Type 2 diabetes testing in Eastern Uganda: a contextualized evaluation of available point-of-care tests

Abstract

Background: Diabetes is one of the major global health threats today with 537 million living with the disease, which contributes 11.3% to global mortality. Sub-Saharan Africa has the highest proportion of undiagnosed diabetes, with 53.6% of all cases. This study aimed to evaluate type 2 diabetes testing in eastern Uganda, comparing the diagnostic performance of the Fasting Plasma Glucose (FPG) and Glycated haemoglobin (HBA1c) tests.

Methods: An exploratory sequential mixed methods study design, involving a qualitative study and three quantitative studies was used. All four studies were conducted at Iganga General Hospital. In-depth interviews were conducted among 17 patients and three health providers. The data was analysed using template analysis. Secondly, 1659 patients underwent FPG and HBA1c POC testing with the OGT testing as the clinical reference standard for test performance. Additionally, 502 participants underwent FPG, HBA1c, OGT and asymptomatic malaria testing and regression modelling was conducted to determine the association between asymptomatic malaria and glycemia. Finally, an economic evaluation was conducted in which the cost-effectiveness of the FPG and HBA1c were compared using decision analytic modelling from a societal perspective.

Results: Patients’ misconceptions about diabetes risk, use of traditional medicine and limited diagnostic facilities at the primary care level contribute to delayed diabetes detection resulted in four diabetes diagnostic pathway typologies. The FPG and HBA1c tests had comparable sensitivity [62.6% (95% CI, 41.5-79.8) versus 69.8% (95% CI, 46.3-86.1), respectively] and specificity [98.6% (95% CI, 95.4-99.6) versus 99.4% (95% CI 98.9-99.7), respectively] for diabetes screening. Similarly, the FPG (AOR 0.33, 95% CI 0.12-0.90) and HBA1c (AOR 0.28, 95% CI 0.10-0.77) tests were comparably less likely to detect diabetes among individuals with asymptomatic malaria compared to those without. However, despite correctly diagnosing a comparable proportion of patients (96.2% vs 96.3%, respectively), the HBA1c test was less costly than the FPG test by $1.27 per test. The ICER was $989.06 per additional patient correctly diagnosed.

Conclusion and implications: Timely diabetes detection requires improved patient sensitization and access to diabetes testing services. Both the HBA1c and FPG POCs have a role in improving diabetes detection, however, their performance is sub-optimal in malaria-endemic regions.