JAMA Health Forum – Health Policy, Health Care Reform, Health Affairs | JAMA Health Forum | JAMA Network
[Skip to Navigation]
Sign In
Figure 1.  Timeline of State Reopenings
Timeline of State Reopenings

Seven states (AR, IA, NE, ND, OK, SD, and WY) did not implement official stay-at-home orders during the study period, although at least some of these states issued orders for nonessential business closures and other guidance to induce social distancing. All states, including the 7 that did not implement stay-at-home orders, had official state reopenings.

Figure 2.  Interrupted Time Series Estimates of Adjusted Change in Rates of COVID-19–Related Hospitalizations and Deaths Associated With State Initial Reopenings
Interrupted Time Series Estimates of Adjusted Change in Rates of COVID-19–Related Hospitalizations and Deaths Associated With State Initial Reopenings

Changes in hospitalizations (A) and deaths (B) relative to the day of initial reopening. The vertical gray bars capture day 0 (day of reopening) through day 12 (end of washout period). The shaded areas represent 95% CIs.

Table 1.  State Reopening Dates in 2020a
State Reopening Dates in 2020a
Table 2.  Adjusted Change in Trends: Rates of COVID-19 Hospitalizations and Deathsa
Adjusted Change in Trends: Rates of COVID-19 Hospitalizations and Deathsa
Table 3.  Adjusted Change in Trends of Hospitalization Rate by Reopening Characteristicsa
Adjusted Change in Trends of Hospitalization Rate by Reopening Characteristicsa
1.
Gupta S, Nguyen TD, Lozano Rojas F, et al. Tracking public and private responses to the COVID-19 epidemic: evidence from state and local government actions. NBER Working Paper Series, No. w27027. National Bureau of Economic Research; 2020.
2.
Courtemanche  C, Garuccio  J, Le  A, Pinkston  J, Yelowitz  A.  Strong social distancing measures in the United States reduced the COVID-19 growth rate.   Health Aff (Millwood). 2020;39(7):1237-1246. doi:10.1377/hlthaff.2020.00608 PubMedGoogle ScholarCrossref
3.
Sen  S, Karaca-Mandic  P, Georgiou  A.  Association of stay-at-home orders with COVID-19 hospitalizations in 4 states.   JAMA. 2020;323(24):2522-2524. doi:10.1001/jama.2020.9176 PubMedGoogle ScholarCrossref
4.
Gupta  S, Montenovo  L, Nguyen  TD,  et al  Effects of Social Distancing Policy on Labor Market Outcomes. National Bureau of Economic Research; 2020. doi:10.3386/w27280
5.
Lee  SY, Park M, Shin Y.  Hit Harder, Recover Slower? Unequal Employment Effects of the Covid-19 Shock. National Bureau of Economic Research; 2021. doi:10.3386/w28354
6.
Wright  AL, Sonin  K, Driscoll  J, Wilson  J.  Poverty and economic dislocation reduce compliance with COVID-19 shelter-in-place protocols.   J Econ Behav Organ. 2020;180:544-554. doi:10.1016/j.jebo.2020.10.008 PubMedGoogle ScholarCrossref
7.
Nguyen TD, Gupta S, Andersen M, et al. Impacts of state reopening policy on human mobility. NBER Working Paper Series, No. w27235. National Bureau of Economic Research; 2020.
8.
University of Minnesota COVID-19 Hospitalization Tracking Project website. Accessed April 30, 2020. https://carlsonschool.umn.edu/mili-misrc-covid19-tracking-project
9.
The New York Times. Coronavirus (Covid-19) data in the United States. Accessed January 17, 2021. https://github.com/nytimes/covid-19-data.
10.
Karaca-Mandic  P, Georgiou  A, Sen  S . Calling all states to report standardized information on COVID-19 hospitalizations. Health Affairs blog. April 7, 2020. Accessed April 17, 2020. https://www.healthaffairs.org/do/10.1377/hblog20200406.532030/full/?cookieSet=1
11.
Karaca-Mandic  P, Sen  S, Georgiou  A, Zhu  Y, Basu  A.  Association of COVID-19–related hospital use and overall COVID-19 mortality in the USA.   J Gen Intern Med. 2020. doi:10.1007/s11606-020-06084-7 PubMedGoogle Scholar
12.
Karaca-Mandic  P, Georgiou  A, Sen  S.  Assessment of COVID-19 hospitalizations by race/ethnicity in 12 states.   JAMA Intern Med. 2021;181(1):131-134. doi:10.1001/jamainternmed.2020.3857 PubMedGoogle ScholarCrossref
13.
Levin  Z, Choyke  K, Georgiou  A, Sen  S, Karaca-Mandic  P.  Trends in pediatric hospitalizations for coronavirus disease 2019.   JAMA Pediatr. 2021;175(4):415-417. doi:10.1001/jamapediatrics.2020.5535 PubMedGoogle ScholarCrossref
14.
Raifman  J, Nocka  K, Jones  D,  et al. COVID-19 US State Policy Database. Inter-university Consortium for Political and Social Research, June 8, 2020. https://www.openicpsr.org/openicpsr/project/119446/version/V3/view
15.
Nguyen T, Simon K. Reopening plans. Accessed January 6, 2021. https://github.com/nguyendieuthuy/ReOpeningPlans
16.
National Centers for Environmental Information. Data access. Accessed January 21, 2021. https://www.ncdc.noaa.gov/data-access
17.
Kapoor R, Rho H, Sangha  K, et al.  God is in the rain: the impact of rainfall-induced early social distancing on COVID-19 outbreaks.  SSRN. Accessed May 19, 2020. https://ssrn.com/abstract=3605549
18.
Liu  M, Thomadsen  R, Yao  S.  Forecasting the spread of COVID-19 under different reopening strategies.   Sci Rep. 2020;10(1):20367. doi:10.1038/s41598-020-77292-8 PubMedGoogle ScholarCrossref
19.
See reopening plans and mask mandates for all 50 states. The New York Times. May 18. 2021. Accessed 23, 2021. https://www.nytimes.com/interactive/2020/us/states-reopen-map-coronavirus.html
20.
Haber  NA, Clarke-Deelder  E, Feller  A,  et al.  Problems with Evidence Assessment in COVID-19 Health Policy Impact Evaluation (PEACHPIE): a systematic strength of methods review.   medRxiv. 2021;2021.01.21.21250243. doi:10.1101/2021.01.21.21250243PubMedGoogle Scholar
21.
Lauer  SA, Grantz  KH, Bi  Q,  et al.  The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application.   Ann Intern Med. 2020;172(9):577-582. doi:10.7326/M20-0504PubMedGoogle ScholarCrossref
22.
Guan  WJ, Ni  ZY, Hu  Y,  et al; China Medical Treatment Expert Group for Covid-19.  Clinical characteristics of coronavirus disease 2019 in China.   N Engl J Med. 2020;382(18):1708-1720. doi:10.1056/NEJMoa2002032PubMedGoogle ScholarCrossref
23.
Kaestner  R.  Did Massachusetts health care reform lower mortality? not according to randomization inference.   Stat Public Policy (Phila). 2016;3(1):1-6. doi:10.1080/2330443X.2015.1102667Google ScholarCrossref
24.
Centers for Disease Control and Prevention. US state and territorial stay-at-home orders: March 15–December 31 by county by day. Accessed April 25, 2021. https://catalog.data.gov/dataset/u-s-state-and-territorial-stay-at-home-orders-march-15-september-14-by-county-by-day-c2aad
25.
Lyu  W, Wehby  GL.  Shelter-in-place orders reduced COVID-19 mortality and reduced the rate of growth in hospitalizations.   Health Aff (Millwood). 2020;39(9):1615-1623. doi:10.1377/hlthaff.2020.00719 PubMedGoogle ScholarCrossref
26.
Andersen  MS, Bento  AI, Basu  A, Marsicano  C, Simon  K.  College openings, mobility, and the incidence of COVID-19.   medRxiv. Published online September 23, 2020:2020.09.22.20196048. doi:10.1101/2020.09.22.20196048Google Scholar
27.
Isphording  IE, Lipfert  M, Pestel  N. School re-openings after summer breaks in Germany did not increase SARS-CoV-2 cases. Social Science Research Network; 2020. Accessed April 14, 2021. https://papers.ssrn.com/abstract=3713631
28.
The association of opening K-12 schools and colleges with the spread of COVID-19 in the United States: county-level panel data analysis.  medRxiv. Accessed April 14, 2021. https://www.medrxiv.org/content/10.1101/2021.02.20.21252131v1
29.
Harris  DN, Ziedan  E, Hassig  S. The effects of school reopenings on COVID-19 hospitalizations. Accessed January 27, 2021. https://www.reachcentered.org/uploads/technicalreport/The-Effects-of-School-Reopenings-on-COVID-19-Hospitalizations-REACH-January-2021.pdf
30.
Department of Health & Human Services. COVID-19 reported patient impact and hospital capacity by state timeseries. May 15, 2021. Accessed February 1, 2021. https://healthdata.gov/Hospital/COVID-19-Reported-Patient-Impact-and-Hospital-Capa/g62h-syeh
31.
Department of Health & Human Services. COVID-19 reported patient impact and hospital capacity by facility. May 17, 2021. Accessed December 23, 2020. https://healthdata.gov/Hospital/COVID-19-Reported-Patient-Impact-and-Hospital-Capa/anag-cw7u
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    1 Comment for this article
    EXPAND ALL
    Habitual Use of Significance Testing Causes the Negative Effect of Reopening to be Undermeasured
    James Gundlach, PhD, University of Texas | Professor Emeritus, Sociology, Auburn University
    Since these authors are using data that is a measure of the total hospitalized population, you are not estimating a population from a sample, you just need to report the data without the 95% confidence intervals. Confidence intervals are statistics designed to estimate possible errors created by drawing a random sample of size N from an infinite population. It tells you the probability that the finding in the sample did not happen in the population the sample was drawn from. Habitually reporting significance testing when you have the best total population data results in throwing away real findings. The reported -0.191 is the actual decline in hospital admissions per 100,000 people that happened to the hospitalization rate before reopening. It was going down and the increase after reopening was the difference between before and after = 1.798 per 100,000, the distance between -.191 and 1.607, not just somewhere between .203 and 3.011. The (mis)use of significance testing leads the measure of the effect of reopening to appear too low and encourages readers to ignore important information. In the same period, 0.0376 per 100 000 people more people died.

    As an aside the large confidence intervals reported are more consistent with a N based on the number of states rather than the actual number of state-day observations in this data set.
    CONFLICT OF INTEREST: None Reported
    READ MORE
    Original Investigation
    June 25, 2021

    US Trends in COVID-19–Associated Hospitalization and Mortality Rates Before and After Reopening Economies

    Author Affiliations
    • 1Department of Economics, Indiana University Purdue University, Indianapolis
    • 2Carlson School of Management, Minneapolis, Minnesota
    • 3Information & Decision Sciences, Carlson School of Management, Minneapolis, Minnesota
    • 4O’Neill School of Public and Environmental Affairs, Indiana University, Bloomington
    • 5Department of Finance, Carlson School of Management, Minneapolis, Minnesota
    JAMA Health Forum. 2021;2(6):e211262. doi:10.1001/jamahealthforum.2021.1262
    Key Points

    Question  Did the trends in COVID-19–related hospitalizations and deaths change after states reopened their economies?

    Findings  In this cross-sectional study of COVID-19–related hospitalizations and deaths across 47 US states between April 16 and July 31, 2020, the daily trend of hospitalizations after state reopenings was higher by 1.607 per 100 000 population. The change in the mortality rate trend was not significant.

    Meaning  These findings suggest that data on COVID-19–related hospitalizations and mortality trends can be used to guide health policy as states make decisions to open or close activities in response to this and future pandemics.

    Abstract

    Importance  After abrupt closures of businesses and public gatherings in the US in late spring 2020 due to the COVID-19 pandemic, by mid-May 2020, most states reopened their economies. Owing in part to a lack of earlier data, there was little evidence on whether state reopening policies influenced important pandemic outcomes—COVID-19–related hospitalizations and mortality—to guide future decision-making in the remainder of this and future pandemics.

    Objective  To investigate changes in COVID-19–related hospitalizations and mortality trends after reopening of US state economies.

    Design, Setting, and Participants  Using an interrupted time series approach, this cross-sectional study examined trends in per-capita COVID-19–related hospitalizations and deaths before and after state reopenings between April 16 and July 31, 2020. Daily state-level data from the University of Minnesota COVID-19 Hospitalization Tracking Project on COVID-19–related hospitalizations and deaths across 47 states were used in the analysis.

    Exposures  Dates that states reopened their economies.

    Main Outcomes and Measures  State-day observations of COVID-19–related hospitalizations and COVID-19–related new deaths per 100 000 people.

    Results  The study included 3686 state-day observations of hospitalizations and 3945 state-day observations of deaths. On the day of reopening, the mean number of hospitalizations per 100 000 people was 17.69 (95% CI, 12.54-22.84) and the mean number of daily new deaths per 100 000 people was 0.395 (95% CI, 0.255-0.536). Both outcomes displayed flat trends before reopening, but they started trending upward thereafter. Relative to the hospitalizations trend in the period before state reopenings, the postperiod trend was higher by 1.607 per 100 000 people (95% CI, 0.203-3.011; P = .03). This estimate implied that nationwide reopenings were associated with 5319 additional people hospitalized for COVID-19 each day. The trend in new deaths after reopening was also positive (0.0376 per 100 000 people; 95% CI, 0.0038-0.0715; P = .03), but the change in mortality trend was not significant (0.0443; 95% CI, −0.0048 to 0.0933; P = .08).

    Conclusions and Relevance  In this cross-sectional study conducted over a 3.5-month period across 47 US states, data on the association of hospitalizations and mortality with state reopening policies may provide input to state projections of the pandemic as policy makers continue to balance public health protections with sustaining economic activity.

    Introduction

    In response to the COVID-19 pandemic, between March and April 2020, US states implemented nonessential business closures and stay-at-home (SAH) orders.1 These immediate policy responses were designed to mitigate transmission of SARS-CoV-2 that could otherwise exhaust hospital and intensive care unit capacity and thereby increase COVID-19–related mortality. These actions, together with voluntary social distancing, appear to have reduced the rates of new COVID-19 cases, deaths,2 and hospitalizations3 but were also associated with substantial increases in unemployment and other economic hardships.4-6

    To alleviate financial harms, several states started reopening at the end of April 2020. By the middle of May 2020, most nonessential businesses had resumed at least some activities nationwide. A recent study reported an increase in human mobility following state reopenings.7 However, the effect of these policies on COVID-19–related hospital use and deaths remains unknown, partly owing to a lack of consistent data sources covering hospitalization data from the early pandemic stages onward.

    We used daily data collected by the University of Minnesota COVID-19 Hospitalization Tracking Project8 since the early days of the pandemic on COVID-19–specific hospitalizations from US states together with daily COVID-19 state-level deaths data tracked by The New York Times.9 In a cross-sectional study using an interrupted time series design, we estimated changes in COVID-19–related hospitalizations and deaths before and after state reopenings. The 2 outcome variables were COVID-19–related hospitalizations and new deaths in state-day observations. This technique compared the trends in the outcome variable by day in the prereopening and postreopening periods.

    Methods
    Data Sources

    Data on COVID-19–related hospitalizations were obtained from the University of Minnesota COVID-19 Hospitalization Tracking Project.8 These data were collected on a daily basis from states’ publicly available Department of Health websites and governor reports and have been used in other studies.3,10-13 The new COVID-19–related deaths per state per day were obtained from The New York Times9 and based on reports from state and local health agencies. Dates of state reopenings were obtained from Raifman et al14 and Nguyen and Simon.15 Data on average daily precipitation and temperature for seasonality controls were obtained from the US Environmental Protection Agency, Western Ecology Division laboratory website16 because these variables were shown to be associated with mobility and virus spread.17,18 The study was determined to be not human participants research and the need for informed consent was waived by the University of Minnesota institutional review board. This report followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies.

    Study Measures and Population

    We examined 2 COVID-19–specific outcome variables: current hospitalizations per capita and new COVID-19–related deaths per capita for each state-day. We sought to evaluate how trends in these outcomes varied before and after state reopenings.

    We collected data on reopenings from The New York Times19 and verified through internet searches as well as details in Raifman et al 2020.14 A state’s initial reopening date was recorded as the date when the state governor’s office declared a state reopened (Table 1).

    Our study population included 3686 state-day observations from 47 US states that reported data on hospitalizations between April 16 and July 31, 2020.

    Statistical Analysis

    To investigate the association of reopening with the outcome variables, we estimated an interrupted time series specification that captured both the changes in levels and the trends for the outcome variables (eMethods in the Supplement provides detailed regression specification). These specifications adjusted for systematic differences between states using state indicators and included calendar date indicators to account for changes that were constant across states but varied over time. The key exposure variables were a daily linear time trend, an indicator for the days after the reopening date (ie, postreopening), and an interaction of the time trend and this postreopening indicator. A positive, statistically significant coefficient estimate on this interaction indicated that the trend in the outcome variables increased after reopenings. Because the median incubation period of novel SARS-Cov-2 is 5 days and the median time between symptom onset to hospitalization is 7 days,3 we excluded these 12 days from our analysis (a washout period) after the reopening day. Our analysis also adjusted for average daily precipitation and average daily temperature in each state to control for any seasonality in human mobility that may have affected COVID-19 transmission. Heteroscedasticity robust SEs were clustered at the state level.

    We conducted several sensitivity analyses to confirm the robustness of our estimates.20 First, we tested alternative washout periods of 8 days (from reopening), corresponding to the 25th percentile of incubation, and 15 days (from reopening), corresponding to the 75th percentile of the incubation period (from infection to hospitalization).21,22 Second, to examine whether we could estimate state reopening associations similar in magnitude to ours by chance, we randomized the timing of state reopenings to alternative pseudo start dates in the preintervention time continuum. We expected that state reopenings take effect only after the actual reopening date, with no significant effect earlier. We randomized the date of state reopenings 1000 times. The P values from the randomization inference exercises were the fraction of estimated coefficients that were as large as those estimated for the true state reopening dates.23 Third, to account for counties with different reopening policies than the state, an alternative specification defined state reopenings as the share of state population living in counties that opened on state reopening dates (population living in counties that reopened/state population) for each state-calendar date observation using county level policies from the Centers for Disease Control and Prevention.24

    All analyses were performed with Stata, version 16.1 (StataCorp LLC). The 95% CIs around estimates reflect 0.025 in each tail or P ≤ .05 and P values are from 2-tailed t tests of the coefficients from regression models.

    Results

    Figure 1 displays each state’s reopening date. Between April 20 (South Carolina, Wisconsin) and June 1, 2020 (Delaware), all states reopened. Details of state reopening dates and study samples are presented in Table 1.

    Unadjusted daily rates of current hospitalization and new deaths varied extensively across states both before and after reopenings. Before state reopenings, the mean hospitalization rate per 100 000 people was 25.52 (interquartile range [IQR], 7.00-37.62), and the corresponding rate after reopenings was 13.08 (IQR, 5.41-16.09). The mean new COVID-19 death rate before reopenings was 0.63 (IQR, 0.10-0.87), and the corresponding rate after reopenings was 0.22 (IQR, 0.03-0.28).

    Interrupted time series estimates are presented in Table 2, which displays the trend and the change in trend of the outcome variables associated with state reopening. Before the state reopenings, the trend in daily hospitalization rate per 100 000 people was not statistically significantly different from 0 (−0.191; 95% CI, −0.720 to 0.339; P = .47). After the reopenings (incorporating the 12-day washout period corresponding to the median effective date),3 the hospitalization trend increased to 1.417 (95% CI, 0.515-2.318; P = .003), which resulted in a statistically significant increase of 1.607 (95% CI, 0.203-3.011; P = .03) in daily time trend of hospitalization rate associated with the state reopening. The mean hospitalization rate on the day of reopening was 17.69 per 100 000 people. By 12 days after reopening, the hospitalization rate increased by 3.96 (95% CI, −0.23 to 8.14), although the increase was not statistically significant. The estimated increase in corresponding rates was 16.70 (95% CI, 4.74-28.66) after 21 days and 26.62 (95% CI, 8.41-44.83) after 28 days of reopening (Figure 2). Overall, the estimated change of 1.607 additional hospitalizations per 100 000 people associated with the reopenings suggested that nationwide reopenings were associated with 5319 additional people hospitalized for COVID-19 in a given day (1.607 multiplied by the US population of 331 002 651, divided by 100 000) (Table 2 and Figure 2).

    The trend in the new daily death rate per 100 000 people was not significantly different from 0 before reopening (−0.0067; 95% CI, −0.0233 to 0.0100; P = .43). Although the trend was positive and statistically significant after reopenings (0.0376; 95% CI, 0.0038-0.0715; P = .03), the difference from before to after reopening was not statistically significant (0.0443; 95% CI, −0.0048 to 0.0933; P = .08) (Table 2). The mean daily new deaths per 100 000 people on the day of reopening was 0.395 (95% CI, 0.255-0.536). Although the death rate started increasing after reopening, increasing by 0.10 (95% CI, −0.08 to 0.28) 12 days after reopening, the increase became statistically significant only after 35 days. The death rate increased by 0.96 (95% CI, 0.03-1.89) 35 days after reopening (Figure 2).

    States varied in the nature of reopening,14 and in additional analyses, we distinguished between states that immediately reopened all economic sectors (outdoor recreation, retail, restaurant, worship, personal care, entertainment, and industry activities) vs those that used a phased approach to reopenings.15 Sixteen states implemented immediate reopenings (n = 1260) of all businesses; 31 states implemented phased reopenings (n = 2426). In 37 states, SAH orders were still in effect at the time of state reopenings (n = 2939); in 10 states, SAH orders had expired on or before the state reopenings (n = 747). In 35 states, public mask mandates were not in effect at the time of state reopenings (n = 2714); 12 states had adopted public mask mandates before or in conjunction with reopenings (n = 972). We found that states with phased reopenings had both higher rates of hospitalization on the day of reopening relative to those with immediate reopenings (20.93; 95% CI, 13.92-9.95 vs 9.95; 95% CI, 7.01-12.90) and a higher change in hospitalization trend after reopening (1.403; 95% CI, −0.033 to 2.840; P = .06 vs −0.659; 95% CI, −2.176 to 0.859 per day, a difference of 2.062; 95% CI, 0.469-3.655; P = .01) (Table 3). States with an SAH order at the time of reopening also had higher hospitalization rates relative to those with an expired order, although the CIs were large (18.93; 95% CI, 12.95-24.90 vs 13.47; 95% CI, 1.67-25.28), and the relative increase in hospitalizations associated with reopening was also higher in states with an SAH order at the time of reopening relative to those with an expired SAH order (1.492; 95% CI, 0.0534-2.931; P = .04 vs −0.011; 95% CI, −1.103 to 1.080 per day, a difference of 1.504; 95% CI, 0.432-2.576; P = .01). We did not find significant differences in the change in hospitalization trend associated with reopening between states with and without a mask mandate at the time of reopening.

    Our results were robust to a battery of sensitivity tests. First, our results were qualitatively similar when using alternative 8-day and 15-day washout periods corresponding to the 25th and 75th percentile of incubation period (from infection to hospitalization) (eTable 1 in the Supplement). Second, when we randomized the timing of state reopenings to alternative pseudo start dates in the preintervention time continuum, the reported P values capturing the fraction of estimated coefficients that were as large as those estimated for the true state reopening dates were generally less than 5% of the cases (eTable 2 in the Supplement). This finding suggests that one is very unlikely to estimate an association with state reopenings similar in magnitude to those we estimated using true state reopening times purely by chance. Third, our results did not qualitatively change if state reopenings were captured as the share of state population living in counties that opened on state reopening dates for each state-calendar date observation; we continued to find statistically significant increases in trends in both hospitalizations and deaths following state reopenings with this alternative reopening policy specification (eTable 3 in the Supplement).

    Discussion

    In this cross-sectional study, we addressed a gap in the literature to examine whether state policies implemented in spring 2020 to protect hospital capacity and minimize deaths due to COVID-19 were associated with hospitalizations and mortality. This gap in knowledge is a shortcoming because the pandemic requires continued reassessment of the optimal level of activities resumption. We found that, prior to reopening, there was a flat trend in current COVID-19–related hospitalizations and new daily deaths regardless of state reopening decisions that was not significantly different from 0; however, the hospitalization and mortality rates were positive after reopening. Earlier research showed that during the closure period (before reopening), reduced mobility was associated with reductions in hospitalizations3 and deaths.25 Our findings that hospitalization and mortality trends were positive after reopenings supports the findings from studies showing reopenings were substantially associated with higher mobility,26-29 emphasizing the health outcomes associated with reopenings.

    When we stratified our analyses by state characteristics, we found that hospitalization rates increased more in states with an active SAH order in place at the time of reopening and in states with phased reopenings. Although our data were not able to offer a definitive explanation for these findings, our findings showed that states that reopened but maintained some interventions to mitigate the spread of COVID-19 (phased reopening, SAH order, and/or mask mandates) had higher levels of hospitalization rates before reopening (Table 3).

    Limitations

    The study has limitations. The cross-sectional study design of our study provides associations and not causally interpretable estimates given the possible nonrandomized nature of state policy decision. There were other data limitations as well. First, states varied in their reporting of COVID-19–related hospitalizations, and some may have included suspected cases in their total. When both suspected and confirmed hospitalization data were available, we included only confirmed cases. Second, some hospitalizations may have included cases in which COVID-19 was a contributing, but not primary, diagnosis. In addition, although it would be informative to also examine other COVID-19–related hospitalization outcomes, such as intensive care unit and ventilator use, these data were available for a considerably smaller subset of states. Nevertheless, our data from the University of Minnesota COVID-19 Hospitalization Tracking Project offer a comprehensive examination of the outcome variables we studied from all states that reported them capturing the early stages of the pandemic, before and after the state reopenings. The Department of Health and Human Services started releasing data on hospital capacity (hospital beds and intensive care unit beds occupied by patients with COVID-19) at the state level starting late July 202030 and at the hospital facility level starting December 2020.31 However, those data sets do not allow capturing the time frame early enough to study state reopenings in April and May.

    Conclusions

    To our knowledge, this is the first study to quantify the association between COVID-19–related hospitalizations, deaths, and state reopenings in the US. Because a major risk of COVID-19 was exceeding the capacity of the health care infrastructure, a better understanding of the projections of COVID-19–related health care use is valuable, especially for the future waves of the pandemic. Our findings provide quantifiable evidence to hospital systems, health care professionals, and policy makers to help project and remain aware of needs for ensuring adequate hospital capacity and care as states continue to further open or close activities.

    Back to top
    Article Information

    Accepted for Publication: April 28, 2021.

    Published: June 25, 2021. doi:10.1001/jamahealthforum.2021.1262

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Gupta S et al. JAMA Health Forum.

    Corresponding Author: Pinar Karaca-Mandic, PhD, Department of Finance, Carlson School of Management, 321 19th Ave S, Room 3-122, Minneapolis, MN 55455 (pkmandic@umn.edu).

    Author Contributions: Drs Gupta and Karaca-Mandic had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: All authors.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: All authors.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: Gupta, Karaca-Mandic.

    Obtained funding: Sen, Karaca-Mandic.

    Administrative, technical, or material support: All authors.

    Supervision: Georgiou, Simon, Karaca-Mandic.

    Conflict of Interest Disclosures: Dr Karaca-Mandic reported receiving grants from the University of Minnesota Office of Academic Clinical Affairs funds to support data collection, which is made publicly available, and grants from United Health Foundation funds to support data collection, which is made publicly available during the conduct of the study; personal fees from Tactile Medical consulting for unrelated work, personal fees from Precision Health Economics consulting for unrelated work, personal fees from Sempre Health consulting for unrelated work, grants from Agency for Healthcare Research and Quality for unrelated research, grants from the American Cancer Society for unrelated research, grants from the National Institute for Health Care Management for unrelated research, and grants from the National Institutes of Health for unrelated research outside the submitted work. No other disclosures were reported.

    Funding/Support: This research used publicly available data from the University of Minnesota COVID-19 Hospitalization Project, which is partially funded by the University of Minnesota Office of Academic Clinical Affairs and United Health Foundation.

    Role of the Funder/Sponsor: The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    References
    1.
    Gupta S, Nguyen TD, Lozano Rojas F, et al. Tracking public and private responses to the COVID-19 epidemic: evidence from state and local government actions. NBER Working Paper Series, No. w27027. National Bureau of Economic Research; 2020.
    2.
    Courtemanche  C, Garuccio  J, Le  A, Pinkston  J, Yelowitz  A.  Strong social distancing measures in the United States reduced the COVID-19 growth rate.   Health Aff (Millwood). 2020;39(7):1237-1246. doi:10.1377/hlthaff.2020.00608 PubMedGoogle ScholarCrossref
    3.
    Sen  S, Karaca-Mandic  P, Georgiou  A.  Association of stay-at-home orders with COVID-19 hospitalizations in 4 states.   JAMA. 2020;323(24):2522-2524. doi:10.1001/jama.2020.9176 PubMedGoogle ScholarCrossref
    4.
    Gupta  S, Montenovo  L, Nguyen  TD,  et al  Effects of Social Distancing Policy on Labor Market Outcomes. National Bureau of Economic Research; 2020. doi:10.3386/w27280
    5.
    Lee  SY, Park M, Shin Y.  Hit Harder, Recover Slower? Unequal Employment Effects of the Covid-19 Shock. National Bureau of Economic Research; 2021. doi:10.3386/w28354
    6.
    Wright  AL, Sonin  K, Driscoll  J, Wilson  J.  Poverty and economic dislocation reduce compliance with COVID-19 shelter-in-place protocols.   J Econ Behav Organ. 2020;180:544-554. doi:10.1016/j.jebo.2020.10.008 PubMedGoogle ScholarCrossref
    7.
    Nguyen TD, Gupta S, Andersen M, et al. Impacts of state reopening policy on human mobility. NBER Working Paper Series, No. w27235. National Bureau of Economic Research; 2020.
    8.
    University of Minnesota COVID-19 Hospitalization Tracking Project website. Accessed April 30, 2020. https://carlsonschool.umn.edu/mili-misrc-covid19-tracking-project
    9.
    The New York Times. Coronavirus (Covid-19) data in the United States. Accessed January 17, 2021. https://github.com/nytimes/covid-19-data.
    10.
    Karaca-Mandic  P, Georgiou  A, Sen  S . Calling all states to report standardized information on COVID-19 hospitalizations. Health Affairs blog. April 7, 2020. Accessed April 17, 2020. https://www.healthaffairs.org/do/10.1377/hblog20200406.532030/full/?cookieSet=1
    11.
    Karaca-Mandic  P, Sen  S, Georgiou  A, Zhu  Y, Basu  A.  Association of COVID-19–related hospital use and overall COVID-19 mortality in the USA.   J Gen Intern Med. 2020. doi:10.1007/s11606-020-06084-7 PubMedGoogle Scholar
    12.
    Karaca-Mandic  P, Georgiou  A, Sen  S.  Assessment of COVID-19 hospitalizations by race/ethnicity in 12 states.   JAMA Intern Med. 2021;181(1):131-134. doi:10.1001/jamainternmed.2020.3857 PubMedGoogle ScholarCrossref
    13.
    Levin  Z, Choyke  K, Georgiou  A, Sen  S, Karaca-Mandic  P.  Trends in pediatric hospitalizations for coronavirus disease 2019.   JAMA Pediatr. 2021;175(4):415-417. doi:10.1001/jamapediatrics.2020.5535 PubMedGoogle ScholarCrossref
    14.
    Raifman  J, Nocka  K, Jones  D,  et al. COVID-19 US State Policy Database. Inter-university Consortium for Political and Social Research, June 8, 2020. https://www.openicpsr.org/openicpsr/project/119446/version/V3/view
    15.
    Nguyen T, Simon K. Reopening plans. Accessed January 6, 2021. https://github.com/nguyendieuthuy/ReOpeningPlans
    16.
    National Centers for Environmental Information. Data access. Accessed January 21, 2021. https://www.ncdc.noaa.gov/data-access
    17.
    Kapoor R, Rho H, Sangha  K, et al.  God is in the rain: the impact of rainfall-induced early social distancing on COVID-19 outbreaks.  SSRN. Accessed May 19, 2020. https://ssrn.com/abstract=3605549
    18.
    Liu  M, Thomadsen  R, Yao  S.  Forecasting the spread of COVID-19 under different reopening strategies.   Sci Rep. 2020;10(1):20367. doi:10.1038/s41598-020-77292-8 PubMedGoogle ScholarCrossref
    19.
    See reopening plans and mask mandates for all 50 states. The New York Times. May 18. 2021. Accessed 23, 2021. https://www.nytimes.com/interactive/2020/us/states-reopen-map-coronavirus.html
    20.
    Haber  NA, Clarke-Deelder  E, Feller  A,  et al.  Problems with Evidence Assessment in COVID-19 Health Policy Impact Evaluation (PEACHPIE): a systematic strength of methods review.   medRxiv. 2021;2021.01.21.21250243. doi:10.1101/2021.01.21.21250243PubMedGoogle Scholar
    21.
    Lauer  SA, Grantz  KH, Bi  Q,  et al.  The incubation period of coronavirus disease 2019 (COVID-19) from publicly reported confirmed cases: estimation and application.   Ann Intern Med. 2020;172(9):577-582. doi:10.7326/M20-0504PubMedGoogle ScholarCrossref
    22.
    Guan  WJ, Ni  ZY, Hu  Y,  et al; China Medical Treatment Expert Group for Covid-19.  Clinical characteristics of coronavirus disease 2019 in China.   N Engl J Med. 2020;382(18):1708-1720. doi:10.1056/NEJMoa2002032PubMedGoogle ScholarCrossref
    23.
    Kaestner  R.  Did Massachusetts health care reform lower mortality? not according to randomization inference.   Stat Public Policy (Phila). 2016;3(1):1-6. doi:10.1080/2330443X.2015.1102667Google ScholarCrossref
    24.
    Centers for Disease Control and Prevention. US state and territorial stay-at-home orders: March 15–December 31 by county by day. Accessed April 25, 2021. https://catalog.data.gov/dataset/u-s-state-and-territorial-stay-at-home-orders-march-15-september-14-by-county-by-day-c2aad
    25.
    Lyu  W, Wehby  GL.  Shelter-in-place orders reduced COVID-19 mortality and reduced the rate of growth in hospitalizations.   Health Aff (Millwood). 2020;39(9):1615-1623. doi:10.1377/hlthaff.2020.00719 PubMedGoogle ScholarCrossref
    26.
    Andersen  MS, Bento  AI, Basu  A, Marsicano  C, Simon  K.  College openings, mobility, and the incidence of COVID-19.   medRxiv. Published online September 23, 2020:2020.09.22.20196048. doi:10.1101/2020.09.22.20196048Google Scholar
    27.
    Isphording  IE, Lipfert  M, Pestel  N. School re-openings after summer breaks in Germany did not increase SARS-CoV-2 cases. Social Science Research Network; 2020. Accessed April 14, 2021. https://papers.ssrn.com/abstract=3713631
    28.
    The association of opening K-12 schools and colleges with the spread of COVID-19 in the United States: county-level panel data analysis.  medRxiv. Accessed April 14, 2021. https://www.medrxiv.org/content/10.1101/2021.02.20.21252131v1
    29.
    Harris  DN, Ziedan  E, Hassig  S. The effects of school reopenings on COVID-19 hospitalizations. Accessed January 27, 2021. https://www.reachcentered.org/uploads/technicalreport/The-Effects-of-School-Reopenings-on-COVID-19-Hospitalizations-REACH-January-2021.pdf
    30.
    Department of Health & Human Services. COVID-19 reported patient impact and hospital capacity by state timeseries. May 15, 2021. Accessed February 1, 2021. https://healthdata.gov/Hospital/COVID-19-Reported-Patient-Impact-and-Hospital-Capa/g62h-syeh
    31.
    Department of Health & Human Services. COVID-19 reported patient impact and hospital capacity by facility. May 17, 2021. Accessed December 23, 2020. https://healthdata.gov/Hospital/COVID-19-Reported-Patient-Impact-and-Hospital-Capa/anag-cw7u
    ×