Mode
Text Size
Log in / Sign up

Diabetes prevalence 22% in Ghanaian TB cohort, associated with older age and higher BMI

Diabetes prevalence 22% in Ghanaian TB cohort, associated with older age and higher BMI
Photo by Towfiqu barbhuiya / Unsplash
Key Takeaway
Consider diabetes screening in TB patients, noting association with older age and higher BMI in this cohort.

A prospective cohort study followed 204 adults (≥20 years) with bacteriologically confirmed pulmonary tuberculosis across 14 health facilities in Greater Accra, Ghana. Participants with diabetes (the exposure) were compared to those with tuberculosis only. The study aimed to assess the association between diabetes and tuberculosis clinical presentation and treatment outcomes.

Among the 204 participants, 45 (22.1%) had diabetes at baseline; 30 of these were newly diagnosed. Diabetes was significantly associated with being age 60 or older (adjusted odds ratio [AOR] 5.7, 95% CI 1.7–19.3), having a BMI of 25 kg/m² or higher (AOR 5.4, 95% CI 1.4–21.9), and having a family history of diabetes (AOR 3.7, 95% CI 1.5–9.4). Cavities on chest X-ray were less frequent in the tuberculosis-diabetes group (51.2%) than in the tuberculosis-only group (72.5%; P = 0.014).

Overall favorable tuberculosis treatment outcomes were reported in 90.7% of participants. Unfavorable outcomes occurred in 6.8% of the tuberculosis-diabetes group versus 10.1% of the tuberculosis-only group (relative risk 0.68, 95% CI 0.2–2.2), a difference that was not statistically significant. Safety and tolerability data were not reported.

Key limitations include the observational design, which cannot prove causation, and the fact that results come from a single cohort. The confidence interval for the association between diabetes and unfavorable treatment outcomes was wide and included 1.0, indicating statistical uncertainty. The study's practice relevance is restrained; it supports existing recommendations for routine diabetes screening in tuberculosis patients for early detection and management, but the clinical implications of the radiographic finding and non-significant outcome difference remain unclear.

Study Details

Study typeCohort
EvidenceLevel 3
PublishedApr 2026
View Original Abstract ↓
BackgroundTuberculosis (TB)-diabetes is a growing public health threat in TB-endemic settings. We aimed to determine diabetes prevalence among TB patients in Greater Accra and its association with TB clinical presentation and treatment outcomes.MethodsWe enrolled 204 adults (≥20 years) with bacteriologically confirmed pulmonary TB across 14 health facilities. At treatment initiation, participants were screened for diabetes using self-report, fasting plasma glucose, and glycated hemoglobin, following standard diagnostic criteria. Baseline characteristics were recorded, and participants were followed to determine TB treatment outcomes. Associations between diabetes, baseline characteristics and treatment outcomes were assessed using bivariate and multivariate analysis in STATA. Adjusted odds ratios (AORs) and risk ratios (RRs) were estimated with 95% confidence intervals (CIs).ResultsThe median age of participants was 40.5 years (IQR 30.5–50.5), and 72.5% were male. The prevalence of baseline diabetes was 22.1% (45/204), including 30 newly diagnosed and 15 previously diagnosed participants on treatment. Among 166 chest X-rays, cavities were less frequent in TB-diabetes than TB-only participants (51.2% vs. 72.5%, P = 0.014). Diabetes was associated with age ≥60 years (AOR 5.7, 95% CI 1.7–19.3), body mass index ≥25 kg/m2 (AOR 5.4, 95% CI 1.4–21.9), and family history of diabetes (AOR 3.7, 95% CI 1.5–9.4). Overall, 90.7% had favorable TB treatment outcomes, with unfavorable outcomes in 6.8% of TB-diabetes and 10.1% of TB-only participants (RR 0.68, 95% CI 0.2–2.2).ConclusionDiabetes was common among TB patients, with many previously undiagnosed. Despite similar treatment outcomes to TB-only participants, routine diabetes screening is recommended for early detection and management.
Free Newsletter

Clinical research that matters. Delivered to your inbox.

Join thousands of clinicians and researchers. No spam, unsubscribe anytime.