Medication Adherence Among Patients With Type 2 Diabetes at the Upper West Regional Hospital, Ghana

The prevalence of type 2 diabetes mellitus (T2DM) has risen rapidly in low- and middle-income countries, including Ghana, and poor adherence to glucose-lowering medications undermines glycemic control and increases risk of complications.

To identify health-system, socio-demographic, and socio-economic factors associated with glucose-lowering medication adherence among adults with type 2 diabetes attending outpatient follow-up at the Upper West Regional Hospital, Ghana.

We conducted a hospital-based cross-sectional study at the Upper West Regional Hospital, Ghana between 20 February and 8 March 2024. Consecutive eligible adults (≥ 18 years) with a clinical diagnosis of T2DM and at least 3 months of self-reported medication use were recruited using convenience sampling. The sample comprised N = 202 participants. Trained research assistants administered an electronic questionnaire (Google Forms) that included the 8-item Morisky Medication Adherence Scale (MMAS-8). Bivariate associations were tested with Chi-square tests, and multivariable binary logistic regression (reporting adjusted odds ratios, AOR, with 95% CIs) was used to identify factors associated with adherence. Analyses were performed in SPSS v20 and two-sided p < 0.05 was considered statistically significant.

Overall, 51.5% of participants were classified as having poor medication adherence. Several socio-economic and health factors were significantly associated with adherence using the chi-square test of associations. In unadjusted logistic regression analysis, participants on drug treatment for < 6 months had higher odds of good adherence compared with those on treatment ≥ 6 months (COR: 3.70 95% CI: (1.50-8.97; p = 0.004) as well as those who spent more than 200 Ghana cedis monthly on drugs (COR: 36.30 95% CI: 10.88-121.1; p = 0.001) compared with those who spent less than 50 Ghana cedis on a monthly basis and participants who were diagnosed between 1 and 3 years (COR: 19.17 95% CI: 6.35-57.86; p = 0.001) and greater than 3 years (COR: 14.52 95% CI: 4.61-45.759; p = 0.001) had better odds of adherence compared to those diagnosed in less than a year. In age, education and monthly income adjusted multivariable models, none of these variables showed a significant association with medication adherence.

A substantial proportion of adults with T2DM attending outpatient follow-up at the Upper West Regional Hospital demonstrated poor self-reported adherence to glucose-lowering medications. Interventions that strengthen patient education, address medication side-effects and affordability, and improve clinician follow-up should be prioritized. The cross-sectional design of this study and further reliance on self-reported medication adherence (MMAS-8) are important limitations.