Rifampicin-resistant tuberculosis (RR-TB) poses a significant threat to global health. In 2021, an estimated 450,000 individuals developed RR-TB, a concerning figure given rifampicin's crucial role in first-line TB treatment regimens. The treatment for RR-TB is more challenging, involving longer durations, more toxic drugs, and lower efficacy compared to first-line treatment for drug-susceptible TB. This results in lower cure rates and higher mortality rates for RR-TB patients. Further complicating the situation is the fact that many individuals diagnosed with TB aren't tested for drug resistance, leading to inappropriate treatment and high rates of treatment failure and recurrence. Consequently, many individuals with RR-TB either die during the disease episode or experience prolonged ill health before a cure is achieved. Evidence shows that TB survivors often experience lasting disability, elevated mortality, income losses, and persistent social and psychological consequences. These post-TB sequelae are believed to be more severe and common in RR-TB survivors due to factors like delayed treatment initiation, lower treatment efficacy, increased lung damage before disease control, and the toxicity of RR-TB drugs. The burden is not evenly distributed; global RR-TB incidence is declining, but the rate of decline and the absolute fraction of incident TB cases with rifampicin resistance vary considerably across different world regions. Similarly, the capacity of national TB programs to identify and treat RR-TB differs greatly. This study aims to quantify the global burden of RR-TB by synthesizing evidence on incidence, detection, and treatment outcomes across 192 countries (representing 99.99% of global TB cases in 2020) and using mathematical modeling to estimate the lifetime burden of disease, including both health losses during the disease episode and among survivors.

The existing literature underscores the significant long-term health consequences associated with tuberculosis (TB) and, more critically, rifampicin-resistant TB (RR-TB). Studies have consistently demonstrated elevated mortality rates and persistent disability among TB survivors, with these effects being more pronounced in those who have survived RR-TB. Previous research highlights the reduced lung function, persistent symptoms, and increased risk of chronic obstructive pulmonary disease (COPD) among individuals who have undergone treatment for RR-TB. The economic burden of TB is substantial due to reduced income and employment opportunities. This existing body of research establishes a compelling rationale for investigating the long-term global impact of RR-TB, encompassing mortality, morbidity, and associated economic costs. This study builds upon previous research which has provided estimates of the overall global burden of disease from TB, but extends this work by focusing specifically on the lifetime burden of RR-TB, taking into account the unique challenges and prolonged health consequences associated with this form of the disease. The researchers leveraged data from various sources including the WHO Global Tuberculosis Programme, UNAIDS, and other epidemiological studies to create a comprehensive model.

This study employed a mathematical model adapted from a previously published model to estimate the lifetime burden of disease due to incident RR-TB in 2020. The model stratified individuals developing TB in 2020 by country, age, sex, HIV status, and rifampicin resistance status. The model tracked the cohort over their lifetimes, recording excess deaths and reductions in health-related quality of life attributable to TB or post-TB sequelae. Data inputs included estimates of the total number of individuals developing TB in 2020 from the WHO Global Tuberculosis Programme, stratified by country, sex, and age group. The fraction of cases receiving treatment was determined using notified TB cases and WHO-estimated treatment coverage. The cohort was further stratified by HIV status using WHO and UNAIDS data. Rifampicin resistance prevalence was incorporated using country-level estimates from drug resistance surveys and routine surveillance, applied to data on the fraction of treated individuals with a history of prior TB. Assumptions about the duration and severity of disability during the TB disease episode were based on WHO estimates, stratified by treatment and HIV status. Disability weights for TB were based on Global Burden of Disease Study estimates. Mortality odds ratios, reflecting differences in survival probabilities based on age, HIV status, and treatment status, were estimated using data from the WHO and a detailed dataset from Brazil on case fatality among notified TB cases. The model accounted for differences in survival between individuals with RR-TB receiving appropriate or inappropriate treatment. Future mortality risks and disability for TB survivors were estimated using data from a systematic review of observational cohort studies, considering both the causal impact of TB and pre-existing factors. The model quantified health outcomes in disability-adjusted life years (DALYs), summing years of life lost due to premature mortality (YLLs) and years lived with disability (YLDs) across four categories: DALYs lost to disability during the TB episode, DALYs lost to mortality during the episode, DALYs lost to post-TB disability, and DALYs lost to post-TB mortality. A comprehensive uncertainty and sensitivity analysis was conducted using second-order Monte Carlo simulation with 1000 parameter sets generated from a Latin hypercube sample, producing 95% uncertainty intervals. Sensitivity analysis included calculating partial-rank correlation coefficients (PRCCs) to assess the influence of individual parameters. The model also assessed the robustness of results under several alternative analytic assumptions.

The study estimated that incident RR-TB in 2020 resulted in 6.9 million DALYs (95% uncertainty interval: 5.5, 8.5) globally, representing 5.4% of lifetime DALYs attributed to all incident TB. The majority (85%) of RR-TB DALYs were due to premature mortality (5.9 million DALYs), with the remainder due to reductions in quality of life (1.1 million DALYs). Over half (56%) of RR-TB DALYs were incurred during the disease episode (3.8 million DALYs), with the remaining accruing from post-TB sequelae among survivors (3.1 million DALYs). An average of 17.3 DALYs were associated with each case of RR-TB, 34.2% more than for rifampicin-susceptible TB. The burden per 100,000 population was highest in former Soviet Union countries and southern African countries. The South-East Asia Region had the largest absolute burden (37.2% of the global total), followed by the African Region (24.8%). The 30 high MDR/RR-TB burden countries identified by the WHO accounted for 85.9% of global RR-TB DALYs. DALYs per incident RR-TB case were generally higher in sub-Saharan Africa. India had the largest absolute burden (1.7 million DALYs, 24.1% of the global total), followed by Russia (0.5 million DALYs). Lesotho had the highest burden per population (773 DALYs per 100,000). The 10 countries of continental southern Africa accounted for 11% of the global burden. DALYs per incident case declined with age, peaking in the 35-44 age group. Men experienced a higher burden than women. Individuals with HIV experienced substantially higher burdens (39.2 times higher than HIV-uninfected individuals), accounting for 15.9% of total RR-TB DALYs. Sensitivity analysis showed that the five most influential parameters were those determining mortality rate ratios and disability weights due to post-TB sequelae, parameters determining the number of RR-TB cases, and case fatality for individuals receiving inappropriate first-line treatment. The magnitude of DALY estimates was generally robust to alternative analytic assumptions, with the greatest differences observed when using an alternative data source for TB-attributable mortality among survivors and when assuming that post-TB disability weights for RR-TB were the same as for RS-TB.

This study's findings highlight the substantial and long-lasting global burden of RR-TB, with a significant portion of the disease burden accruing among survivors. The high burden in former Soviet Union countries is likely due to historical treatment practices that led to high resistance levels, while in southern African countries, the dual burden of TB and HIV plays a significant role. The elevated burden associated with HIV infection emphasizes the need for integrated TB and HIV programs. This research underscores the importance of improved diagnostic tools, more effective treatments (such as the new shorter regimens), and robust post-treatment care to mitigate the long-term health and economic consequences of RR-TB. The findings are crucial for informing resource allocation and the development of targeted interventions to reduce the global burden of RR-TB.

This study demonstrates the substantial global burden of RR-TB, extending beyond immediate mortality and morbidity to encompass long-term health consequences among survivors. The findings highlight geographical disparities and emphasize the disproportionate impact on individuals with HIV. This necessitates the development of improved diagnostic tools, more effective shorter-course treatment regimens and improved post-treatment care, along with policy analyses to optimize resource allocation for RR-TB prevention, treatment, and rehabilitation. Future research should focus on establishing the causal effect of TB on long-term health outcomes, independent of confounding factors, and on developing effective rehabilitation interventions.

The study acknowledges several limitations. Uncertainty surrounding RR-TB outcomes in pediatric populations exists due to challenges in TB detection and limited data on post-TB outcomes for this group. Assumptions about healthcare interventions and access for RR-TB survivors might not hold true in the future. The model's reliance on epidemiological data highlights the importance of ongoing surveillance of drug resistance patterns. The methods used to separate the causal impact of TB from other factors influencing health outcomes require several assumptions, highlighting the need for additional empirical research. Data gaps were addressed through imputation, introducing an element of uncertainty.