Using web-based familial risk information for diabetes prevention: a randomized controlled trial

BACKGROUND It has been suggested that family history information may be effective in motivating people to adopt health promoting behaviour. The aim was to determine if diabetic familial risk information by using a web-based tool leads to improved self-reported risk-reducing behaviour among individuals with a diabetic family history, without causing false reassurance among those without a family history. METHODS An online sample of 1,174 healthy adults aged 35-65 years with a BMI ≥ 25 was randomized into two groups receiving an online diabetes risk assessment. Both arms received general tailored diabetes prevention information, whilst the intervention arm also received familial risk information after completing a detailed family history questionnaire. Separate analysis was performed for four groups (family history group: 286 control versus 288 intervention group; no family history: 269 control versus 266 intervention group). Primary outcomes were self-reported behavioural outcomes: fat intake, physical activity, and attitudes towards diabetes testing. Secondary outcomes were illness and risk perceptions. RESULTS For individuals at familial risk there was no overall intervention effect on risk-reducing behaviour after three months, except for a decrease in self-reported saturated fat intake among low-educated individuals (Beta (b) -1.01, 95% CI -2.01 to 0.00). Familial risk information resulted in a decrease of diabetes risk worries (b -0.21, -0.40 to -0.03). For individuals without family history no effect was found on risk-reducing behaviour and perceived risk. A detailed family history assessment resulted in a greater percentage of individuals reporting a familial risk for diabetes compared to a simple enquiry. CONCLUSIONS Web-based familial risk information reduced worry related to diabetes risk and decreased saturated fat intake of those at greatest need of preventative care. However, the intervention was not effective for the total study population on improving risk-reducing behaviour. The emphasis on familial risk does not seem to result in false reassurance among individuals without family history. Additionally, a detailed family history questionnaire identifies more individuals at familial risk than a simple enquiry. TRIAL REGISTRATION NTR1938.