Predictors of Self-care Practices among Patients with Type 2 Diabetes in Ghana: The Role of Illness Perception and Diabetes Knowledge

International audience Background: Type 2 diabetes mellitus requires continuous self-care across diet, physical activity, blood-glucose monitoring, foot care and medication use. Evidence from Ghana and sub-Saharan Africa shows persistent gaps in diabetes self-management, partly because biomedical advice is mediated by patients’ knowledge, beliefs, illness representations, health literacy and everyday resource constraints. Although diabetes knowledge and illness perception have each been linked to self-care, evidence on their simultaneous contribution among patients receiving tertiary diabetes care in Ghana remains limited.Aim: This study assessed diabetes knowledge, illness perception and self-care practices among adults with type 2 diabetes receiving care at Komfo Anokye Teaching Hospital (KATH), Kumasi, Ghana.Methods: A facility-based cross-sectional study was conducted among 200 adults with type 2 diabetes attending the diabetic clinic at KATH. Participants were recruited using convenience sampling. Data were collected with a structured questionnaire covering socio-demographic characteristics, diabetes knowledge, illness perception and self-care practices. Diabetes knowledge items were adapted from the Michigan Diabetes Knowledge Test, illness perception was assessed using the Brief Illness Perception Questionnaire approach, and self-care was assessed using items consistent with the Summary of Diabetes Self-Care Activities. Data were analysed using IBM SPSS Statistics version 25. Descriptive statistics, chi-square tests, Pearson correlations and multiple linear regression were conducted at the 5% significance level.Results: Most participants were female (66.5%), married (51.5%) and aged 61–70 years (31.5%). Overall, 75.0% had a low level of self-care practice. Domain-specific adherence was lowest for exercise (21.5%), followed by diet (31.0%), blood-glucose testing (39.5%) and foot care (50.5%), while medication adherence was high (92.0%). Age (χ²=11.67, p=0.02) and educational level (χ²=10.72, p=0.03) were significantly associated with self-care practice. Diabetes knowledge correlated positively with overall self-care practice (r=0.219, p<0.01). Illness perception was not significantly correlated with overall self-care (r=-0.043, p>0.05), although small, significant correlations were observed with exercise (r=0.162, p<0.05) and blood-glucose testing (r=-0.147, p<0.05). In regression analysis, diabetes knowledge independently predicted self-care practice (B=0.055, t=2.284, p=0.024), while illness perception did not. The model explained 3.8% of the variance in overall self-care.Conclusion: Self-care practice among adults with type 2 diabetes at KATH was generally low despite high reported medication adherence. Diabetes knowledge was associated with self-care and was the only significant predictor in the limited regression model; however, the model’s low explanatory power indicates that knowledge alone accounted for little of the variation in self-care. The findings support the strengthening of context-sensitive diabetes education, particularly for exercise, dietary practice, blood-glucose monitoring and foot care, while recognising the likely influence of social, economic, behavioural and health-system factors not measured in this study.