key: cord-0888621-2d1pdcom
authors: Mahajan, Shiwani; Caraballo, César; Li, Shu-Xia; Dong, Yike; Chen, Lian; Huston, Sara K.; Srinivasan, Rajesh; Redlich, Carrie A.; Ko, Albert I.; Faust, Jeremy S.; Forman, Howard P.; Krumholz, Harlan M.
title: SARS-CoV-2 Infection Hospitalization Rate and Infection Fatality Rate Among the Non-Congregant Population in Connecticut
date: 2021-02-20
journal: Am J Med
DOI: 10.1016/j.amjmed.2021.01.020
sha: e0f3e05707305fb297aff8ae61f9ed70050a00b1
doc_id: 888621
cord_uid: 2d1pdcom

BACKGROUND: Infection fatality rate and infection hospitalization rate, defined as the proportion of deaths and hospitalizations, respectively, of the total infected individuals, can estimate the actual toll of COVID-19 on a community as the denominator is ideally based on a representative sample of a population, which captures the full spectrum of illness, including asymptomatic and untested individuals. OBJECTIVE: To determine the COVID-19 infection hospitalization rate and infection fatality rate among the non-congregate population in Connecticut between March 1 and June 1, 2020. METHODS: The infection hospitalization rate and infection fatality rate were calculated for adults residing in non-congregate settings in Connecticut before June 2020. Individuals with SARS-CoV-2 antibodies were estimated using the seroprevalence estimates from the recently conducted Post-Infection Prevalence study. Information on total hospitalizations and deaths was obtained from the Connecticut Hospital Association and the Connecticut Department of Public Health. RESULTS: Before June 1, 2020, nearly 113,515 (90% CI 56,758–170,273) individuals were estimated to have SARS-CoV-2 antibodies and there were 7792 hospitalizations and 1079 deaths among the non-congregate population. The overall COVID-19 infection hospitalization rate and infection fatality rate was 6.86% (90% CI, 4.58%–13.72%) and 0.95% (90% CI, 0.63%–1.90%) and there was variation in these rate estimates across subgroups; older individuals, men, non-Hispanic Black individuals, and those belonging to 2 of the counties had a higher burden of adverse outcomes, though the differences between most subgroups were not statistically significant. CONCLUSIONS: Using representative seroprevalence estimates, the overall COVID-19 infection hospitalization rate and infection fatality rate were estimated to be 6.86% and 0.95%, respectively, among community residents in Connecticut.

Post-Infection Prevalence study. Information on total hospitalizations and deaths was obtained from the Connecticut Hospital Association and the Connecticut Department of Public Health.

Results: Before June 1, 2020, nearly 113,515 (90% CI 56,758-170,273) individuals were estimated to have SARS-CoV-2 antibodies and there were 7792 hospitalizations and 1079 deaths among the non-congregate population. The overall COVID-19 infection hospitalization rate and infection fatality rate was 6.86% (90% CI, 4.58%-13.72%) and 0.95% (90% CI, 0.63%-1.90%) and there was variation in these rate estimates across subgroups; older individuals, men, non-Hispanic Black individuals, and those belonging to 2 of the counties had a higher burden of adverse outcomes, though the differences between most subgroups were not statistically significant.

Conclusions: Using representative seroprevalence estimates, the overall COVID-19 infection hospitalization rate and infection fatality rate were estimated to be 6.86% and 0.95%, respectively, among community residents in Connecticut.

Key words: SARS-CoV-2, covid-19, seroprevalence, infection hospitalization rate, infection fatality rate

 Using seroprevalence estimates, we found that through June 2020, Connecticut's COVID-19 infection hospitalization rate and infection fatality rate were 6.86% (90% CI, 4.58%-13.72%) and 0.95% (90% CI, 0.63%-1.90%) among the non-congregate population, and these estimates varied across subgroups.

 Representative seroprevalence studies provide important information regarding infections in a community and can provide robust estimates of the infection hospitalization and fatality rate, when combined with hospitalization and death data.

Accurate estimation of the hospitalization and fatality rate is important to guide public health strategies during infectious disease outbreaks. Although the case fatality rate, defined as the proportion of deaths of the confirmed cases, is a commonly used metric, it will be biased based on the availability of testing, especially early in the outbreak. 1 Moreover, since COVID-19 symptoms range widely, mild or asymptomatic infections may be untested. Thus, the number of infections confirmed by testing will underestimate the total infections, inflating the estimated fatality rate.

Infection fatality rate, defined as the proportion of deaths of the total infected individuals, can estimate the actual toll of the disease as the denominator is ideally based on a representative sample of a population, which captures the full spectrum of illness, including asymptomatic and untested individuals. For hospitalizations, the infection hospitalization rate is a comparable measure. Accordingly, with support from the US Centers for Disease Control and Prevention from the Coronavirus Aid, Relief, and Economic Security (CARES) Act, 2 we conducted a statewide SARS-CoV-2 seroprevalence study-the Post-Infection Prevalence study (PIP)-in Connecticut 3 and assessed the SARS-CoV-2 infection hospitalization and fatality rates using the statewide all-payer databases and statewide mortality data.

Based on the PIP study, 3 the seroprevalence of SARS-CoV-2 antibodies was 4.0% (90% confidence interval [CI] 2.0%-6.0%) among a representative population of adults residing in non-congregate settings (i.e., excluding adults residing in a long-term care facility, assisted living facility, nursing home, and a prison or jail) in Connecticut before June 2020. We used this The infection hospitalization rate and the infection fatality rate were calculated as the number of individuals who were hospitalized and died, respectively, due to COVID-19 divided by the total estimated number of individuals who had COVID-19 using the seroprevalence estimates (details in eMethods 2). We estimated the infection hospitalization and fatality rates for the overall population and by age, sex, race/ethnicity, and region subgroups. The margin of error for our estimates was calculated at the 90% confidence level in accordance with the design of the PIP study; however, estimates at 95% CI have also been provided. Due to sample size limitations, the upper end of the CI was non-estimable (NE) when stratifying by some sociodemographic characteristics. Statistical analyses were performed using R version 4.0.2.

This study was exempted from review by the Institutional Review Board at Yale University because of the public health surveillance activity exclusion.

Of the 2.8 million individuals residing in the non-congregate settings in Connecticut, 113,515 (90% CI 56,758-170,273) had SARS-CoV-2 antibodies ( Table 1) The overall COVID-19 infection hospitalization rate and infection fatality rate was 6.86% (90% CI, 4.58%-13.72%) and 0.95% (90% CI, 0.63%-1.90%) among the non-congregate population. There was variation in infection hospitalization rate and infection fatality rate estimates across subgroups and older individuals, men, non-Hispanic Black individuals, and those belonging to New Haven and Litchfield counties had a higher burden of adverse outcomes, though the differences between most subgroups were not statistically significant (Table 2) .

Population estimates and the estimates for the infection hospitalization and fatality rates at the 95% CI are presented in eTable 1 and eTable 2.

Using seroprevalence estimates, we found that, through June 2020, Connecticut's COVID-19 infection hospitalization rate and infection fatality rate were 6.86% and 0.95%, respectively, and these estimates varied across subgroups. Our estimates are distinctive because they reflect people living in the community and are based on a methodology that sought to obtain a representative estimate for the denominator and brought together multiple streams of data.

There has been continued controversy about the infection fatality rate and the literature is replete with widely varied estimates. [4] [5] [6] [7] [8] However, most studies do not have a representative sample or separate out special populations, such as those in nursing homes. 5, 9 Moreover, infection fatality rate is not an inherent characteristic of the disease, but rather a confluence of the pathogen virulence, sociodemographic and clinical characteristics of the population, health care availability and quality, therapeutic availability, and accurate counting and reporting of COVID-19-related deaths. As such, an overall infection fatality rate may not be very informative given the heterogeneity across regions.

The COVID-19 infection hospitalization rate is not well described and most studies report the case hospitalization rate, which also varies widely in the literature. [10] [11] [12] The Centers for Disease Control and Prevention estimated a US COVID-19 case hospitalization rate of 14.0% for infections before June 2020. 12 As expected, this case hospitalization rate estimate was higher than Connecticut's estimated infection hospitalization rate in our study (6.9%), as infection hospitalization rate includes the total estimated infections rather than detected positive cases only. Moreover, our estimate excluded individuals from congregate settings, which had a higher burden of adverse outcomes.

Our subgroup findings are notable, even though the differences between subgroups were not statistically significant. Prior studies have noted that age and sex are associated with disease severity, however, they have been hampered by including nursing home residents and biased by testing patterns. We had the opportunity to identify hospitalizations and deaths among the noncongregate population in Connecticut and show that, even in the community, these associations remain. Prior studies have shown that Black individuals have had disproportionately higher infection rates, even as some studies indicate that hospital mortality does not vary by race/ethnicity. 13 Our findings highlight that the burden of COVID-19 among Black subpopulations is not just about infection rates but also worse outcomes. The lower infection hospitalization rate for Hispanic individuals was in accordance with previously reported low hospital admission rates among Hispanic individuals testing positive for SARS-CoV-2 in Baltimore-Washington, DC 14 and may be associated with the lower insurance rates among the Hispanic subpopulation in Connecticut. 15 Limitations of our study include potential underestimation of COVID-19-related hospitalizations due to limited testing availability; underestimation of total infections due to the decrease in antibody concentration over time or poor sensitivity of serology tests; lack of power to detect statistical differences between subgroups due to the small sample size; and that our findings may not be generalizable to other regions or across time. Nevertheless, this study shows that representative seroprevalence studies provide important information regarding infections in a community and can provide robust estimates of the infection hospitalization and fatality rates, when combined with hospitalization and death data.

In conclusion, using representative seroprevalence estimates, we estimate an infection hospitalization rate and infection fatality rate of 6.86% (90% CI 4.58%-13.72%) and 0.95% (90% CI 0.63%-1.90%), respectively, for COVID-19 infections through June 1, 2020, among the non-congregate population in Connecticut.

Association for their support. COVID-19 Note: the above diagnosis codes can be in any position, i.e. they do not have to be the principal diagnosis code.

 Assessment of the number of hospitalizations and deaths among the non-congregate populations-residents were identified as living in a congregate facility if they resided in a long-term care facility, assisted living facility, nursing home, and a prison or jail. Information on COVID-19-related deaths among the residents from non-congregant settings was provided directly by the Connecticut Department of Public Health. For hospitalizations among the non-congregate population, we excluded patients admitted from skilled nursing facilities using the admission source fields from the hospitalization data obtained from the Connecticut Hospital Association.

 Calculation of IHR and IFR-IHR and IFR were calculated as the number of individuals who were hospitalized and died, respectively, due to COVID-19 divided by the total estimated number of individuals who had COVID-19 using the seroprevalence estimates. The lower (upper) bound of the IHR/IFR was calculated by dividing the number of hospitalizations/deaths by the lower (upper) bound of the denominator estimation. eTable 1. Estimated number of individuals with SARS-CoV-2-specifc antibodies at 95% confidence interval among the non-congregate population in Connecticut, between March 1 and June 1, 2020, by sociodemographic characteristics. 

Covid-19 fatality is likely overestimated

HHS announces CARES Act funding distribution to States and Localities in support of COVID-19 response

Seroprevalence of SARS-CoV-2-specific IgG antibodies among adults living in Connecticut: Post-Infection Prevalence (PIP)

Infection fatality ratios for COVID-19 among noninstitutionalized persons 12 and older: results of a random-sample prevalence study

Estimation of SARS-CoV-2 infection fatality rate by real-time antibody screening of blood donors

Infection fatality risk for SARS-CoV-2 in community dwelling population of Spain: nationwide seroepidemiological study

The infection fatality rate of COVID-19 inferred from seroprevalence data. medRxiv

Estimating the infection and case fatality ratio for coronavirus disease (COVID-19) using age-adjusted data from the outbreak on the Diamond Princess cruise ship

Estimating the infection-fatality risk of SARS-CoV-2 in New York City during the spring 2020 pandemic wave: a model-based analysis

Hospitalization and mortality among Black patients and White patients with Covid-19

Factors associated with hospitalization and disease severity in a racially and ethnically diverse population of COVID-19 patients. medRxiv

Coronavirus disease 2019 case surveillance-United States

Clinical characteristics and outcomes for 7,995 patients with SARS-CoV-2 infection

SARS-CoV-2 positivity rate for Latinos in the Baltimore-Washington, DC region

Health disparities in Connecticut: causes, effects, and what we can do

Connecticut Health Foundation

79%)

Abbreviations: CI, confidence interval

non-estimable *Total COVID-19-related deaths includes both confirmed and probable COVID-19 deaths