Abstract
BACKGROUND: Frailty is frequent in older adults with type 2 diabetes and may be associated with negatively affect treatment adherence and metabolic control. OBJECTIVE: This study aimed to assess frailty levels in older adults with type 2 diabetes and to examine their associations with treatment adherence and metabolic parameters. METHODS: This descriptive correlational study was conducted with 384 type 2 diabetes. Data were collected using “Personal Information Form”, “Edmonton Frail Scale (EFS)”, “Medication Adherence Report Scale (MARS)”, and “Metabolic Parameters Monitoring Form”. Descriptive statistics, correlation analyses and multiple linear regression analyses were performed. RESULTS: Frailty levels were significantly higher among women, those with lower education, longer disease duration, insulin or OAD+insulin users, and individuals with complications. Frailty was observed in 78.6% of participants, with 32.8% classified as severe. The mean MARS score was 14.3, indicating moderate adherence, which was lower among those with complications and OAD+insulin therapy. EFS scores showed significant associations with metabolic parameters and a negative correlation with MARS, suggesting reduced adherence with increasing frailty. Multiple regression confirmed the model predicting frailty was significant (F = 16.170, p < 0.001), explaining 28% of the variance (R²=0.28). CONCLUSION: Frailty is highly prevalent among older adults with type 2 diabetes and is associated with lower medication adherence and adverse metabolic profiles. These findings underscore the importance of routine frailty screening and individualized, multidisciplinary care in diabetes management. IMPLICATIONS OF THE STUDY: Early identification of frailty in older adults with type 2 diabetes is essential to improve treatment adherence and metabolic outcomes. Integrating frailty screening into routine diabetes care can guide personalized interventions and support multidisciplinary management to reduce complications and enhance overall well-being.