Yamaoka K and Tango T (2005) Efficacy of lifestyle education to prevent type 2 diabetes: a meta-analysis of randomized controlled trials. Diabetes Care 28: 2780–2786

Type 2 diabetes is increasingly common and causes significant morbidity and reduction in quality of life. Prevention of this disease is, therefore, highly desirable. A Japanese team has conducted the first meta-analysis of the impact of lifestyle education (i.e. diet and exercise) on risk of type 2 diabetes (determined by 2 h plasma glucose and disease incidence).

The meta-analysis included eight randomized, controlled trials with a follow-up period ≥6 months that were identified via MEDLINE and the Educational Resources Information Center database. All patients within the trials (n = 2,026) were at high risk of diabetes and received lifestyle, or solely dietary, intervention.

According to a RANDOM-EFFECTS MODEL, the mean 2 h plasma glucose levels in patients who received 1 year of lifestyle intervention were 0.84 mM lower in patients than in control subjects (95% CI 0.39–1.29). The incidence of type 2 diabetes in those receiving lifestyle education was approximately half that of controls (relative risk = 0.55; 95% CI 0.44–0.69). Similar trends were seen when a range of statistical models were used, and when other subsets of patients (e.g. those receiving treatment for >1 year) were analyzed.

Although confined to English-language articles, a factor the authors suggest might introduce bias, this study provides important evidence of benefit and highlights the importance of lifestyle education for the prevention of type 2 diabetes.