key: cord-0846222-xoc566qj
authors: Couture, Alexia; Lyons, B Casey; Mehrotra, Megha L; Sosa, Lynn; Ezike, Ngozi; Ahmed, Farah S; Brown, Catherine M; Yendell, Stephanie; Azzam, Ihsan A; Katić, Božena J; Cope, Anna; Dickerson, Kristen; Stone, Jolianne; Traxler, L Brannon; Dunn, John R; Davis, Lora B; Reed, Carrie; Clarke, Kristie E N; Flannery, Brendan; Charles, Myrna D
title: Severe Acute Respiratory Syndrome Coronavirus 2 Seroprevalence and Reported Coronavirus Disease 2019 Cases in US Children, August 2020–May 2021
date: 2022-01-30
journal: Open Forum Infect Dis
DOI: 10.1093/ofid/ofac044
sha: 06dba5439e875a3ad7e2603ac79f4dfc3f2fbc7e
doc_id: 846222
cord_uid: xoc566qj

BACKGROUND: Case-based surveillance of pediatric coronavirus disease 2019 (COVID-19) cases underestimates the prevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections among children and adolescents. Our objectives were to estimate monthly SARS-CoV-2 antibody seroprevalence and calculate ratios of SARS-CoV-2 infections to reported COVID-19 cases among children and adolescents in 8 US states. METHODS: Using data from the Nationwide Commercial Laboratory Seroprevalence Survey, we estimated monthly SARS-CoV-2 antibody seroprevalence among children aged 0–17 years from August 2020 through May 2021. We calculated and compared cumulative incidence of SARS-CoV-2 infection extrapolated from population-standardized seroprevalence of antibodies to SARS-CoV-2, cumulative COVID-19 case reports since March 2020, and infection-to-case ratios among persons of all ages and children aged 0–17 years for each state. RESULTS: Of 41 583 residual serum specimens tested, children aged 0–4, 5–11, and 12–17 years accounted for 1619 (3.9%), 10 507 (25.3%), and 29 457 (70.8%), respectively. Median SARS-CoV-2 antibody seroprevalence among children increased from 8% (range, 6%–20%) in August 2020 to 37% (range, 26%–44%) in May 2021. Estimated ratios of SARS-CoV-2 infections to reported COVID-19 cases in May 2021 ranged by state from 4.7–8.9 among children and adolescents to 2.2–3.9 for all ages combined. CONCLUSIONS: Through May 2021 in selected states, the majority of children with serum specimens included in serosurveys did not have evidence of prior SARS-CoV-2 infection. Case-based surveillance underestimated the number of children infected with SARS-CoV-2 more than among all ages. Continued monitoring of pediatric SARS-CoV-2 antibody seroprevalence should inform prevention and vaccination strategies.

As of January 2022, children and adolescents aged 0-17 years accounted for 16% of >45 million individual confirmed and probable cases of coronavirus disease 2019 (COVID-19) reported since January 2020 by jurisdictional health departments to the United States (US) Centers for Disease Control and Prevention (CDC) [1] . Although children and adolescents experienced lower risk of severe COVID-19 and COVID-19related mortality than older adults, children remain at risk of severe complications of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection including hospitalization, multisystem inflammatory syndrome in children, post-COVID-19 conditions, and death [2] [3] [4] [5] [6] . Furthermore, as a result of COVID-19 vaccination of older persons, children and adolescents account for an increasing proportion of reported COVID-19 cases and SARS-CoV-2 testing [7] . Comirnaty (COVID-19 messenger RNA [mRNA] vaccine), produced by Pfizer-BioNTech, was granted Emergency Use Authorization (EUA) by the US Food and Drug Administration (FDA) and recommended for persons aged ≥16 years by the Advisory Committee on Immunization Practices (ACIP) in December 2020 [8] . Following early vaccination of priority groups and high-risk individuals [9] , uptake of COVID-19 vaccines among adolescents aged 16-17 years increased during March-April 2021 once widespread community vaccination became available [10] . After the Pfizer-BioNTech COVID-19 vaccine was approved for adolescents aged 12-15 years in May 2021 and children aged 5-11 years in November 2021, ACIP recommended vaccination for these age groups [11, 12] . As of January 2022, >16 million 12-to 17-year-olds and 7 million 5to 11-year-olds had received at least 1 dose of COVID-19 vaccine, reaching an estimated 64% of 16-to 17-year-olds and 26% of 12-to 15- year-olds in the US [10] .

Analyses of case surveillance data from health departments have used different age groups in the case definitions for estimating COVID-19 incidence among children and adolescents [7] . Estimates of SARS-CoV-2 infections among children and adolescents based on case-based surveillance data may be less reliable than those for older age groups because children may be less likely to present with symptoms and be tested for SARS-CoV-2 infection. Since July 2020, CDC has partnered with commercial laboratories to conduct repeated cross-sectional seroprevalence studies using residual clinical serum specimens to test for the presence of SARS-CoV-2 antibodies [13, 14] . In unvaccinated individuals, detection of specific antibodies to SARS-CoV-2 proteins in serum indicates previous infection [15] . Laboratory assays that detect nucleocapsid (N) antibodies identify individuals with a history of SARS-CoV-2 infection regardless of immunization status because currently approved or authorized vaccines in the US elicit antibodies to SARS-CoV-2 spike protein [14] .

While commercial laboratory serosurveillance has provided seroprevalence estimates across the age spectrum, including persons aged 0-17 years [13] , differences in seroprevalence among pediatric age groups have not been evaluated. This study looked at data from early in the pandemic through May 2021 to estimate numbers of children and adolescents who had been infected with SARS-CoV-2 prior to widespread COVID-19 vaccination among children aged 5-15 years. To investigate trends in estimated SARS-CoV-2 infections among children and adolescents, we analyzed commercial laboratory serosurvey data from 8 US states and compared cumulative incidence of SARS-CoV-2 infection inferred from seroprevalence with total confirmed and probable COVID-19 cases reported during the same period among pediatric age groups. The objectives of this analysis were to estimate pediatric SARS-CoV-2 antibody seroprevalence monthly from August 2020 through May 2021 in selected states and to estimate ratios of SARS-CoV-2 infections to COVID-19 cases among children and adolescents reported through case-based surveillance.

Seroprevalence Survey have been previously described [13, 14] . In brief, CDC partnered with 3 commercial laboratories to test for the presence of SARS-CoV-2 antibodies in convenience samples of residual sera submitted for routine clinical testing, including but not limited to lipid testing, sodium level testing, and hormone testing. Serum specimens submitted for COVID-19 or SARS-CoV-2 antibody testing were excluded. Commercial laboratories assigned patient identifiers to de-duplicate serum specimens within each serosurvey round; de-identified laboratory results were provided to CDC with patient demographics (age in years, sex, and residential zip code). Patient race and ethnicity were not provided.

Commercial laboratories used 1 of 3 immunoassays authorized under EUA: Elecsys Anti-SARS-CoV-2 assay (Roche, Indianapolis, Indiana) and ARCHITECT SARS-CoV-2 immunoglobulin G (IgG) assay (Abbott, Chicago, Illinois), both targeting the viral N protein; and VITROS Anti-SARS-CoV-2 IgG Assay (Ortho-Clinical Diagnostics, Raritan, New Jersey) targeting the viral spike protein. Proportions of sera tested by each immunoassay varied by state and calendar month ( Figure  1 ). From August to October 2020, evidence of prior SARS-CoV-2 infection was defined as a positive test by any immunoassay, according to interim guidelines for COVID-19 antibody testing for public health serosurveillance [16] . To compare trends in pediatric seroprevalence after COVID-19 vaccination was recommended for adolescents aged ≥16 years, we restricted analysis to states in which commercial laboratories tested all sera collected after November 2020 using a single antinucleocapsid antibody immunoassay (Elecsys, Roche). We also restricted analysis to states in which commercial laboratories tested ≥100 residual specimens from individuals aged 0-17 years for each month from August 2020 to May 2021 for estimates of pediatric seroprevalence by month.

Using de-identified patient data from the commercial laboratory serosurvey, we estimated SARS-CoV-2 seroprevalence among persons of all ages, children and adolescents aged 0-17 years, and for 3 pediatric age groups: 0-4, 5-11, and 12-17 years. We calculated weighted estimates of seropositivity for each state and month of specimen collection, using individual weights to account for differences in distributions of age, sex, and metropolitan vs rural residence among patients included in serosurveys compared to the state population aged 0-17 years [13] . As previously described, we estimated populationweighted seroprevalence and 95% confidence intervals (CIs) accounting for sensitivity and specificity of the assays using an iterative poststratification process known as raking and layered bootstrapping, with adjusted seroprevalence defined as the mean of the bootstrap distribution and upper and lower confidence limits set at the 2.5th and 97.5th percentiles [17] . We estimated cumulative SARS-CoV-2 infections among persons of all ages and children aged 0-17 years, as well as by pediatric age groups, by multiplying monthly mean seroprevalence estimates, and upper and lower confidence limits, by the respective age-specific population in each state [18] .

For COVID-19 case-based surveillance, CDC receives individual-level data (including age, sex, date of illness onset, and specimen collection) for probable and laboratoryconfirmed COVID-19 cases from jurisdictional health departments through the COVID-19 Case Report Form and the National Notifiable Diseases Surveillance System [19, 20] . For each reported case patient, date of illness was defined as either symptom onset (if reported), the earliest clinical date on case report forms, or date reported to CDC. For this analysis, we included individual-level case data from 8 states (California, Illinois, Nevada, New Jersey, North Carolina, Ohio, South Carolina, and Tennessee) with ≥90% concordance between daily aggregate COVID-19 case counts and cumulative individual case reports from March 2020 through May 2021 and case patient age recorded for >90% of individual case reports.

To calculate ratios of cumulative infections to reported cases, we divided estimated infections (with upper and lower confidence limits) for each month by the number of confirmed and probable COVID-19 cases in persons of all ages and children aged 0-17 years reported by each state health department since January 2020. To allow for delayed antibody response to SARS-CoV-2 infection, cumulative numbers of infections estimated from monthly seroprevalence were divided by COVID-19 cases reported through the 15th of the corresponding month from 15 August 2020 through 15 May 2021. Analyses were conducted in SAS software, version 9.4 (SAS Institute) and R software, version 3.6.3 (R Core Team).

For the 8 states included in the analysis, a total of 41 583 residual serum specimens were tested from persons aged 0-17 years during August 2020 through May 2021. By state, total numbers of specimens tested from individuals aged 0-17 years ranged from 3235 for Nevada to 7750 for Illinois (Table 1) . Fewer specimens were tested each month from August to October 2020 (range, 2644-3734) compared to November 2020 through May 2021 (range, 4246-4932). Overall, the age groups 0-4, 5-11, and 12-17 years accounted for 3.9%, 25.3%, and 70.8% of specimens tested, respectively. In total, 23 542 (56.6%) specimens tested were from females and 18 041 (43.4%) from males. Higher proportions of pediatric serum specimens tested were obtained from females than males for all states (range, 53.1%-59.2% female) and calendar months (range, 54.2%-58.8% female). Proportions of pediatric specimens tested by each of the 3 serologic assays varied by state and calendar month ( Figure 1 ). Figure 1 shows weighted seroprevalence estimates based on residual serum specimens from children aged 0-17 years. In August 2020, median seroprevalence in the 8 states was 8%, ranging from 6% in North Carolina to 20% in New Jersey. Overall, estimated seroprevalence was <10% in 6 of 8 states. By May 2021, all 8 states had estimated pediatric seroprevalence >25%. Median seroprevalence in May 2021 was 37%, ranging from 26% in North Carolina to 44% in New Jersey.

In Figure 2 , population-weighted pediatric seroprevalence estimates from August 2020 through May 2021 are presented with cumulative incidence of reported COVID-19 cases per 100 000 Seroprevalence estimates by pediatric age group are shown in Figure 3 . Comparisons between age groups were limited by small numbers of specimens from children aged 0-4 years, but seroprevalence increased from August 2020 to May 2021 among all pediatric age groups (P < .005 from nonparametric test for all). Table 2 compares estimated numbers of SARS-CoV-2 infections based on population-weighted seroprevalence estimates in May 2021 from the commercial laboratory serosurvey with cumulative numbers of COVID-19 cases (confirmed and probable) reported by each state health department among persons of all ages and children aged 0-17 years. Compared to ratios in the total population (ranging from 2.2 in South Carolina and Tennessee to 3.9 in Ohio), estimated ratios of SARS-CoV-2 infections to reported cases in all states were higher among persons aged 0-17 years (ranging from 4.7 in North Carolina to 8.9 in Ohio). Pediatric infection-to-case ratios varied between states. Overall, pediatric age-specific infection-to-case ratios 8 of SARS-CoV-2 infections to reported COVID-19 cases were highest during August to October 2020. From November 2020 through May 2021, infection-to-case ratios remained relatively stable (Supplementary Figure 1) . Ratios of infections to reported cases tended to be higher in children aged 0-11 years than in those aged 12-17 years. Infection-to-case ratios were similar among children aged 0-4 years and 5-11 years in months when ≥30 specimens from children aged 0-4 years were tested (Supplementary Figure  2) . Infection-to-case ratios varied more across states than between pediatric age groups in the same state. In August 2020, we observed the highest ratios in New Jersey for all pediatric age groups (0-4 years: 87.0; 5-11 years: 98.2; 12-17 years: 30.0). In May 2021, we observed the highest ratios in Ohio among children aged 0-4 and 5-11 years (14.3 and 13.1, respectively) and in California among adolescents aged 12-17 years (5.6).

As of May 2021, monthly aggregated commercial laboratory seroprevalence survey data from 8 US states indicated that less than half, 26%-44% depending on the state, of children and adolescents were seropositive for SARS-CoV-2 antibodies against the viral nucleocapsid protein, a marker of prior infection rather than COVID-19 vaccination. In most states, we estimated from 5 to 9 times more SARS-CoV-2 infections in children and adolescents during the COVID-19 pandemic than reported through case-based surveillance, with substantial state-to-state variability. Underestimation of SARS-CoV-2 infections in children was greater than in the total population, with an estimated 2-4 infections for each reported COVID-19 case among persons of all ages [14] . Estimating cumulative incidence of COVID-19 among children and adolescents from casebased surveillance alone would substantially underestimate the proportion of children infected with SARS-CoV-2. Children and adolescents remain at risk of SARS-CoV-2 infection and COVID-19-related complications. Continued SARS-CoV-2 testing and reporting of pediatric COVID-19 cases are important for evaluation of COVID-19 vaccine effectiveness among adolescents and documenting burden of disease in younger children. Ongoing serosurveys with pediatric specimens will be needed in addition to case-based surveillance to monitor trends in SARS-CoV-2 infections in children and adolescents.

The national commercial laboratory seroprevalence survey has provided population-based, cross-sectional data from July 2020 to July 2021 to monitor changes in seroprevalence over time that can be used to estimate cumulative incidence of SARS-CoV-2 infection [13] [14] [15] . Results from seroprevalence surveys conducted in 10 geographic locations during March to May 2020 indicated substantial underestimation of SARS-CoV-2 infections and COVID-19 cases early in the pandemic, highlighting the importance of repeated serosampling [15] . In early seroprevalence surveys, ratios of estimated numbers of people infected to reported COVID-19 cases varied widely by state and geography, suggesting not only differences in COVID-19 epidemiology and timing of pandemic waves, but also differences in access to SARS-CoV-2 diagnostic testing and healthcare seeking [21] . Comparisons of SARS-CoV-2 infection rates between population age groups may also reflect differential exposure risks by age and differences in prevalence of asymptomatic infections [21] [22] [23] [24] . Our analyses identified fewer differences when comparing seroprevalence by pediatric age group within a state than between states. In the nationwide commercial laboratory serosurvey [13, 14] , residual serum specimens from persons aged 0-17 years made up a smaller proportion of specimens tested than specimens from older age groups, limiting comparisons of seroprevalence among children vs adults in the same state. While aggregation of individual-level data for this analysis by month of specimen collection provided seroprevalence estimates for pediatric age groups, smaller numbers of specimens from children aged <5 years resulted in less precise estimates for this important age group.

National commercial laboratory serosurveys have also contributed substantially to our understanding of the COVID-19 pandemic by providing comparable estimates of cumulative rates of SARS-CoV-2 infection by age group and across geographic areas. Seroprevalence data add to case-based surveillance data [7, 25] , which are subject to differences in SARS-CoV-2 testing, healthcare seeking, and reporting practices. In general, while timing of SARS-CoV-2 infection of seropositive individuals is unknown, individual-level data on patient age, sex, and urban/ rural residence for specimens randomly sampled for testing allows for analyses of trends by population demographics [13] . In addition to analyses by calendar month, trends in seroprevalence may be analyzed following waves of COVID-19 cases or among age groups targeted for vaccination at different times. For pediatric age groups, ongoing cross-sectional serosurveys may help to identify associations between timing of school and community mitigation measures and incidence of SARS-CoV-2 infections in school-aged children and adolescents [26] .

Findings from this analysis are subject to several limitations. First, convenience samples of residual clinical specimens are not representative of the general population and weighted estimates accounted only for age, sex, and rural vs urban residence of children in the sample compared to the state population. Important differences have been reported in seroprevalence and infection-to-case ratios by race/ethnicity, socioeconomic status, and health insurance [27, 28] , but these data were not available. Second, we restricted analyses to only 8 states with >90% completeness of individual case patient age in COVID-19 surveillance data. The 8 states included in this analysis included different geographic regions and variable timing and intensity of pandemic waves. Third, sample sizes for younger children were small, limiting comparisons between pediatric age groups. Fourth, sensitivity of immunoassays for identifying past SARS-CoV-2 infections may be influenced by illness severity, kinetics of antibody responses in children, types of antibody measured (pan-immunoglobulin vs IgG-specific assays), and antigen targets of immunoassays [23, [29] [30] [31] [32] [33] [34] [35] . While the presence of cross-reactive antibodies to other human coronaviruses may affect SARS-CoV-2 antibody seroprevalence estimates at low levels of infection [36, 37] , antibody cross-reactivity would have little impact on trends in seroprevalence over time. SARS-CoV-2 antibody waning below qualitative immunoassay cutoff values for seropositivity may have underestimated SARS-CoV-2 seroprevalence and numbers of children previously infected. However, the Roche Elecsys immunoassay, which detects total anti-SARS-CoV-2 nucleocapsid immunoglobulin, has been shown to be highly sensitive at least 6 months following infection in nonhospitalized patients [33] . Fifth, although commercial laboratories removed duplicate serum specimens from each convenience sample, residual sera from few individuals may have been included in >1 round. Finally, this analysis included data through the end of May 2021, before increased COVID-19 cases due to SARS-CoV-2 Delta variant viruses.

Overall, these findings provide evidence of higher rates of COVID-19 among children than detected by case-based surveillance. Through May 2021, the majority of children had no serologic evidence of past SARS-CoV-2 infection, leaving them at risk for future infection. This finding reinforces the need for multiple layers of mitigation strategies to reduce the risk of viral transmission in schools, childcare centers, and other congregate settings. Use of masks, physical distancing, increased ventilation, and combined measures have demonstrated effectiveness [33, [38] [39] [40] [41] [42] [43] [44] [45] [46] . Among age groups where vaccines are FDA approved, vaccination is a critical component of strategies to protect children and reduce potential for transmission in households, schools, and communities. With increased COVID-19 incidence in some jurisdictions due to the highly transmissible SARS-CoV-2 Omicron variant, estimation of SARS-CoV-2 infection rates through population-level serosurveys can contribute to understanding the impact of the COVID-19 pandemic on children and adolescents [4] .

Demographic trends of COVID-19 cases and deaths in the US reported to CDC

Epidemiology of COVID-19 among children in China

COVID-19 in children: initial characterization of the pediatric disease

Hospitalizations associated with COVID-19 among children and adolescents-COVID-NET, 14 states

Hospitalization of adolescents aged 12-17 years with laboratory-confirmed COVID-19-COVID-NET, 14 states

Characteristics and outcomes of children with coronavirus disease 2019 (COVID-19) infection admitted to US and Canadian pediatric intensive care units

Children and COVID-19: state-level data report

The Advisory Committee on Immunization Practices' interim recommendation for allocating initial supplies of COVID-19 vaccine-United States

The Advisory Committee on Immunization Practices' updated interim recommendation for allocation of COVID-19 vaccine-United States

Centers for Disease Control and Prevention. COVID data tracker: vaccination demographic trends

The Advisory Committee on Immunization Practices' interim recommendation for use of Pfizer-BioNTech COVID-19 vaccine in adolescents aged 12-15 years-United States

The Advisory Committee on Immunization Practices' interim recommendation for use of Pfizer-BioNTech COVID-19 vaccine in children aged 5-11 years-United States

Comparison of estimated SARS-CoV-2 seroprevalence through commercial laboratory residual sera testing and a community survey

Commercial laboratory seroprevalence survey data

Seroprevalence of antibodies to SARS-CoV-2 in 10 sites in the United States

Centers for Disease Control and Prevention. Interim guidelines for COVID-19 antibody testing. 2021

Changes in severe acute respiratory syndrome coronavirus 2 seroprevalence over time in ten sites in the United States

Centers for Disease Control and Prevention. CDC Wonder: bridged-race population estimates. 2020

Information for health departments on reporting cases of COVID-19

COVID-19 appendices

The challenge of clearly counting COVID-19 cases in children

Age-dependent effects in the transmission and control of COVID-19 epidemics

Clinical and immunological assessment of asymptomatic SARS-CoV-2 infections

Natural history of asymptomatic SARS-CoV-2 infection

National trends of cases of COVID-19 in children based on US state health department data

Severe acute respiratory syndrome coronavirus 2 antibody prevalence in blood in a large school community subject to a coronavirus disease 2019 outbreak: a cross-sectional study

SARS-CoV-2 cumulative incidence and period seroprevalence: results from a statewide population-based serosurvey in California

Prevalence of SARS-CoV-2 antibodies in New York City adults

Duration of effective antibody levels after COVID-19

A systematic review of antibody mediated immunity to coronaviruses: kinetics, correlates of protection, and association with severity

SARS-CoV-2 antibodies remain detectable 12 months after infection and antibody magnitude is associated with age and COVID-19 severity. medRxiv

Antibody responses to SARS-CoV-2 in patients with COVID-19

SARS-CoV-2 antibody magnitude and detectability are driven by disease severity, timing, and assay

Typically asymptomatic but with robust antibody formation: children's unique humoral immune response to SARS-CoV-2. medRxiv

Decline in SARS-CoV-2 antibodies after mild infection among frontline health care personnel in a multistate hospital network-12 states

Serologic testing of US blood donations to identify severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-reactive antibodies

Severe acute respiratory syndrome coronavirus 2-specific antibody responses in coronavirus disease patients

Guidance for implementing COVID-19 prevention strategies in the context of varying community transmission levels and vaccination coverage

Pilot investigation of SARS-CoV-2 secondary transmission in kindergarten through grade 12 schools implementing mitigation strategies

COVID-19 cases and transmission in 17 K-12 schools-Wood County

Mask use and ventilation improvements to reduce COVID-19 incidence in elementary schools-Georgia

COVID-19 transmission in US child care programs

Low SARS-CoV-2 transmission in elementary schools-Salt Lake County

Characteristics of COVID-19 cases and outbreaks at child care facilities-District of Columbia

Limited secondary transmission of SARS-CoV-2 in child care programs-Rhode Island

Minimal SARS-CoV-2 transmission after implementation of a comprehensive mitigation strategy at a school-New Jersey

Acknowledgments. We thank the individuals involved in COVID-19 response at state and local health departments in California, Connecticut, Illinois, Kansas, Massachusetts, Minnesota, Nevada, New Jersey, North Carolina, Ohio, Oklahoma, South Carolina, Tennessee, and Washington. We also thank employees of commercial laboratories conducting the national serosurvey, and members of the Centers for Disease Control and Prevention Patient consent. This activity was reviewed by the CDC and determined to be consistent with non-human participant research activity. Informed consent was waived, as data were de-identified.Data sharing. De-identified individual participant data will not be made available.Financial support. This work was supported by the CDC (Atlanta, Georgia).Potential conflicts of interest. All authors: No reported conflicts of interest.All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Author contributions. A. Couture., B. C. L., C. R., K. E. N. C., B. F., and M. D. C. conceptualized and designed the study, drafted the initial manuscript, and reviewed and revised the manuscript. J. S., M. L. M., L. S., N. E., F. S. A., C. M. B., S. Y., I. A. A., B. J. K., A. Cope, K. D., L. B. T., J. D., and L. B. D. coordinated data collection and reviewed data for accuracy and critically reviewed and revised the manuscript for important public health content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.