Grant: Travel Awards - Institute Community Support

Diabetes has been a public health concern in Canada for decades. About 30% of Canadians live with diabetes, while approximately 90% of them have type 2 diabetes (T2D). T2D management is closely related to socioeconomic factors. In democratic societies, voting is important to political representativeness in policy-making that can influence population health and health inequities. However, barriers to voting are also disproportionately affecting certain populations. The populations more at risk for T2D are often the ones with limited voting participation based on the literature review. Given these inequities, it is important to explore the relationship between voting rates and T2D outcomes. We aim to map the prevalence of T2D and voting turnout, to determine the association between voting rates and prevalence of T2D, and to predict the incidence of T2D by voting turnout. First, I will visualize the geographic distributions of voting rates and location of healthcare facilities in Canada using ArcGIS Pro 3.3. I will use inverse distance weighting to present the accessibility to healthcare facilities and compare it with voting rate heat maps. Second, I will use data from the Canadian Community Health Survey to estimate odds ratios for T2D with voting rates as the exposure variable while controlling for confounders. Interactions with sex/gender and cultural/racial background will be analyzed. Third, I will employ random forest algorithms to predict T2D incidence based on voting turnout and other risk factors. I will also compare the outputs from the machine learning techniques with the outcomes from traditional epidemiological methods to decide which is better in predicting outcomes and inequities in population health. This study connects voting turnout to T2D outcomes, informing strategies for equitable healthcare and guiding interventions for formerly neglected groups. This study will be part of my PhD thesis on chronic disease management in Canada.