Development and validation of the Short Version of Diabetes Obstacles Questionnaire (DOQ) to assess obstacles in managing Type 2 diabetes among patients of Estonia

Abstract

The aim of this study was to develop the Short Version of Diabetes Obstacles Questionnaire (SDOQ), a less time consuming measure, to assess the obstacles in managing type 2 diabetes among patients in Estonia. The SDOQ is based on the 78-item Diabetes Obstacles Questionnaire previously validated in Estonia (DOQ; Kongi, 2001; Hearnshaw, Dale, Sturt, Vermeire, and Van Royen, 2007). 267 respondents participated in the study (137 in Estonia, 130 in Slovenia). General practitioners invited 5 consecutive diabetic patients to participate in the study, who completed the DOQ and the Problem Areas in Diabetes Score (PAID; Welch, Jacobson, and Polonsky, 1997). Demographic and medical data was collected in addition. Statistical analyses for estimating dimensionality and homogeneity of the measure, and correlations with the PAID and glycemic control were performed in SPSS version 15.0. Further analysis of dimensionality of the SDOQ subscales, and measurement invariance cross culturally was estimated on data of Estonian and Slovenian respondents with single- and multi-group Confirmatory Factor Analysis (CFA) with program Lisrel 8.80 (Jöerskog & Sörbrom, 2006). CFA resulted in the 40-item measure with 6 subscales. The subscales demonstrated acceptable internal consistency (Cronbach’s alpha .67-.86 in Estonia, and .66- .85 in Slovenia) and CFA models showed acceptable fit levels. Most subscales were significantly correlated with the PAID (Pearson r from 0.15 to 0.46 in Estonia, and .07-.45). Acceptable measurement invariance was only partly confirmed. Therefore, future research is needed to confirm the cross-cultural validity of the SDOQ. In conclusion, the SDOQ is a valid instrument to measure obstacles to treatment adherence and diabetes self-management living in Estonia.