Alternative Methods for Grouping Race and Ethnicity to Monitor COVID-19 Outcomes and Vaccination Coverage

Area of Science:

• Public Health
• Epidemiology
• Health Equity

Background:

• Population-based analyses of COVID-19 data by race and ethnicity are crucial for identifying and monitoring health disparities.
• Accurate race and ethnicity data are essential for ensuring equitable access to protective measures like vaccines, but this information is often missing in reported COVID-19 data.
• Existing data collection standards (OMB's Statistical Policy Directive No. 15) include specific race and ethnicity categories.

Purpose of the Study:

• To compare the current method of grouping persons by race and ethnicity in COVID-19 data with two alternative methods.
• To assess how different methods of handling incomplete race and ethnicity data impact the understanding of COVID-19 disparities.
• To evaluate the effect of alternative methods on reported COVID-19 case counts, vaccination coverage, and incidence/coverage rates across racial and ethnic groups.

Main Methods:

• Compared the current CDC method (prioritizing ethnicity) with two alternative methods (Method A and Method B) using available COVID-19 case and vaccination data.
• Method A assumed non-Hispanic ethnicity when ethnicity data were missing.
• Method B allowed for non-mutually exclusive race and ethnicity categories, differing from the current method and Method A.

Main Results:

• Alternative methods (A and B) generally resulted in higher counts of COVID-19 cases and fully vaccinated persons across racial categories (American Indian/Alaska Native, Asian, Black, Native Hawaiian/Other Pacific Islander, White).
• Method B showed the largest relative increase in cases for American Indian/Alaska Native persons (58.5%) and vaccinated persons for Native Hawaiian/Other Pacific Islander persons (51.6%).
• The rate ratios for COVID-19 cases compared to White persons varied by method, with Asian persons consistently showing ratios <1 and other groups >1.

Conclusions:

• Alternative methods for analyzing incomplete race and ethnicity data in COVID-19 surveillance can lead to different conclusions regarding health disparities.
• The choice of method significantly impacts the observed disparities, particularly for American Indian/Alaska Native and Native Hawaiian/Other Pacific Islander populations.
• Further examination of potential biases and expert consultation are needed to develop optimal methods for tracking disparities with incomplete data.