Risk factors and titers of COVID-19 infection in a longitudinal statewide seroepidemiology cohort

BACKGROUND

Virginia is a large state in the USA, yet it remains unclear what percentage of the population has had natural COVID-19 infection and whether risk factors for infection have changed over time.

METHODS

Using a longitudinal cohort, from December 2021-July 2022 we performed follow up serology and a questionnaire on 784 individuals from across Virginia who had previously participated in a statewide COVID-19 seroepidemiology study in 2020. Children were also invited to participate and an additional 62 children also completed the study. Serology was performed using Roche nucleocapsid and spike serological assays.

RESULTS

The majority of participants were white (78.6%), over 50 years old (60.9%), and reported having received COVID-19 vaccine (93.4%). 28.6% had evidence of prior COVID-19 infection (nucleocapsid positive). Reweighted by region, age, and sex to match the Virginia census data, the seroprevalence of nucleocapsid antibodies was estimated to be 30.6% (95% CI: 24.7, 36.6). We estimated that 25-53% of COVID-19 infections were asymptomatic. Infection rates were lower in individuals > 60 years old and were higher in Blacks and Hispanics. Infection rates were also higher in those without health insurance, in those with greater numbers of household children, and in those that reported a close contact or having undergone quarantine for COVID-19. Participants from Southwest Virginia had lower seropositivity (16.2%, 95% CI 6.5, 26.0) than other geographic regions. Boosted vaccinees had lower infection rates than non-boosted vaccinees. Frequenting indoor bars was a risk factor for infection, while frequently wearing an N95 mask was protective, though the estimates of association were imprecise. Infection rates were higher in children than adults (56.5% vs. 28.6%). Infection in the parent was a risk factor for child infection. Spike antibody levels declined with time since last vaccination, particularly in those that were vaccinated but not previously infected. Neutralizing antibody positivity was high (97-99%) for wild type, alpha, beta, gamma, delta, and omicron variants. Neutralizing antibody levels were higher in the follow-up survey compared to the first survey in 2020 and among individuals with evidence of natural infection compared to those without.

CONCLUSIONS

In this longitudinal statewide cohort we observed a lower-than-expected COVID-19 infection rate as of August 2022. Boosted vaccinees had lower infection rates. Children had higher infection rates and infections tracked within households. Previously identified demographic risk factors for infection tended to persist. Even after the omicron peak, a large number of Virginians remain uninfected with COVID-19, underscoring the need for ongoing vaccination strategies.

Assessment of Seroprevalence of SARS-CoV-2 and Risk Factors Associated With COVID-19 Infection Among Outpatients in Virginia

Importance

Data from seroepidemiologic surveys measuring severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) exposure in diverse communities and ascertaining risk factors associated with infection are important to guide future prevention strategies.

Objective

To assess the prevalence of previous SARS-CoV-2 infection across Virginia and the risk factors associated with infection after the first wave of coronavirus disease 2019 (COVID-19) infections in the US.

Design, Setting, and Participants

In this statewide cross-sectional surveillance study, 4675 adult outpatients presenting for health care not associated with COVID-19 in Virginia between June 1 and August 14, 2020, were recruited to participate in a questionnaire and receive venipuncture to assess SARS-CoV-2 serology. Eligibility was stratified to meet age, race, and ethnicity quotas that matched regional demographic profiles.

Main Outcomes and Measures

The main outcome was SARS-CoV-2 seropositivity, as measured by the Abbott SARS-CoV-2 immunoglobulin G assay.

Results

Among 4675 adult outpatients (mean [SD] age, 48.8 [16.9] years; 3119 women [66.7%]; 3098 White [66.3%] and 4279 non-Hispanic [91.5%] individuals) presenting for non-COVID-19-associated health care across Virginia, the weighted seroprevalence was 2.4% (95% CI, 1.8%-3.1%) and ranged from 0% to 20% by zip code. Seroprevalence was notably higher among participants who were Hispanic (10.2%; 95% CI, 6.1%-14.3%), residing in the northern region (4.4%; 95% CI, 2.8%-6.1%), aged 40 to 49 years (4.4%; 95% CI, 1.8%-7.1%), and uninsured (5.9%; 95% CI, 1.5%-10.3%). Higher seroprevalence was associated with Hispanic ethnicity (adjusted odds ratio [aOR], 3.56; 95% CI, 1.76-7.21), residence in a multifamily unit (aOR, 2.55; 95% CI, 1.25-5.22), and contact with an individual with confirmed COVID-19 infection (aOR, 4.33; 95% CI, 1.77-10.58). The sensitivity of serology results was 94% (95% CI, 70%-100%) among those who reported receiving a previous polymerase chain reaction test for COVID-19 infection. Among 101 participants with seropositive results, 67 individuals (66.3%) were estimated to have asymptomatic infection. These data suggested a total estimated COVID-19 burden that was 2.8-fold higher than that ascertained by PCR-positive case counts.

Conclusions and Relevance

This large statewide serologic study estimated that 2.4% of adults in Virginia had exposure to SARS-CoV-2, which was 2.8-fold higher than confirmed case counts. Hispanic ethnicity, residence in a multifamily unit, and contact with an individual with confirmed COVID-19 infection were significant risk factors associated with exposure. Most infections were asymptomatic. As of August 2020, the population in Virginia remained largely immunologically naive to the virus.