Health Disparities
Racial/ethnic and socioeconomic disparities in antibiotic prescription for lactation mastitis among commercially insured individuals Ashley Judge* Ashley Judge Christina Ludema
Lactation mastitis, an acute infection resulting in painful swelling of the breast, affects approximately 10% of breastfeeding people. The current US standard of treatment recommends the provision of antibiotic treatment 24-48 hours after the onset of symptoms. Few studies have examined demographic and socioeconomic characteristics associated with antibiotic prescription. Administrative datasets, like insurance claims, are an important data source for questions about provision of care, but use algorithms to generate demographic classifications. We examined the prevalence of antibiotic prescription after a lactation mastitis diagnosis by race/ethnicity and socioeconomic characteristics using Optum Clinformatics Data Mart healthcare claims dataset among individuals aged 15 to 49 diagnosed between January 1, 2007 to December 31, 2019. We used log-binomial regression to estimate risk ratios and 95% confidence intervals (CI) adjusted for age and stratified by race/ethnicity and socioeconomic characteristics. Seventy percent of race/ethnicity in Optum is imputed and prior research shows that Black individuals are especially likely to be misclassified using this algorithm. We perform a quantitative bias analysis using parameters from a prior validation study to estimate the impact of misclassification on our results comparing estimates of antibiotic prescription among Black and White participants. Among 9433 individuals diagnosed with mastitis, the median age was 31, 71.9% of individuals were White, 51.8% had some college education, 71.3% were a homeowner, 47.8% had an income over $100,000, 29.7% had a net worth over $500,000, and 57% were prescribed an antibiotic. Asian and Black individuals were 0.92 (53.1% vs 58.3%; [95% CI 0.86, 0.97]) and 0.89 (52.3% vs 58.3%; [0.82, 0.96]) less likely to have been prescribed an antibiotic than White individuals. Estimates further stratified by socioeconomic status revealed few statistically significant results. After accounting for misclassification, Black compared to White individuals were 0.82 less likely to have been prescribed an antibiotic.