Abstract Epidemiology of the U.S. COVID-19 outbreak focuses on individuals’ biology and behaviors, despite centrality of occupational environments in the viral spread. This demonstrates collusion between epidemiology and racial capitalism because it obscures structural influences, absolving industries of responsibility for worker safety. In an empirical example, we analyze economic implications of race-based metrics widely used in occupational epidemiology. In the U.S., White adults have better average lung function and worse hearing than Black adults. Both impaired lung function and hearing are criteria for Worker’s compensation, which is ultimately paid by industry. Compensation for respiratory injury is determined using a race-specific algorithm. For hearing, there is no race adjustment. Selective use of race-specific algorithms for workers’ compensation reduces industries’ liability for worker health, illustrating racial capitalism operating within public health. Widespread and unexamined belief in inherent physiological inferiority of Black Americans perpetuates systems that limit industry payouts for workplace injuries. We see a parallel in the epidemiology of COVID-19 disparities. We tell stories of industries implicated in the outbreak and review how they exemplify racial capitalism. We call on public health professionals to: critically evaluate who is served and neglected by data analysis; and center structural determinants of health in etiological evaluation.

Most outcomes related to influenza-like illness were significantly lower in intervention-school households than in control-school households. (ClinicalTrials.gov number, NCT00192218.)

Identifying the exposures or interventions that exacerbate or ameliorate racial health disparities is one of social epidemiology’s fundamental goals. Introducing an interaction term between race and an exposure into a statistical model is commonly utilized in the epidemiologic literature to assess racial health disparities and the potential viability of a targeted health intervention. However, researchers may attribute too much authority to the interaction term and inadvertently ignore other salient information regarding the health disparity. In this article, we highlight empirical examples from the literature demonstrating limitations of over-reliance on interaction terms in health disparities research; we further suggest approaches for moving beyond interaction terms when assessing these disparities. We promote a comprehensive framework of three guiding questions for disparity investigation, suggesting examination of the group-specific differences in 1) outcome prevalence, 2) exposure prevalence, and 3) effect size. Our framework allows for better assessment of meaningful differences in population health and the resulting implications for interventions, demonstrating that interaction terms alone do not provide sufficient means for determining how disparities arise. The widespread adoption of this more comprehensive approach has the potential to dramatically enhance understanding of the patterning of health and disease and the drivers of health disparities.

Decades of historical practices like housing discrimination in Detroit have lasting impacts on communities. Perhaps the most explicit example is the practice of redlining in the 1930s, whereby lenders outlined financially undesirable neighborhoods, populated by minority families, on maps and prevented residents from moving to better resourced neighborhoods. Awareness of historical housing discrimination may improve research assessing the impacts of current neighborhood characteristics on health. Using the Detroit Neighborhood Health Study (DNHS), we assessed the association between two-year changes in home foreclosure rates following the 2007-2008 Great Recession, and residents’ five-year self-rated health trajectories (2008-2013); and estimated the confounding bias introduced by ignoring historical redlining practices in the city. We used both ecological and multilevel models to make inference about person- and community-level processes. In a neighborhood-level linear regression adjusted for confounders (including percent redlined); a 10 percentage-point slower foreclosure rate recovery was associated with an increase in prevalence of poor self-rated health of 0.31 (95% CI: −0.02-0.64). At the individual level, it was associated with a within-person increase in probability of poor health of 0.45 (95% CI: 0.15-0.72). Removing redlining from the model biased the estimated effect upward to 0.38 (95% CI: 0.07-0.69) and 0.56 (95% CI: 0.21-0.84) in the neighborhood and individual-level models, respectively. Stratum-specific foreclosure recovery effects indicate stronger influence in neighborhoods with a greater proportion of residents identifying as white and a greater degree of historic redlining. These findings support theory that structural discrimination has lasting influences on current neighborhood health effects, and suggests that historical redlining specifically may increase vulnerability to contemporary neighborhood foreclosures. Community interventions should consider historical discrimination in conjunction with current place-based indicators to more equitably improve population health.

BackgroundThe pathway from evidence generation to consumption contains many steps which can lead to overstatement or misinformation. The proliferation of internet-based health news may encourage selection of media and academic research articles that overstate strength of causal inference. We investigated the state of causal inference in health research as it appears at the end of the pathway, at the point of social media consumption.MethodsWe screened the NewsWhip Insights database for the most shared media articles on Facebook and Twitter reporting about peer-reviewed academic studies associating an exposure with a health outcome in 2015, extracting the 50 most-shared academic articles and media articles covering them. We designed and utilized a review tool to systematically assess and summarize studies’ strength of causal inference, including generalizability, potential confounders, and methods used. These were then compared with the strength of causal language used to describe results in both academic and media articles. Two randomly assigned independent reviewers and one arbitrating reviewer from a pool of 21 reviewers assessed each article.ResultsWe accepted the most shared 64 media articles pertaining to 50 academic articles for review, representing 68% of Facebook and 45% of Twitter shares in 2015. Thirty-four percent of academic studies and 48% of media articles used language that reviewers considered too strong for their strength of causal inference. Seventy percent of academic studies were considered low or very low strength of inference, with only 6% considered high or very high strength of causal inference. The most severe issues with academic studies’ causal inference were reported to be omitted confounding variables and generalizability. Fifty-eight percent of media articles were found to have inaccurately reported the question, results, intervention, or population of the academic study.ConclusionsWe find a large disparity between the strength of language as presented to the research consumer and the underlying strength of causal inference among the studies most widely shared on social media. However, because this sample was designed to be representative of the articles selected and shared on social media, it is unlikely to be representative of all academic and media work. More research is needed to determine how academic institutions, media organizations, and social network sharing patterns impact causal inference and language as received by the research consumer.