Dorien Brouwer

115 Motivational interviewing to support lifestyle behavior change PART 2 · Self-efficacy, measured with the self-efficacy scale, a 7-item scale with scores that range from 1 to 5. Higher values indicate more confidence to carry out the behavior necessary to reach the desired goal. [22, 43, 44] Questions are formulated as: I think I am able to quit smoking / choose healthy food/ care for enough physical activity. Cronbach’s alpha was 0.84. Outcome measures Primary outcome measure was lifestyle behavior change after 6 months, defined as smoking cessation and/or increase of physical activity of 30 minutes a day and/or increasing score of 5 points on the Food Frequency Questionnaire. These outcome measures were based on the food and exercise recommendations in the Netherlands. Secondary outcome measures included change in self-efficacy and intention to change lifestyle behavior, change in weight, waist circumference, blood pressure, cholesterol and blood glucose after six months and feasibility of motivational interviewing on the nurse- led outpatient clinic. A sample of 136 patients was expected to have a power of 80% to detect a difference in lifestyle behavior change (quitting smoking and/ or a 30-minute increase in exercise and/ or a 5-point increase in Food Frequency Questionnaire) between the intervention group and control group of 25% after 6 months assuming a significance level of α = 0.05 and a change in lifestyle behavior of 30% in the control group. Statistical analysis Statistical analyses were performed with STATA 12.1 statistical package (Statacorp, College Station, Texas). Analyses were done by intention-to-treat principles. Differences in characteristics were analyzed with Chi2 tests (proportions), ttests and in case of unequally distributed data with Mann-Whitney U tests. Differences in primary and secondary outcomes between control and intervention group were compared with Chi2 tests and further analyzed by multivariable logistic or linear regression in which we adjusted for age, sex, baseline self-efficacy, baseline smoking status and history of TIA (because of unbalanced trial arms). Differences between primary outcomes in subgroups were analyzed with Chi2 tests. These subgroups contained patients with high self-efficacy scores (22 or higher based on median self-efficacy of 22) or low (<22) self-efficacy scores and patients who received all three counselling sessions and patients who had one or