key: cord-0723913-oxf71vyi authors: Tsai, A. C.; Harling, G.; Reynolds, Z. C.; Gilbert, R. F.; Siedner, M. J. title: COVID-19 transmission in the U.S. before vs. after relaxation of social distancing measures date: 2020-07-16 journal: nan DOI: 10.1101/2020.07.15.20154534 sha: 730362ab32e7aa1b2cd0d097dfc1f6a3f0098e34 doc_id: 723913 cord_uid: oxf71vyi Background: All fifty U.S. states and the District of Columbia implemented social distancing measures that interrupted transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and reduced morbidity and mortality from coronavirus disease 2019 (COVID-19). All 51 jurisdictions have since begun to relax these measures, albeit in an uncoordinated fashion. Methods: We obtained state-level data on implementation and relaxation of social distancing measures. We fitted mixed-effects linear regression models with a random effect for state, specifying the time-varying, state-specific effective reproduction number (Rt, the expected number of secondary infections generated by each index case given date-specific population susceptibility and risk behavior). Explanatory variables included time in days, implementation period (1-21 days prior to first relaxation of any statewide social distancing measure vs. time from relaxation through May 28, 2020), and a time-by-period product term. Results: During the three weeks prior to relaxation, the estimated Rt was declining by an average of 0.0016 units per day (95% confidence interval [CI], -0.0025 to -0.0008). At the time of initial relaxation, states had reported a median 7,628 cases (range, 337-361,313) and 289 COVID-attributable deaths (range, 7-28,663), and the median estimated Rt was 1.0 (range, 0.8-1.3) (i.e., the median rate of epidemic growth across states was estimated to be zero). After the first relaxation of statewide social distancing measures, the estimated Rt began increasing by an average of 0.0047 units per day (95% CI, 0.0037 to 0.0058) compared with the pre-relaxation period. If these trends continue, the estimated state-level Rt would reach an average value of 1.05 (95% CI, 1.03-1.07) by 30 days after relaxation and 1.15 (95% CI, 1.11-1.18) by 60 days. A similar upward trajectory in the estimated Rt was observed after relaxation of statewide restrictions on internal movement. Conclusions: Relaxation of U.S. statewide social distancing measures was associated with a subsequent reversal of an earlier downward trend in transmission of SARS-CoV-2, particularly in states with less active epidemics at the time of relaxation. All fifty U.S. states and the District of Columbia implemented social distancing measures that interrupted transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and reduced morbidity and mortality from coronavirus disease 2019 (COVID-19) (1, 2). All 51 jurisdictions have since begun to relax these measures, albeit in an uncoordinated fashion. Using state-level data, we undertook a longitudinal pretest-posttest comparison group study to determine the extent to which relaxation of social distancing measures in the U.S. has led to a recrudescence in transmission. We obtained state-level data on implementation and relaxation of social distancing measures, as described previously (2) (Supplementary Appendix). We fitted mixed-effects linear regression models with a random effect for state, specifying the time-varying, statespecific effective reproduction number (R t , the expected number of secondary infections generated by each index case given date-specific population susceptibility and risk behavior, provided by Abbott et al. (3)) as the dependent variable. Explanatory variables included time in days, implementation period (1-21 days prior to first relaxation of any statewide social distancing measure vs. time from relaxation through May 28, 2020), and a time-by-period product term. In sensitivity analyses, we used an alternate model of R t estimation (4), and fitted stratified models by number of deaths and R t at the time of relaxation. As a secondary exposure, we examined relaxations of statewide restrictions on internal movement ("lockdowns"). Between April 20 and June 1, 2020, all 51 jurisdictions had relaxed at least one statewide social distancing measure. The Figure displays a scatterplot of the estimated R t for each state by day, along with a smoothed line derived from locally weighted scatterplot smoothing. During the three weeks prior to relaxation, the estimated R t was declining by an average of 0.0016 units per day (95% confidence interval [CI] , -0.0025 to -0.0008) (see Table) . At the time of initial relaxation, states had reported a median 7,628 cases (range, 337-361,313) and 289 COVIDattributable deaths (range, 7-28,663), and the median estimated R t was 1.0 (range, 0.8-1.3) (i.e., the median rate of epidemic growth across states was estimated to be zero). After the first relaxation of statewide social distancing measures, the estimated R t began increasing by an average of 0.0047 units per day (95% CI, 0.0037 to 0.0058) compared with the pre-relaxation period. If these trends continue, the estimated state-level R t would reach an average value of 1.05 (95% CI, 1.03-1.07) by 30 days after relaxation and 1.15 (95% CI, 1.11-1.18) by 60 days. A similar upward trajectory in the estimated R t was observed after relaxation of statewide restrictions on internal movement. The estimated R t increased faster after relaxation in states with a lower estimated R t at the time of relaxation. Similar trajectories were obtained with alternative estimates of R t (4). . CC-BY-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) Scatterplot of the estimated R t for each state by day, along with a smoothed line derived from locally weighted scatterplot Our strategy to identify all statewide social distancing measures being relaxed involved the triangulation of data from state government and third-party sources, as described previously in Siedner MJ, Harling G, Reynolds Z, Gilbert RF, Venkataramani A, Tsai AC. Social distancing to slow the U.S. COVID-19 epidemic: an interrupted time-series analysis. medRxiv Epub 8 Apr 2020. doi:2010. 1101/2020.2004.2003.20052373 . First, during March 25-June 5, 2020 we searched the web sites for all 50 U.S. state and commonwealth governments and the District of Columbia to identify all statewide social distancing measures that were implemented to address the spread of coronavirus infection (including executive orders, legislation, state of emergency or public health emergency declarations, major disaster declarations, restrictions on travel or internal movement, limitations on mass gatherings, and closures of schools or workplaces) that were subsequently relaxed or rescinded. Our searches included the web sites for the respective state governments, state departments of education, state departments of public health, and any state-sponsored COVID-specific web sites. To ensure the completeness of our search, we then cross-referenced our findings against several third-party sources: • Cohn A. These states have partially reopened after coronavirus shutdown. The Hill, April 28, 2020. Available at: https://thehill.com/homenews/state-watch/495012-these-stateshave-reopened-after-coronavirus-shutdown. Last accessed June 5, 2020. • Elassar A. This is where each state is during its phased reopening. CNN, May 27, 2020. Available at: https://www.cnn.com/interactive/2020/us/states-reopen-coronavirus-trnd/. Last accessed June 5, 2020. • Martineau P. Which states are reopening, which remain on lockdown, and why. Wired, May 6, 2020. Available at: https://www.wired.com/story/which-states-reopeninglockdown/. Last accessed June 5, 2020. Times. Available at: https://www.nytimes.com/interactive/2020/us/states-reopen-mapcoronavirus.html. Last accessed June 5, 2020. • Treisman R. Which states are reopening? A state-by-state guide. NPR, June 4, 2020. Available at: https://www.npr.org/2020/03/12/815200313/what-governors-are-doing-totackle-spreading-coronavirus. Last accessed June 5, 2020. In the event we identified any discrepancies, official state government sources were assumed to have precedence. Relaxation of social distancing policies were categorized as follows: reopening public schools, including child care facilities and camps; reopening workplaces, including professional offices and curbside retail; reopening the service industry, including barber shops, bowling alleys, cafes, gyms, in-store retail, libraries, pet groomers, museums, restaurants, salons, and zoos; allowing public events, including concerts, conferences, religious assembly, sporting events, and theatres; reopening outdoor recreational facilities, including beaches, campgrounds, golf courses, parks, swimming pools, and team athletics; lifting restrictions on internal movement (i.e., shelter-inplace orders, stay-at-home orders, "lockdowns"); and lifting mandatory quarantines for interstate travel. Appendix Table. Dates of relaxation of statewide social distancing measures, by type of measure and by state r n o r -a n n o u n c e s -p h a s e -t h r e e -o f -r e o p e n -a l a s k a -r e s p o n s . . . 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 July 16, 2020. 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 July 16, 2020. 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 July 16, 2020. 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 July 16, 2020. h t t p s : / / g o v e r n o r . s c . g o v / s i t e s / d e f a u l t / f i l e s / D o c u m e n t s / E x e c u t i v e -O r d e r s / 2 0 2 0 -0 4 -2 0 % 2 0 F I L E D % 2 0 E x e c u t i v e % 2 0 O r d e r % 2 0 N o . % 2 0 2 0 2 0 -2 8 % 2 0 -% 2 0 M o d i f i c a t i o n % 2 0 o f % 2 0 R e s t r i c t i o n s % 2 0 f o r % 2 0 P u b l i c % 2 0 B e a c h e s % 2 0 % 2 6 % 2 0 W a t e r s % 2 0 % 2 6 % 2 0 I n c r e m e n t a l % 2 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 July 16, 2020. g o v e r n o r -a b b o t t -i s s u e s -e x e c u t i v e -o r d e r -t e r m i n a t i n g -a i r -t r a v e l -r e s t r i c t i o n s -r e l a t e d -t o c o v i d -1 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 July 16, 2020. . 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 July 16, 2020. 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 July 16, 2020. 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 July 16, 2020. . https://doi.org/10.1101/2020.07.15.20154534 doi: medRxiv preprint The effect of largescale anti-contagion policies on the COVID-19 pandemic