Reviewed by Rebekah Moehring, MD, MPH, FSHEA, Duke University
Anyone working in infectious diseases epidemiology has learned long ago that infections are a social disease, with marginalized communities bearing a heavy burden of suffering while affluent areas have better access to care and resources, and therefore recovery. COVID-19 has painfully reminded us how disparities of race and socioeconomic status are a recurring theme in the American experience.
A brief report from the February JAMA Network Open described an assessment of Massachusetts and Boston Area public health testing databases. The study evaluated weekly measures of 1) testing intensity (# of tests per week per 100,000 population) and 2) epidemic intensity (weekly test positivity) among different communities in the state. Investigators defined a “testing gap” as the difference in rank (among communities) from measure 1 and 2. Each community was defined as in the American Community Survey. Then, the analysts modeled the association of the magnitude of testing gap with time (by week), social vulnerability index domains (which include socioeconomic status, minority race/ethnicity, and language), and university student population. Among >4.2 million tests evaluated, there was increased testing gap in less socioeconomically vulnerable localities, vacation regions, and areas near universities. Thus, more affluent areas and universities benefitted from more SARS-CoV-2 testing resources while those areas with social vulnerabilities did not.
Contrasting these findings, investigators from Kaiser Permanente (KP) in Northern California performed a large, retrospective cohort study of their health system’s population to evaluate the association of race/ethnicity on SARS-CoV-2 testing, infection incidence, and clinical outcomes including hospitalization and death. Bear in mind that the study population included over 3.4 million during the first phase of the pandemic (February to May 2020), and all study participants were enrollees in KP health insurance, which impacts the findings and generalizability. Testing prevalence overall was 2.6% and incidence was 105.9 per 100,000. White persons made up 42% of the population and had the lowest testing and infection incidence. While non-white persons’ testing rates were marginally higher, infection rates were significantly higher with geographic clusters of infections occurring in communities with higher proportions of non-white population. In adjusted analyses, race was the most important predictor of infection. In clinical outcomes, however, race was associated with increased hospitalization risk, but not mortality. Thus, among a large population with adequate access to care in an integrated health system, race was a predictor of who got sick, but clinical outcomes were not significantly different after adjustment for other predictors of poorer outcomes such as comorbid conditions and age. This finding supports prior literature suggesting clinical outcome disparities associated with race that persist despite the similar access to care are likely mediated by excess comorbidity.
Taken together, these two studies illustrate a difference in access to testing resources and infection incidence associated with race/ethnicity. Thus, access to testing resources and integrated care for communities affected by racial disparities is a well-documented target for public health efforts in this and future pandemic responses.
References:
