Data Analysis
The impact of the intervention across all three behaviours was assessed using 2 (intervention, control) × 2 (pre, post-intervention) MANOVAs. The integrated model constructs were considered as predictor variables for low-fat food/meal consumption, carbohydrate counting, and physical activity. In another MANOVA analysis, BMI, weight, TG, and LDL-c were considered as predictors to show the effects of the intervention on clinical and biochemical outcomes. Mediation analyses were also conducted to assess whether any of the (significant) extended TPB variables mediated the impact of the intervention on behaviour. α was set at 0.05 as the significance level. Baseline differences for age, gender, education level, job, marital status, years since diagnosis, weight, height, BMI, TG, and LDL-c were examined by Chi-square and independent t-tests.