Discussion
This systematic review and meta-analysis determined the pooled proportion of non-adherence to anti-diabetic medications and synthesized the associated factors with medication non-adherence among individuals with diabetes. Based on the MMAS-8, the MMAS-4, and self-care tools, the proportion of non-adherence to anti-diabetic medication was reported at different ranges.  We found the pooled proportion of non-adherence to anti-diabetic medications using the MMAS-8, MMAS-4 and cut-off at 80 or 90% scale was 43.4%, 29.5% and 29.1%, respectively Evidence from previous research reported that the rates of non-adherence to prescribed anti-diabetic regimens among individuals with diabetes ranged from 9% to ˃80% 5, which is in line with current findings. In addition, another review on adherence to medication among individuals with type 2 diabetes reported that average adherence to anti-diabetic medication ranges from 36% to 93%4. However, earlier reviews were predominantly focused on adherence with diabetes medication4,57, factors affecting medication adherence among patients with diabetes58,59, and economic consequences of mediation adherence in diabetes60.
Failure to adherence to prescribed medication regime is seen as a serious issue resulting in negative impact on individuals’ health and health care system. There is, thus, a need to identify specific barriers that will help in adopting suitable techniques to overcome them and improve medication adherence. Previous studies reported various factors of medication non-adherence among individuals with diabetes, such as the severity of disease, demographic factors and socioeconomic status58. This study attempted to collate factors affecting non-adherence from each of the five groups, including disease, therapy, healthcare, patient, and socioeconomic factors, which would have wide generalizability. Results from current study advocates that patient-related factors were consistently related to non-adherence. Patients having high BMI and low concerns were related to non-adherence. Further, non-adherence was consistently associated with socioeconomic factors, such as low education and being employed in our review. Similar findings were also reported in previous studies where association between education and non-adherence reported25. It has been reported that low education level can interfere with knowledge of the disease and medication to some extent26. Considering potential gap between education and disease awareness, educational research is urgently warranted to identify effective strategies to assist individuals with diabetes in medication adherence.
Patient’s knowledge about disease and associated factors could play a pivotal role for better compliance with treatment strategy. This study found that non-adherence was related with condition related characteristics. Individuals suffering from depression was consistently associated with non-adherence to anti-diabetic medication in current study which is in accordance with findings from prior studies61,62. For instance, a study by Grenard and colleague reported that individuals with depression were 1.76 more likely to be non-adherent to taking medication for chronic diseases than patients without depression62. Therapy-related factor was yet another factor consistently associated with non-adherence in our review. Drug regimen complexity, i.e., taking multiple daily doses was found to be a critical factor for non-adherence to medication. This finding is consistent with that found in a previous review that showed adherence is inversely associated with the number of medication doses per day 63. This suggests that clinicians should be aware of the fact that access to simplified dosage regimens by patients may be important aspect in maximizing therapeutic success.
Lastly, healthcare system factors were consistently related to non-adherence in our review. We found that poor doctor-patient relationship was associated with medication non-adherence in individuals with diabetes and aligns with findings from a prior review64, suggesting that doctor-patient relationship quality is a potentially important point of intervention to improve medication adherence. Indeed, the provider-patient relationship plays a major role in keeping the patient well informed about their medications, which directly influences the adherence 65. In the present study, dissatisfaction with communication was another healthcare system factor associated with non-adherence. This finding is consistent with a prior study about aspects of patient-provider communication and medication adherence in diabetes66. Future studies should investigate whether improving communication skills among providers with poorer patient communication ratings could improve their anti-diabetic medication adherence and outcomes. In addition, some factors such as poor knowledge, self-efficacy, young adulthood age, unmarried, low income, having more comorbid conditions, being hypertensive, having adverse side effects and complex medication regime had inconsistent relationships with non-adherence to anti-diabetic medications across studies. Similar results was also reported in the previous studies15,37,67.
We used a comprehensive search strategy to identify all eligible studies and attempted to increase the quality of the included studies by using well-defined eligibility criteria. The overall quality of the included studies was acceptable and meeting all the specified criteria. However, current study has some limitation as it was restricted to papers written in the English language only, contributing to potential publication bias. It was not possible to quantitatively pool data from all 43 studies that were identified for the systematic search and this could reduce the generalizability of findings. There was a high heterogeneity between the included studies due to differences in assessment methodologies. Further, we were unable to obtain pooled estimates of the strength of association of each factor. Nevertheless, our extensive summary of the factors associated with non-adherence will be useful when designing interventions to combat non-adherence in future research.