Psychometric evaluation of the summary of diabetes self-care activities measure in Ghanaian adults living with type 2 diabetes

Abstract Aims We evaluated the content validity, factorial structure, internal consistency, construct validity, and floor and ceiling effects of the SDSCA among Ghanaian persons with type 2 diabetes. Methods The summary of diabetes self-care activities measure (SDSCA) was administered to 187 adults living with type 2 diabetes from three diabetes clinics. Results A confirmatory factor analysis maintained the four factor structure of the SDSCA. However, two items, 3 (fruit and vegetable servings) and 4 (red meat or full-fat dairy products) had factor loadings of 0.26 and 0.16 respectively. The model also had a statistical power of 0.72 (below acceptable criteria). Modification of the model by removing item 4 resulted in an improved revised model with a power of 0.82. Construct validity was found for the exercise and diet subscales of the SDSCA but not for the self-monitoring of blood glucose and foot care subscales. The internal consistency of the SDSCA measure was 0.68, below acceptable criteria for internal consistency. No floor effects were present but the exercise subscale had ceiling effects. Conclusion The SDSCA measure had content validity, maintained its multidimensionality and met the criteria for floor effects but not for construct validity, internal consistency and ceiling effects. The SDSCA measure may require improvements to evaluate self-care behaviours of adult type 2 diabetes patients in Ghana and probably in other sub-Saharan countries.