key: cord-0989642-92z77upt authors: Suh, H. H.; Blaisdell, L.; Meehan, J.; Browne, L. title: Effectiveness of Non-Pharmaceutical Interventions on Child and Staff COVID-19 Cases in US Summer Camps date: 2021-02-22 journal: nan DOI: 10.1101/2021.02.18.21250271 sha: 7b72b855d5e5e21b4055ac119319d9f3f3e3e3dd doc_id: 989642 cord_uid: 92z77upt Background. Most camps remained closed during Summer 2020, due to concerns regarding child transmission of SARS-CoV-2 and limited information about the effectiveness of non-pharmaceutical interventions (NPIs) within child congregate settings. Methods. We surveyed US camps about on-site operations, camper and staff demographics, COVID-19 cases amongst campers and staff, and NPI usage as related to pre-camp quarantines, facial coverings, physical distancing, cleaning, and facility modifications. For all NPIs, save quarantines, responses were provided on a 5-point Likert scale format. Results. Within 486 on-site camps, a range of NPIs were instituted, most often related to reduced camper interactions, staff face coverings, cleaning, and hand hygiene. Camper facial coverings were less common, with campers always wearing masks at ~34% of the camps. Approximately 15% of camps reported 1+ confirmed COVID-19 case in either campers or staff, with three camps reporting a COVID outbreak. In both single and multi-NPI analyses, the risk of COVID-19 cases was lowest when campers always wore facial coverings. While less effective, constant use of staff facial coverings and targeted physical distancing measures, but not pre-camp quarantine, also reduced COVID-19 risks. Conclusions. We found constant facial coverings, especially for campers, and targeted physical distancing measures to reduce risks of SARS-CoV-2 transmission within summer camps. Our findings provide valuable guidance for future operations of camp and other child congregate settings with regard to efficient and effective NPI usage to mitigate SARS-CoV-2 infection. Each year, summer camps in the United States (US) host more than 26 million children and employ 1.5 million staff of all ages, race/ethnicities, genders, and socio-economic position [1] . Alongside the closing of schools, camps were massively disrupted across the US as a result of the coronavirus disease 2019 (COVID-19) pandemic. Many camp programs, including approximately 82% of US overnight camps [1] , did not open during Summer 2020, due to a lack of understanding of (1) the degree to which SARS-CoV-2 is transmitted in within child congregate settings and (2) the non-pharmaceutical interventions (NPIs) needed to minimize this transmission. To date, few studies have been conducted that examine SARS-CoV-2 transmission and NPI effectiveness in camp settings, with evidence to date anecdotal in nature or based on small sample sizes. This evidence shows that camps that operated during Summer 2020 experienced varying degrees of success in mitigating SARS-CoV-2 transmission, possibly due to differences in the use of non-pharmaceutical interventions (NPI). SARS-CoV-2 transmission, for example, was low within four Maine overnight camps that implemented prearrival quarantine, pre-and post-arrival testing and symptom screening, cohorting, face covering use, physical distancing, enhanced hygiene measures, cleaning and disinfecting, and maximal outdoor programming [2] . In contrast, an overnight camp in Georgia experienced significant SARS-CoV-2 transmission among campers and staff [3] , with an overall attack rate of 44%. Attack rates increased with length of time spent at the camp. Transmission was attributed to large number of campers sleeping in the same cabin and frequent singing and cheering without the regular use of facial coverings. Similarly, an outbreak of COVID-19 occurred at a boys' overnight summer school retreat in Wisconsin [4] . Among 152 attendees, 116 (76%) were classified as having confirmed or probable COVID-19. At this retreat, organizers required documentation of a negative prearrival RT-PCR result, 7-day prearrival quarantine, and outdoor programming, but did not implement other recommended NPIs. In this paper, we use data from a nationwide survey of camps operating during Summer 2020 to estimate the prevalence of COVID-19 cases amongst campers and staff and its relation to individual and multiple NPIs instituted at these camps. A total of 1,193 single and multi-site camps completed the survey, representing 1,489 camps, as shown in Table 1 . Camps were located in 49 states and the District of Columbia, with 26% located in the Midwest, 23% in the Northeast, 22% in the South, and 16% in the West. The racial and socioeconomic distribution of campers mirrored that of the entire camp population [1] . Approximately 73% of campers identified as White, 8 .3% Black, 6.0% Hispanic, 4.2% Asian, and 6.0% bi-or multi-racial, while about 48% of campers were from middle income, 34% from high, and 18.5% from low income households. 486 of the responding camps, serving about 90,000 campers, operated on-site during summer 2020, including 59 overnight, 206 day, and 220 combination day, overnight, and rental camps (with 1 missing response). Day and overnight single site camps operated in all regions of the US, with lower rates of operation in the West and highest rates of operation in the Northeast (Figure 1 ). Rates of operation for the single-site camps were highest during the middle of the summer in all regions of the US, as shown by the mid-summer peak in the number of camps, children, and staff at these single-site camps. NPI usage and COVID-19 cases across on-site camps are shown in Table 2 . Most camps reported constant use of NPIs related to staff facial coverings (69%), reduced capacity (89%), smaller cohort sizes (86%), increased cleaning (95%), and more frequent handwashing (96%). Constant physical distancing (66%) and face coverings among campers (33%) was reported by a smaller number of camps. Correspondingly, approximately one-third of camps required campers and staff to quarantine at home prior to attending camp, with fewer camps requiring staff to quarantine at camp and only 18 requiring campers to quarantine at camp. Modifications to sleeping arrangements were fairly typical amongst overnight and combination overnight/day/rental camps, while altered dining and bathroom arrangements were more common at day and combination camps. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. For single site camps, the number of confirmed COVID-19 cases each week was low, absent the 3 observed outbreaks (Figure 1 ). Camper and staff COVID-19 cases occurred primarily in mid-summer, corresponding to weekly trends in the number of campers and staff and the overall US case rates. The higher number of confirmed cases in mid-summer in the South and West reflected community rates at that time, while those in the Northeast and Midwest largely reflected the three COVID-19 outbreaks. Case rates were higher amongst campers in day as compared to overnight camps and camps with multiple program types, but the opposite pattern was observed for staff, with case rates for staff higher in overnight as compared to other camps. The risk of COVID-19 cases was significantly reduced when campers or staff always wore facial coverings ( programs modified to increase physical distance. While statistically significant, risk reduction for staff was less pronounced when programs were All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. CI: 0.59, 3.52); staff: 3.77, 9 5% CI: 1.63, 8.72), suggesting that physical distancing behavior is needed for risk reduction rather than simply reduced numbers. Since NPIs are often implemented together, we also examined RRs for specific pairwise NPI combinations ( We found an approximate 80% reduction in risks for both campers (RR: 0.22; 95% CI: 0.07, 0.65) and staff (RR: 0.18; 95% CI: 0.08, 0.44), when camps always implemented both camper facial coverings and a physical distancing NPI, as compared to when neither was always implemented. This reduction in risks was larger than that when only physical distancing NPIs were always implemented. When two targeted physical distancing NPIsphysical distancing and modified programs to increase physical distancing -were always implemented together, RRs were significantly lower as compared to camps where neither were always implemented. However, risks for camps implementing both physical distancing NPIs were similar to camps implementing just one physical distancing NPI. The additional use of pre-camp home quarantines did not reduce risks beyond that afforded by physical distancing or facial coverings alone. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Our findings are consistent with evidence showing lower incidence rates in children overall [6] and within congregate settings [7] , reduced risks of SARS-CoV-2 infection with the use of facial coverings and physical distancing [2, 8, 9, 10] and with anecdotal evidence of SARS-CoV-2 outbreaks in camps where camper and often staff face coverings was not universal [3, 4, 11] . For example, in a study of COVID-19 cases and transmission in 17 Wisconsin schools, authors reported low transmission rates in schools that required student masking and cohorting [10] . Correspondingly, In a study of almost 2 million confirmed COVID-19 cases from 190 countries [12], for example, the individual and combined effectiveness of mandatory public face coverings, quarantine, social distancing, and traffic restrictions on the COVID-19 effective reproduction number (R t ) was examined. The study showed the greatest reductions in R t for physical distancing measures and the simultaneous implementation of 2+ NPIs, with individual implementation of any of the examined NPIs, including mandatory facial coverings, also reducing R t of COVID-19. As in this study, we also found physical distancing measures to effectively reduce risks. However, we found camper facial coverings to offer greater risk reduction, likely reflective of (1) our study being within a child congregate setting, where children and staff spend considerable time together and physical distancing measures are harder to enforce and (2) the ability of camper masks when always worn to reduce risks of SARS-CoV-2 transmission during the period before a COVID-19 case can be identified. NPIs related to decreased capacity and pre-camp quarantining were not found by themselves to be significant COVID-19 risk reduction measures, which particularly in the case of pre-camp quarantining was surprising given evidence of its effectiveness in other settings. For example, successful operation of four Maine overnight camps during Summer 2020 was attributed in part to implementation of strict prearrival quarantines and pre-and All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 February 22, 2021. post-arrival testing, which identified asymptomatic COVID-19 cases prior to the start of camp, and with subsequent isolation and contact quarantining, prevented further SARS-CoV-2 transmission [2] . These camps also instituted face coverings, targeted physical distancing measures, and other NPIs. Our results suggest that face coverings and targeted physical distancing measures in particular were important contributors to the lack of COVID-19 cases at the Maine camps. Our study has several limitations. First, our findings are based on aggregate data reported by camps, and as such, measures of NPI usage and COVID-19 cases were not independently verified. As a result, COVID-19 cases were likely underestimated due to under-reporting and lack of detection of asymptomatic cases. It is also possible that NPI usage was underestimated or overestimated. Second, our findings may be affected by selection bias, resulting from the self-selection of camps voluntarily participating in our survey. Demographics of camps participating in the survey mirrored that of the overall camp population, suggesting that selection bias was minimal. Third, completion rates of several survey questions were below 75%, suggesting the potential for reporting bias in regards to what information they chose to report. These limitations are balanced by our substantial study strengths, including our nationwide cohort of day and overnight camps and the associated data documenting COVID-19 cases and the usage and adherence to numerous NPIs. These data allowed us to characterize COVID-19 risks to both campers and staff and to assess the effectiveness of NPIs, including facial coverings, physical distancing, quarantine, and cleaning measures. Our findings show rates of COVID-19 cases for campers and staff were relatively low -even in areas with high community COVID-19 rates -and further demonstrate the importance of strict face covering and targeted physical distancing measures to reduce SARS-CoV-2 infection in campers and staff. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. a Comparisons made using subset of data for which there is complete data for each examined NPI. RRs calculated comparing 1 or 2 NPIs to neither NPI. = All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 February 22, 2021. Weekly number of single site camps and associated campers, staff, and cases: by camp type, region p All rights reserved. No reuse allowed without permission. (which was not certified by peer review) 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 February 22, 2021. ; Preventing and mitigating SARS-CoV-2 transmission -four overnight camps SARS-CoV-2 transmission and infection among attendees of an overnight camp-Georgia COVID-19 outbreak at an overnight summer school retreat -Wisconsin COVID-19 Transmission in US Child Care Programs Incidence and secondary transmission of SARS-CoV-2 infections in schools A large COVID-19 outbreak in a high school 10 days after schools' reopening, Israel COVID-19 cases and transmission in 17 K-12 schools -Wood County