse outcomes. In contrast to young children aged 0–4 years, young people commonly develop infectious forms of tuberculosis and frequently have a much wider range of social contacts outside of the household [ 4 ]. Consequently, as well as suffering disease, adolescents and young adults with tuberculosis contribute to ongoing transmission. In high tuberculosis transmission settings, the incidence of tuberculosis increases rapidly during adolescence to peak in early adulthood [ 5 ]. Historical data suggest that the risk of infection with Mycobacterium tuberculosis is highest during adolescence and young adulthood, and that between the ages of 12 and 24 years there may also be a transient increase in the risk of progression to disease after infection compared with children or older adults [ 5 , 6 ]. Furthermore, many comorbidities relevant to tuberculosis emerge or are exacerbated during the adolescent period, including infection with HIV, diabetes, risky substance use (including tobacco use) and mental health conditions. Finally, many adolescent girls and young women face the health challenges associated with pregnancy and childbirth, which may increase their risk of both developing tuberculosis and experiencing adverse outcomes [ 7 , 8 ].

Adolescents face specific age-related challenges in accessing appropriate care as they transition between child and adult health services, particularly in tuberculosis-endemic settings where dedicated adolescent health services are usually absent [ 1 , 2 ]. A study of adolescents aged 12–18 years in Kenya reported that their prevalence of tuberculosis was six-fold higher than indicated by case notifications [ 9 ]. A recent study from Botswana reported higher rates of loss to follow-up in 10–19-year-old adolescents on treatment for tuberculosis than among adults [ 10 ]. Adolescents and young adults are a recognised key population in the global HIV epidemic and HIV-related deaths are increasing among adolescents, in contrast to falling mortality in every other age group [ 11 ]. Although the importance of the adolescent age group in HIV control is increasingly appreciated, tuberculosis among adolescents remains neglected.

This at-risk age group requires increased attention for case detection, care and prevention within the global End TB Strategy [ 12 ]. A critical first step to addressing the issue of tuberculosis in at-risk groups is to “know your epidemic” [ 13 ]. The World Health Organization (WHO) first published estimates of the burden of tuberculosis in “children” (children and young adolescents aged 0–14 years) only as recently as 2012 [ 14 ]. In recognition of the limitations of data availability and interpretation, there have been ongoing efforts to improve the accuracy of these estimates using various statistical and modelling approaches [ 15 – 17 ]. It is currently estimated that there were 1.0 million incident cases of tuberculosis and 210 000 tuberculosis-related deaths in children and young adolescents in 2015 [ 17 ].

In contrast, despite being a highly dynamic period for health in general and tuberculosis in particular, there are no published estimates of tuberculosis incidence among adolescents and young adults globally. The majority of the literature on tuberculosis divides the population into two age groups, 0–14 years (referred to as “children”) and ≥15 years (referred to as “adults”), ignoring adolescents and young adults as a distinct group. The aim of this study was to estimate the global incidence of tuberculosis among young people aged 10–24 years in 2012 from the global tuberculosis notification dataset.

Data sources

While a majority of national tuberculosis programmes routinely report all notified child tuberculosis cases (0–14 years) each year to the WHO, and increasingly disaggregate data into the age groups of 0–4 and 5–14 years, the adolescent age group remains obscured by current reporting practices. Young adolescents (10–14 years) are included in the “child” age group, while older adolescents and young adults aged 15–24 years are categorised as “adults” [ 18 ]. This does not allow a direct accounting of tuberculosis notifications for adolescents and young adults aged 10–14, 15–19 and 20–24 years or a direct observation of the epidemiological changes that occur with tuberculosis across these years. In order to estimate the global incidence of tuberculosis in adolescents and young adults, this study used the WHO global tuberculosis surveillance dataset for 2012 [ 19 ], supplemented with more detailed surveillance data from Brazil, Indonesia, South Africa, Romania and Estonia, disaggregated into 5-year age groups. These countries were chosen based on both data availability and epidemiological diversity. The analysis was restricted to 183 countries that reported notification data from 2012 to the WHO in 2013 and for which the United Nations Population Division provides age-disaggregated population estimates (in 10-year age groups).

Tuberculosis data for 2012 were used as that was the most recent completed dataset available at the outset of this project, and changes to the WHO recording and reporting framework for tuberculosis with the 2013 revision [ 20 ] prevent the application of the same estimation method presented in this article to more recent data on tuberculosis case notifications.

This analysis was exempt from ethical review as it utilised routinely collected, aggregated data only.

Key variables

Six key notification variables in the global tuberculosis surveillance dataset were used in this analysis: the numbers of notifications for each of 1) sputum smear-positive tuberculosis, 2) smear-negative (including smear not done) tuberculosis and 3) extrapulmonary tuberculosis, for each of the 0–14- and 15–24-year-old age groups. Where a value was missing for a specific country, feasible potential values were estimated based on available data as described in detail later. The resulting actual or estimated numbers of notifications for each country were then disaggregated into 5-year age bands (0–4, 5–9, 10–14, 15–19 and 20–24 years) based on the observed distributions by age in the supplementary data. The resulting estimates were then adjusted for underreporting and underdiagnosis based on the estimates of case detection rates (CDRs) available for each country for 2016. CDR estimates from 2016 were used as these have undergone important revisions in recent years for some high-burden countries, with improved accuracy following national prevalence surveys, and these revisions apply retrospectively [ 17 ].

Analyses were conducted in Stata version 13 (StataCorp, College Station, TX, USA). Uncertainty intervals (UIs) were generated using data simulation bootstrapping techniques, generating a dataset with 1000 iterations of each country, assuming the key variables followed the distributions given in table 1 . Each step of the process is now described in detail, with the relevant equations provided at the end of this section.

View inline

Definitions of key parameters

Estimation of incidence among young adolescents aged 10–14 years

Disaggregation of the 15–24-year-old age group into 15–19 and 20–24 years.

The numbers of tuberculosis cases in the 15–19- and 20–24-year-old age groups were estimated at the country level using a three-step process. First, the number of sputum smear-positive notifications for the 15–24-year-old age group in each country was extracted from the global surveillance dataset. Where this value was missing for a country, it was estimated from the number of all new smear-positive notifications and the age structure of the population in the country, obtained from published United Nations Population Division estimates [ 3 ]. A linear model was used to estimate the proportion of all smear-positive cases that affect the 15–24-year-old age group, using the proportion of the national population aged 15–24 years as the explanatory variable. This estimation was necessary for seven countries with a combined total population of 156 million people (a large majority in Ethiopia, Peru and Mozambique; supplementary table S1 ).

Second, the numbers of notifications for smear-negative and extrapulmonary tuberculosis were extracted from the dataset. These variables were considered together because, when missing, they were almost always missing concurrently (the number of missing values for the key variables is summarised by country in supplementary table S1 ). Where the number of smear-negative and/or extrapulmonary tuberculosis notifications for the 15–24-year-old age group was not reported, the number was estimated based on the observed ratio of smear-negative plus extrapulmonary notifications to smear-positive tuberculosis notifications in the global dataset. This estimation was necessary for 50 countries ( supplementary table S1 ).

Third, the proportion of all tuberculosis notifications from the 15–24-year-old age group that pertained to the 15–19-year-old age group was estimated from the observed proportions in the supplementary data; the median value was used for the point estimates and the two extreme values for the uncertainty interval. The remainder of the actual or estimated notifications were assumed to pertain to the 20–24-year-old age group.

Accounting for underreporting of tuberculosis

We adjusted actual and estimated notifications to adjust for underdetection and underreporting of tuberculosis using the reciprocal of the estimated CDR for each country as estimated in 2016 [ 21 ], with uncertainty incorporated during the bootstrapping process. This step was necessary for all countries.

Accounting for uncertainty

Each of the estimated parameters described was randomly generated for each of the 1000 iterations of each country in the bootstrapping process and allowed to vary according to the distributions listed in table 1 . The normal distribution was used for the ratios and proportions derived from the global surveillance dataset, as this distribution was supported by the observed data for these two variables. For other parameters the uniform distribution was used, to better account for the uncertainty around estimates based on few observations (those from the supplementary dataset) and around the CDRs for each country; the limits of the uniform distributions were based on the most extreme observations in the supplementary dataset and the limits estimated by the WHO for the CDRs. Uncertainty intervals for our results were then sampled from the 2.5th and 97.5th percentile of the distributions for the estimates for each country, and point estimates from the median of all estimates for the country. These were then summed across each of the six WHO regions in order to provide regional estimates, which were rounded to the nearest 1000.

We estimate that 1.78 million (UI 1.23–3.00 million) adolescents and young adults aged 10–24 years developed tuberculosis during 2012. Estimated case numbers were lowest in the 10–14-year-old age group and highest in the 20–24-year-old age group. Figure 1 shows estimates with uncertainty intervals for 5-year age bands by WHO region. The WHO South East Asian Region and the WHO African Region had the highest total caseloads ( table 2 ).

Download figure
Open in new tab
Download powerpoint

Estimated tuberculosis incidence for adolescents and young adults, by 5-year age band and World Health Organization region, 2012. Vertical lines represent uncertainty intervals around best estimates of tuberculosis incidence.

Estimates of new cases of tuberculosis among adolescents and young adults aged 10–24 years, by World Health Organization region, 2012

The observed proportion of all tuberculosis in the 0–14-year-old age group which affected young adolescents aged 10–14 years ranged from 0.107 in South Africa to 0.426 in Brazil, while the proportion of all tuberculosis in the 15–24-year-old age group that affected the 15–19-year-old age group in the supplementary data ranged from 0.298 in Estonia to 0.376 in Indonesia ( supplementary table S2 ).

Visual inspection via scatter plot confirmed that the relationship between the proportion of the population aged 15–24 years and the proportion of smear-positive tuberculosis notifications that pertain to 15–24 year olds in the global dataset was linear in nature and the R 2 -value indicated that variation in the former variable explained 34.7% of variation in the latter. According to the linear model, as the percentage of the population aged 15–24 years increases 1%, the proportion of all smear-positive tuberculosis notifications that pertain to that age group increases by 1.16% (95% CI 1.09–1.23). This value was used to estimate the expected number of smear-positive tuberculosis notifications in the 15–24-year-old age group for seven countries that did not report this figure.

The observed ratio of smear-negative and extrapulmonary notifications to smear-positive notifications among the 15–24-year-old age group was 1.28:1 (95% CI 1.15–1.42). This figure was used to estimate the expected number of smear-negative and extrapulmonary notifications for the 15–24-year-old age group for 50 countries that did not report this figure. Of these countries, 27 were in the WHO African Region, with the remainder distributed across the other five WHO regions ( supplementary table S1 ).

This study provides original global estimates of the large burden of tuberculosis in adolescents and young adults aged 10–24 years. We estimate that 1.78 million young people developed tuberculosis worldwide in 2012. Numbers of cases increase markedly by age group across adolescence and into young adulthood, with the lowest numbers in 10–14 year olds and the highest numbers in 20–24 year olds. These estimates were determined using the most comprehensive source of data available, i.e. the global tuberculosis surveillance dataset [ 19 ], supplemented with data from several large tuberculosis-endemic countries from different epidemiological and geographical contexts that are broadly representative of most settings globally.

The marked increase in the estimated tuberculosis caseload between early adolescence and young adulthood is consistent with historical observation [ 5 ], and with age-stratified tuberculosis estimates published by the Global Burden of Diseases group in 2013 [ 22 ]. Historical data from the pre-HIV era further confirm that this phenomenon is an independent feature of tuberculosis epidemiology [ 5 ]. However, the concomitant increase in HIV prevalence throughout adolescence in high HIV incidence settings [ 23 ] likely exacerbates this phenomenon in many modern tuberculosis epidemics, particularly those in the African region. The fact that tuberculosis risk increases so markedly throughout this period suggests that adolescence may be an important window for preventive interventions for tuberculosis.

The requirements for reporting tuberculosis data to the WHO have expanded over time as tuberculosis control strategies have widened beyond an initial focus on those cases most likely to contribute to transmission, i.e. cases of sputum smear-positive tuberculosis. In 2006, the WHO revised tuberculosis registers to record and report all new tuberculosis cases by type, disaggregated by the age groups of 0–4, 5–14 and >15 years, with new sputum smear-positive pulmonary tuberculosis cases in adults to be further disaggregated into 10-year age groups ( i.e. 15–24, 25–34, …, ≥65 years) [ 24 ], although as observed during this analysis, implementation by countries varied. In 2013, the WHO requested that countries transition to reporting all new and relapse tuberculosis cases together (regardless of site of disease), disaggregated by age group (0–5, 5–14, 15–24, 25–34 years, etc. ) [ 19 ]. However, current reporting practices still do not allow direct determination of the numbers of notifications in adolescents and young adults (aged 10–24 years). As case-based electronic tuberculosis surveillance systems are more widely adopted and implemented, more nuanced analysis of tuberculosis notifications in these age groups will become possible in a variety of settings.

There are a number of important limitations to the accuracy of these estimates. Many young people at risk of tuberculosis live in countries with at least some missing data. It was necessary to estimate the expected number of smear-negative and extrapulmonary notifications in the 15–24-year-old age group for 50 countries in total, including several large, high tuberculosis burden countries ( supplementary table S1 ).

We chose to estimate smear-negative and extrapulmonary notifications from smear-positive tuberculosis notifications for several reasons. First, there is greater diagnostic certainty for smear-positive tuberculosis than for the other forms of the disease. Second, all but seven countries reported smear-positive notifications with full age disaggregation in 2012, providing a high-quality indicator of the incidence of this form of disease in the 15–24-year-old age group after adjustment for underreporting. As the relative prevalence of smear-positive disease appears to be largely biological [ 25 ], the ratio between forms of disease was relatively constant across countries.

Finally, CDRs in several high tuberculosis burden countries are still based on expert opinion because national tuberculosis prevalence surveys have not yet been completed for all priority countries [ 21 ]. CDRs for several large countries have been revised downward in recent years, resulting in higher estimates of the global tuberculosis burden [ 21 ]. If case detection in any other large, high tuberculosis burden countries is presently overestimated, then the point estimates of tuberculosis burden given here will be conservative. Likewise, if case detection is lower in adolescents than in adults, as has been documented in some settings [ 9 , 26 ], we will also have underestimated the tuberculosis burden in this age group. CDRs may vary with age given the age-related challenges for bacteriological confirmation in young adolescents (aged 10–14 years) as well as barriers to health service access for young people in some settings. At present there are no comprehensive global data on age-specific CDRs for tuberculosis, which prevents adjustment for potential age-specific CDRs. An additional challenge is that young adolescents are not usually included in national prevalence surveys. However, as estimates of the uncertainty around each of the key parameters have been incorporated into this analysis, we believe that the likely extent of uncertainty is captured by the intervals around our point estimates.

In conclusion, this study provides original estimates of the large burden of tuberculosis in adolescent and young adults globally in 2012, showing a sharp age-related increase in case numbers. The limitations of the accuracy of these estimates are recognised and highlight the pressing need for further studies of the epidemiology of tuberculosis among young people in a range of settings, and for continued strengthening of tuberculosis surveillance systems. Nonetheless, this work provides an important platform for increasing attention to and advocacy for the needs of this neglected and vulnerable group, who must be considered in scaling up efforts to end the global tuberculosis epidemic.

Supplementary material

Supplementary Material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary tables ERJ-02352-2017_Supplementary_tables

Acknowledgements

Age-specific tuberculosis data were provided by the national tuberculosis programmes of Brazil, Indonesia and South Africa. The European Centres for Disease Control provided age-specific data pertaining to Romania and Estonia. Dennis Petrie (Monash University, Clayton, Australia) provided guidance on the generation of confidence intervals.

Author contributions: S.M. Graham and K.J. Snow conceived the study. K.J. Snow conducted the literature search and created the figures. K.J. Snow and C. Sismanidis designed the study and analysed the data. S.M. Graham, S.M. Sawyer and J. Denholm interpreted the data. All authors contributed to writing the manuscript. C. Sismanidis is a staff member of the World Health Organization. The author alone is responsible for the views expressed in this publication and they do not necessarily represent the decisions or policies of the World Health Organization.

This article has supplementary material available from erj.ersjournals.com

Support statement: K.J. Snow is supported by the Australian National Health and Medical Research Council's Centre for Research Excellence in Tuberculosis, which funded this work. The funder had no role in the design or conduct of the study. Funding information for this article has been deposited with the Crossref Funder Registry .

Conflict of interest: None declared.

Received November 14, 2017.
Accepted December 13, 2017.
Sawyer SM ,
Bearinger LH , et al.
Patton GC ,
Santelli JS , et al.
United Nations Population Division
Marais BJ ,
Schaaf HS , et al.
Comstock GW ,
Livesay VT ,
Woolpert SF
Mathad JS ,
Rendell NL ,
Batjargal N ,
Jadambaa N , et al.
Mitchell E , et al.
Lowenthal ED ,
Arscott-Mills T , et al.
World Health Organization
Gardiner E ,
Coghlan R , et al.
Jenkins HE ,
Tolman AW ,
Yuen CM , et al.
Murray CJ ,
Ortblad KF ,
Guinovart C , et al.
Gillespie A ,
Porth T , et al.
Lestari BW ,
Arisanti N ,
Siregar AYM , et al.

Table of Contents
Index by author

Thank you for your interest in spreading the word on European Respiratory Society .

NOTE: We only request your email address so that the person you are recommending the page to knows that you wanted them to see it, and that it is not junk mail. We do not capture any email address.

Citation Manager Formats

EndNote (tagged)
EndNote 8 (xml)
RefWorks Tagged
Ref Manager

Respiratory infections and tuberculosis
Tweet Widget
Facebook Like
Google Plus One

Tuberculosis

Blood transcriptomic signatures for symptomatic TB
Long-term outcomes of TB and COVID-19 co-infection
Drivers determining TB disease screening yield in screening programmes

Loading metrics

Open Access

Peer-reviewed

Methods and Resources

Methods and Resources report novel methods, substantial improvements to current methodologies, or informational datasets.

See Journal Information »

A noninvasive BCG skin challenge model for assessing tuberculosis vaccine efficacy

Roles Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing

Affiliation Department of Infectious Disease, Imperial College London, London, United Kingdom

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review & editing

Roles Data curation, Formal analysis, Investigation, Methodology, Resources, Software

Roles Investigation, Methodology

Affiliation Department of Life Sciences, Centre for Bacterial Resistance Biology, Imperial College London, London, United Kingdom

Roles Conceptualization, Formal analysis, Investigation, Methodology, Resources, Software, Writing – review & editing

Affiliation Stanford Photonics Research Center, Stanford University, Stanford, California, United States of America

Roles Data curation, Formal analysis, Methodology, Software, Writing – review & editing

Affiliation Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Supervision, Writing – review & editing

Roles Conceptualization, Funding acquisition, Project administration, Supervision, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Nitya Krishnan,
Miles Priestman,
Iria Uhía,
Natalie Charitakis,
Izabella T. Glegola-Madejska,
Thomas M. Baer,
Albin Tranberg,
Alan Faraj,
Ulrika SH Simonsson,
Brian D. Robertson

Published: August 19, 2024
https://doi.org/10.1371/journal.pbio.3002766
Peer Review
Reader Comments

This is an uncorrected proof.

We report here on the characterisation in mice of a noninvasive bacille Calmette-Guérin (BCG) skin challenge model for assessing tuberculosis (TB) vaccine efficacy. Controlled human infection models (CHIMs) are valuable tools for assessing the relevant biological activity of vaccine candidates, with the potential to accelerate TB vaccine development into the clinic. TB infection poses significant constraints on the design of a CHIM using the causative agent Mycobacterium tuberculosis (Mtb). A safer alternative is a challenge model using the attenuated vaccine agent Mycobacterium bovis BCG as a surrogate for Mtb, and intradermal (skin) challenge as an alternative to pulmonary infection. We have developed a unique noninvasive imaging system based on fluorescent reporters (FluorBCG) to quantitatively measure bacterial load over time, thereby determining a relevant biological vaccine effect. We assessed the utility of this model to measure the effectiveness of 2 TB vaccines: the currently licenced BCG and a novel subunit vaccine candidate. To assess the efficacy of the skin challenge model, a nonlinear mixed effect model was built describing the decline of fluorescence over time. The model-based analysis identified that BCG vaccination reduced the fluorescence readout of both fluorophores compared to unvaccinated mice ( p < 0.001). However, vaccination with the novel subunit candidate did not alter the fluorescence decline compared to unvaccinated mice ( p > 0.05). BCG-vaccinated mice that showed the reduced fluorescent readout also had a reduced bacterial burden in the lungs when challenged with Mtb. This supports the fluorescence activity in the skin as a reflection of vaccine induced functional pulmonary immune responses. This novel noninvasive approach allows for repeated measurements from the challenge site, providing a dynamic readout of vaccine induced responses over time. This BCG skin challenge model represents an important contribution to the ongoing development of controlled challenge models for TB.

Citation: Krishnan N, Priestman M, Uhía I, Charitakis N, Glegola-Madejska IT, Baer TM, et al. (2024) A noninvasive BCG skin challenge model for assessing tuberculosis vaccine efficacy. PLoS Biol 22(8): e3002766. https://doi.org/10.1371/journal.pbio.3002766

Academic Editor: Matthew K. Waldor, Brigham and Women’s Hospital, UNITED STATES OF AMERICA

Received: October 24, 2023; Accepted: July 25, 2024; Published: August 19, 2024

Copyright: © 2024 Krishnan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Relevant data are within the paper and its Supporting Information files. Flow cytometry FCS files are available from https://doi.org/10.5281/zenodo.12794251 . Python scripts are available from https://doi.org/10.5281/zenodo.12781429 .

Funding: BDR received funding from Aeras grant “Human Challenge Model for TB”. BDR and TMB received funding from the Bill and Melinda Gates Foundation grant OPP1180610. The sponsors played no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Abbreviations: BCG, bacille Calmette-Guérin; CFU, colony-forming unit; CHIM, controlled human infection model; FOCEI, first-order conditional estimation method with interaction; IAV, inter-animal variability; ID, intradermally; Mtb, Mycobacterium tuberculosis ; OFV, objective function value; PsN, Perl-speaks-NONMEM; RFU, relative fluorescence unit; SCM, stepwise-covariate modelling; TB, tuberculosis; VPC, visual predictive check; YFP, yellow fluorescent protein

Introduction

Despite advances in modern medicine and better living conditions for many people, tuberculosis (TB) remains a major killer of the poor and disadvantaged around the world. Notwithstanding major efforts in the diagnosis, treatment, and prevention of TB—which have saved an estimated 54 million lives between 2000 and 2017—these have had little impact on transmission, and there are still nearly 10 million newly diagnosed cases and around 1.5 million deaths every year [ 1 ]. The current bacille Calmette-Guérin (BCG) vaccine is routinely given to neonates in endemic countries and high-risk populations and protects against the disseminated forms of TB disease to which children are susceptible. However, BCG works poorly in adolescents and adults who are the significant drivers of TB in high incidence communities [ 2 , 3 ], meaning that large-scale vaccination programmes have had negligible impact on transmission. Consequently, we need new vaccines with efficacy in all ages and all populations to successfully control TB and reach the WHO End-TB targets of a 95% decrease in deaths and a 90% decrease in incidence by 2035 (both relative to 2015) [ 4 ].

We have learnt much about the pathogenesis and immunology of TB using the mouse model. However, TB vaccine discovery has been hampered by animal models that poorly reflect the stages of infection and disease found in humans. Candidate vaccines that show good efficacy in a range of animal models do not always perform as well in human trials, where a lack of biomarkers for vaccine efficacy make incident TB the primary outcome, which requires several years of follow-up post-vaccination [ 5 , 6 ]. As the number of vaccine candidates completing preclinical studies increases [ 7 – 10 ], there is a need to prioritise those that proceed to expensive large-scale clinical trials without relying solely on the current suite of animal model data.

Human infection/challenge models have a long history in experimental medicine [ 11 ]. The malaria sporozoite challenge model has a safe history of use to assess new drugs and vaccines with a relatively short antimalarial treatment at the conclusion of the study [ 12 , 13 ]. A human challenge model for TB creates several challenges, including pulmonary infection, lengthy 6-month drug treatment, and the potential for latent or asymptomatic infection. The establishment of a pulmonary human infection model using BCG has demonstrated its feasibility for advancing the understanding of pulmonary immunobiology during TB infection [ 14 ]. A strategy to avoid issues associated with Mycobacterium tuberculosis (Mtb) as a challenge agent is to use Mycobacterium bovis BCG, which has a long history of safe use and has been used as a surrogate [ 14 – 18 ]. Our study utilises BCG as the basis for a fluorescent reporter strain (FluorBCG) that can be introduced intradermally, with the fluorescent signal measured noninvasively through the skin using a sensitive and cost-effective imaging system. This system reproducibly measures an accelerated loss of fluorescent signal over time in vaccinated compared to nonvaccinated animals. In this report, we describe the development and characterisation of the system and demonstrate its ability to detect relevant vaccine effects in the mouse model of TB. This skin-based challenge model has the potential to be used as an early indicator of vaccine efficacy and provide useful data to inform and advance vaccine development. This feasibility study is a step towards developing a human challenge model for TB.

Materials and methods

Animal studies.

All animal procedures were performed under the Animal Scientific Procedures Act (1986) under the licence issued by the UK Home Office (PPL 70/7160). Six- to 8-week-old female BALB/c mice (Charles River, United Kingdom) were maintained in Biosafety Containment Level 3 facilities (BSL3) according to institutional protocols.

Bacterial strains and growth conditions

Experiments with Mtb H37Rv (a kind gift from Dr Christophe Guilhot, Institut de Pharmacologie et de Biologie Structurale, France) were carried out in BSL3 facilities according to institutional protocols. BCG-Pasteur and BCG Pasteur Δ panCD strains were a kind gift from Dr Nathalie Cadieux, Aeras Global TB Vaccine Foundation. Mycobacterial strains were cultured in Difco Middlebrook 7H9 liquid medium (Becton Dickinson, UK) supplemented with 0.2% glycerol (Sigma, UK), 0.05% Tween 80 (Sigma, UK), and 10% oleic acid-albumin-dextrose-catalase (OADC, USBiological, UK). For growth on solid medium, Middlebrook 7H10 plates were supplemented with 0.5% glycerol and 10% OADC. BCG Pasteur Δ panCD was grown in medium supplemented with 24 μg/ml calcium pantothenate (Sigma-Aldrich, UK).

Construction of the fluorescent mycobacterial strain

We constructed a recombinant fluorescent mycobacterium, BCG Pasteur Δ panCD [Psmyc Turbo635asv-YFP] (FluorBCG), expressing dual fluorophores (one unstable to increase detection of growth/death). The pCB22 backbone plasmid (6572 bp; Dr Nathalie Cadieux, Aeras Global TB Vaccine Foundation) is a shuttle vector containing E . coli and mycobacterial origins of replication, a hygromycin resistance cassette, and the panCD operon driven by the constitutive BCG_3667c promoter; the plasmid complements the panCD auxotrophy in the host BCG Δ panCD strain ensuring the plasmid is stably maintained without antibiotic selection. Turbo635 has been previously used for in vivo imaging of mycobacteria [ 19 ], and Turbo635asv has a C-terminal degradation tag added to promote protein turnover [ 20 ]. The monomeric superfolder (msf) yellow fluorescent protein (YFP) is a derivative of GFP [ 21 ]. An artificial operon was constructed with both genes driven by the Psmyc mycobacterial strong promoter [ 22 ] and cloned into the XbaI-PacI sites in the pCB22 MCS downstream of the panCD operon ( S1 Fig ). The fluorescent BCG reporter strain was obtained by electroporation of the plasmid into BCG Pasteur Δ panCD as previously described [ 23 ].

A dual fluorescent reporter strain of Mtb was also constructed by transforming the same Psmyc Turbo635asv-YFP reporter plasmid into an H37Rv panCD mutant kindly provided by Prof Bill Jacobs, Albert Einstein College of Medicine to create Fluor-Mtb.

Immunisation and challenge

Mice were vaccinated subcutaneously with 1 × 10 4 BCG Pasteur and rested for 4 weeks before challenge. In Mtb challenge groups, each mouse was infected with 1 × 10 3 colony-forming units (CFUs) of strain H37Rv in 35 μl via the intranasal route. For the fluorescent strain challenge, either 5 × 10 6 CFU of FluorBCG or 5 × 10 6 CFU of Fluor-Mtb was injected intradermally (ID) with a 30G insulin needle (Easy Touch, United States of America) into the skin on the dorsal side of the mouse ear. Ears were imaged immediately post-challenge to provide a reading for baseline fluorescence; this was the day 0 time point. Imaging of the ears was repeated at regular intervals until day 28. Four weeks post-challenge with Mtb, the bacterial burden was determined in the lungs and spleen. Lungs and spleen were removed aseptically, homogenised in PBS containing 0.05% Tween 80, and serial dilutions of the organ homogenates were plated on Middlebrook 7H10 agar plates. The number of CFU was enumerated 21 days later. A summary of experiments and treatments is provided in S1 Table .

ChAdOx1.PPE15 strain and immunisations

ChAdOx1.PPE15 was a kind gift from Dr Elena Stylianou and Prof Helen McShane, University of Oxford [ 24 ]. Female BALB/c mice were immunised intranasally with 1 × 10 8 infectious units of the virus in a final volume of 35 μl. A summary of experiments and treatments is provided in S1 Table .

Portable imager

The portable imager instrument was designed to provide quantitative images of the fluorophore emission from intradermal injections of FluorBCG in the mouse ears. FluorBCG is excited by green (Cree XPE2, Green) and amber (Cree XPE2, Amber) LEDs, which are chosen to have emission maxima near the peak excitation of the YFP and Turbo635 fluorophores, respectively. The LEDs broad band emissions are narrowed using dichroic filters centred at 505 nm (YFP) and 590 nm (Turbo-365) with passbands of 20 nm. The Nikon D5300 camera exposure settings are typically ISO 400 and shutter speed 1/30 second. The images are captured in RAW format and then are converted to 48-bit TIFF images for quantitation. The images are analysed using custom software that integrates the fluorescent intensities over a user-specified region, typically at 4 mm circular region centred on the intradermal injection location.

Construction details of the portable imager

The imager consists of a Nikon 5300 SLR colour camera with a 24.2-megapixel sensor with a 3.89-μm pixel pitch, and a 14-bit resolution depth. The image is formed on the camera sensor using an image relay system consisting of a 75-mm focal length, 25-mm diameter objective lens and a 100-mm focal length, 50-mm diameter tube lens. The field of view of the imaging system is approximately 18 mm by 12 mm. The mouse ear is located at the focus of the objective lens and the camera sensor is located at the focal plane of the tube lens. The mouse is located in a sealed chamber with a transparent cover. The mouse ears are gently held flat in the chamber and are positioned to be at the image plane of the camera. The filtered LED broad band emission of YFP (505 nm) and Turbo635 (590 nm) is partially collimated by a 25-mm focal length, 25-mm diameter lens located approximately 25 mm from the LED source. Typical LED intensities at the image plane are 20 mW/cm 2 at 505 nm and 25 mW/cm 2 at 590 nm. The fluorescent emission from the bacteria passes through a dichroic filter placed between the objective lens and the tube lens with passbands centred at 542 nm and 639 nm, with passband widths of 27 nm and 42 nm, respectively.

Python pipeline

A semiautomated, user-friendly python pipeline was created for data analysis. The pipeline requires python 3.7.0 to be installed and packages: numpy 1.20.1, matplotlib 3.0.2, pandas 0.23.4, scipy 1.6.2, and xlsxwriter 1.3.8. Inputs for the pipeline are the text files generated from the custom-built imager containing the average intensity for YFP and Turbo635 fluorescence. The imager software has established analytical routines for measuring the average fluorescence over a defined area and storing this information as text files rather than images, allowing for a less memory intensive pipeline input. It is possible to collect background measurements of fluorescence and subtract these values from the measurements of the injected area to decrease background noise. While different visualisations can be generated using the pipeline, the initial processing steps are consistent. First, as images of each ear are taken separately, these fluorescence intensity values are averaged for each mouse and the standard deviation for all mice at each time point is calculated. The pipeline can visualise any number of mice and time points but assumes there is a maximum of 2 fluorescence colours. In addition to plotting the average fluorescence intensity value for all mice at each time point, the pipeline can also plot the normalised average values, where each time point after the first recorded fluorescence value is calculated as a percentage of the initial value. The code generated for the python pipeline is available here: https://doi.org/10.5281/zenodo.12781429 .

Isolation of cells from murine ear

Mouse ears were removed postmortem using dissecting scissors and cut into small pieces to facilitate tissue digestion. The tissue pieces were incubated in a digestion cocktail of 300 μg/ml Liberase TM (Roche, UK) and 50 U/ml of DNaseI (Sigma-Aldrich, UK) at 37°C with slow agitation for 90 min [ 25 ]. The digested skin fragments were passed through a nylon 100 μm cell strainer (Falcon, Thermo Fisher Scientific, UK) to obtain single-cell suspensions. Cell viability was >90% as determined by Trypan Blue (Sigma-Aldrich, UK) exclusion.

Flow cytometry and cell sorting

Red blood cells were lysed using RBC lysing buffer (Sigma-Aldrich, UK). Cells were subsequently washed, enumerated, and the cell concentration adjusted to 1 × 10 7 cells/ml in PBS (Thermo Fisher Scientific, UK). The single-cell suspension was incubated with TruStain FcX anti-mouse CD16/CD32 antibody (BioLegend, UK) for 10 min at 4°C. Following Fc block, cells were stained using horizon fixable viability stain 510 (Becton Dickinson, UK) for 15 min at room temperature, washed, and then resuspended in staining buffer (PBS +1 mM EDTA (Invitrogen, UK) +0.1% BSA (SigmaAldrich, UK)). Cells were then surface stained with fluorochrome-conjugated antibodies summarised in Table 1 . Following antibody staining, cells were washed, resuspended in PBS, filtered through a Falcon tube with an integrated cell strainer (Fisher Scientific, UK) before being analysed and sorted using the FACSAria III (Becton Dickinson, UK) cell sorter housed in a biosafety cabinet. Flow cytometry data were acquired using the FACSDiva software. The gating strategy used to identify the immune subsets in the ear are presented as supporting data ( S5 Fig ). Post-acquisition analysis was performed using FlowJo 10.7.1 (Becton Dickinson).

PPT PowerPoint slide
PNG larger image
TIFF original image

All antibodies were purchased from BioLegend, UK.

https://doi.org/10.1371/journal.pbio.3002766.t001

Statistical analysis

To assess the efficacy of the skin challenge, 2 nonlinear mixed effect models were built to, empirically, describe the decline of fluorescence over time. The model was based on data collected from 5 experiments with a total of 100 mice ( S1 Table ), including 45 vaccinated with BCG, 45 unvaccinated, and 10 vaccinated with ChAdOx1.PPE15. Eighty mice received the FluorBCG reporter, and 20 Fluor-Mtb. There were 1,030 observations in total. Among these, 455 observations were from unvaccinated mice, 455 observations were from BCG-vaccinated mice, and 120 observations were from ChAdOx1.PPE15-vaccinated mice. Forty-five mice underwent pulmonary challenge with Mtb. This subset included 20 mice vaccinated with BCG, 20 unvaccinated mice, and 5 mice vaccinated with ChAdOx1.PPE15. The median and range of log10_fluoroscence from the YFP reporter was 8.754 (min: 6.326, max: 9.476). Similarly, for the Turbo635 reporter, the median and range was 7.575 (min: 6.847, max: 8.836). Twenty animals received the Fluor-Mtb reporter, of which 10 were vaccinated with BCG, and 10 unvaccinated. A total of 220 observations were collected from the Fluor-Mtb reporter animals.

Separate models were developed for each fluorophore using log transformed data. Different structural models were evaluated (Eqs 1 and 2 ) to find a base structural model to empirically describe the 2 endpoints.

An additive residual error model was used. Thereafter, all combinations of inter-animal variability (IAV) were explored on the different parameters of the base model. The individual parameters were assumed to be log-normally distributed and covariance between different IAV were explored. Furthermore, Box-cox transformation of the IAV was evaluated on all parameters.

The final base model was taken forward to study the effects of available covariates (vaccination status, pulmonary challenge status, reporter, and baseline fluorescence) using stepwise-covariate modelling (SCM) [ 26 ]. The SCM was configured to include a covariate on a parameter if it would result in an objective function value (OFV) drop corresponding to a statistical significance change of p ≤ 0.05 during the forward step. For the backward step, it was configured to only remove covariates if the removal corresponded to a statistically significant increase of the OFV using a stricter criterion of p ≤ 0.01. Only the reporter covariate was allowed to be implemented on the first intercept parameter, while all covariates were allowed to be implemented on the remaining parameters.

Parameter estimation was performed using nonlinear mixed effects modelling in NONMEM (version 7.5.1; ICON plc, North America, Gaithersburg) [ 27 ]; the first-order conditional estimation method with interaction (FOCEI) was used [ 28 ]. Perl-speaks-NONMEM (PsN) (version 5.3.1, Department of Pharmacy, Uppsala University, Sweden) was utilised to run models, generate visual predictive checks (VPCs), and SCM runs [ 29 ].

Model evaluation was performed by comparing the OFV of nested hierarchical models, where a decrease in OFV of 3.84 can be considered statistically significant at a 5% level for one degree of freedom change using a χ 2 distribution. In addition, stratified VPCs, goodness of fit plots, precision of model parameters, shrinkage, scientific plausibility, and model stability were considered in the model selection and evaluation procedure. Graphical and numerical analysis of the data was performed in R (version 4.2.3, R Foundation for Statistical Computing, Vienna, Austria). All model evaluation plots were generated in R package Xpose4 (version 4.7.2; Department of Pharmacy, Uppsala University, Sweden). Documentation and comparison between models were performed using Pirana (Version 23.1.1, build 1, Certara, Princeton, USA).

The final model for each fluorophore was used to simulate the typical predictions (no IAV or residual error) over 30 days to visualise the covariate effects found on the measured endpoints fluorescence over time.

Fluorophore selection

There are 2 issues to consider when imaging fluorescent reporters in vertebrates. Firstly, the ability of light to penetrate tissue is a function of its wavelength, with red light above 600 nm showing less light absorption by haemoglobin [ 30 ]. Secondly, fluorescent proteins are stable, so signals take time to decay after the cell stops producing the fluorophore. To address these issues, we decided to construct a dual reporter strain, FluorBCG, including an unstable version of a red fluorophore that would ensure tissue penetration and be degraded faster as the bacteria stop growing and die, plus a stable fluorophore protein as an additional marker of bacteria should degradation of the unstable fluorophore take it below detectable levels. We surveyed the available fluorophores for those reported to be bright and stable in mycobacteria. We have previously demonstrated the utility of the red fluorophore Turbo635 for imaging in murine systems [ 19 ] and made an unstable derivative by the addition of the tripeptide ASV to the C-terminus [ 20 ]. We selected a YFP reported as bright and stable in mycobacteria [ 21 ]. A YFP is a monomeric superfolder derivative of GFP that folds faster and more efficiently with superior solubility and brightness [ 31 ]. Fig 1A shows each fluorophore was detected by the custom imager, but Turbo635ASV signals were noticeably dimmer. Quantification showed 10-fold less Turbo635ASV RFUs compared to YFP ( S2 Fig ).

( A ) Experimental schema ( B ) and ( D ) raw YFP fluorescence ( C ) and ( E ) normalised YFP fluorescence from both control and BCG vaccinated mice post-ID skin challenge with fluorescent BCG (FluorBCG). Representative data from 2 experiments are shown. Data represent mean fluorescence ± SD from n = 5 mice (average of 2 ears per mouse). The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; ID, intradermal; YFP, yellow fluorescent protein.

https://doi.org/10.1371/journal.pbio.3002766.g001

Utilising fluorescence output as a measure of vaccine efficacy in a murine skin challenge model

We chose the skin of the mouse ear as a suitable model to establish proof-of-concept for noninvasive monitoring of vaccine responses using a fluorescence-based readout. The mouse ear is thin with little fur or pigmentation, making it ideal for imaging and avoiding the autofluorescence associated with fur. A dose escalation study was carried out to determine the bacterial dose that would provide the optimum signal to noise ratio for imaging. Female BALB/c mice were dosed with 3 different concentrations of fluorescent BCG (FluorBCG): low dose of 5 × 10 4 CFU, mid-dose of 5 × 10 5 CFU, and high dose of 5 × 10 6 CFU. The high dose inoculum provided the optimum signal to noise ratio for YFP and was well tolerated in mice with no adverse effects ( S2 Fig ). Female BALB/c mice were then vaccinated with BCG and 4 weeks later challenged ID in each ear with 5 × 10 6 CFU of FluorBCG. Control mice were unvaccinated. Fluorescence readings were taken at the indicated time points until 28 days post-challenge ( Fig 1A ). A time-course experiment captures the YFP fluorescence dynamics observed in the control and BCG-vaccinated groups ( Fig 1B and 1D ). Data were normalised to day 0 to allow for operator-dependent variations in the initial intradermal inoculation in the ear ( Fig 1C and 1E ). The initial YFP readout (days 0 to 2) was similar in both vaccinated and control groups. From day 6 to day 21, there is a marked difference in groups, with a lower YFP readout in the vaccinated compared to control mice. After day 21, the YFP output declines in both vaccinated and control groups ( Fig 1B–1E ). In the control group, the increase in YFP fluorescence on day 5 and day 6 could represent a short growth period. In contrast, in the BCG-vaccinated group, the fluorescence readout declines throughout the measurement period. There are some variations in signal, but these are mirrored in both vaccinated and control groups suggesting a measurement anomaly on those days. The YFP fluorescence kinetics in the vaccinated and control groups are paralleled in the Turbo635 output, with the only major difference being a 10-fold decrease in relative fluorescence units (RFUs) ( S3 Fig ).

Fluorescence output related to viable BCG burden in the ears

Fluorescent proteins have relatively long half-lives, which could result in false-positive signals from residual fluorescent proteins when bacteria are no longer viable. To investigate BCG replication kinetics with fluorescence output, a time-course experiment was performed measuring the YFP readout ( Fig 2A and 2B ) and quantifying bacterial load ( Fig 2C and 2D ) over 21 days. Both vaccinated and control mice had similar fluorescence readout on days 0 and 2; however, BCG-vaccinated mice displayed lower YFP output between days 7 and 21 ( Fig 2A ). The trend was mirrored in the readout from the Turbo635 channel ( S4 Fig ). The effect of BCG immunisation is more evident post-normalisation, with the fluorescence readout being lower in the vaccinated mice in comparison to the controls ( Fig 2B ).

Mice were imaged and fluorescence from ( A ) YFP channels were quantified and normalised to day 0 ( B ) following intradermal skin challenge with fluorescent BCG. Data represent mean fluorescence ± SD from n = 5 mice. The bacterial load was quantified in the ears ( C ) and in the lymph nodes ( D ). The ears and proximal auricular draining lymph nodes from n = 5 mice were pooled, and CFUs were quantified at various time points. Data from one of 2 experiments are shown ( A and B ) and average of 2 experiments ( C and D ). Error bars represent mean ± SD. The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; CFU, colony-forming unit; YFP, yellow fluorescent protein.

https://doi.org/10.1371/journal.pbio.3002766.g002

To link bacterial viability with fluorescence output, bacterial numbers were enumerated in the ear at days 2, 7, and 21 post-challenge ( Fig 2C ). There is a decrease in the CFUs after day 2, with a further decrease in the vaccinated animals between days 7 and 21, which parallels the onset of adaptive immunity.

It is well documented that BCG is trafficked by host cells from the site of injection to the draining lymph nodes [ 32 – 37 ].To confirm movement of FluorBCG from the ear, bacterial load was quantified in the proximal auricular draining lymph nodes at day 21 post-challenge and showed that the BCG-vaccinated mice had lower bacteria load than the control group, although this was not statistically significant ( Fig 2D ).

Antigen presenting cells dominate the local immune environment in the mouse ear following BCG vaccination

To study the cellular milieu in mouse ears after FluorBCG challenge, a panel of defined markers was used to characterise the different cellular subsets by flow cytometry. To obtain single-cell suspensions, mouse ears were enzymatically digested and passed through a cell strainer to obtain viable cells. The gating strategy used for flow cytometry was adapted from Yu and colleagues‘ study ( S5 Fig ) [ 38 ]. The ear skin consists of several types of innate immune cells including neutrophils, macrophages, dendritic cells, T cells, and mast cells. The local environment of the mouse ear at day 7 following FluorBCG challenge of control animals is dominated by neutrophils, Langerhans cells, and macrophages ( Fig 3A ). In response to intradermal injection of FluorBCG in control animals, neutrophils in the ear increased to a maximum of 5.79% on day 7, decreasing to 1.74% by day 14 ( Fig 3A and 3B ). In vaccinated mice challenged with FluorBCG, the frequency of neutrophils at day 7 (3.13%) is lower in comparison to control animals (5.79%), dropping further by day 14 ( Fig 3A and 3B ). The marker Langerin (CD207) identifies the main subsets of dendritic cells that reside in the mouse ear. Langerhans cells are CD207 + , and, although conventionally thought of as belonging to the dendritic cell cohort, recent evidence has redefined these cells to be more like tissue-resident macrophages that have acquired dendritic cell-like functions. Langerhans cells possess both in situ functions and migratory functions that promote antigen presentation and T cell priming [ 39 ]. A high frequency of Langerhans cells is present in both control and vaccinated mice with an increase overall in vaccinated mice between day 7and day 14 ( Fig 3A and 3B ). Localisation of Langerhans cells in the mouse ear corroborates the known prevalence of this subset in the skin epidermis [ 39 ]. Another important subset of cells in the skin are the dermal dendritic cells, which are subdivided into CD207 − CD103 +/− and CD207 + CD103 + . Both these subsets of dendritic cells are up-regulated in response to BCG vaccination, with the vaccinated mice exhibiting a large influx of CD207 + dermal dendritic cells ( Fig 3A and 3B ) in comparison to the control mice. In response to BCG vaccination, there is an influx of macrophages that persist in the skin until day 14 ( Fig 3A and 3B ).

Mice were immunised with BCG and post-4-week rest, mice were challenged ID in the ear with FluorBCG and cellular infiltrates at the site of challenge were measured at day 7 ( A ) and day 14 ( B ). Gates in contour plots contains single-cell populations—neutrophils (CD45 + Ly6G + ), dendritic cells subdivided into Langerhans cells* (CD45 + MHC-II + CD207 + CD103 − ), dermal DCs (CD45 + MHC-II + CD207 − CD103 + ) and dermal langerin + DCs (CD45 + MHC-II + CD207 + CD103 + ); monocytes (CD45 + CD11b + CD64 int ) and macrophages (CD45 + MHC-II + F4/80 + CD64 + ). *Langerhans cells, although classified under the dendritic cells group here, are more like tissue-resident macrophages, which acquire a phenotype-like dendritic cells. The data underlying this figure can be found at https://doi.org/10.5281/zenodo.12794251 . BCG, bacille Calmette-Guérin; DC, dendritic cell; ID, intradermally.

https://doi.org/10.1371/journal.pbio.3002766.g003

These results suggest that the response to FluorBCG intradermal challenge leads to an inflammatory influx dominated by migratory dendritic cells, neutrophils, and macrophages.

Next, we investigated if viable FluorBCG can be recovered from selected subsets of immune cells from the ear. Neutrophils, macrophages, and dendritic cells were gated for the presence of YFP + BCG cells, single-cell sorted using a FACS Aria sorter, and plated to determine CFUs. Neutrophils (CD45 + Ly6G + ) accounted for a minor population of cells in the ear ( Fig 4A and 4B ) but had a high BCG burden, which was comparable to the macrophage and dendritic cells present in the ear at day 7 ( Fig 4E ) with a reduced bacterial load on day 14 ( Fig 4F ). The reduction in viable bacterial numbers in neutrophils noted at day 14 ( Fig 4F ) does not match with the expression of YFP + bacteria in neutrophils ( Fig 4D ); however, the frequency of YFP+ neutrophils does match with frequency of YFP+ macrophages and dendritic cell subsets. In contrast, macrophages were the dominant population in the mouse ear ( Fig 4A and 4B ), but only 12% of this subset carried YFP + BCG ( Fig 4C and 4D ). Overall, reduction of the viable bacterial load was observed by day 14 in neutrophils, macrophages, and dendritic cells ( Fig 4F ).

Bar graphs ( A , B ) depicting immune cell profiles of neutrophils (expressed as a frequency of CD45+ cells), macrophages, and dendritic cells (expressed as a frequency of the myeloid cell population). Immune cellular subsets were further gated on YFP+ cells to isolate populations that phagocytosed fluorescent BCG ( C , D ). These gated populations were sorted and plated to quantify viable BCG ( E , F ). Flow cytometry data are an average of 2 independent experiments with n = 5 mice per group. Error bars represent mean ± SD. The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; CFU, colony-forming unit; YFP, yellow fluorescent protein.

https://doi.org/10.1371/journal.pbio.3002766.g004

Fluorescence readout from the skin relates to protective immunity in the lung

The ideal route of challenge in humans would be the pulmonary route as it mimics the natural route of infection with Mtb but detecting challenge bacteria in vivo in the lung poses obstacles. Exposure via the skin is a more feasible approach, and detection of the bacteria can be achieved by noninvasive methods. It has been reported that the efficacy of BCG vaccination against intradermal BCG skin challenge has comparable outcomes to an aerosol Mtb challenge, highlighting the viability of using a BCG-based skin challenge as an alternative to a pulmonary challenge [ 34 ]. To test whether the reduction in skin fluorescence in response to BCG vaccination matched an immune response in the lungs, mice were BCG vaccinated, challenged with FluorBCG in the ear, and given a pulmonary Mtb challenge by the intranasal route. In response to BCG vaccination, there was a reduced YFP fluorescence readout overall from the skin of vaccinated mice ( Fig 5A and 5B ) and Turbo635 ( S6 Fig ). In the lungs and spleen, BCG vaccination significantly reduced the bacterial burden in comparison to the control group ( Fig 5C and 5D ). The group of control mice that received a fluorescent BCG challenge (Unvaccinated_FluorBCG_H37Rv) in the ear had a statistically lower bacterial load in the lungs and spleen in comparison to the control mice (Unvaccinated_H37Rv) that did not receive a fluorescent BCG inoculation ( Fig 5C and 5D ), suggesting the single-dose BCG challenge in the ear affords a degree of protection. The results show that the reduced fluorescence readout from FluorBCG in the skin provides a sensitive and reproducible measure of a relevant biological effect that is shown to reflect traditional measures of TB vaccine efficacy in the lung.

BALB/c mice were immunised with a single dose of BCG and after a 4-week rest period, mice were either challenged ID in the ear with 5 × 10 6 CFU of FluorBCG and IN with 1 × 10 3 H37Rv or only challenged IN with 1 × 10 3 H37Rv (H37Rv). Fluorescence intensities from the site of ID challenge in the ear are represented as raw outputs from the YFP channel ( A ) and fluorescence values normalised to day 0, YFP ( B ). Lungs and spleen were harvested 4 weeks post-challenge and processed to quantify the bacterial burden ( C and D ). Representative data from duplicate experiments are displayed. Data represent mean ± SD from n = 5 mice. The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; CFU, colony-forming unit; ID, intradermally; IN, intranasally; YFP, yellow fluorescent protein.

https://doi.org/10.1371/journal.pbio.3002766.g005

Evaluating the skin challenge model using a novel TB subunit vaccine candidate

The skin challenge model was utilised to assess the efficacy of a novel vaccine candidate, ChAdOx1.PPE15. This subunit vaccine candidate comprises a replication-deficient, recombinant chimpanzee adenovirus vector (ChAdOx1), which expresses the mycobacterial antigen PPE15 [ 24 ]. Subunit vaccines are primarily designed to boost protective immune responses conferred by the BCG vaccine, but it has been shown that administering ChAdOx1.PPE15 as a single intranasal dose followed by a Mtb challenge resulted in a significant reduction of bacteria load in the lungs compared to controls [ 16 , 24 ]. We immunised BALB/c mice with a single intranasal dose of ChAdOx1.PPE15 and 4 weeks later challenged them with intranasal Mtb and intradermal FluorBCG in the ear. The fluorescence output from immunised mice was stable until day 16 after which there was a decline in the YFP readout. A diminished fluorescence output from the ChAdOx1.PPE15 immunised group in comparison to the control group was noted ( Fig 6A ). However, the YFP ( Fig 6B ) and Turbo635 ( S7 Fig ) normalisation data did not indicate a vaccine effect in mice immunised with only ChAdOx1.PPE15 ( Fig 6B ). Normalisation of data to day 0 is critical as it removes any confounding variables such as differences in initial dosing. Mice immunised with ChAdOx1.PPE15 also failed to demonstrate a protective effect in the lungs and spleen ( Fig 6C and 6D ), which corroborates the fluorescence data ( Fig 6A and 6B ), and mirrors the poor responses reported [ 24 ]. A reduction in bacterial load in both lungs and spleen were primarily noted in the groups that received a dose of FluorBCG in the ears ( Fig 6C and 6D ). Comparison of ChAdOx1.PPE15 vaccinated with BCG vaccinated groups showed no overall difference in terms of RFU ( S10 Fig ). BCG performed better than ChAdOx1.PPE15 at controlling Mtb H37Rv CFU in spleen but not lungs ( S11 Fig , panels A and B); ChAdOx1.PPE15 showed better control of lung CFU but not spleen CFU in the groups that also received intradermal challenge with FluorBCG ( S11 Fig , panels C and D).

BALB/c mice were immunised with a single intranasal dose of ChAdOx1.PPE15, and, after a 4-week rest period, mice were either challenged ID in the ear with 5 × 10 6 CFU of FluorBCG and challenged IN with 1 × 10 3 H37Rv (H37Rv). Fluorescence intensities from the site of ID challenge in the ear are represented as raw outputs from the YFP channel ( A ) and fluorescence values normalised to day 0, YFP ( B ). Lungs and spleen were harvested 4 weeks post-challenge and processed to quantify the bacterial burden ( C and D ). Data represent mean ±SD from n = 5 mice. The data underlying this figure can be found in S1 Data . CFU, colony-forming unit; ID, intradermally; IN, intranasally; YFP, yellow fluorescent protein.

https://doi.org/10.1371/journal.pbio.3002766.g006

Statistical analysis of aggregated experimental data

We built nonlinear mixed-effects models of the aggregated data collected from 5 independent experiments. Based on the graphical and numerical analysis, the baseline fluorescence of YFP in 1 mouse was excluded as it was approximately 100 times lower compared to the median value.

The final model for YFP fluorescence consisted of a biexponential structural model that provided a better fit to the data than a monoexponential model. In addition, a biexponential model was deemed more suitable to describe the initial rapid decline of fluorescence for some groups of mice, which was confirmed by comparing the ability of the mono- and biexponentials models to describe the data through stratified VPCs. IAV on the second intercept and the slope were statistically significant; no covariance between IAV parameters were supported by the data. Box-cox transformation on individual error (eta) distributions were not supported. An additive residual error model was found to be adequate. Covariates in the SCM procedure were implemented as percentage change of typical parameter values. Statistically significant covariates were vaccination (1.52% decrease, p < 0.001), baseline fluorescence (10 th-quantile = 2.2% decrease and 90 th-quantile = 1.36% increase, p < 0.001), Mtb pulmonary challenge (1.31% increase, p < 0.001), and the reporter (3.43% decrease, p < 0.001) on the second intercept, reporter on first intercept (24.2%, increase, p < 0.001), and reporter on the second SLOPE (52.4% decrease, p < 0.001). However, the covariate effect of vaccination on the second intercept was not statistically significantly different from unvaccinated for the ChAdOx1.PPE15 vaccine, and, as such, the group of mice treated with ChAdOx1.PPE15 was treated as unvaccinated.

The structural base model for Turbo635 fluorescence also consisted of a biexponential model. IAV on the first and second intercept were statistically significant, and no covariance between IAV parameters was found to be significant. Box-cox transformation on ETA distributions were not supported, and an additive residual error model was found to be adequate. Statistically significant covariates were vaccination (1.47% decrease, p < 0.001), baseline fluorescence (10 th-quantile = 2.3% decrease and 90 th-quantile = 3.11% increase, p < 0.001), and reporter (4.93% decrease, p < 0.001) on the second intercept, reporter (12.3% increase, p < 0.001) on the first intercept, reporter (45.2% decrease, p < 0.001) and baseline fluorescence (10 th-quantile = 22.3% decrease and 90 th-quantile = 30.3% increase, p < 0.001) on the second slope. Similar to the YFP model, the ChAdOx1.PPE15-vaccinated groups was not statistically significantly different from unvaccinated mice.

The final models were considered suitable to describe the data in all different experimental groups based on stratified VPCs ( S8 and S9 Figs). In Figs 7 and 8 , typical predictions can be found for all the possible covariate combinations for both the models. The final parameter estimates for the 2 models are provided in Tables 2 and 3 .

Log10 RFU based on Turbo635 versus time for different combinations of covariate effects. The typical predictions were performed using the final model for Turbo635 (one prediction of the dependent variable per time point, using a fine time grid, over 28 days). This is done for each possible covariate effect. The plots can be recreated by using the parameter values in Table 2 and the covariate relationships described for Turbo635 in the results section.

https://doi.org/10.1371/journal.pbio.3002766.g007

Log10 RFU based on YFP versus time for different combinations of covariate effects. The typical predictions were performed using the final model for YFP (one prediction of the dependent variable per time point, using a fine time grid, over 28 days). This is done for each possible covariate effect. The plots can be recreated by using the parameter values in Table 3 and the covariate relationships described for YFP in the results section.

https://doi.org/10.1371/journal.pbio.3002766.g008

https://doi.org/10.1371/journal.pbio.3002766.t002

https://doi.org/10.1371/journal.pbio.3002766.t003

We report on a proof-of-concept noninvasive, skin-based challenge model for detection and quantification of relevant biological activity reflecting traditional measures of TB vaccine efficacy in the mouse (viable bacterial loads in the lung). Although the utility of the BCG skin challenge model has previously been demonstrated in humans [ 16 ], to our knowledge, this is the first reported noninvasive, skin-based challenge study to demonstrate the feasibility of using a fluorescence-based readout as a measure of vaccine efficacy. We also show that the fluorescence output from the skin after challenge with fluorescent BCG serves as a reliable indicator of vaccine-induced immunity in the lung. The fluorescence-based quantitative measurement of protective efficacy described in this study solves several problems associated with vaccine assessment in humans. It uses a proven safe challenge organism, and it allows for repeated noninvasive measurement at the challenge site, providing high-quality temporal data on vaccine responses in vivo. It uses a low-cost imaging platform that can be used in the field, and the technology and approach have potential applications in other areas.

The maximal difference in fluorescence signal between the control and vaccinated groups were observed between day 4 and day 21, providing a window in which new vaccine candidates can be assessed for their efficacy using this noninvasive approach. This defined interval for determining vaccine efficacy will cover the period of collective input from both the innate and adaptive immune system, thus providing a more comprehensive output of vaccine protective immunity. This time interval is consistent with the optimum period for detecting live BCG in the skin [ 34 , 40 ]. Although in this first-generation approach, there was some variability in the fluorescence readout, making it harder to achieve statistical significance at specific time points. However, a trend for reduced fluorescence output is noted in the groups of mice vaccinated with BCG compared to control. The relatively long half-life of fluorescent proteins does not seem to impact their utility in this system. The addition of the tripeptide ASV degradation tag to Turbo635 did not change the fluorescent dynamics in this system over time compared to YFP but instead reduced the overall fluorescent signal. Whether Turbo635-ASV is advantageous in other systems remains to be tested and there is potential for fluorescence optimisation.

We used the dorsal surface of the mouse ear as the site for intradermal injection and imaging. Use of the mouse ear as a surrogate for intradermal vaccination site has several advantages: (1) autofluorescence is low, so background fluorescence levels are negligible; (2) ease of accessibility; and (3) repeated imaging of the same area requires no advance preparation, such as hair removal. Using a microfine, 30G insulin needle, we could precisely deliver the dose needed while minimising disruption of the local microenvironment.

Live BCG was detectable in the skin till day 21, although BCG counts were not quantified at later time points in this study. In comparison, Minassian and colleagues reported that BCG could be detected until 12 weeks in the mouse ear by culture. However, a significant decline in BCG CFUs was observed after 4 weeks post-immunisation [ 34 ]. Furthermore, in a clinical study, live BCG could be cultured from skin punch biopsy specimens up to 4 weeks [ 16 ] and 2 months [ 41 ] after the BCG challenge. We observed a slight increase in YFP output from unvaccinated mice between day 2 and day 7; however, this increase in fluorescence readout did not correspond to an increase in CFU counts between day 2 and day 7. Hence, we conclude there was no replication of BCG at the immunisation site between day 2 and day 7. However, a delicate equilibrium may exist between bacterial replication, migration, and bacterial death in the ears, a critical point that needs to be addressed in future experiments. Chambers and colleagues [ 35 ] demonstrated that early accumulation of inflammatory cells at the immunisation site reflected the clearance of live BCG from the footpad of mice to the draining lymph nodes. In contrast, dead BCG persisted at the immunisation site for extended periods [ 35 ]. We have shown that bacteria migrate out of the ears to the draining lymph nodes with no significant difference in CFU counts between vaccinated and control mice at day 21 post-FluorBCG challenge in the ear. A decline in CFUs was observed by day 7 in both control and BCG-vaccinated mice. By day 21, a reduction in bacterial load was noticed in mice immunised with BCG, which matches the onset of the adaptive immune response and likely to involve more efficient priming of T cells in BCG-vaccinated mice [ 36 ].

Two empirical models were successfully developed to describe both fluorophores over time with good predictive performance for both FluorBCG and Fluor-Mtb. For Turbo635, the impact of covariate effects on the fluorescence decay was investigated. The vaccinated population with the FluorBCG reporter exhibits a decline compared to the unvaccinated group ( Fig 7A ), in addition to reaching a lower fluorescence level. For the Fluor-Mtb reporter, both the unvaccinated and vaccinated groups demonstrate a decline compared to the FluorBCG reporter, with the vaccinated group reaching a lower fluorescence level ( Fig 7C ) and displaying a decline compared to the unvaccinated group. For the identified covariate effects using YFP data, it was observed that the vaccinated population with the FluorBCG reporter exhibits a decline compared to the unvaccinated group and reaching a lower fluorescence level ( Fig 8C ). Additionally, the effect of pulmonary infection with Mtb leads to an earlier stabilisation of fluorescence levels and a decline for both unvaccinated and vaccinated ( Fig 8D and 8B ). In the case of the Fluor-Mtb reporter, both the vaccinated and unvaccinated groups show a decline compared to the FluorBCG reporter, with the vaccinated group ( Fig 8G and 8H ) reaching an even lower fluorescence level and displaying a decline compared to the unvaccinated group ( Fig 8E and 8F ). Similar to BCG, the effect of pulmonary infection results leads to an earlier stabilisation of fluorescence levels and a decline for both unvaccinated and vaccinated.

In the typical prediction plots, a rapid decline for the Fluor-Mtb-reporter was also noticed, and from Tables 2 and 3 , there is a difference in parameter values between the 2 reporters (FluorBCG and Fluor-Mtb). This could be attributed to the Fluor-Mtb reporter being more antigenically complete, thereby facilitating a quicker initial immunological response. However, the reasons behind the reduction in the second intercept value for the YFP reporter due to pulmonary challenge are uncertain. To the best of our knowledge, there are no physiological explanations for this phenomenon compared to the Turbo635 reporter. It is possible that there are minor outlier values within the experiments involving pulmonary infection or that the inclusion of the Mtb-challenge (MTBCH) factor accounts for variances stemming from factors other than the actual pulmonary infection in the YFP model. The effect of the baseline fluorescence was statistically significant for both the second slope and the second intercept in the case of Turbo635, while only for the second intercept in the case of YFP. This discrepancy may possibly be attributed to a higher variance in the readouts from the Turb635 fluorophore. The empirical models described the data well. However, the evaluations of model performance were performed on the same data that were used for model building. The typical prediction plots (Figs 7 and 8 ) are specific to the mouse model, which has been shown to be a useful screening tool for early-stage TB vaccines, and, although these specific data have yet to be confirmed in human studies, current evidence is supportive. The models analysed all data over time from all experiments simultaneously using a mixed effects approach, which is suitable for longitudinal datasets. An alternative would be to select a datapoint of interest and perform group-wise statistical comparisons. However, such an approach is not as informative and does not utilise the full set of generated experimental data.

Our study demonstrates the presence of migratory immune cells, particularly an accumulation of dendritic cells at the site of FluorBCG administration in both control and BCG-vaccinated mice. The skin milieu is complex with multiple migratory immune cell subsets [ 42 , 43 ], so we utilised additional cell surface markers to study the different populations of prominent dendritic cell populations in the ear. In this study, the marked decrease in the fluorescence readout in BCG-vaccinated mice from day 7 likely corresponds to an increase in the presence of migratory immune cells such as dermal langerin+ dendritic cells and, to a lesser extent, the Langerhans cells in the ear. BCG vaccination influences the early appearance of migratory dendritic cells [ 35 ]. Migratory immune cells are likely to phagocytose and carry BCG to the draining lymph nodes, although the migration of free bacteria to the lymph nodes cannot be discounted. The dermal langerin+ dendritic cells are known to migrate to lymph nodes in response to infection and inflammation and presents antigens to T cells [ 44 ]. These cells are also early responders and are continually replenished from the blood [ 45 ]. The presence of a large pool of FluorBCG at the administration site is likely to indicate continuous migration and replenishment of dendritic cells, or persistence in the ear. We observed that the population of dermal langerin+ dendritic cells are maintained in the ear until day 14. Neutrophils were observed in both control and vaccinated mouse skin, with the neutrophil population being smaller in the BCG-vaccinated group. Neutrophils are present in the skin under steady-state conditions, and they are part of the early responders to appear in the skin following sterile injury [ 46 ]. Neutrophils can be detected as early as 4 hours following intradermal BCG vaccination [ 33 ]. Neutrophils have also been shown to migrate to the draining lymph nodes in response to specific microbial stimuli [ 47 ]. Although neutrophils were not the dominant population of cells in the ear, they can phagocytose BCG, as the bacterial load in neutrophils was comparable to the bacterial load enumerated from dendritic cells and macrophages on day 7. Abaide and colleagues also reported on the colocalisation of neutrophils with BCG in the ear and demonstrated that neutrophils could phagocytose BCG present in the skin dermis [ 33 ].

By day 14, there was a smaller population of neutrophils present in the ear, with the bacterial load in neutrophils lower than the CFUs in macrophages and dendritic cells. However, the frequency of YFP+ neutrophils present in the skin was similar to YFP+ macrophages and YFP+ dendritic cells. The fluorescence readout is indicative that BCG is present in the neutrophils. However, the lower bacterial load by culture from these cells suggests that at day 14, the neutrophils are predominantly harbouring nonviable BCG. The role of neutrophils in the direct clearance of mycobacteria is not clear [ 33 , 48 – 50 ]; however, they are known to function cooperatively with dendritic cells and macrophages to orchestrate the killing of mycobacteria [ 51 ]. Moorlag and colleagues demonstrated that BCG vaccination could induce trained immunity in neutrophils, which contributed to a more efficient response to C . albicans infection by the up-regulation of reactive oxygen species and enhanced expression of degranulation markers [ 52 ]. Further work needs to be carried out to confirm if neutrophils play a direct role in mycobacterial killing due to increased neutrophil antimicrobial factors or an indirect role by establishing an early inflammatory response following BCG vaccination in our skin challenge model.

The skin-based challenge model described in this study has a unique advantage for detecting and measuring the protective efficacy by a noninvasive method. Furthermore, repeated sampling of the injection site with minimal discomfort could be carried out, eliminating the need for invasive skin sampling procedures. This study demonstrates that the skin-based fluorescence output is a measurable indicator of the vaccine protective response in the lung. A reduction in the fluorescence output from BCG-vaccinated group compared to the control mirrors a significant decrease in bacterial load in both the lungs and spleen of BCG-vaccinated mice. Control mice that received FluorBCG ID in the ear also demonstrate a vaccine effect that was expected and agrees with previously published data [ 36 ]. Additionally, a BCG boosting effect was observed in response to the intradermal administration of FluorBCG in vaccinated mice, which resulted in further reduction of CFU counts in the lungs and spleen. In our study, we have confirmed and extended on the results demonstrating that protection against a skin-based challenge is associated with a protective vaccine response in the lung [ 37 ].

This skin challenge model study is a feasibility study that will ultimately aid the development of a human challenge model for assessment of TB vaccines. Such a model needs to be tested for its utility of measuring responses from vaccines other than BCG. We used a subunit vaccine ChAdOx1.PPE15 to test the skin challenge model with a vaccine other than BCG. In our study, ChAdOx1.PPE15 was not effective in conferring a protective immune response, which was confirmed by both the fluorescence readout from the skin and CFU data from the lungs and spleen of BALB/c mice. This observation agrees with data published by Stylianou and colleagues, demonstrating that in CB6F1 mice, ChAdOx1.PPE15 was only effective in boosting BCG vaccination, and administration of the ChAdOx1.PPE15 vaccine on its own demonstrated a moderate vaccine efficacy in CB6F1 mice but not in BALB/c mice [ 24 ]. Significant protective immune responses were observed in the groups that were administered FluorBCG. BCG is a vaccine that induces strong protective immune responses in mice, so any effects of ChAdOx1.PPE15 will likely to be masked unless it can significantly boost the BCG immune response.

Conclusions

Several animal models exist for TB, but none encompasses the complexity of the spectrum of clinical disease, representing a significant hurdle in translational TB research. For TB, where prognostic animal models are lacking, human challenge trials will be beneficial to identify correlates of immune protection and test vaccine efficacy. TB, a chronic disease caused by a virulent bacterium, does not immediately fit the criteria as a disease for which a human challenge trial can be designed easily. Using alternative approaches to test the efficacy of vaccines is therefore warranted. This skin-based challenge model is one such approach to stratify and select vaccine candidates based on their immune response to accelerate the most relevant candidates along the clinical trial pipeline.

In summary, in this vaccination-challenge model, serially tracking fluorescently labelled bacteria in the skin using a portable imager is a novel noninvasive strategy for measuring vaccine efficacy. This novel detection method has unique advantages for clinical implementation, and this feasibility study establishes the first steps towards developing a safe, tractable, relevant human challenge model for TB.

Supporting information

S1 fig. pcb22-turbo635-asv-yfp plasmid map..

https://doi.org/10.1371/journal.pbio.3002766.s001

S2 Fig. Optimisation of BCG dose.

BALB/c mice were infected ID with 3 concentrations of FluorBCG: FluorBCG_High of 5 × 10 6 CFU, FluorBCG_Mid of 5 × 10 5 CFU, and FluorBCG_Low of 5 × 10 4 CFU. ( A ) Raw YFP fluorescence. ( B ) Raw Turbo-635 fluorescence from unvaccinated mice. Data represent mean fluorescence ± SD from n = 3 mice (mean of 2 ears per mouse). The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; CFU, colony-forming unit; ID, intradermally; YFP, yellow fluorescent protein.

https://doi.org/10.1371/journal.pbio.3002766.s002

S3 Fig. Measuring the Turbo-635 signal from the skin in response to a fluorescent BCG challenge.

( A , C ) Raw Turbo-635 fluorescence. ( B , D ) Normalised Turbo-635 fluorescence from both control and vaccinated mice post-ID skin challenge with fluorescent BCG. Representative data from one of 2 experiments are shown. Data represent mean fluorescence ± SD from n = 5 mice (average of 2 ears per mouse). The data underlying this figure can be found in S1 Data .

https://doi.org/10.1371/journal.pbio.3002766.s003

S4 Fig. Turbo-635 fluorescence output from BCG-vaccinated mice.

Mice were imaged and fluorescence from ( A ) Turbo-635 channels were quantified and normalised to day 0 ( B ) following intradermal skin challenge with fluorescent BCG. Data from one of 2 experiments are shown representing mean fluorescence ± SD from n = 5 mice. The data underlying this figure can be found in S1 Data .

https://doi.org/10.1371/journal.pbio.3002766.s004

S5 Fig. Gating strategy for flow cytometry.

Total cells were isolated from murine ear, and single cells were identified using the FSC-A and FSC-H plot. The viability of the skin cell suspension was determined using the SSC-A and viability plot. Viable cells were gated to identify the CD45 + cells and then neutrophils (CD45 + Ly6G + ), dendritic cells subdivided into Langerhans cells* (CD45 + MHC-II + CD207 + CD103 − ), dermal DCs (CD45 + MHC-II + CD207 − CD103 + ) and dermal langerin + DCs (CD45 + MHC-II + CD207 + CD103 + ); monocytes (CD45 + CD11b + CD64 int ) and macrophages (CD45 + MHC-II + F4/80 + CD64 + ).

https://doi.org/10.1371/journal.pbio.3002766.s005

S6 Fig. Fluorescence output from the ear and bacterial burden in the lungs.

BALB/c mice were immunised with a single dose of BCG, and, after a 4-week rest period, mice were challenged ID in the ear with 5 × 10 6 CFU of FluorBCG and IN with 1 × 10 3 H37Rv (FluorBCG+H37Rv). Fluorescence intensities from the site of ID challenge in the ear are represented as raw outputs from the Turbo-635 channel ( A ) and fluorescence values normalised to day 0 ( B ). Representative data from duplicate experiments are displayed. Data represent mean ± SD from n = 5 mice. The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; CFU, colony-forming unit; ID, intradermally; IN, intranasally.

https://doi.org/10.1371/journal.pbio.3002766.s006

S7 Fig. Testing the skin challenge model with a TB subunit vaccine.

BALB/c mice were immunised with a single dose of ChAdOx1.PPE15, and, after a 4-week rest period, mice were challenged ID in the ear with 5 × 10 6 CFU of FluorBCG and IN with 1 × 10 3 H37Rv (FluorBCG+H37Rv). Fluorescence intensities from the site of ID challenge in the ear are represented as raw outputs from the Turbo-635 channel ( A ) and fluorescence values normalised to day 0 ( B ). Representative data from duplicate experiments are displayed. Data represent mean ± SD from n = 5 mice. The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; CFU, colony-forming unit; ID, intradermally; IN, intranasally.

https://doi.org/10.1371/journal.pbio.3002766.s007

S8 Fig. VPC based on the final model for TURBO635 data stratified by all existing covariate effects based on 2,000 simulations.

The solid line is the median of the observed data, the shaded area is the 95% confidence interval for the median, and the open blue circles are the observations. The VPC was generated in NONMEM using the PsN VPC command. It can be regenerated by utilising the software specified in the statistical analysis method section and the S2 Data .

https://doi.org/10.1371/journal.pbio.3002766.s008

S9 Fig. VPC based on the final model for YFP data stratified by all existing covariate effects based on 2,000 simulations.

https://doi.org/10.1371/journal.pbio.3002766.s009

S10 Fig. Measuring the skin fluorescence signal from BCG- and ChAdOx.PPE15-vaccinated mice in response to a fluorescent BCG challenge.

BALB/c mice were either immunised with a single dose of BCG or a single intranasal dose of ChAdOx1.PPE15, and, after a 4-week rest period, mice were challenged ID in the ear with 5 × 10 6 CFU of FluorBCG. ( A , C ) Raw YFP fluorescence. ( E , G ) Raw Turbo-635 fluorescence. ( B , D ) Normalised YFP fluorescence from both BCG- and ChAdOx.PPE15-vaccinated mice post-ID skin challenge with fluorescent BCG ( F , H ) Normalised Turbo-635 fluorescence from both BCG- and ChAdOx.PPE15-vaccinated mice post-ID skin challenge with fluorescent BCG. Data represent the mean fluorescence ± SD from n = 5 mice (an average of 2 ears per mouse). The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; CFU, colony-forming unit; ID, intradermally; YFP, yellow fluorescent protein.

https://doi.org/10.1371/journal.pbio.3002766.s010

S11 Fig. Comparing the skin challenge model using a novel vaccine candidate and BCG.

BALB/c mice were either immunised with a single dose of BCG or a single intranasal dose of ChAdOx1.PPE15, and, after a 4-week rest period, mice were either challenged ID in the ear with 5 × 10 6 CFU of FluorBCG and challenged IN with 1 × 10 3 H37Rv (H37Rv) or only challenged with 1 × 10 3 H37Rv (H37Rv). Lungs ( A , C ) and spleen ( B , D ) were harvested 4 weeks post-challenge and processed to quantify the bacterial burden. Data represent mean ± SD from n = 5 mice. * p > 0.05, ** p > 0.01. The data underlying this figure can be found in S1 Data . BCG, bacille Calmette-Guérin; CFU, colony-forming unit; ID, intradermally; IN, intranasally.

https://doi.org/10.1371/journal.pbio.3002766.s011

S1 Table. Summary of experiment names and treatments.

Each experiment had 5 mice per group and was measured for both ears; mean ears are tabulated. VC-2, VC-4, and VC-5 had 12 RFU readings taken over 28 days. Flow1 and Flow2 had 5 RFU readings taken over 21 days and were also used for flow cytometry and bacterial load measurements in ears and lymph nodes.

https://doi.org/10.1371/journal.pbio.3002766.s012

S1 Data. Numerical data for plots in all figures.

https://doi.org/10.1371/journal.pbio.3002766.s013

S2 Data. Details of software and statistical analysis methods to recreate S8 Fig .

https://doi.org/10.1371/journal.pbio.3002766.s014

S3 Data. Details of software and statistical analysis methods to recreate S9 Fig .

https://doi.org/10.1371/journal.pbio.3002766.s015

Acknowledgments

Thanks to the TB-Human Challenge Team for their input and discussions. Special thanks to Barry Walker for his constant support for the project.

View Article
Google Scholar
PubMed/NCBI

Remote Access
Save figures into PowerPoint
Download tables as PDFs

Infectious Diseases: A Case Study Approach

34: Tuberculosis

David Cluck

Download Chapter PDF

Disclaimer: These citations have been automatically generated based on the information we have and it may not be 100% accurate. Please consult the latest official manual style if you have any questions regarding the format accuracy.

Download citation file:

Search Book

Jump to a Section

Patient presentation.

Full Chapter
Supplementary Content

Chief Complaint

“I have a cough that won’t go away.”

History of Present Illness

A 63-year-old male presents to the emergency department with complaints of cough/shortness of breath which he attributes to a “nagging cold.” He states he fears this may be something worse after experiencing hemoptysis for the past 3 days. He also admits to waking up in the middle of the night “drenched in sweat” for the past few weeks. When asked, the patient denies ever having a positive PPD and was last screened “several years ago.” His chart indicates he was in the emergency department last week with similar symptoms and was diagnosed with community-acquired pneumonia and discharged with azithromycin.

Past Medical History

Hypertension, dyslipidemia, COPD, atrial fibrillation, generalized anxiety disorder

Surgical History

Appendectomy at age 18

Family History

Father passed away from a myocardial infarction 4 years ago; mother had type 2 DM and passed away from a ruptured abdominal aortic aneurysm

Social History

Retired geologist recently moved from India to live with his son who is currently in medical school in upstate New York. Smoked ½ ppd × 40 years and drinks 6 to 8 beers per day, recently admits to drinking ½ pint of vodka “every few days” since the passing of his wife 6 months ago.

Sulfa (hives); penicillin (nausea/vomiting); shellfish (itching)

Home Medications

Albuterol metered-dose-inhaler 2 puffs q4h PRN shortness of breath

Aspirin 81 mg PO daily

Atorvastatin 40 mg PO daily

Budesonide/formoterol 160 mcg/4.5 mcg 2 inhalations BID

Clonazepam 0.5 mg PO three times daily PRN anxiety

Lisinopril 20 mg PO daily

Metoprolol succinate 100 mg PO daily

Tiotropium 2 inhalations once daily

Venlafaxine 150 mg PO daily

Warfarin 7.5 mg PO daily

Physical Examination

Vital signs.

Temp 100.8°F, P 96, RR 24 breaths per minute, BP 150/84 mm Hg, pO 2 92%, Ht 5′10″, Wt 56.4 kg

Slightly disheveled male in mild-to-moderate distress

Normocephalic, atraumatic, PERRLA, EOMI, pale/dry mucous membranes and conjunctiva, poor dentition

Bronchial breath sounds in RUL

Cardiovascular

NSR, no m/r/g

Soft, non-distended, non-tender, (+) bowel sounds

Genitourinary

Get free access through your institution, pop-up div successfully displayed.

This div only appears when the trigger link is hovered over. Otherwise it is hidden from view.

Please Wait

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Publications
Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

Advanced Search
Journal List
v.14(9); 2022 Sep

Miliary Tuberculosis: A Case Report Highlighting the Diagnostic Challenges Associated With the Condition

Usman ilyas.

1 Internal Medicine, Icahn School of Medicine at Mount Sinai, Queens Hospital Center, Jamaica, USA

Abrahim Mahmood

2 Internal Medicine, New York Institute of Technology College of Osteopathic Medicine, Flushing, USA

Amee M Pansuriya

3 Internal Medicine, New York Institute of Technology College of Osteopathic Medicine, Old Westbury, USA

Zaryab Umar

End-stage renal disease requiring chronic dialysis is an immunocompromised state which increases the risk of tuberculosis development and its spread. Due to the high frequency of non-specific or “decoy” symptoms at presentation and frequent extrapulmonary involvement, diagnosis of tuberculosis is a significant challenge. Therefore, it is correctly labeled as ‘Tuberculosis; the great imitator’ as it can mimic various other disease processes, causing confusion and testing of subsystems involved in the disease process, which come back as abnormal, leading to a vicious cycle. Missing the diagnosis leads to grave consequences, especially in a patient with a miliary form of tuberculosis, as the prognosis with any delay in treatment is poor. High diagnostic suspicion is required to promptly diagnose and treat the condition, especially in a resource-rich setting where tuberculosis is uncommon. Here, we report a patient with miliary tuberculosis who presented with a chief complaint of chronic diarrhea and fecal continence, with prior recent negative interferon-gamma release assay testing. Due to every organ system involved, multiple subspecialties were on board, with a broad differential in mind, including malabsorption syndromes, neoplasia, infections, amyloidosis, and autoimmune disorders, and therefore, numerous tests were performed. However, despite all efforts, the diagnosis was delayed significantly, leading to the unfortunate demise of the patient. The case report sheds light on unique clinical features of miliary tuberculosis, diagnostic findings, and a reminder to always keep tuberculosis high in the differential in an appropriate clinical setting.

Introduction

Miliary tuberculosis (TB) continues to pose a great diagnostic challenge due to multiple modes of presentation, depending on the extent and severity of both pulmonary and extrapulmonary organ involvement [ 1 ]. Although miliary TB is rare, the risk of disease rises with immunocompromised states such as acquired immune deficiency syndrome, diabetes, end-stage renal disease (ESRD) on dialysis, and usage of corticosteroid or other immunosuppressive agents [ 2 ]. Gastrointestinal symptoms such as abdominal pain, nausea, vomiting, and diarrhea are uncommon clinical findings in miliary TB [ 3 - 5 ]. Here, we present a unique case of miliary TB in an ESRD patient with primarily gastrointestinal complaints at admission. Additional clinical and laboratory findings noted through the course of hospital stay included pancytopenia (including profound symptomatic thrombocytopenia), clotting factor abnormalities, hypoxemia, ascites, transaminitis, and recurrent fevers, which in the setting of a negative quantiferon tuberculosis test and positive preliminary cultures for Mycobacterium avium complex led to an incorrect diagnosis and treatment, leading to a death that could have been prevented.

Case presentation

The patient was a 49-year-old man with a past medical history of type 2 diabetes, hypertension, ESRD on thrice weekly dialysis, coronary artery disease (drug-eluting stent in right coronary artery for 80% stenosis), heart failure with a reduced ejection fraction of 30-35%, and pulmonary hypertension with pulmonary artery systolic pressure of 65.76 mmHg who presented with complaints of six months of diarrheal episodes described as one time during the day and one time during the night, with stool being brown-green and of sticky and watery consistency. Over the last three months, he had a progressive loss of control over his bowel movements causing spontaneous episodes without him being able to anticipate them. Therefore, he had to resort to wearing adult diapers. He endorsed dry cough and subjective fevers as well. A day before presenting to the emergency department, he had an episode of falling when he became dizzy upon getting up from the sofa. He felt lightheaded, therefore, could not prevent the fall but denies losing consciousness. His physical examination in the emergency department showed bilateral coarse crackles, soft, mildly distended abdomen with positive bowel sounds and bilateral 2+ pedal edema. On admission, lab results showed pancytopenia (Tables (Tables1, 1 , ,2). 2 ). A chest x-ray is shown in Figure Figure1 1 .

Lab with reference range and units	1 year before admission	6 months before admission	Day of admission	Day 7 of admission	Day 14 of admission	Day 21 of admission	Day 28 of admission	Day 35 of admission	Day 42 of admission
Hemoglobin (14.0-18.0 g/dL)	11	10.9	8.8	7.9	6.8	7.7	9.2	8.5	8.0
Platelet (150-450 x10^3/mcL)	162	144	72	30	36	54	13	9	22
White blood cell count (4.8-10.8 x10^3/mcL)	9.09	5.36	2.84	4.99	4.44	6.26	10.89	15.65	9.04
Absolute neutrophil count (2.1-7.60x10^3/mcL)	5.19	4.2	2.39	4.26	3.65	5.58	10.34	15.25	8.7
Absolute lymphocyte count (1.00-4.9x10^3/mcL)	1.37	0.41	0.17	0.24	0.27	0.16	0.24	0.12	0.04
Absolute monocyte count (0.1-1.1x10^3/mcL)	0.74	0.56	0.23	0.41	0.4	0.39	0.11	0.15	0.19
Absolute eosinophil count (0.1-0.4x10^3/mcL)	1.75	0.17	0.02	0.03	0.08	0.09	0.03	0.00	0.00
Absolute basophil count (0.0-0.2x10^3/mcL)	0.03	0.01	0.01	0.01	0.02	0.02	0.10	0.01	0.02

Lab	Patient’s value	Reference range and units
Sodium	124	136-145 mmol/L
Potassium	4.1	3.5-5.1 mmol/L
Chloride	84	98-108 mmol/L
Bicarbonate	21	22-29 mmol/L
Blood urea nitrogen	56	6-23 mg/dL
Creatinine	6.44	0.50-1.20 mg/dL
Calcium	9.7	8.6-10.3 mg/dL
Albumin	2.4	3.5-5.2 g/dL
Total protein	6.6	6.6-8.7 g/dL
Total bilirubin	0.9	0.0-1.2 mg/dL
Direct bilirubin	0.3	0.0-0.3 mg/dL
Alanine aminotransferase	32	0-41 U/L
Aspartate aminotransferase	15	5-40 U/L
Alkaline phosphatase	183	40-129 U/L
Prothrombin time	12.8	10-13 seconds
International normalized ratio	1.1	1.4 ratio
Activated partial thromboplastin time	34	25.1-36.5 seconds
Procalcitonin	2.18	0.02-0.10 ng/mL
Hemoglobin A1C	6.3	4-5.6%
Erythrocyte sedimentation rate	41	0-10 mm/hr
C-Reactive protein	98.3	0-5 mg/L

An external file that holds a picture, illustration, etc.
Object name is cureus-0014-00000029339-i01.jpg

He was admitted with the suspicion of diarrhea secondary to infectious versus malabsorptive etiology with malabsorption higher on the differential, given the chronic nature of his symptoms and lack of fever or leukocytosis. However, a complete workup, including stool routine examination, stool bacterial culture, blood cultures, Giardia, Clostridium difficile, human immunodeficiency virus, Strongyloides, stool leukocytes, stool calprotectin, stool ova and parasites, celiac workup, stool fat, and stool elastase, was sent and came back negative (Table (Table3 3 ).

Lab	Patient’s value	Reference range and units
Fecal leukocyte count with wright stain	Negative	Negative
Ova and parasite screen stool	Negative	Negative
Giardia antigen	Negative	Negative
Clostridium difficile antigen	Negative	Negative
Clostridium difficile toxin A/B	Negative	Negative
Immunoglobulin A	297	84-499 mg/dL
Transglutaminase antibody Immunoglobulin A	<1.2	0-3.9 U/mL
Fecal occult blood test	Negative	Negative
Stool fats neutral, and total	<60 and <100 droplets /high power field	<60 and <100 droplets / high power field
Calprotectin, fecal	103	0-120 ug/g
Pancreatic elastase, stool	412	>200

A computed tomography (CT) scan of the abdomen and pelvis showed mild abdominopelvic ascites. On the fourth day of hospitalization, he had a sharp drop in hemoglobin from 8.8 to 7.1 with mild nasal bleeding. Heparin prophylaxis was stopped, intermittent pneumatic compression devices were placed, and a fecal occult blood test was sent. The patient also started desaturating to 88% oxygen on room air, for which 3 liters of oxygen per nasal cannula was started, maintaining oxygen saturation above 94%. Chest CT with contrast is shown below (Figures (Figures2, 2 , ,3 3 ).

An external file that holds a picture, illustration, etc.
Object name is cureus-0014-00000029339-i02.jpg

Hypoxic respiratory failure secondary to fluid overload was assumed based on CT findings. An echocardiogram showed a left ventricle ejection fraction of 30-35% with grade 3 left ventricular diastolic dysfunction, pulmonary artery systolic pressure of 65.76 mmHg, and severely elevated central venous pressure of 11-20 mmHg; therefore, more aggressive dialysis with fluid removal was planned. A nuclear medicine ventilation-perfusion scan ruled out pulmonary embolism (Figure (Figure4 4 ).

An external file that holds a picture, illustration, etc.
Object name is cureus-0014-00000029339-i04.jpg

A gastroenterology consult recommended a colonoscopy, considering his diarrheal episodes and lower gastrointestinal bleeding. For thrombocytopenia with platelet count drop from 72,000 per mcL to 35,000 per mcL in five days, heparin-induced thrombocytopenia workup with platelet factor 4 was sent due to a 4T score for heparin-induced thrombocytopenia of 4 with intermediate probability and came back negative. Heparin use, including for dialysis, was avoided. Piperacillin tazobactam was stopped as it was considered the likely cause of recurrent hypokalemia and thrombocytopenia. Peripheral smear showed polychromasia and burr cells but no schistocytes. An antinuclear antibody screen for autoimmune conditions was negative. Complete anemia workup results are shown in Table Table4 4 .

Lab	Patient’s value	Reference range and units
Vitamin B12	1568	232-1245 pg/mL
Serum Folate	2.5	>4.7 ng/mL
Lactate dehydrogenase	340	135-225 U/L
Haptoglobin	75	34-200 mg/dL
Reticulocyte count	2.41	0.5-1.5%
Reticulocyte absolute	0.0757	0.0221-0.0963x10^6/mcL
Red blood cell count	3.14	4.7-6.1x10^6/mcL
Heparin-platelet factor 4 antibody	<0.6	0.0-0.9 u/mL
Platelet antibodies Human leukocyte antigen Class I Antibody, Platelet IIB/IIIA Antibody, Platelet IB/IX Antibody, Platelet IA/IIA Antibody, Glycoprotein IV Antibody	Negative	Negative
Antinuclear antibodies	Negative	<1:80
Complement C3, serum	68	81-157 mg/dL
Complement C4, serum	29	13-39 mg/dL
Serum Protein Electrophoresis
Total Protein	5.8	6-8.3g/dL
Albumin	2.3	3.6-5.5 g/dL
Alpha 1 Globulin Fraction	0.5	0.1-0.4 g/dL
Alpha 2 Globulin Fraction	0.5	0.5-1.0 g/dL
Beta Globulin Fraction	0.6	0.5-1.0 g/dL
Gamma Globulin Fraction	1.9	0.6-1.6 g/dL
Monoclonal spike (M-spike)	Two gamma migrating paraproteins identified (one IgG Lambda and one IgG Kappa band)
Immunoglobulin Free Light Chain
Immunoglobulin Kappa	18.52	0.33-1.94 mg/dL
Immunoglobulin Lambda	30.95	0.57-2.63 mg/dL
Kappa Lambda Ratio	0.60	0.26-1.65 Ratio

On the sixth admission day, he had an episode of altered mental status in the setting of a high-grade fever of 102.9 °F off antibiotics. The fever came down with acetaminophen. Repeat blood cultures and stool for acid-fast bacilli were sent. CT of the head was normal. Serum procalcitonin was elevated to 5.73 ng/mL (normal range 0.02-0.10 ng/mL). Vancomycin and cefepime were started, and he remained afebrile on antibiotics, with procalcitonin trending down to 2.73 ng/mL. He also had recurrent self-limiting nasal bleeds. The goal was to give cryoprecipitate if fibrinogen was below 120 mg/dL (normal range 200-393 mg/dL) and platelets transfusion if platelet count was below 20,000/mcL without bleed and 30000/mcL with bleed. Colonoscopy with biopsy was unremarkable for both gross and microscopic pathology.

A bone marrow biopsy (Figure (Figure5) 5 ) was done to find the cause of thrombocytopenia along with ruling out malignancy. Flow cytometry of the bone marrow aspirate was unremarkable; however, chromosome analysis showed an interstitial deletion of the long arm of chromosome 20 (20q deletion).

An external file that holds a picture, illustration, etc.
Object name is cureus-0014-00000029339-i05.jpg

Due to frequent nosebleeds with low platelet count and low fibrinogen requiring platelets and cryoprecipitate, respectively, coagulation factors and mixing studies were obtained (Table (Table5), 5 ), keeping in mind that although factor 8 deficiency is most common in amyloidosis, low levels of all factors have been reported previously.

Lab	Patient’s value	Reference range and units
Activated partial thromboplastin time (APTT)	45.6	25.1-36.5
Protime	15.1	10-13 seconds
International Normalized Ratio	1.3
Fibrinogen	152	200-393 mg/dL
Dilute Russel Viper Venom time(DRVVT)
Screen/Confirm ratio	0.80	0.00-1.21 Ratio
DRVVT Interpretation	LA Neg	LA neg
DRVVT Screen	49.5	<44 seconds
DRVVT Confirmatory	46.0	<44 seconds
Factor XII Assay	32	45-150%
Factor XI Assay	40	70-145%
Factor X Assay	56	70-170%
Factor IX Assay	57	80-165%
Factor VIII Assay	202	60-125%
Cancer antigen 125 (CA-125)	41	<38 U/mL
Carcinoembryonic antigen (CEA)	7.4	0.0-3.8 ng/mL
Cancer Antigen 19-9	449	<35 U/mL
Beta-2 Microglobulin	72.8	0.8-2.2 mg/mL

A rectal fat pad biopsy (Figure (Figure6) 6 ) was performed to rule out amyloidosis, considering abnormal serum protein electrophoresis.

An external file that holds a picture, illustration, etc.
Object name is cureus-0014-00000029339-i06.jpg

After completion of 10 days of cefepime and vancomycin, antibiotics were stopped. However, he spiked a fever of 102 ℉ again in the 48 hours, and antibiotics were resumed. A repeat serum procalcitonin came back as 9.86 ng/mL (normal range 0.02-0.10 ng/mL). His diarrhea restarted as well. Stool studies for Microspora, Isospora, Cryptosporidium, and Cyclospora, and blood cultures with blood tests for blood parasites, Babesia, Ehrlichia, and Anaplasma were sent and all came back negative. In the fourth week of admission, his acid-fast bacilli stool came back positive for Mycobacterium avium complex (MAC), however, it was deemed a contaminant at that time.

On the 30th day of admission, he was found to have an altered mental status. Glucose was undetectable with a blood pressure of systolic 80 by diastolic 40 mmHg and mean arterial pressure below the 60s. After 50% dextrose, glucose increased to 184, and blood pressure normalized with one albumin with 1 liter of normal saline. Dextrose water (5%) intravenous was started. The CT head was negative. Hypoglycemia secondary to poor oral intake and chronic condition was suspected with altered mental status due to hypoglycemia. He remained on the medical floor as he was not deemed a candidate for the intensive care unit (ICU). Hydrocortisone 100 mg every eight hours was started. Although his mental status improved temporarily, it later declined, and he became hypothermic. On the 31st day, he had a cardiac arrest. Advanced cardiac life support protocol was followed, with the return of spontaneous circulation achieved after 29 minutes of high-quality cardiopulmonary resuscitation with an initial rhythm of asystole. The patient was transferred to the ICU.

Suspecting a disseminated MAC infection, he was started on meropenem with foscarnet added after cytomegalovirus deoxyribonucleic acid (DNA) polymerase chain reaction of a blood sample returned as positive. He remained intubated for 15 days, with his code status changed to do not resuscitate after a family discussion regarding goals of care. Repeat blood and fungal cultures were sent. Electroencephalogram showed a predominant posterior rhythm consisting of diffuse and generalized bursts of the delta, theta, and sharply contoured activity followed by periods of relative cerebral inactivity, consistent with the hypoxic injury. Brainstem reflexes were absent with a Glasgow coma scale of 3/15. He had recurrent fever spikes of up to 104℉. On the 45th day, his vital signs began to decline despite maximal support until a flat line was seen on an electrocardiogram with fixed and dilated pupils. Approximately three weeks after his sad demise, acid-fast stool and blood tests returned positive for the rare Mycobacterium tuberculosis complex detected by a DNA probe and already detected Mycobacterium avium complex. A subsequent complete report documented a moderate mixed infection with pan-sensitive Mycobacterium tuberculosis and Mycobacterium avium complex.

John Jacobus Manget coined the word "miliary" in miliary tuberculosis in the year 1700. He used this word to describe the firm, small white nodules on the surface of an affected lung, similar to the appearance of millet seeds. Miliary TB is a rare manifestation of TB, and less than 2% of TB cases demonstrate miliary disease [ 6 ]. It is described as lymphohematogenous dissemination of M. tuberculosis, usually from central foci with subsequent infiltration into various organs, including the liver, bone marrow, spleen, lungs, and meninges, forming small caseating granulomas, about 1-3 mm in diameter [ 1 ]. Immunosuppression in the form of impaired cell-mediated immunity allows for the development and spread of TB [ 2 , 6 ]. Immunocompromised states leave the host more susceptible to the infection; examples of conditions include HIV/AIDS, diabetes, usage of corticosteroids or other immunosuppressive agents, or renal failure [ 2 , 6 ]. Patients like ours with ESRD on dialysis are an immunocompromised population and consequently are six to 25 times more likely to develop TB than the general population [ 7 ]. However, a history of TB is only present in 10-20% of the patients [ 1 ].

Miliary TB in a clinical setting typically presents insidiously with nonspecific symptoms related to the location of the bacterial infiltration. Common symptoms include weight loss, malaise, fever, and anorexia with or without respiratory symptoms (cough, dyspnea, pleuritic chest pain) [ 1 ]. Uremia is commonly associated with fatigue, malnutrition, and other nonspecific complaints, thereby concealing underlying TB disease [ 7 ]. Morning temperature spikes are a finding of diagnostic significance, mainly if it is of unknown origin, as they are only seen in three conditions: tuberculosis, typhoid fever, and periarteritis nodosa [ 1 ]. Several papers have reported the "damp shadow sign," or the sweat-drenched silhouette on the patient's bed, as a common clinical finding due to night sweats [ 8 ].

Gastrointestinal symptoms such as abdominal pain, nausea, vomiting, and diarrhea are uncommon clinical findings in miliary TB, as three retrospective series reported gastrointestinal complaints in approximately 21%, 12%, and 32% of patients [ 3 - 5 ]. However, our patient presented primarily with diarrheal symptoms for six months and gradually lost bowel movements over three months. Such a presentation prompted an extensive workup for an infectious vs. malabsorptive etiology.

The nonspecific symptoms of miliary TB require a high degree of suspicion to recognize the disease and order appropriate workup. Common lab abnormalities seen in TB are nonspecific. Erythrocyte sedimentation rate was elevated in approximately two-thirds of the cases in one study [ 2 ]. Arterial blood-gas analysis shows hypoxemia with a high alveolar-arterial gradient and hypocapnia [ 2 ]. Polyclonal gammopathy and hyponatremia are very common, with the latter likely from parenchymal lung disease with the inappropriate secretion of antidiuretic hormone [ 2 ]. Both were seen in our patient. Although thrombocytosis and leukocytosis are frequently present in pulmonary TB, miliary TB is associated with anemia of chronic disease, leukopenia, and thrombocytopenia [ 2 ]. Pancytopenia and coagulation abnormalities raised suspicion of malignancy as a source of fever in our patient with subsequent extensive workup.

Chest radiography classically demonstrates faint, reticulonodular, homogenous infiltrates with a relatively uniform distribution throughout the lungs [ 9 - 11 ]. In the early stages of the disease, however, radiograph findings may appear normal, making it challenging to diagnose miliary TB from chest radiographs alone [ 12 ]. Chest radiography has a sensitivity of 60-70% in miliary TB, which is influenced by the size of miliary calcifications, the astuteness of the radiologist, and the extent of underlying disease, which may make detection of the infiltrates more complex, leading to delayed or missed diagnosis with fatal consequences [ 1 , 2 ]. This form of the disease, termed cryptic miliary TB, is a diagnostic challenge even in areas where tuberculosis is endemic with high physician awareness [ 2 ]. In our resource-developed setting with low physician awareness, the absence of miliary calcifications and predominantly gastrointestinal symptoms on the presentation made diagnosis possible only retrospectively. High-resolution computed tomography (HRCT) of the chest is a more sensitive imaging test for miliary TB than plain chest radiography [ 13 ]. Disseminated nodules can be identified; however, these findings are sensitive and not specific to miliary TB. Imperative to add in the differential diagnosis include infectious (histoplasmosis, cryptococcosis, brucellosis, toxoplasmosis) and non-infectious diseases (sarcoidosis, lymphoma, metastatic disease), as they present with similar clinical findings.

A higher proportion of patients with miliary tuberculosis exhibit tuberculin anergy compared to those with pulmonary or extrapulmonary tuberculosis [ 6 ]. In addition to cross-reactivity with nontuberculous Mycobacteria and tuberculin positivity in Bacille Calmette-Guérin (BCG) vaccination, anergy makes tuberculin skin test a nondiagnostic test in miliary tuberculosis [ 6 ]. Similarly, in one study, interferon-gamma release assays such as quantiferon TB gold showed a negative result in 28.8% of patients with extrapulmonary tuberculosis, with sensitivity further being affected by HIV status, diabetes mellitus, neutropenia, immunosuppression, and severe TB disease [ 14 ]. This explains the two negative quantiferon TB gold tests in our patient three months before and at the presentation time.

Miliary TB can be diagnosed with either microbiological or histopathological tests. A microbiological examination of sputum is performed, which, if present, demonstrates a positive acid-fast stain due to the presence of mycolic acid in the cell wall of Mycobacterium [ 6 ]. The current clinical practice recommends three sputum specimens for acid-fast bacilli smear microscopy in all patients suspected of having pulmonary TB [ 15 ]. Sputum induction is advised in those unable to expectorate sputum [ 15 ]. In addition, performing a diagnostic nucleic acid amplification test on the initial respiratory specimen is recommended [ 15 ]. Mycobacterium blood cultures could be obtained if hematogenous dissemination is suspected. However, it is essential to identify the species as positive blood cultures can also be seen in Mycobacterium avium, which is usually present in patients with advanced HIV infection [ 6 ]. We were initially misled towards the wrong diagnosis as Mycobacterium avium complex on the preliminary test was deemed a contaminant. In instances where obtaining a sputum culture is difficult (such as with our patient), a stool culture offers an alternative method for diagnosing TB, which can be of particular utility when the patient with suspected TB presents with lower gastrointestinal symptoms [ 16 ]. Other smear and culture examination sources include gastric lavage, pleural, peritoneal, or pericardial fluid, cerebrospinal fluid, urine, pus from a cold abscess, or bronchoscopic secretions [ 6 ]. Tissue biopsy of the affected site(s) can be histopathologically examined, where specimens in the setting of TB demonstrate granulomatous inflammation [ 15 ]. The presence of caseating granulomas (granulomas with central "cheese-like" necrosis) is characteristic, but its absence does not exclude miliary TB. This was the case in our patient, who presented with noncaseating granulomas on bone marrow and abdominal fat pad biopsy. Ultimately the gold standard for diagnosing TB is either culture isolating TB or detecting nucleic acids specific to TB (e.g., nucleic acid amplification tests) [ 17 ].

Miliary TB can be a challenge to diagnose. The difficulty in identifying the disease can delay prompt treatment, which plays a role in the high rates of morbidity and mortality from this disease. If untreated, miliary TB is fatal within 12 months [ 6 ]. Factors such as changes in mental status, meningitis, female sex, dyspnea, advanced age, and a failure to initiate treatment quickly are all associated with a worse prognosis [ 4 , 7 ]. As a consequence, it is imperative to begin treatment early. The availability of antibiotics has markedly improved clinical outcomes. The general approach to treating miliary TB mirrors the traditional six-month regimen of pulmonary TB, consisting of isoniazid, rifampin, and pyrazinamide for two months, followed by isoniazid and rifampin for four months. Ethambutol can be added if resistance to rifampin is suspected, and its continuation is recommended until sensitivity to rifampin is demonstrated [ 18 ]. Unfortunately, our patient's nonspecific and uncharacteristic presentation made diagnosing and promptly treating his TB difficult, ultimately resulting in the patient's demise.

Conclusions

TB, despite diagnostic advances, remains a difficult diagnostic task, as it can mimic multiple other disease processes, have a varied clinical presentation, give rise to a variety of different abnormalities in laboratory tests, and unique radiographic abnormalities, which can be difficult to interpret especially in the background of other illnesses. The challenge is an even further uphill climb in resource-rich settings, where tuberculosis cases are rare. However, this case again emphasizes the dire need for all physicians to be aware of the pathophysiology, clinical presentation, diagnostic approach, and treatment of TB, as it is still a lingering problem in developed countries, especially in the immunocompromised population.

The content published in Cureus is the result of clinical experience and/or research by independent individuals or organizations. Cureus is not responsible for the scientific accuracy or reliability of data or conclusions published herein. All content published within Cureus is intended only for educational, research and reference purposes. Additionally, articles published within Cureus should not be deemed a suitable substitute for the advice of a qualified health care professional. Do not disregard or avoid professional medical advice due to content published within Cureus.

The authors have declared that no competing interests exist.

Human Ethics

Consent was obtained or waived by all participants in this study

This paper is in the following e-collection/theme issue:

Published on 26.8.2024 in Vol 26 (2024)

Evaluation of a Natural Language Processing Approach to Identify Diagnostic Errors and Analysis of Safety Learning System Case Review Data: Retrospective Cohort Study

Authors of this article:

Azade Tabaie 1, 2 , PhD ; Alberta Tran 3 , RN, CCRN, PhD ; Tony Calabria 3 , MA, CPHQ, CSSBB ; Sonita S Bennett 1 , MSc ; Arianna Milicia 4 , BSc ; William Weintraub 5, 6 , MACC, MD ; William James Gallagher 6, 7 , MD ; John Yosaitis 6, 8 , MD ; Laura C Schubel 4 , MPH ; Mary A Hill 9, 10 , MS ; Kelly Michelle Smith 9, 10 , PhD ; Kristen Miller 4, 6 , MSPH, MSL, CPPS, DrPH

1 Center for Biostatistics, Informatics, and Data Science, MedStar Health Research Institute, Washington, DC, United States

2 Department of Emergency Medicine, Georgetown University School of Medicine, Washington, DC, United States

3 Department of Quality and Safety, MedStar Health Research Institute, Washington, DC, United States

4 National Center for Human Factors in Healthcare, MedStar Health Research Institute, Washington, DC, United States

5 Population Health, MedStar Health Research Institute, Washington, DC, United States

6 Georgetown University School of Medicine, Washington, DC, United States

7 Family Medicine Residency Program, MedStar Health Georgetown-Washington Hospital Center, Washington, DC, United States

8 MedStar Simulation Training & Education Lab (SiTEL), MedStar Institute for Innovation, Washington, DC, United States

9 Institute of Health Policy, Management & Evaluation, University of Toronto, Toronto, ON, Canada

10 Michael Garron Hospital, Toronto, ON, Canada

Corresponding Author:

Azade Tabaie , PhD
Center for Biostatistics, Informatics, and Data Science
MedStar Health Research Institute
3007 Tilden Street NW
Washington , DC , 20008
United States
Phone: 1 202-244-9810
Email: [email protected]

IMAGES

(PDF) CASE STUDY ON PULMONARY TUBERCULOSIS
CASE STUDIES IN TUBERCULOSIS / case-studies-in-tuberculosis.pdf / PDF4PRO
(PDF) Oral manifestation of tuberculosis: A case-report
Journal of Tuberculosis Research
(PDF) A Scarce Presentation of Mycobacterium tuberculosis , Case Report
(PDF) Unusual presentation of extrapulmonary tuberculosis: A case

COMMENTS

Verification of Diagnosis in Tuberculosis: A Case Report and Discussion
Introduction. Tuberculosis (TB) is caused by strains of Mycobacterium tuberculosis (M. Tuberculosis). TB is a primarily pulmonary infection spread by airborne droplet transmission. The development and spread of drug-resistant strains of M. tuberculosis greatly jeopardize TB control efforts. In the US, 91 cases of MDR-TB were reported in 2014.
A case-report of a pulmonary tuberculosis with lymphadenopathy
Clinical and radiological manifestations of tuberculosis (TB) are heterogeneous, and differential diagnosis can include both benign and malignant diseases (e.g., sarcoidosis, metastatic diseases, and lymphoma). Diagnostic dilemmas can delay appropriate therapy, favoring Mycobacterium tuberculosis transmission.We report on a case of TB in an immunocompetent, Somalian 22-year-old boy admitted in ...
Tuberculosis in older adults: case studies from four countries with
The case studies are a part of a broader endeavour, including a narrative review and analysis of epidemiological trends, to understand and document TB management among older adults in the region. The findings from the narrative review (see supplementary materials for methods) were used to support the country-specific experiences reported in the ...
Living with tuberculosis: a qualitative study of patients' experiences
Although tuberculosis (TB) is a curable disease, treatment is complex and prolonged, requiring considerable commitment from patients. This study aimed to understand the common perspectives of TB patients across Brazil, Russia, India, China, and South Africa throughout their disease journey, including the emotional, psychological, and practical challenges that patients and their families face.
Community-based active case-finding interventions for tuberculosis: a
We systematically reviewed the literature for studies that reported the effects of active case-finding interventions on tuberculosis epidemiological indicators. Our literature search was an update of a 2013 systematic review by Kranzer and colleagues, 3 which covered the period between Jan 1, 1980, and Oct 13, 2010, with additional searches by ...
A Path to Recovery
Ensuring Continuity of Tuberculosis Care during Social Distancing through Integrated Active Case Finding at COVID-19 Vaccination Events in Vietnam: A Cohort Study, Tropical Medicine and Infectious ...
Tuberculosis active case-finding: looking for cases in all the right
report findings of a systematic review summarising the evidence for population-level effectiveness of active case-finding for tuberculosis. The authors appraised 36 studies published between Jan 1, 1980, and April 13, 2020, of adult populations from 16 countries that were exposed to different active case-finding interventions.
Severe Pleural Tuberculosis: a Case Report and Management ...
INTRODUCTION: Tuberculosis (TB) is a highly infectious disease caused by Mycobacterium tuberculosis (MTB) that primarily affects the respiratory system. It is the second leading cause of infectious disease-related deaths. In the US, there were 7,882 cases of active TB in 2021. Pleural tuberculosis, a complication of active TB that can be life ...
PDF Tuberculosis active case-finding: looking for cases in all the right
Three outcomes were evaluated: case notification rates (which were expected to increase3), prevalence of active tuberculosis, and tuber culosis infection among young children as an indicator of transmission (both expected to decrease in the long term).3 Case notification rates were assessed in 28 studies, of which six were randomised trials.
Treatment of Highly Drug-Resistant Pulmonary Tuberculosis
Patients with highly drug-resistant forms of tuberculosis have limited treatment options and historically have had poor outcomes. In an open-label, single-group study in which follow-up is ongoing ...
Risk factors for tuberculosis: A case-control study in Addis Ababa
Background Tuberculosis remains a major public-health problem in the world, despite several efforts to improve case identification and treatment compliance. It is well known cause of ill-health among millions of people each year and ranks as the second leading cause of death from infectious disease worldwide. Despite implementation of the World health organization recommended strategy, the ...
A Trial of Mass Isoniazid Preventive Therapy for Tuberculosis Control
Tuberculosis is a leading cause of death in adults globally and was responsible for an estimated 1.4 million deaths in 2011. 1,2 Human immunodeficiency virus (HIV) infection, exposure to silica ...
Role of families in tuberculosis care: A case study : Muller Journal of
Role of families in tuberculosis care: A case study : Muller Journal of Medical Sciences and Research ... This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as ...
Risk factors for tuberculosis: A case-control study in Addis Ababa
Smoking has also been identified as important risk factor for developing TB by four times (AOR = 4.43; 95% CI: 2.10, 9.3). BCG was found to be protective against TB reducing the risk by one-third (AOR = 0.34; 95% CI: 0.22, 0.54). Conclusion: This study showed that TB is more common among the most agile and economically active age group, and ...
Tuberculosis
Tuberculosis does not accept case report submissions or clinical case studies. You can submit manuscripts of this type to the companion journals IDCases and Clinical Microbiology Newsletter . Tuberculosis is a speciality journal focusing on basic experimental research on tuberculosis.
A case-control study of lifestyle risk factors associated with
The aim of this study was to identify the subtle influences of exposure and individual lifestyles on the risk of developing tuberculosis. A retrospective case-control study (with matching by sex, age, postcode and ethnicity) of all tuberculosis cases notified over a 7-yr period in Liverpool, UK, was carried out. Multiple logistic regression showed that, before diagnosis, cases were 7.4 times ...
A case study of a patient with multidrug-resistant tuberculosis
Abstract. In this case study, a nurse presents her reflections on the challenges of supporting a patient through his treatment journey for multidrug-resistant tuberculosis. The patient has significant comorbidities and social issues, such as diabetes and homelessness. There was also a language barrier. All these aspects made the management of ...
Tuberculosis in older adults: case studies from four countries with
The Western Pacific Region has one of the fastest-growing populations of older adults (≥ 65 years) globally, among whom tuberculosis (TB) poses a particular concern. This study reports country case studies from China, Japan, the Republic of Korea, and Singapore reflecting on their experiences in managing TB among older adults. Across all four countries, TB case notification and incidence ...
Global, regional, and national burden of HIV-negative tuberculosis
Tuberculosis (TB) is a major infectious disease with significant public health implications. Its widespread transmission, prolonged treatment duration, notable side effects, and high mortality rate pose severe challenges. This study examines the epidemiological characteristics of TB globally and across major regions, providing a scientific basis for enhancing TB prevention and control measures ...
Nationwide surveys of awareness of tuberculosis in India ...
Tuberculosis remains a major challenge in India, with an estimated 2.69 million cases each year. Although men are more affected than women, gender differences and related factors affect awareness ...
Permissive lung neutrophils facilitate tuberculosis immunopathogenesis
Author summary While tuberculosis (TB) remains a global threat to public health, the immunologic factors contributing to TB susceptibility are not yet fully defined. To enhance our understanding of TB immunopathogenesis, we utilized TB-susceptible male Nox2-/- mice to dissect the immunologic factors accelerating TB pathogenesis. In this study, we observed that male Nox2-/- mice infected with ...
The incidence of tuberculosis among adolescents and young adults: a
Historical data show that the risk of tuberculosis increases dramatically during adolescence, and young people face unique challenges in terms of case detection and effective treatment. However, little is known about the burden of tuberculosis among young people in the modern era. This study aimed to provide the first estimates of the global and regional incidence of tuberculosis among young ...
Tuberculosis of the Stomach Mimicking Gastric Submucosal Tumor
Journal List; Case Rep Gastroenterol; v.18(1); Jan-Dec 2024; ... All data generated or analyzed during this study are included in this article and its online supplementary material. Further inquiries can be directed to the corresponding author. ... Intestinal tuberculosis: case series of three patients. Respir Med Case Rep. 2020; 29:100942.
A noninvasive BCG skin challenge model for assessing tuberculosis
A strategy to avoid issues associated with Mycobacterium tuberculosis (Mtb) as a challenge agent is to use Mycobacterium bovis BCG, which has a long history of safe use and has been used as a surrogate [14-18]. Our study utilises BCG as the basis for a fluorescent reporter strain (FluorBCG) that can be introduced intradermally, with the ...
Tuberculosis
Read chapter 34 of Infectious Diseases: A Case Study Approach online now, exclusively on AccessPharmacy. AccessPharmacy is a subscription-based resource from McGraw Hill that features trusted pharmacy content from the best minds in the field. ... Tuberculosis. In: Cho JC. Cho J.C.(Ed.), Ed. Jonathan C. Cho. eds. Infectious Diseases: A Case ...
PDF Case Studies in Tuberculosis
Case Studies in Tuberculosis: Training in Nurse Case Management. Introduction Dear Healthcare Professional, Prior to reviewing the content of this book, it is highly recommended that you complete the Centers for Disease Control and Prevention (CDC) Self-Study Modules on Tuberculosis (TB). The modules contain
Application of large language models in engineering education: A case
This study explores the innovative use of large language models (LLMs) in system modeling and simulation courses. Specifically, LLMs were applied to assist in MATLAB programming tasks, allowing students to learn MATLAB commands and programming techniques more conveniently.
Miliary Tuberculosis: A Case Report Highlighting the Diagnostic
Case presentation. The patient was a 49-year-old man with a past medical history of type 2 diabetes, hypertension, ESRD on thrice weekly dialysis, coronary artery disease (drug-eluting stent in right coronary artery for 80% stenosis), heart failure with a reduced ejection fraction of 30-35%, and pulmonary hypertension with pulmonary artery systolic pressure of 65.76 mmHg who presented with ...
Journal of Medical Internet Research
Background: Diagnostic errors are an underappreciated cause of preventable mortality in hospitals and pose a risk for severe patient harm and increase hospital length of stay. Objective: This study aims to explore the potential of machine learning and natural language processing techniques in improving diagnostic safety surveillance. We conducted a rigorous evaluation of the feasibility and ...
Shifting air travel demand: a case study on enabling experimentation in
Abstract. This paper presents the first empirical evaluation of Climate Perks, an employee benefit scheme designed to reduce air travel demand by offering additional paid annual leave to those travelling without flying for vacation purposes.

Secondary Logo

Role of families in tuberculosis care

Reaction of the family toward the TB patient (response of families to a social pathology)

Perspective of family strength in TB care

Contribution of family strength in TB care

Limitations of family based care for TB

Way ahead and further research

Financial Support and Sponsorship

Save citation to file

Add to My Bibliography

A case study of a patient with multidrug-resistant tuberculosis

Similar articles

Related information

Tuberculosis in older adults: case studies from four countries with rapidly ageing populations in the western pacific region

TB epidemiology

Health system structure and TB services

TB case finding

TB prevention

Summary and conclusion

Cross-cutting challenges

Data Availability

Change history

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Consent for publication

Additional information

Electronic supplementary material

Supplementary Material 1

About this article

Share this article

BMC Public Health

Global, regional, and national burden of HIV-negative tuberculosis, 1990–2021: findings from the Global Burden of Disease Study 2021

Conclusions

Graphical Abstract

Date source

Case definition

Statistical analysis

Incidence and temporal trend

Death and temporal trend

DALY and temporal trend

Age-gender characteristics

Association between ASRs and SDI

Risk factors for ASMRs and age-standardized DALY rates

Projecting disease burden

Efficient and innovative diagnostic technologies and strategies are needed to control the spread of TB

Effective TB control requires low-toxicity anti-TB drugs, effective vaccines, and an efficient primary healthcare system

Low-SDI regions require increased focus on MDR-TB and XDR-TB

Preventing latent TB infection (LTBI) can curb TB transmission

Control strategies developed under the guidance of the One Health approach will better curb TB transmission

Availability of data and materials

Abbreviations

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding authors

Ethics declarations

Consent for publication

Competing interests

Supplementary Information

Rights and permissions

About this article

Share this article

Infectious Diseases of Poverty

Nationwide surveys of awareness of tuberculosis in India uncover a gender gap in tuberculosis awareness

Conclusions

Plain Language Summary

Similar content being viewed by others

Drivers of sex differences in the South African adult tuberculosis incidence and mortality trends, 1990–2019

Identification and Prediction of Tuberculosis in Eastern China: Analyses from 10-year Population-based Notification Data in Zhejiang Province, China

Long-term trends of tuberculosis incidence and mortality in four central African countries

Study design

Data sources

Sample characteristics

Participants

Independent