Introduction

India has the highest tuberculosis (TB) burden (26%) and accounts for 34% of the TB deaths in the world.1 The recently conducted national TB prevalence survey reported the prevalence of pulmonary TB (PTB) as 316 per 100 000 population in the country and 386 per 100 000 population in the state of Madhya Pradesh, central India.2 Though the indigenous people comprise <5% of the global population, the burden of TB among them is high and varies greatly.3 The Saharia are one among the three particularly vulnerable tribal groups (PVTGs) in the state of Madhya Pradesh and work mainly as agriculture labourers. In contrast to the other two PVTGs, an extremely high TB prevalence of 3294 per 100 000 population is reported among them.4 Though the National TB Elimination Programme (NTEP) is being implemented in the area, there are several challenges such as remotely located Saharia habitats, poor access to healthcare facilities, and their beliefs and practices related to TB, making the situation more complex.5 Despite the high TB burden, the efforts to bring down the prevalence of TB seem to be inadequate in this key population.6 NTEP, in addition to the Tribal Action Plan, has collaborated with the Ministry of Tribal Affairs, Government of India, for the Tribal TB initiative for vulnerability mapping and active case finding (ACF). Nine districts of the state of Madhya Pradesh have been identified for this activity based on predominance of the tribal population.7 But the Saharia dominated districts are not included in this list as the Saharia tribe contributes to a small portion to the district’s population. This again makes them more disadvantaged, and it becomes more important to put in special efforts for controlling TB in this disadvantaged population.

Our earlier study carried out in a defined geographical area adopting community-based approaches to reach the population and to improve access to services demonstrated a decline in TB prevalence in this community.8 In view of this, an intervention study was conducted covering the total Saharia population of all seven Saharia-dominated districts during 2018–2021 to improve case finding and treatment outcomes through community-based approaches.

MethodsStudy area and population

The Saharia are one of the PVTGs in the central Indian state of Madhya Pradesh and have very poor socioeconomic indicators.9 The intervention study was carried out among the Saharia population (more than 0.5 million) in seven Saharia-dominated districts, namely, Shivpuri, Sheopur, Ashoknagar, Gwalior, Bhind, Datia and Morena, in collaboration with state and district health authorities and NTEP (figure 1).

Study procedures and intervention

In addition to regular NTEP activities, we implemented an intervention package to improve TB case finding and treatment outcomes. The intervention package included symptom screening by trained village TB volunteers; mapping and strengthening of health facilities; training and capacity building of NTEP and project staff; sensitisation and involvement of Panchayati Raj institutions (rural local self-government in India) and advocacy, communication, and social mobilisation activities (details submitted as online supplemental file 1). Throughout the study period, monthly house-to-house visits were performed by the village TB volunteers to identify TB presumptive cases. Sputum specimens were collected from symptomatic individuals and transported to the nearest NTEP Gene Xpert/TrueNAT lab. Whenever required, an X-ray was done at the nearest health facility. Notification of TB cases, treatment initiation, and follow-up were done as per NTEP guidelines by the project and NTEP supervisors.

Baseline and endline TB disease prevalence surveys

A preintervention and postintervention survey design was adopted to assess the impact of the intervention. The sample size to observe a reduction of 20% in the original prevalence (P1=1500/100 000) to the assumed endpoint prevalence (P2=1200/100 000), with a 95% confidence level and 80% power was estimated as 18 500 individuals. The baseline TB prevalence survey was carried out during January–May 2019 in 144 selected villages from all the seven districts. The detailed methodology is mentioned elsewhere.10 The endline survey was carried out in 128 selected villages in these districts during September–November 2021. During both surveys, the estimated sample size was proportionally distributed in the Saharia population in each district and then in the administrative blocks. Villages were selected through the probability-proportional-to-size sampling technique from all blocks, and all households belonging to the Saharia tribe in the selected village were included in the survey. Every eligible individual aged ≥15 years in the selected village was screened for symptoms of TB. Two sputum specimens, one spot and one overnight, were collected from eligible individuals and were processed for microscopy and culture. Following the NTEP guideline and algorithm, the positive specimens were further subjected to drug susceptibility testing. An individual was labelled as a case of PTB if any one sputum specimen tested positive by microscopy/molecular testing and/or culture examination. All the detected cases were linked to the NTEP for anti-TB treatment.

Data management and analyses

All completed forms and laboratory reports were scrutinised and checked at the block/district level. The data entry operators entered data in a predesigned and pretested data entry software developed on the CSPro V.7.0 platform. TB prevalence was calculated as TB cases tested positive by microscopy/molecular testing and/or culture per 100 000 population. The extended Mantel-Haenszel χ2 for linear trend with a value of p<0.05 at 1 degree of freedom was used to study the decline in TB positivity during the intervention period. The z-test was used to measure the differences in disease between the two disease surveys. IBM SPSS V.26.0 was used for carrying out different statistical analyses.

ResultsCoverage and registration

The Saharia population in all the 1814 Saharia villages of 32 administrative blocks in seven districts of Madhya Pradesh was covered under the study. TB case positivity was recorded as 41.3% in November–December 2018, which declined to 15.0% in 2019, 11.27% in 2020 and 9.3% in January–March 2021. This shows a significant linear decline (extended Mantel Haenszel Chi Square for linear trend=459.43, p<0.001 at 1 degree of freedom) in TB positivity during the intervention period. Overall, during the study period a total of 68862 presumptive cases were tested, and among them 8797 TB cases were detected with a case positivity of 12.8%. Of these, 371 cases were drug-resistant, including 302 multidrug-resistant TB, 12 extensively drug-resistant TB and 57 H-mono-resistant TB cases. The treatment success rate remained above 90% throughout the study period (table 1).

Table 1

Yearwise tuberculosis (TB) cases and resistance detected

TB prevalence during the baseline and endline surveys

During the baseline survey (January–May 2019), of the 21 828 registered eligible individuals (≥15 years of age), 20114 individuals (92.1%) were screened for PTB. Of these, 2893 (14.4%) were symptomatic and 273 were microbiologically confirmed PTB cases. The baseline prevalence was 1357 (95% CI 1206 to 1527) per 100 000 population.10 In the endline survey (September–November 2021), of the 22 805 registered individuals, 21946 individuals (96.2%) were screened, 1631 (7.4%) were symptomatic and 165 were microbiologically confirmed. The prevalence in the endline survey was 752 (95% CI 646 to 875) per 100 000 population. The reduction in the prevalence of PTB was statistically significant (p<0.001) (figure 2). The prevalence in both baseline and endline disease surveys increased with age and was significantly higher among men than women. In the endline survey, men were about 2.8 times more likely to have TB compared with women, similarly, older individuals (55+ years and above) were about 5.7 times more likely to have TB positivity compared with younger individuals (15–24 years).

Figure 2

Prevalence in baseline and endline surveys.

Discussion

This is the first large-scale intervention study covering the entire Saharia tribal population residing in the seven Saharia-dominated districts in the central Indian state of Madhya Pradesh. The study results show high TB case detection and a significant reduction in TB prevalence in this vulnerable population. Reducing TB burden requires preventing its transmission in the community. The best way to achieve this is by searching actively for cases, testing with good diagnostic methods, initiating patients promptly on treatment, and supporting and following them till the completion of the treatment.11 The present study demonstrated a significant decline in the prevalence of PTB from 1357 to 752 per 100 000 population. This might be due to innovative community-based approaches introduced in the study area during the study. The intervention package provided diagnostic and treatment services closer to the tribal community through the active involvement of village TB volunteers from the same community. The high treatment success rate achieved in the present study could be due to regular follow-up for treatment adherence by village TB volunteers and the project staff. A study in an urban Ugandan community in 2019, involving an intensive case finding campaign, showed that the burden of prevalent TB as measured by systematic screening had decreased by 45% in 2021.12 Another study by Marks et al in Vietnam also showed a similar finding.13

The other important aspect of this study is the economic benefit of ACF through community-based interventions. A study carried out in the state of Tamil Nadu during October 2016 to March 2018 showed that there was a 78% reduction in out-of-pocket expenditure for treatment by patients detected through ACF as compared with passive case finding in public health facilities (₹255 vs ₹1163).14 Thus, ACF significantly averted catastrophic costs due to TB among patients and it could ensure financial protection of TB patients and limit their risk of poverty.15 In our study the estimated cost for screening a population of more than 0.5 million for about two and half years was just ₹212 ($2.6) per person, but it was ₹1547 ($18.9) for identification of a presumptive TB case and ₹3268 ($39.8) to detect a confirmed PTB case in a tribal community scattered in the seven districts. In interventions such as the ACF, understanding incremental benefits and costs is challenging. In addition, the ACF is basically a provider-initiated activity with the primary objective of detecting TB cases early in targeted groups and initiating treatment promptly. It is expected to increase coverage and to prioritise by focusing on clinically, socially and occupationally vulnerable populations. The ACF remains one of the key approaches that can address the challenges of TB among tribal communities.

The findings of the present study have immense potential to improve case detection and treatment outcomes, reducing TB burden and financial burden in remote hilly and tribal areas. Working closely with the existing health system and linkage with the TB control programme in the study area are key to the success of this intervention package. This should promote sustainability and has the potential for scale-up in other resource-poor and high TB burden areas.

Limitations

The lack of a control group is a major limitation of the study. As the study was carried out in one of the PVTGs, taking some tribal villages as the control group was not feasible. Hence, a preintervention and postintervention survey design was adopted. The high case detection and subsequent decline in TB prevalence from this study could be due to improved access and utilisation of TB services. However, the study design does not establish causality. Furthermore, the roles of confounding factors and seasonality have not been considered while interpreting the results. The intervention package was comprehensive, and it is not possible to pinpoint specific interventions for the observed impact. Despite the above limitations, the study has robust strength to show high TB case detection and subsequent decline in TB prevalence using community participatory approaches and would be useful in TB programmes, especially in vulnerable and disadvantaged populations.

Ethics statementsPatient consent for publication

Not applicable.

Ethics approval

This study involves human participants. The Institutional Ethics Committee of ICMR-NIRTH, Jabalpur, approved the study (NIRTH/IEC/2273/2016). Participants gave informed consent to participate in the study before taking part.

Acknowledgments

The authors thank Dr Balram Bhargava, former secretary, Department of Health Research, Ministry of Health and Family Welfare, Government of India, and director general, Indian Council of Medical Ressearch (ICMR), for his support, insights and guidance. The authors also thank the principal secretary Health, the principal secretary Tribal Development, the health commissioner and the state tuberculosis officer, Government of Madhya Pradesh, for their constant support in undertaking this work in the Saharia tribe. The authors also thank Dr R Gangakhedkhar, former head, Epedemiology & Communicable Diseases division, ICMR, for his interest and constant support to project activities. The authors also thank Dr Aparup Das, former director ICMR-NIRTH, Jabalpur, for his support to the study. The authors also thank the district tuberculosis officers, the WHO/National TB Elimination Programme consultants, the district tribal welfare authorities, the block medical officers and the peripheral field staff for their contributions. The authors also thank the study participants and the field and laboratory staff, data entry operators, and secretarial staff involved in the study.