Are behavioural approaches feasible and effective in the treatment of type 2 diabetes? A propensity score analysis...

Lifestyle changes are considered first line treatment in type 2 diabetes, but very few data are available in the "real world" of diabetes units. We aimed to measure the effectiveness of moderate and high intensity interventions on weight loss, metabolic control and insulin use. We report a prospective cohort study, carried out in 822 consecutive subjects with type 2 diabetes, first seen in a 4-year period in a diabetes unit of an academy hospital.

Subjects were treated with either a sole prescriptive diet (Diet), or received an additional short-course Elementary Nutritional Education (4 group sessions-ENE) or an intensive Cognitive Behavioural Therapy (12-15 group sessions-CBT). The results were adjusted for the propensity score to be assigned different treatments, derived from logistic regression on the basis of age, gender, BMI, HbA1c, diabetes duration and insulin use at baseline. Main outcome measures were weight loss and weight loss maintenance, metabolic control, and secondary failure to insulin use. Both structured programmes produced a larger weight loss, and the adjusted probability of achieving the 7% weight loss target was increased. Similarly, both programmes favoured metabolic control, irrespective of insulin use. After adjustment for propensity score, both ENE (hazard ratio, 0.48; 95% CI, 0.27-0.84) and CBT (hazard ratio, 0.36; 95% CI, 0.16-0.83) were associated with a reduced risk of de novo insulin treatment.

Structured behavioural programmes aimed at lifestyle changes are feasible and effective in the "real world" setting of a diabetes unit for the treatment of type 2 diabetes.