Bicycling and Mortality Among Individuals With Diabetes in the European Prospective Investigation Into Cancer and Nutrition (EPIC)

AIMS The primary aim of the study is to study the relationship between overall bicycling and all-cause mortality and secondarily cardiovascular disease mortality among individuals with diabetes from European countries. A secondary aim will be to study the relation of change in bicycling to all-cause mortality and cardiovascular disease mortality.

METHODS Study design and setting The study is a nested cohort study in European Prospective Investigation into Cancer and Nutrition (EPIC) cohort. In EPIC, 29 centers in 10 western European countries have collected information on nutrition, lifestyle, anthropometrics and medical history from more than 521 000 study individuals participating in this prospective cohort study. Dietary, medical history and lifestyle were assessed by questionnaires. Blood samples from the participants were taken at the study centers. Baseline (first examination) information was collected 1992-2000 and second examination for exposure has since been obtained at least once in every cohort. Exposure data has subsequently, to the two examinations, been linked to information on vital status and on cause of death were obtained through record linkages with national, regional or local registers, regional health departments, physicians or hospitals, active follow-up or health insurance.

Study population This nested prospective cohort study, sampled individuals with prevalent diabetes in the EPIC cohorts at the baseline assessments in the years 1992-2000. Self-reported and confirmed diabetes (validated by a second source (at least 1), including repeated self-report, contact with physician, linkage to register later point, intake of diabetes medicine, registration of diabetic chiropody, baseline glycated hemoglobin>=6.0%, five annual blood glucose measurements or two blood glucose measurements per year for five consecutive years) cases were included in the analyses. No information is available to distinguish between type 1 and type 2 diabetes across the population but type 1 is rare by comparison. The nested cohort for this study comprised 12,848 individuals with self-reported and/or confirmed diabetes. Mean (SD) follow-up time was 14.6 (4.8) years. Among the 12,848 prevalent diabetes cases, 5,809 cases were confirmed at baseline as described above.

For further detail, please see the predefined statistical analysis plan (SAP). It can be accessed through the url: http://aktivsundhed.dk/images/docs/SAPMRL.pdf