key: cord-0276851-0a79w9j0 authors: Hoover, C. M.; Skaff, N. K.; Blumberg, S.; Fukunaga, R. title: Aligning staffing schedules with testing and isolation strategies reduces the risk of COVID-19 outbreaks in carceral and other congregate settings: A simulation study date: 2021-10-25 journal: nan DOI: 10.1101/2021.10.22.21265396 sha: 389340907ebddab2c77923edb9d798be242ded05 doc_id: 276851 cord_uid: 0a79w9j0 COVID-19 outbreaks in congregate settings remain a serious threat to the health of disproportionately affected populations such as people experiencing incarceration or homelessness, the elderly, and essential workers. An individual-based model accounting for individual infectiousness over time, staff work schedules, and testing and isolation schedules was developed to simulate community transmission of SARS-CoV-2 to staff in a congregate facility and subsequent transmission within the facility that could cause an outbreak. Systematic testing strategies in which staff are tested on the first day of their workweek were found to prevent up to 16% more transmission events than testing strategies unrelated to staff schedules. Testing staff at the beginning of their workweek, implementing timely isolation following testing, limiting test turnaround time, and increasing test frequency in high transmission scenarios can supplement additional mitigation measures to aid outbreak prevention in congregate settings. Throughout the COVID-19 pandemic, outbreaks in congregate settings such as skilled 2 nursing facilities (1), homeless shelters (2-5), and carceral (e.g., prisons and jails) facilities (6) 3 have been devastating. Staff have inadvertently served as a conduit for introducing SARS-CoV-4 2, the virus that causes COVID-19, from the community to people in congregate settings (6-8). 5 As such, routine testing of staff and subsequent isolation of infectious staff is essential to 6 mitigate case importation among resident populations and staff-to-staff transmission. Prior 7 analyses suggest that routine SARS-CoV-2 screening testing is one approach to reduce 8 transmission in homeless shelters (9), in healthcare settings (10), and during airline travel (11) . 9 In correctional and detention facilities, preventing spillover from the community to 10 facility staff and subsequently into resident populations remains one of many challenges to limit 11 SARS-CoV-2 transmission (12). Having a robust and responsive testing and isolation strategy At this time, the CDC Interim Guidance for SARS-CoV-2 Testing in Correctional and 22 Detention Facilities (17) does not specify when staff should be tested during the workweek to 23 . 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 October 25, 2021. ; https://doi.org/10.1101/2021.10.22.21265396 doi: medRxiv preprint minimize the spread of SARS-CoV-2 via rapid identification and isolation of new staff cases. 24 The timing of systematic testing in relation to work schedules and variable infectiousness 25 profiles could have profound importance for designing optimal systematic testing strategies and 26 for informing downstream activities to prevent transmission, such as rapid identification and 27 isolation of positive staff cases. Testing early in the work week may miss recently acquired 28 infections and lead to staff working around the time of their peak infectiousness. However, 29 testing later in the work week risks missing infectious individuals who are then allowed to work 30 several days prior to being tested and isolated. 31 This study examines the relationship between work schedules, testing schedules, and 32 within-facility transmission. An analytic framework to estimate the effect of variable testing 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) Fig 1a) . The viral dynamics of SARS-CoV-2 make control efforts challenging, as high 52 infectiousness in the absence of symptoms is common (20-22 removing a larger slice from the overall infectiousness triangle by reducing (Fig 1a) . . 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. Figure 1d shows that such delays have a 87 . 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. . D) Probability isolation occurs as a function of testing frequency, , delay in obtaining test results, 110 , and days from exposure to isolation , i.e. ≤ , demonstrating that delays in obtaining test results 111 substantially reduce the probability of prompt isolation, particularly among most frequent testing 112 scenarios. 113 . 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. with each time step corresponding to a work shift as described below. Staff move through susceptible (S), exposed (E), infected (I), and recovered (R) states, 129 with the infected state corresponding to time when > . Recovered staff are assumed to 130 remain in state R and not return to state S due to the relatively short time frame of the simulation. Parameters for newly exposed staff are drawn to determine , , and , workweek. Testing on Wednesday and Thursday was also common across work schedules. Test 182 results were usually returned on the same day or the day after specimen collection and almost all 183 test results were received within 2 days of specimen collection. week (x-axis; i.e. darker shades of red indicate that staff with the specified schedule more commonly worked on that 190 day). The size of the black circles represents the mean proportion of the total number of tests administered to each 191 group that were given on the specified day. Systematic testing strategies were found to consistently outperform random testing 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 October 25, 2021. The horizontal gray line in Figure 3 demonstrates a potential threshold number of 212 infections to avoid exceeding at ℐ = 18.00. This threshold corresponds to an average of one 213 transmission event within the simulated facility every ten days. Implementing a systematic-214 rather than random-testing strategy can be sufficient to prevent ℐ from exceeding such a 215 threshold without changing the frequency in many transmission scenarios (e.g. compare circles 216 to squares of the same color in Figure 3 ) though in the highest transmission scenarios, greater 217 than twice-weekly testing may be needed. quartile of expected infections in the simulated facility below a threshold of 1 every ten days across 237 transmission scenarios conveyed by the within-facility basic reproduction number (ℛ), community 238 prevalence (CP), and delays between testing and isolation of infectious workers. Discussion (1120 words) 263 This study builds on previous modeling and simulation analyses to demonstrate that 264 systematic testing strategies with limited delays between test administration and isolation of 265 infectious individuals can limit SARS-CoV2 transmission. Building on this, testing schedules 266 that are aligned with working schedules are found to prevent more transmission events than 267 random testing strategies or those with a delay between testing and isolation. A major benefit of 268 such strategies is that they do not require higher testing frequency, only a change in timing of 269 when testing occurs. As such, there may be substantial value in implementing systematic rapid 270 testing at the beginning of the work week for staff working in facilities at high risk for SARS- CoV-2 transmission such as carceral facilities, skilled nursing facilities, and homeless shelters. . 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. interventions were implemented, we expect qualitative trends in the expected number of 314 transmission events to persist between testing strategies and frequencies across different 315 transmission scenarios. Third, we do not distinguish between staff-to-staff and staff-to-resident 316 transmission events within a simulated facility, but rather record the total number of transmission 317 events assuming ℛ remains constant rather than decreasing due to susceptible depletion. . 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 modeling and simulation framework presented here is applicable beyond COVID-19 332 in congregate settings in which outbreaks may be due to community importation of a pathogen. 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 October 25, 2021. ; ( < ), for instance, may be more effectively controlled at lower cost via 341 symptom screening and subsequent isolation (18). In conclusion, these results suggest that aligning the timing of testing with regular 343 working schedules for staff in congregate settings, in addition to timely implementation of 344 prevention strategies (e.g., isolation) can improve the efficacy of systematic screening testing. . 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. . 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 October 25, 2021. ; https://doi.org/10.1101/2021.10.22.21265396 doi: medRxiv preprint Nursing home staff networks and COVID-19 Among Three Affiliated Homeless Service Sites -King County, Washington, 2020. MMWR Morb 373 Mortal Wkly Rep Prevalence of SARS-CoV-2 Infection in Residents of a 378 Large Homeless Shelter in Boston Infection Prevalence in Homeless Shelters -Four U.S. Cities COVID-19 in 383 Correctional and Detention Facilities -United States Community-Associated 386 Outbreak of COVID-19 in a Correctional Facility Serial 389 Laboratory Testing for SARS-CoV-2 Infection Among Incarcerated and Detained Persons in a 390 Correctional and Detention Facility -Louisiana Comparison of 393 infection control strategies to reduce COVID-19 outbreaks in homeless shelters in the United States: 394 a simulation study Frequency of Routine Testing for 396 COVID-19) in High-risk Healthcare Environments to Reduce 397 Routine asymptomatic testing strategies for airline travel during the COVID-19 pandemic: a 401 simulation study. The Lancet Infectious Diseases SARS-CoV-2 Infection Among Correctional Staff in the Federal 403 Bureau of Prisons Interim Public Health Recommendations for Fully 405 Vaccinated People MMWR Morb Mortal Wkly Rep Covid-19 Breakthrough 412 Infections in Vaccinated Health Care Workers Breakthrough SARS-CoV-2 415 Infections in Prison after Vaccination Comparing nonpharmaceutical interventions for 420 containing emerging epidemics Test sensitivity is 422 secondary to frequency and turnaround time for COVID-19 screening. Science Advances Number 426 4-April 2021 -Emerging Infectious Diseases journal -CDC Substantial undocumented infection 429 facilitates the rapid dissemination of novel coronavirus (SARS-CoV-2) Temporal dynamics in viral shedding and 432 transmissibility of COVID-19 Incubation period of COVID-434 19: a rapid systematic review and meta-analysis of observational research Impact of Social Distancing 437 Measures on Coronavirus Disease Healthcare Demand Performance 441 Characteristics of a Rapid Severe Acute Respiratory Syndrome Coronavirus 2 Antigen Detection 442 Assay at a Public Plaza Testing Site in San Francisco. The Journal of Infectious Diseases Public Health Response Using the BinaxNOWTM Rapid Severe Acute Respiratory Syndrome 446 Coronavirus 2 (SARS-CoV-2) Antigen Detection Assay During Community-Based Testing Clinical Infectious Diseases Multidisciplinary assessment of 450 the Abbott BinaxNOW SARS-CoV-2 point-of-care antigen test in the context of emerging viral 451 variants and self-administration. Sci Rep Core Team. R: A Language and Environment for Statistical Computing Austria: R Foundation for Statistical Computing Welcome to the 455 tidyverse Carnell R. triangle: Provides the Standard Distribution Functions for the Triangle Distribution The Composer of Plots Mitigating outbreaks in 461 congregate settings by decreasing the size of the susceptible population. medRxiv Outbreaks of COVID-19 464 variants in US prisons: a mathematical modelling analysis of vaccination and reopening policies The Lancet Public Health Frequency of surveillance testing necessary to reduce transmission of 468 SARS-CoV-2 Dynamic contact networks of 471 patients and MRSA spread in hospitals. Sci Rep Influenza, Varicella, and Mumps Outbreaks in US Migrant Detention Centers Evaluating 476 strategies for control of tuberculosis in prisons and prevention of spillover into communities: An 477 observational and modeling study from Brazil