Summary
A medRxiv preprint combining 82 population surveys found that bacteriologically confirmed pulmonary TB prevalence rose with age across 30 low- and middle-income countries. Notification-to-prevalence ratios fell in older age groups, which the authors say suggests poorer case detection.
A systematic review of tuberculosis (TB) surveys in low- and middle-income countries found that bacteriologically confirmed pulmonary TB prevalence increased with age among adults. The medRxiv preprint, posted on October 19, 2026, also found that notification-to-prevalence ratios declined in older age groups—a pattern the authors interpret as suggesting poorer case detection among older people.
The analysis combined 82 surveys from 30 countries, covering 4.6 million participants aged 15 and older. The surveys were reported between 1993 and 2025. The researchers used Bayesian meta-regression to estimate how TB prevalence varied across age groups and settings. Prevalence describes existing disease in a population; it is not a measure of how many new infections occur over time.
Prevalence increased across adulthood
Across the pooled data, each 10-year increase in age was associated with a 30% increase in bacteriologically confirmed TB prevalence (95% credible interval: 22% to 38%). The model estimated prevalence at 796 per 100,000 among people aged 75 and older, compared with 127 per 100,000 among those aged 15–19. That is an estimated 6.4-fold difference, with a 95% credible interval of 4.0 to 10.4.
The direction of the age pattern was positive in every country included. The estimated increase per decade varied: the interquartile range across countries was 23% to 40%. Age gradients were lower in Africa than in other regions and were also lower for smear-positive TB than for bacteriologically confirmed TB. In countries where HIV prevalence was at least 5%, the age gradient was flatter above age 40. The authors also report that age gradients declined slowly over the period covered by the surveys.
These are pooled, model-based estimates of prevalence patterns, not evidence that age itself causes TB. They show how the existing burden was distributed across adult age groups in the surveyed settings.
Notifications lagged further behind prevalence with age
The researchers also compared TB notifications with estimated prevalence by age. Notifications are cases recorded by health systems; prevalence surveys estimate existing disease in the population. The notification-to-prevalence ratio declined by 10% for each 10-year increase in age (95% credible interval: 7% to 14%). The authors treat this ratio as an indicator of diagnostic delay and interpret its decline as a sign that TB cases may be detected less completely among older people.
The ratio is a population-level indicator, not a direct measurement of how long an individual waits for diagnosis. Taken together, the two patterns suggest that older adults carry a higher burden of existing TB while a smaller share of that burden is reflected in notifications. The findings support examining age-specific patterns in TB surveillance and case detection, rather than treating adults aged 15 and older as a single group.
The report is a preprint, and its estimates cover adults in low- and middle-income countries; they do not describe age patterns in children.