Globally, in 2017, 10 million people had tuberculosis (TB) and around 1.3 million died from TB [1]. A majority (around 46%) of the cases occurred in the South-East Asia region [2]. Approximately 3.6% of the new cases and 17% of the previously treated cases developed multidrug-resistant TB (MDR-TB) in 2017 [1]. The prevalence of MDR-TB among new and previously treated cases was 1.6% and 29% respectively, in the South-East Asia region [2]. MDR-TB is difficult to treat, requiring longer treatment duration and hospitalization, and the drugs used are less effective, costly and produce severe side effects [3]. The treatment outcome is also not always favorable [3, 4]. Thus, the prevention of initial infection, treatment and management of drug-susceptible TB adequately are keys to containing the spread of MDR-TB.

Bhutan is a small mountainous landlocked country in South Asia with a population of 727,145 [5]. Health care is delivered free of charge by the state through its three-tiered health care system that comprises of 186 Primary Health Centers (PHC) at the local level, 50 hospitals and three referral hospitals. The National Tuberculosis Control Program (NTCP), established in 1986 is responsible for the prevention and control of TB in Bhutan. All hospitals are designated as TB reporting and treatment centers. Individuals with symptoms of TB are referred from PHC to these centers for diagnosis, registration, and initiation of treatment. All confirmed TB cases are routinely tested for drug resistance and those cases found resistant to both isoniazid and rifampicin are classified as MDR-TB cases [6, 7].

Bhutan adopted the Directly Observed Treatment Short Course (DOTS) strategy in 1997. Tuberculosis patients are admitted in the treatment centers for the first two weeks for initiation of DOTS and are then followed-up by the respective health centers [6, 7]. For MDR-TB, cases are admitted in the treatment centers for the initiation of DOTS until two consecutive monthly culture results are negative [7]. A DOT provider, usually a family member is identified to provide DOTS for the remaining phase of the treatment.

Bhutan has been able to make commendable progress in TB control. This is reflected in the decline of TB prevalence from 536 in 2005 [8] to 190 per 100,000 population in 2014 [9], and high treatment success rate (more than 90% since 2005) [6, 8]. However, the drug resistance survey conducted in 2014 reported alarming rates of MDR-TB, 10.1% and 37.2% among new and previously treated cases respectively, which is much higher than the global rates [10]. MDR-TB cases have also increased over recent years [1, 6]. This suggests high transmission rates and poses a massive challenge towards achieving the Sustainable Development Goals by 2030 and the END TB targets in 2035 [11]. Urgent efforts are needed to control the MDR-TB epidemic.

Data on factors determining MDR-TB risk is limited in Bhutan. A descriptive study that recruited 19 MDR-TB patients from a treatment center of Bhutan showed that a majority of the cases did not comply with the DOT approach, had previous TB treatment history, and contact with MDR-TB cases [12]. However, this study did not assess socioeconomic and environmental factors such as education, income, housing conditions, residential area, and other lifestyle factors. Furthermore, the study was limited to MDR-TB cases in just one treatment center, did not conduct regression analysis to identify the most important factors associated with MDR-TB. More studies with better approaches that also investigates a range of plausible factors are needed to inform prevention and control policies adequately. The objective of this nationwide case-control study was to examine the factors associated with MDR-TB. The findings are expected to provide evidence for targeted policy investments that can help reduce the burden of MDR-TB in Bhutan.

A total of 197 participants (79 cases and 118 controls) from all the 20 districts (nationwide) in the country formed the final study sample. Of 101 MDR-TB cases registered with the NTCP and currently on treatment, 8 of them were less than 18 years of age, one could not communicate well, 2 refused to participate, 3 were mentally unsound, and 8 patients could not be traced due to the short study period. The mean age was 32.4 (SD: 2.44) years in the cases and 33.7 (SD: 3.22) years in the controls. Most of the participants were married, unemployed, and below the age of 49 years (Table 1). In both the case and control groups, a majority were 18–29 years of age, female, resided in urban areas, were frequent travelers, had middle-secondary level education, and had a family income of more than Ngultrum 10,000 per month (One USD = approximately Nu.70/-).

The bivariate analysis revealed sleep duration of less than 8 hours per day, in-county travel by public transport, history of TB treatment, number of TB treatment, living in multistoried houses, and having residence surrounded by buildings to be associated with MDR-TB (Table 2). The multivariate logistic regression analysis showed sleep duration, history of TB treatment, travel by public transport, and number of rooms to be the predictors of MDR-TB (Table 2). TB patients who slept less than 8 hours a day were more likely to have MDR-TB than those who slept more than or equal to 8 hours (AOR: 2.77, 95%CI: 1.11–6.92). Those who traveled in public transport within the country (AOR: 2.96, 95%CI: 1.36–6.48) were also at a higher odd of developing MDR-TB than those who travelled in private cars. Relative to those not previously treated for TB, those who were previously treated for TB had higher odds of having MDR-TB (AOR: 5.90, 95%CI: 2.55–13.64). Having more than 2 rooms in the house also emerged to be marginally associated with having MDR-TB (p = 0.049). The goodness-of-fit test did not show any evidence for the lack of fit of the final model (p = 0.269). The VIF values of all the variables were less than 10 in the final model suggesting that multi-collinearity was not a concern in the analysis.

The results of this study showed that sleep duration, travel by public transport, previous TB treatment, number of TB treatment, residential house type, and having residence surrounded by buildings were associated with MDR-TB in the bivariate analysis. In the multivariate analysis, sleep duration of less than 8 hours per day, travelling in public transport, previous TB treatment, and a greater number of rooms were associated with higher odds of developing MDR-TB. To our knowledge, this is the first nationwide case-control study that examined the risk factors associated with MDR-TB in Bhutan. The findings may help inform planning and policy interventions for effective control and prevention of MDR-TB in Bhutan.

Consistent with the evidence in the literature [13, 15, 19–21], those who were previously treated for TB had higher odds of developing MDR-TB. The number of times treated for TB also emerged to be positively associated with MDR-TB in the bivariate analysis. Incomplete treatment due to poor adherence to TB treatment and erratic drug supply, and inappropriate treatment regimens in previous treatment episodes can result in the emergence of MDR strains to any TB drugs [16, 19, 21, 22]. Inadequate financing and technical capacity, and poor TB program supervision also contributes to poor case management, leading to increased morbidity [3]. In Bhutan, suboptimal practice and adherence to DOTS, inadequate supervision and human resources, and poor infection control were highlighted as some of the programmatic challenges [9]. Rigorous patient follow-up and monitoring, improving patient counselling and communication, alternative supervision and choice of DOTS provider, and family member education may help promote compliance [23]. Future studies that investigate factors influencing retreatment and treatment failure, and deeper exploration of DOTS provision may help inform MDR-TB prevention efforts.

The results also indicated that those who frequently travel in public transport such as buses and taxis were more likely to develop MDR-TB than those travelling via private vehicles. The results suggest the free movement of infectious sources in such public spaces and late case detection leading to increased transmission. Our finding is in agreement with a study that reported a positive association between public transport workers and MDR-TB risk [14] and another one that showed higher TB risk among public transport users [24]. Likewise, a study also found higher odds of MDR-TB among frequent travelers [17]. Crowding, proximity to the index case and longer duration of exposure can amplify the risk of transmission in such settings. Furthermore, generally people from the low-income groups who cannot afford to own cars may be the ones using public transport, and these individuals are also more likely to have TB/MDR-TB [19, 22, 25]. Those in the lower socioeconomic group also have low education and poor housing conditions and may be unemployed that exacerbates their MDR-TB risk [15, 25]. Developing partnership with public transport providers aimed to raise awareness on infection control and motivate referral of symptomatic individuals may reinforce early case detection efforts.

Our analysis also revealed that those who slept for less than 8 hours a day were more likely to have MDR-TB. In contrast, the association was not robust in a few studies [25, 26]. A prolonged period of inadequate sleep can lead to chronic stress, which in turn affects the immunity through upregulation of stress hormones and downregulation of immunity. The dysregulation of body physiological system perpetuated by exposure to stressors and unhealthy behaviors including poor sleep also called “allostatic load,” can also expedite the development of chronic diseases [27, 28], that in turn can potentially increase susceptibility to TB through immune system impairment. This may partly help explain the association between sleep and MDR-TB. Studies have also shown that those with MDR-TB were more likely to endure life pressure or stress and have poor mental health [26, 29]. It is possible that patients may experience sleep difficult due to the physical and psychological trauma and social stressors associated with MDR-TB [29]. The relationship between sleep and MDR-TB warrants further investigation with better study design and data.

The results also showed having a greater number of rooms to be marginally associated with MDR-TB. The greater number of rooms among those with larger family size in this study sample can be a possible explanation for this finding. The proportion of having more than three rooms were 47.1%,47.6% and 40.3% among those with 3, 3–5 and more than 5 family members respectively in this study sample. Greater family size and household crowding has been shown to be associated with TB risk [30, 31]. Larger family size may have increased contacts among the household members, including with index cases, resulting in a higher propensity of exposure and transmission. Although, these findings, including for sleep duration seems plausible, they may also be statistical artefacts representing other correlated factors.

The absence of strong statistical associations for other factors, however, does not mean that they are not important. Variations in the epidemiology of MDR-TB and these factors may have led to the non-robust association. Besides, the small sample might have also reduced the power of the study to detect significant associations. Furthermore, the influence of factors such as social support and connectedness, substance abuse [32], mental health [22], HIV status [20, 32], clinical and laboratory manifestation of TB (such as quantity of bacilli in sputum smear/ smear positivity during course of treatment, lung cavity features, pleural effusion) [16, 21] were not assessed in our study. Future studies need to consider these factors to gain a deeper understanding of the factors influencing MDR-TB in Bhutan.

This nationwide case-control study revealed previous TB treatment, use of public transport for travelling, inadequate sleep, and number of rooms to be associated with a higher risk of having MDR-TB. The findings suggest that identifying the source of infection through early case detection and effective management of drug-susceptible cases, implementing strategies to enhance adherence and improve infection control are essential in halting and reducing the rising MDR-TB burden.