SARS-CoV-2 Period Seroprevalence and Related Factors, Hillsborough County, Florida, USA, October 2020–March 2021

Hillsborough County, Florida, USA, October 2020-March 2021

Anna R. Giuliano; Shari Pilon-Thomas; Michael J. Schell; Martha Abrahamsen; Jessica Y. Islam; Kimberly Isaacs-Soriano; Kayoko Kennedy; Christopher W. Dukes; Junmin Whiting; Julie Rathwell; Jonathan A. Hensel; Leslie N. Mangual; Ernst Schonbrunn; Melissa Bikowitz; Dylan Grassie; Yan Yang

Disclosures

Emerging Infectious Diseases. 2022;28(3):556-563.

Abstract and Introduction

Abstract

Estimating the actual extent of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic is challenging because virus test positivity data undercount the actual number and proportion of persons infected. SARS-CoV-2 seroprevalence is a marker of past SARS-CoV-2 infection regardless of presence or severity of symptoms and therefore is a robust biomarker of infection period prevalence. We estimated SARS-CoV-2 seroprevalence among residents of Hillsborough County, Florida, USA, to determine factors independently associated with SARS-CoV-2 antibody status overall and among asymptomatic antibody-positive persons. Among 867 participants, SARS-CoV-2 period prevalence (October 2020–March 2021) was 19.5% (asymptomatic seroprevalence was 8%). Seroprevalence was 2-fold higher than reported SARS-CoV-2 virus test positivity. Factors related to social distancing (e.g., essential worker status, not practicing social distancing, contact with a virus-positive person, and length of contact exposure time) were consistently associated with seroprevalence but did not differ by time since suspected or known infection (<6 months vs. ≥6 months).

Introduction

In late 2019, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in China, ultimately leading to a global pandemic.^[1] Since January 2020, the United States has observed a dramatic rise in the incidence of SARS-CoV-2 infection, for which no endogenous immunity exists,^[2] leading to >70.6 million cases of SARS-CoV-2 and ≈860,000 deaths in the United States.^[3] Although these data provide an estimate of the infection burden, challenges exist in estimating the actual extent of the pandemic. US public health data record the number of residents that test positive for SARS-CoV-2 RNA, rates of hospitalizations, and deaths from coronavirus disease (COVID-19) among those who undergo viral testing. Missing is the proportion of the population that was ever positive for SARS-CoV-2, including those who were symptomatic but did not undergo testing and those with no or mild symptoms, where the person did not recognize COVID-19 symptoms and therefore did not undergo testing.^[4–6] Complicating the estimate of SARS-CoV-2 prevalence is the fact that early in the pandemic in the United States, the availability of test reagents varied on any given day at any location and recommendations for testing eligibility changed. Test positivity data likely undercounted the actual number and proportion of persons who were infected with SARS-CoV-2.^[7,8] As such, the period prevalence of SARS-CoV-2 remains unknown for most communities.

Antibodies to SARS-CoV-2 begin to be detected 7 days after symptom onset^[9] and IgG antibodies are detectable within 2 weeks after onset of infection.^[10] SARS-CoV-2 seroprevalence is a marker of past SARS-CoV-2 infection regardless of presence or severity of symptoms and therefore is a robust biomarker of infection period prevalence.

As of June 1, 2021, Florida had the third-highest number of confirmed SARS-CoV-2 cases in the United States, 2,283,315 cases (10.6% of residents), resulting in 95,210 hospitalizations and 36,869 deaths.^[11] Hillsborough County (≈1.47 million residents), where the city of Tampa is located, is one of the most populous counties in Florida. As of June 1, 2021, a total of 142,013 test-confirmed SARS-CoV-2 cases had occurred among Hillsborough County residents (9.7% of the population). The goals of this study were to estimate SARS-CoV-2 seroprevalence among Hillsborough County residents and to determine the demographic and behavioral factors independently associated with SARS-CoV-2 antibody status overall and among asymptomatic antibody-positive persons.

1 2 3 4 5

Next Section

Table 1. Factors associated with severe acute respiratory syndrome coronavirus 2 antibody prevalence among residents, univariate and multivariable models, Hillsborough County, Florida, USA, October 2020–March 2021*
Characteristic	Unadjusted OR (95% CI)	Adjusted OR† (95% CI)
Age group, y
18–44	Referent	Referent
45–54	0.88 (0.56–1.37)	1.13 (0.64–1.98)
55–64	0.71 (0.46–1.12)	0.86 (0.48–1.55)
≥65	0.71 (0.43–1.18)	1.64 (0.80–3.35)
Sex
M	Referent	Referent
F	0.95 (0.67–1.35)	0.88 (0.57–1.38)
Marital status
Married or living together	Referent	Referent
Single, never married	0.91 (0.60–1.37)	1.05 (0.61–1.80)
Divorced, separated, or widowed	0.49 (0.28–0.86)	0.40 (0.20–0.77)
Living with chronic disease‡	0.88 (0.61–1.28)	0.56 (0.34–0.93)
Essential worker status
Not essential worker	Referent	Referent
Hospital, clinic, grocery store, public services	1.89 (1.26–2.84)	2.40 (1.42–4.07)
Financial services, banking, or other	0.91 (0.55–1.52)	0.70 (0.37–1.31)
Mean hours/week interacting with virus-positive contact	1.02 (1.01–1.02)	1.01 (1.001–1.01)
Relationship to virus-positive contact
No known contact with virus-positive person	Referent	Referent
Family member	8.79 (5.62–13.77)	4.62 (2.49–8.58)
Friend or other	5.52 (3.43–8.90)	4.22 (2.44–7.30)
Co-worker	2.89 (1.51–5.53)	2.04 (0.95–4.40)
Does not avoid groups of people	1.65 (1.12–2.43)	1.71 (1.06–2.76)
Ever had coronavirus disease symptoms	9.24 (6.33–13.48)	9.14 (5.93–14.08)

*Based on a 863-person sample size. The following variables were included in the backward selection modeling approach: age, sex, race, ethnicity, marital status, smoking status, living with chronic disease, lung problem, work environment during pandemic, practiced mask use since start of pandemic, practiced social distancing since start of pandemic, mean hours/week interacting with virus-positive contact, traveled out of state after February 2020, relationship to virus-positive contact, avoid groups of people, only going outside the home for essential trips, and ever had coronavirus disease symptoms. OR, odds ratio.
†Results from the full model, which included all variables, listed in Appendix Table (https://wwwnc.cdc.gov/EID/article/28/3/21-1495-App1.pdf).
‡Including cancer, heart disease, diabetes, autoimmune disease, kidney disease, liver disease, and being immunosuppressed.

Table 2. Factors associated with severe acute respiratory syndrome coronavirus 2 antibody prevalence among asymptomatic participants, univariate and multivariable models, Hillsborough County, Florida, USA, October 2020–March 2021*
Characteristic	Unadjusted OR (95% CI)	Adjusted OR (95% CI)†
Age group, y
18–44	Referent	Referent
45–54	0.95 (0.47–1.92)	1.04 (0.49–2.21)
55–64	0.48 (0.21–1.11)	0.57 (0.23–1.38)
≥65	0.43 (0.16–1.18)	0.69 (0.24–2.03)
Sex
M	Referent	Referent
F	1.11 (0.60–2.06)	0.91 (0.47–1.78)
Essential worker status
Not essential worker	Referent	Referent
Hospital, clinic, grocery store, public services	2.44 (1.28–4.65)	2.28 (1.13–4.60)
Financial services, banking, or other	0.76 (0.28–2.02)	0.55 (0.19–1.58)
Mean hours/week interacting with virus-positive contact	1.01 (1.01–1.02)	1.01 (1.00–1.02)
Relationship to virus-positive contact
No known contact with virus-positive person	Referent	Referent
Family member	4.23 (1.95–9.16)	2.67 (0.94–7.59)
Co-worker	2.32 (0.83 −6.52)	1.89 (0.62–5.73)
Friend or other	4.40 (2.10–9.21)	3.72 (1.71–8.11)
Does not avoid groups of people	3.03 (1.66–5.56)	2.90 (1.53–5.50)

*Based on a 602-person sample size. The following variables were included in the backward selection modeling approach: age, sex, race, ethnicity, marital status, smoking status, living with chronic disease, lung problem, work environment during pandemic, practiced mask use since start of pandemic, practiced social distancing since start of pandemic, mean hours/week interacting with virus-positive contact, traveled out of state after February 2020, relationship to virus-positive contact, avoid groups of people, only going outside the home for essential trips, and ever had coronavirus disease symptoms. OR, odds ratio.
†Results from the full model, which included all variables, listed in Appendix Table (https://wwwnc.cdc.gov/EID/article/28/3/21-1495-App1.pdf).