Non-adherence to self-care and associated factors among diabetes adult population in Ethiopian: A systemic review with meta-analysis

Background: Self-care practice among people with diabetes is not well-implemented in Ethiopia. So far, in Ethiopia, several observational studies have been done on self-care practice and its determinants in people with diabetes. However, a comprehensive review that would have a lot of strong evidence for designing intervention is lacking. So, this review with a meta-analysis was conducted to bridge this gap. Methods: A systematic review of an observational study is conducted following the PRISMA checklist. Three reviewers have been searched and extracted from the World Health Organization’s Hinari portal (SCOPUS, African Index Medicus, and African Journals Online databases), PubMed, Google Scholar and EMBASE. Articles’ quality was assessed using the Newcastle-Ottawa Scale by two independent reviewers, and only studies with low and moderate risk were included in the final analysis. The review presented the pooled proportion of non-adherence to self-care practice in people with diabetes and the odds ratios of risk factors hindering to self-care practice after checking for heterogeneity and publication bias. The review has been registered in PROSPERO with protocol number CRD 42020149478. Results: We included 21 primary studies (with 7,134 participants) in this meta-analysis. The pooled proportion of non-adherence to self-care in the diabetes population was 49.91 (95% CI: 44.73–55.08, I2 = 89.8%). Male (Pooled Odds Ratio (POR): 1.84 95%CI; 1.04–2.64, I2 = 15.0%), having private glucometer (POR: 2.71; 95%CI: 1.46–3.95, I2 = 0.0%), short-term Diabetes Mellitus (DM) duration (POR: 3.69; 95%CI: 1.86–5.52, I2 = 0.0%), DM complication (POR: 2.22; 95%CI: 1.48–2.95, I2 = 0.0%), treatment satisfaction (POR: 1.8; 95% CI: 1.15–2.44, I2 = 0.0%), received diabetes self-management education (POR: 2.71; 95% CI: 1.46–3.95, I2 = 0.0%) and poor self-efficacy (POR: 3.09; 95% CI: 1.70–4.48, I2 = 0.0%) were statistically significant factors of non-adherence to self-care practice. Conclusions: The overall pooled proportion of non-adherence to self-care among adult diabetes in Ethiopia was high. Further works would be needed to improve self-care practice in the diabetes population. So, factors that were identified might help to revise the plan set by the country, and further research might be required to health facility fidelity and each domain of self-care practice according to diabetes self-management guideline.