Background. Thailand has experienced one of the most rapid rates of population ageing, rising to more than 10% of the population. Type 2 diabetes is one of the most common chronic diseases of older adults. Self-efficacy is an important predictor for health behaviors of patients with diabetes and therefore has important theoretical and practical application for health promotion in this population. A systematic review of the literature, however, revealed that no valid and reliable instrument for assessing self-efficacy in diabetes management of Thai older patients with type 2 diabetes is currently available.

Purpose. The purpose of this study was to develop and evaluate the psychometrics of the Thai Version of the Diabetes Management Self-Efficacy Scale (T-DMSES) for older adults with type 2 diabetes.

Methods. This study was carried out in two phases. Phase 1 consisted of a multi-step process of back-translation method, using consensus meetings to translate and modify the 20-item English version Diabetes Management Self-Efficacy Scale to Thai and test semantic equivalence of translation. The investigator evaluated the psychometric properties of the T-DMSES in phase 2 by examining the content, construct, convergent and criterion validity, internal consistency and stability testing. The sample for Phase 2 comprised 209 older adults with type 2 diabetes (M = 67.4, SD =5.5) from diabetic clinics in eight hospitals from four parts of Thailand.

Results. There was semantic equivalence of translation between the original English version and the back-translated version of the T-DMSES with .88 of Intraclass Correlation Coefficient (ICC). The results revealed an acceptable content validity with the scale-level content validity index of .96 and the item-level content validity indexes of .80 or higher. Exploratory and confirmatory factor analysis supported the T-DMSES consisting of three subscales with 19 items. The convergent validity of the T-DMSES correlated well with the validated measure of the Thai version of the General Self-efficacy Scale in measuring self-efficacy with r = .36 (p < .01). Criterion-related validity showed that the T-DMSES significantly correlated with the Thai version of the Ronsenbergs’ Self-Esteem Scale with r = .43 (p < .01). High internal consistency was found with Cronbach’s alpha = .95 and test-retest reliability (ICC = .69) were acceptable.

Conclusion. The T-DMSES is a brief and psychometrically sound measure to evaluate self-efficacy towards management of diabetes by older adults with type 2 diabetes in Thai population. Based on the T-DMSES with three subscales, health care providers can structure interventions to improve knowledge and skill in areas where self-efficacy is low. Future study will include using the T-DMSES to predict diabetic self-care behaviors in older adults with type 2 diabetes.