BRFSS 2023 Flu Shot & Insurance Status Analysis

This project explores the relationship between insurance status and preventive health behavior, specifically flu shot uptake, using the 2023 Behavioral Risk Factor Surveillance System (BRFSS) dataset.

Data Source

• Source: CDC BRFSS 2023 Data
• File: LLCP2023.XPT (fixed-width .XPT format provided by the CDC)

Key Research Questions

• Does insurance coverage influence preventive health behaviors such as flu vaccination and routine checkups?
• Are there demographic patterns by sex, age, education, income, or race?

Objective

Analyze flu shot uptake based on:

• Insurance coverage
• Routine checkup history
• Age
• Income
• Education
• Race and sex

Tools & Packages

• tidyverse, haven, janitor, broom, caret, pscl, ggplot2

File Structure

• brfss2023_flu_insurance_analysis.R: Script with all steps from import to logistic regression
• brfss2023_flu_insurance_analysis.qmd: Renderable Quarto report
• .gitignore: Git configuration
• README.md: This is a documentation file

Key Steps

1. Data Import & Cleaning

   • BRFSS .XPT file loaded using haven::read_xpt()
   • Cleaned and filtered using dplyr, janitor, and tidyverse tools

2. Exploratory Analysis

   • Frequency tables
   • Cross-tabulations (e.g., flu shot by insurance)
   • Bar plots and histograms

3. Logistic Regression

   • Predictors: insurance, age, income, education, race, sex
   • Outcome: flu shot uptake
   • Evaluated with pseudo R², confusion matrix, accuracy

Key Findings

• Among those who received a flu shot, 61% were insured, while 39% were uninsured..
• Preventive care usage correlates with education and income.
• Logistic regression confirms insurance status as a statistically significant predictor of flu shot uptake (p < 0.001).

These results reinforce the role of access to insurance in supporting public health outcomes.

Libraries Used

library(tidyverse)
library(haven)
library(janitor)
library(broom)
library(caret)
library(pscl)