Abstract
BACKGROUND: Urban and rural settings differ in key determinants of tuberculosis (TB) burden, including transmission dynamics, social and structural determinants, and healthcare access. However, understanding of urban and rural TB burden is limited, hindering implementation of public health interventions to end TB. METHODS AND FINDINGS: We conducted a systematic review and meta-analysis of urban and rural differences in adult pulmonary TB prevalence in low- and middle-income countries. We searched PubMed, Embase, Global Health, the Cochrane Library, Africa Index Medicus, LILACS, and SciELO for community-representative prevalence surveys conducted between 1st January 1993 and 14th October 2025. Studies focussing solely on symptomatic or healthcare-seeking individuals and those conducted in congregate settings like prisons, universities, and health facilities were excluded. Risk of bias was assessed using a tool for prevalence surveys. Bayesian multilevel meta-regression was used to estimate pooled urban-to-rural prevalence ratios (PR) for bacteriologically-confirmed and smear-positive TB overall, and by World Health Organization (WHO) region. We also investigated time trends in the urban-to-rural prevalence ratio, and associations between urban-to-rural prevalence ratios and survey-level risk of bias (not low versus low), TB screening algorithm (whether used symptom screening for sputum eligibility), national TB incidence, percentage of population living in urban areas, and representativeness of prevalence surveys (national versus sub-national). To estimate the number of people with prevalent TB in urban and rural areas in study countries, and how these have changed between 2000 and 2024, we fitted a Bayesian multivariate model to WHO incidence and case detection ratio data and combined these estimates with assumptions about the duration of treated and untreated TB and the distribution of urban and rural populations. We included 47 surveys conducted between 2000 and 2024, encompassing 2,454,443 participants. The pooled urban-to-rural PR of bacteriologically-confirmed TB was 1.09 (95% credible interval [CrI]: 0.90, 1.30) and was 1.24 (95% CrI: 0.94, 1.61) for smear-positive TB. However, there were substantial differences between WHO regions: averaged across the 24 year study period the African Region had higher urban bacteriologically-confirmed prevalence (PR 1.18, 95% CrI: 0.91, 1.52), while the Western Pacific Region (PR 0.85, 95% CrI: 0.64, 1.07) and South-East Asia Region (PR 0.86, 95% CrI: 0.67, 1.08) had broadly similar urban and rural prevalence. Time trends indicated an increase in the overall bacteriologically-confirmed urban-to-rural prevalence ratio between 2000 and 2024, with a mean PR increase of 2.4% (95% CrI: -0.8%, 6.0%) per year. We estimated that, for 2024 in the 26 represented study countries (combined population: 2.24 billion [48.3%] urban; 2.40 billion [51.7%] rural), 49% (6.6 million, 95% CrI: 4.2, 12.0 million) of prevalent TB was in urban areas, and 51% (6.8 million, 95% CrI: 4.2, 12.0 million) in rural areas. Within countries, there were striking changes in the urban and rural distribution between 2000 and 2024, with the share of urban cases increasing in nearly all countries. The main limitations include lack of unified definitions for urban and rural areas, and absence of data for some global regions (e.g., Americas and Europe). CONCLUSION: Between 2000 and 2024, TB epidemics have become increasingly urbanised, both in proportional and absolute terms, although with considerable variation in timing across countries and regions. Public health approaches tailored to urban and rural TB epidemiology and demography will be required to end TB.