Global, Regional and National Trends in Latent Tuberculosis Infection Prevalence From 1990 to 2023, With Projections to 2050

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background: Latent tuberculosis infection (LTBI) is the major reservoir for future tuberculosis (TB) disease. Updated estimates of long-term trends, demographic redistribution, and future burden are needed to inform prevention-oriented TB policy in the era of the World Health Organization (WHO) End TB Strategy.Methods: We analyzed The Global Burden of Disease (GBD) 2023 LTBI prevalence estimates, summarizing age-standardized prevalence rates and prevalent cases globally and by sex, age group, socio-demographic index (SDI) quintile, WHO region, and country. Temporal trends were assessed using log-linear joinpoint regression, with national AAPC estimated for 2010–2023 and sensitivity analyses under alternative maximum joinpoint specifications. Global projections for 2024–2050 were generated using a Bayesian age–period–cohort model with second-order random-walk (RW2) priors linked to UN population projections; model diagnostics included convergence checks, posterior predictive checks, Pareto-smoothed importance sampling leave-one-out (PSIS-LOO), and comparison of Poisson and negative-binomial likelihoods. We also performed a fine-age sensitivity analysis that disaggregated population denominators into <20, 20–39, 40–54, 55–69, and ≥70 years while preserving the GBD broad-age prevalence rates. GBD-derived 95% uncertainty intervals (UIs), joinpoint-derived 95% confidence intervals (CIs), and Bayesian model-derived 95% credible intervals (CrIs) are distinguished throughout; joinpoint and APC model intervals are conditional on GBD point estimates and do not propagate upstream GBD estimation uncertainty.Results: Globally, the age-standardized LTBI prevalence rate declined from 30,421.65 (95% uncertainty interval [UI] 27,223.03–33,584.80) per 100,000 in 1990 to 22,664.54 (20,437.19–25,107.66) in 2023, corresponding to an AAPC of –0.87% (95% CI –0.95 – –0.79). Over the same period, prevalent cases increased from 1594.58 (1415.40–1777.61) million to 1876.45 (1696.79–2063.84) million, with an AAPC of +0.48% (0.43–0.54), demonstrating a clear rate-volume gap. Declines in age-standardized prevalence rates were observed across all SDI quintiles and WHO regions, but national trends were heterogeneous; 202 of 204 countries had negative AAPC from 2010 to 2023, whereas the United States and Sri Lanka had non-negative AAPC estimates. Decomposition analyses showed that the 281.87 million net increase in global prevalent cases from 1990 to 2023 was driven mainly by population growth (+725.19 million cases) and population ageing (+66.94 million), partly offset by declining age-specific prevalence (−510.26 million). In the negative-binomial APC model favored by diagnostics, the projected crude global prevalence rate was 23,127.50 (95% CrI 22,573.02–23,737.67) per 100,000 in 2024 and 23,653.08 (20,518.82–27,139.87) in 2050, corresponding to 1.88 (1.83–1.93) billion and 2.28 (1.98–2.61) billion prevalent cases, respectively.Conclusions: Despite sustained declines in age-standardized LTBI prevalence rates since 1990, the absolute number of people living with LTBI has increased and is projected to remain large through 2050. These findings suggest that TB elimination strategies cannot rely on declining standardized rates alone; they require stronger prevention-oriented policies that reduce progression to active disease, particularly in ageing and other high-risk populations.