Abstract

BackgroundA healthy lifestyle is important for maintaining health and preventing complications in patients with type 2 diabetes, and yet, few instruments are available to measure this. AimThe aim of the present study was to examine the psychometrics of a recently developed tool that can be used to screen for a health-promoting lifestyle in patients with type 2 diabetes. MethodsData were collected from outpatients attending diabetes clinics. The Type 2 Diabetes and Health Promotion Scale (T2DHPS), EQ-5D, medical records and a demographic questionnaire were administered to 368 participants. Forward-backward translation of the original English version was used to develop a Persian version. Internal consistency of the scale was assessed by Cronbach's alpha and item-to-total correlation. Acceptability was measured by assessing floor and ceiling effects for each item and subscale. The item scaling test was used to examine the convergent and discriminant validity of the measure. Pearson correlation was used to determine the predictive validity of the scale. An explanatory factor analysis and known-group method were used to establish construct validity. ResultsAdjusted item-total correlations were higher than 0.20. Cronbach's alpha for the 28-item scale was 0.88 and for subscales ranged from 0.53 to 0.94. Correlations between the total score and subscale scores were significant (<0.01) and adequate (r's0.53). There were significant relationships between the T2DHPS and both the EQ-5D and indicators of glycaemic control. Convergent and discriminant validity of the scale was established. Significant differences in lifestyle dimensions were present between different groups of patients, demonstrating known-group validity. A six-factor solution was obtained that explained 54.6 of the total variance. ConclusionThe T2DHPS is a valid and reliable tool for investigating lifestyle behaviours in patients with type 2 diabetes. Further studies to establish the psychometric properties of the scale in other languages and cultures are suggested.