This project investigates the potential relationship between poverty rates and heart attack prevalence in Minnesota counties. Heart disease is a leading cause of death in the United States, and poverty is a known risk factor for poor health outcomes. We aimed to explore whether poverty rates correlate with higher heart attack rates across Minnesota counties.
Heart disease's impact on communities and its potential link to socio-economic factors necessitate a deeper investigation. Understanding this relationship can help tailor public health interventions and policies to address heart disease more effectively, particularly in underprivileged areas.
- Poverty Data: Gathered by county and year using the Census API, including demographic details (race, gender, age).
- Heart Attack Data: Obtained from the Centers for Disease Control and Prevention (CDC), focusing on heart attack death data.
- Cleaning Process: Due to limitations in the homogeneity of health data, we focused on overall poverty rates and county types based on Office of Management and Budget (OMB) definitions.
Our analysis aimed to identify any correlation between poverty rates and heart disease prevalence. Despite an R-squared value of 0.0311, indicating a low correlation, this result does not negate the possibility of a relationship. The analysis was hindered by time constraints and data distribution challenges, as highlighted by the Shapiro-Wilk test indicating non-normality in heart attack data distribution.
- Data Representativeness: The focus on heart attacks might limit insights compared to analyzing broader health conditions.
- Data Measurement: Variations in poverty and health statistics definitions could affect outcomes. A per capita basis might be more appropriate for analyzing smaller counties.
- Geographic Scope: Limiting the study to Minnesota may affect the generalizability of findings due to regional variations and a smaller dataset.
- Statistical Expertise: Further analysis could benefit from collaboration with statistical experts, particularly in interpreting r-squared and p-values.
The initial analysis did not find a strong correlation between poverty rates and heart attack rates in Minnesota. However, the absence of statistically significant results does not imply a lack of relationship. Further research, possibly with contributions from public health professionals, could provide a more comprehensive understanding of the link between socio-economic status, environmental factors, and health outcomes.
- Importing data via the Census API may take an hour or more due to the volume of the dataset.
Eric Johnson, Paul Moses, Waynei Mebrahtu, Mitchell Lor, and Lucinda Hodgson