Effect of Preventive Messages Tailored to Family History on Health Behaviors: The Family Healthware Impact Trial

Abstract

We wanted to determine the impact of automated family history assessment and tailored messages for coronary heart disease, stroke, diabetes, colorectal, breast, and ovarian cancer on preventive behaviors compared with a standard preventive message. The study was a cluster-randomized clinical trial that included 41 primary care practices, the majority in the Midwest, using Family Healthware, a self-administered, Web-based tool that assesses familial risk for the diseases and provides personalized risk-tailored messages. Patients in the control group received an age- and sex-specific health message related to lifestyle and screening. Smoking cessation, fruit and vegetable intake, physical activity, aspirin use, blood pressure, and cholesterol and blood glucose screening were assessed at baseline and 6 months after the intervention. Of 4,248 participants, 3,344 (78%) completed the study. Participants were white (91%), female (70%), and insured (97%), and had a mean age of 50.6 years (range 35-65 years). Intervention participants were more likely to increase daily fruit and vegetable consumption from 5 or fewer servings a day to 5 or more servings a day (OR = 1.29; 95% confidence interval [CI], 1.05-1.58) and to increase physical activity (OR = 1.47; 95% CI, 1.08-1.98) to 5 to 6 times a week for 30 minutes or more a week. The absolute differences in proportion were 3% and 4%, respectively. Intervention participants were less likely to move from not having cholesterol screening in the last 5 years to having their cholesterol measured within 5 years (OR = 0.34; 95% CI, 0.17-0.67), with an absolute difference of 15%. Messages tailored to an individual's familial risk for 6 common diseases modestly increased self-reported physical activity and fruit and vegetable intake but reduced the likelihood of receiving cholesterol screening.