key: cord-331871-colmj7uk
authors: Feehan, A. K.; Fort, D.; Garcia-Diaz, J.; Price-Haywood, E.; Velasco, C.; Sapp, E.; Pevey, D.; Seoane, L.
title: Point prevalence of SARS-CoV-2 and infection fatality rate in Orleans and Jefferson Parish, Louisiana, May 9-15, 2020
date: 2020-06-24
journal: nan
DOI: 10.1101/2020.06.23.20138321
sha: 
doc_id: 331871
cord_uid: colmj7uk

Using a novel recruitment method to reduce selection bias with paired molecular and antibody testing for SARS-CoV-2 infection, we determined point prevalence in a racially diverse municipality. Infections were highly variable by ZIP and differed by race. Overall census-weighted prevalence was 7.8% and the calculated infection fatality rate was 1.63%.

Using a novel recruitment method to reduce selection bias with paired molecular and antibody testing for SARS-CoV-2 infection, we determined point prevalence in a racially diverse municipality. Infections were highly variable by ZIP and differed by race. Overall censusweighted prevalence was 7.8% and the calculated infection fatality rate was 1.63%.

Seroprevalence studies around the world have estimated the spread of the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus, but none have been performed in New Orleans, Louisiana, USA, an early epicenter of the outbreak. Additionally, has been widely reported to disproportionately affect Black patients. While the absolute number of deaths have been reported by race, we do not know the infection fatality rate (IFR) which requires knowing how many people are at risk (e.g. infected). This study was designed to estimate SARS-CoV-2 infections in Orleans and Jefferson Parishes (O/JP) and the COVID-19 related IFR by race.

The protocol was approved by the Ochsner IRB and designed to enroll and test up to 3,000 subjects at 10 sites throughout O/JP between May 9 and May 15, 2020. To recruit a representative sample for this high-throughput method, a novel two-step system developed by Public Democracy (Arlington, VA) considered more than 50 characteristics, including social determinants of health and US Census population data, to establish a pool of potential participants reflective of the demographics of the Parishes, from which a randomized subset of 150,000 was selected.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 24, 2020. . https://doi.org/10.1101/2020.06.23.20138321 doi: medRxiv preprint Over 25,000 volunteers were recruited from this subset through dynamic, cross-device digital ads. This volunteer pool was stratified by the same attributes and then randomly issued a text message inviting subjects to private testing locations. Invitations were adjusted daily based on response rates to ensure we achieved our a priori goals for a representative sample.

Volunteers checked in with a QR code or phone number to discourage walk-ups. Ochsner Health did not turn uninvited people away but excluded these from analysis if they did not fit criteria. Family members of participants (234), or people who lived in ineligible ZIP codes (34) were excluded.

Six people withdrew consent. Digital ads, consent forms, and surveys were created in English, Spanish, and Vietnamese. Participants were offered free taxi service to and from the test sites.

Verbal consent was electronically documented, and subjects were asked a short list of questions followed by a blood draw and nasopharyngeal (NP) swab.

US Food and Drug Administration-Emergency Use Authorization approved tests were used. Real-time reverse transcriptase polymerase chain reaction (PCR) tests of NP swabs were performed on the Abbott m2000 RealTime system. Qualitative Immunoglobulin G (IgG) blood tests were performed on the ARCHITECT i2000SR. The IgG test meets criteria described by the CDC to yield high positive predictive value, which was validated by Ochsner Health laboratory and by others. (1,2) Study participants with either or both positive tests were assessed as having been infected with SARS-CoV-2. Census-weighted values were calculated for prevalence and seroprevalence. The positive-testing population included early-stage infections (PCR+ only) as well as people recovering (PCR+ and IgG+) and recovered (IgG+ only). Early-stage infections were excluded from IFR estimation as the outcomes of their infections would not yet be registered as official deaths. Therefore, weighted seroprevalence (anyone with an IgG+ result)

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 24, 2020. . https://doi.org/10.1101/2020.06.23.20138321 doi: medRxiv preprint was used to calculate "presumed recovered." IFR was calculated by dividing cumulative deaths by race, reported by the Louisiana Department of Health, by "presumed recovered" individuals.

Among the 2,640 analyzed, the sample was 63.5% female, 61.8% white, average age of 50.6 years, and average household size of 2.55 people. Among the 183 who tested positive, 49%

were Black. The raw prevalence of SARS-CoV-2 in the sample population is 6.9% (7.8%, census-weighted) with 2% positive for active viral shedding (PCR+, with or without IgG). By race, prevalence was highest (10.3%) in Black subjects followed by multiracial (9.4%), Asian (6.4%), and white (5.9%). Hispanic prevalence was 7.5%. 2018 population estimates are indicated in Table 1 and multiplied by weighted seroprevalence (percent IgG+) to generate the number of "presumed recovered." Reported deaths are divided by "presumed recovered" to calculate the IFR, which was 1.63% overall. The IFR was statistically similar for white (1.58%), Black (1.72%), and multiracial (1.40%), but Asian IFR was significantly lower (0.61%). Prevalence studies help to understand infection spread, especially when testing resources are limited. This representative, high minority enrollment study found an overall prevalence of PCR+ and/or IgG+ tests to be 7.8%. Hispanics and Asians had higher prevalence compared to whites, and this study confirms a recent report of over-representation of Black individuals with COVID-19 infection in the New Orleans area. (4) The overall IFR was 1.63%, which agrees with a recent estimate of 1.3% IFR (0.6%, 2.1%) for the US (3). The similar IFR among most racial . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 24, 2020. . https://doi.org/10.1101/2020.06.23.20138321 doi: medRxiv preprint groups indicates that viral spread at least partially explains the increased number of deaths among minorities.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 24, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 24, 2020. f Hispanic ethnicity is a separate analysis and numbers were not subtracted from race. Hispanic deaths were not reported by the state as of May 16, 2020.

.

It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 24, 2020. 

Interim Guidelines for COVID-19 Antibody Testing

The authors would like to especially thank the labs at the Ochsner Medical Center