Towards a better understanding of self-management interventions in type 2 diabetes: A meta-regression analysis

Abstract Aims Attributes that operationally conceptualize diabetes self-management education (DSME) interventions have never been studied previously to assess their impact on relevant outcomes of interest in people with type 2 diabetes (T2D). The aim of this study was to determine the impact of existing interventions classified by their delivery of skills or information related attributes on immediate (knowledge), intermediate (physical activity), post-intermediate (HbA1c), and long-term (quality of life) outcomes in people with T2D. Methods PubMed, Embase, PsycINFO, and Cochrane Library/Cochrane CENTRAL as well as the grey literature were searched to identify interventional studies that examined the impact of DSME interventions on the four different outcomes. Eligible studies were selected and appraised independently by two reviewers. A meta-regression analysis was performed to determine the impact of delivery of the skills- and information-related attributes on the chosen outcomes. Results 142 studies (n = 25,511 participants) provided data, of which 39 studies (n = 5278) reported on knowledge, 39 studies (n = 8323) on physical activity, 99 studies (n = 17,178) on HbA1c and 24 studies (n = 5147) on quality of life outcomes. Meta-regression analyses demonstrated that skills-related attributes had an estimated effect suggesting improvement in knowledge (SMD [standardized mean difference] increase of 0.80; P = 0.025) and that information-related attributes had an estimated effect suggesting improvement in quality of life (SMD increase of 0.96; P = 0.405). Skill- and information-related attributes did not have an estimated effect suggesting improvement in physical activity or in HbA1c. Conclusions The study findings demonstrate that the skills and information related attributes contribute to different outcomes for people with T2D. This study provides, for the first time, preliminary evidence for differential association of the individual DSME attributes with different levels of outcome.