A risk index for COVID-19 severity is associated with COVID-19 mortality in New York City

New York City was once considered the epicenter of the coronavirus disease (COVID-19) epidemic in the United States. From the first reported case on March 1, 2020, to April 2020, nearly 600 people were dying each day, new daily cases numbered in the thousands, and new hospitalizations were roughly 1500 per day [1]. The mortality count peaked on April 7, where there were 598 reported deaths, 6042 cases and 1578 people hospitalized. Reflecting on this time has highlighted racial and socioeconomic disparities in COVID-19 testing patterns that are indicative of gaps in initial COVID-19 response policy [2‐5]. Analysis of COVID-19 case identification and mortality rates in New York City has also underscored variations in the amount of COVID-19 testing performed over time, delays in receiving testing results, and lags in the reporting of death data [6, 7]. To provide a more complete picture of mortality data, the New York City Department of Health and Mental Hygiene COVID-19 Response Team issued a publication on the topic and reported that over 5000 excess deaths occurred from the COVID-19 pandemic from March 11th through May 2nd [8]. Within NYC, cumulative mortality rates are highest in Hispanic/Latino and African American/Black populations, as well as older age groups [1].

Clinical studies across the US and NYC have shown that people with underlying health conditions and risk factors, including diabetes, chronic lung disease, and cardiovascular disease, are at higher risk for severe outcomes and mortality from COVID-19 [1, 9‐11]. However, little is known about how the presence of these chronic risk factors correlates with mortality at the population level [12‐14]. Such information is key to developing effective public health interventions to mitigate COVID-19 spread and adverse outcomes, and population studies can guide policy at the community-level. In addition, population level studies are enriched with minorities and the whole spectrum of socioeconomic status, thus reflecting reality more closely. A study of US data looked at the prevalence of risk factors for COVID-19 severity according to CDC guidelines, and asserted that people living in low-income households and those who are Black or American Indian are more likely to have chronic conditions that place them in a higher risk category for COVID-19 complications [12]. The study however did not correlate this high-risk profile with mortality data. Within NYC, the distribution of behavioral risk factors and chronic conditions is highly variable from neighborhood to neighborhood; higher obesity rates in Brooklyn and the Bronx have been posited as being partially responsible for the increased COVID-19 mortality in these areas [15]. A separate study reported older age and severe obesity as risk factors for COVID-19 mortality and worse in-hospital outcomes in a Bronx cohort [16]. However, the research published so far only partially tackles a more complex reality when studying COVID-19. Minorities and underserved populations have higher prevalence of chronic conditions that put them in danger of a more serious infection; at the same time, they also have higher risk of exposure to the virus, as they frequently live in neighborhoods with higher population density and proportions of essential workers, with fewer opportunities to exercise social distancing [17‐19]. Here we study risk factors associated with COVD-19 mortality at the neighborhood level in NYC. We hypothesized that areas with the highest prevalence of risk factors, minority residents and lower SES experienced the highest mortality in NYC.

This is the first analysis to incorporate CDC-defined risk factors for severe COVID-19 illness, NYC mortality data, racial and ethnic composition, and SES to predict COVID-19 death rates across NYC at the neighborhood level. Here we identify that areas with fewer white residents, more Hispanic residents, older residents, and with higher prevalence of a greater number of risk factors had increased COVID-19 mortality. These findings expand on existing literature that has partially addressed the potential risk of severe COVID-19 illness deriving from the complex intersection of chronic risk factors, SES and race [12] and can be useful in plans to address and reduce excess mortality in NYC⁵. In the present analysis, the COVID-19 death rate was highest in areas with the highest prevalance of risk factors, including hypertension, obesity, diabetes, asthma, and older age. Within NYC, these areas of greatest risk correspond to the Bronx, southeastern and southern Brooklyn, and parts of southern Queens; areas that are also of lower socioeconomic status with larger resident minority populations. Past neighboorhood analysis of NYC communities has reported that these areas have limited capacity to social distance, as measured by NYC subway data, and therefore have higher infection risk [29]. It is possible that these areas contain a large proportion of essential workers, such as healthcare, transportation and public-service workers, who are unable to work from home. It has been cited that social distancting, including working from home, is a privilege [18], and this is not a reality in communities characterized by higher proportions of comorbidites, risk factors for COVID-19, and adverse social determinants of health, including potential overcrowding and food insecurity. Further disparities in SES and healthcare access would only exacerabate adverse circumstances in these communities and contribute to increased risk of infection and mortality.

It has been noted that blanket public health recommendations in regards to social distancing for COVID-19 do not take local contexts and populations into account [17], and may be applied less effectively in socioeconomically disadvatanged and minorty communities due to several factors, including overcrowded households and higher proportions of essential workers. Instead, understanding local contexts is central in creating effective public health solutions. One avenue to begin to rectify disparities is to perform increased testing in these communities [2], and more specifically in sub-groups at higher risk of serious COVID-19 complications. However, there remains a gap in appropriate testing and attention given to these vulnerable communites [2]. Here, we emphasize that many of the same communities that are more suceptible to increased COVID-19 spread also face higher risks of severe illness and mortality if infected. The data point to the need for more aggressive actions in preventing COVID-19 spread and detecting early infections in specific areas of NYC.

One limitation of this analysis is that the data used are aggregate ZCTA-level data, which limits our ability to draw individual-level conclusions. For instance, we cannot comment on the interaction between being non-white and having chronic risk factors on the likelihood of mortality, information that could help identify the most vulnerable populations. This study was ecological in nature and there are more individual level factors that may influence COVID-19 mortality that could not be taken into account here, although an accepted and comprehensive risk factor measure from the CDC was utilized. Another limitation is the cross-sectional nature of the analysis. Future studies should investigate these relationships with mortality using a larger study period, representing a larger number of COVID-19 related deaths, to validate these findings.

This analysis identifies that there are behavioral and racial/ethnic disparities in COVID-19 mortality, and highlights the most vulnerable areas in NYC that require increased focus from COVID-19 response policies. This analysis represents the first to our knowledge to investigate risk factors for COVID-19 mortality at the population-level in New York City, and supports the need for repeated and widespread testing in these communities with many risk factors for severe illness and mortality from COVID-19. Additionally, this analysis shines light on vulnerable communities that require equitable resources to rebound from COVID-19 related hardships and mortality in these communities. Future analyses could explore how the presence of comorbidities influences the immune response following COVID-19 infection, as this could ultimately be a major determinant of mortality outcomes [30]. Future vaccination efforts should understand the vulnerability to COVID-19 and mortality in NYC communities and direct attention to these areas accordingly. A successful public health response to future COVID-19 outbreaks will need to provide more attention to communities at the greatest risk in order to minimize the threat of serious illness and mortality in NYC.