Risk of TB in individuals with type 2 DM based on the TPI score [TBN 076]

Who

Adult patients >18 years with type 2 diabetes mellitus, with or without pulmonary TB, attending the Internal Medicine Outpatient Department at Fatmawati General Hospital. Final sample: 109 participants, comprising 39 cases with diabetes and TB and 70 controls with diabetes only. Patients with immunocompromised conditions, autoimmune disease receiving major immunosuppressive care, or incomplete records were excluded.

What

The study evaluated the tuberculosis predictive index (TPI) score for identifying TB risk among patients with diabetes. High TPI score was significantly associated with TB: 82.1% of diabetes-TB patients had high TPI scores versus 40.0% of diabetes-only patients. The association was significant, with OR 6.8, 95% CI 2.6–17.6, p<0.001.

Among individual factors, TB-like symptoms showed the strongest association with TB risk: OR 13.3, 95% CI 5.1–34.3, p<0.001. Low BMI <18.5 kg/m² was also associated with TB risk: OR 3.3, 95% CI 1.0–11.0, p=0.039. Poor housing ventilation ≤10% of floor area was associated with increased TB risk: OR 3.2, 95% CI 1.4–9.8, p=0.008.

When

Medical records from 2021–2024 were reviewed. Data collection occurred from May to August 2024, with questionnaires during July–August 2024.

Where

Fatmawati General Hospital, Internal Medicine Outpatient Clinic, Indonesia.

Why

The study addressed limitations of conventional TB screening among patients with diabetes, especially atypical presentation, latent or early TB detection difficulties, and the added complexity of poor glycemic control. The objective was to assess whether the TPI score could improve TB risk stratification in patients with diabetes.

How

Observational case-control study using consecutive sampling. Data came from medical records and patient questionnaires. The TPI score included age, sex, TB contact history, HbA1c, TB-like symptoms, BMI, diabetes duration, house ventilation, and psychological well-being. TB-like symptoms required ≥3 symptoms such as prolonged cough, hemoptysis, fever, night sweats, weight loss, or reduced appetite.

Level of evidence: observational analytic case-control evidence; useful for association and risk stratification, but not causal inference.

Major limitations: potential recall bias for subjective symptoms, reliance on medical records, single-center setting, and case-control design limiting causal conclusions.

Source: Audina DP, Aritonang RS, Mokoagow MI. Risk of tuberculosis in individuals with type 2 diabetes mellitus based on the tuberculosis predictive index score: a case-control study in Indonesia. Osong Public Health and Research Perspectives. 2025 Jun 11;16(4):406. https://benangmerah.net/record/95/risk-of-tuberculosis