Rosella et al. (2011) applied the Diabetes Population Risk Tool (DPoRT) model in a cohort study to determine the risk factors for diabetes mellitus occurrence in Canada whereby it was noted that elevated BMI, increasing age, male gender, hypertension and less than secondary education are risk factors for diabetes. The method entailed first acquiring data from Canadian Community Health Survey administered by Statistics Canada. The individuals who participated in the survey were selected by applying a multilevel stratified cluster sampling approach that looked at administrative and geographic regions as primary sampling units. The samples for each region were calculated to be proportional to the official population reports, with the persons living in reserves, remote areas, institutions, and members of the armed forces excluded from the study. Once the sample population had been identified, the survey was administered through telephones with the information collected from subjects who consented to provide their demographic and medical data. The collected data was then modeled and related for predicted and observed diabetes occurrence rates with the results validated by comparing them to the results of two previous research studies. As it stands, DPoRT is intended to act as a tool for predicting diabetes occurrence within a population based on demographic information minus clinical data.
