Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2020 Oct 13;15(10):e0240648. doi: 10.1371/journal.pone.0240648

State-level variation of initial COVID-19 dynamics in the United States

Easton R White 1,2,*, Laurent Hébert-Dufresne 3,4
Editor: Jeffrey Shaman5
PMCID: PMC7553297  PMID: 33048967

Abstract

During an epidemic, metrics such as R0, doubling time, and case fatality rates are important in understanding and predicting the course of an epidemic. However, if collected over country or regional scales, these metrics hide important smaller-scale, local dynamics. We examine how commonly used epidemiological metrics differ for each individual state within the United States during the initial COVID-19 outbreak. We found that the detected case number and trajectory of early detected cases differ considerably between states. We then test for correlations with testing protocols, interventions and population characteristics. We find that epidemic dynamics were most strongly associated with non-pharmaceutical government actions during the early phase of the epidemic. In particular, early social distancing restrictions, particularly on restaurant operations, was correlated with increased doubling times. Interestingly, we also found that states with little tolerance for deviance from enforced rules saw faster early epidemic growth. Together with other correlates such as population density, our results highlight the different factors involved in the heterogeneity in the early spread of COVID-19 throughout the United States. Although individual states are clearly not independent, they can serve as small, natural experiments in how different demographic patterns and government responses can impact the course of an epidemic.

Introduction

The global COVID-19 outbreak caused by the SARS-CoV-2 virus began in Wuhan, China in late 2019 [1]. As of Sep 19th, over 30 million cases and nearly one million deaths have been reported across the globe. There have been several sets of efforts to track the progression of the outbreak across the world and within countries. Early in the outbreak, the Johns Hopkins University Center for Systems Science and Engineering (CSSE) started compiling data from various sources, including the US Center for Disease Control and the World Health Organization, to present a global picture of COVID-19 cases and deaths [2]. Later in the outbreak, large distributed teams were necessary to compile and curate epidemiological case data at the individual level [3]. These efforts have allowed for international scientific research and political decision-making. Although data are collected at local scales (e.g. within hospitals), in an emerging pandemic, data are typically reported at the country or regional level. This allows for interesting comparisons between countries [4–6] and for information from an earlier affected country to be used to slow the outbreak in other places. For instance, South Korea was able to “flatten their outbreak curve” through early and widespread testing as well as strict quarantine policies [7]. However, country-level analyses still hide more local dynamics that are important to the overall epidemic progression [8, 9]. For example Lin et al. (2020) found that, in China, traffic control and social distancing measures did not work effectively everywhere. Instead, these measures depended on income and population size [9].

Spatial heterogeneity is important for population dynamics generally [10–12] and in particular for understanding the progression of infectious disease dynamics [13, 14]. Spatial heterogeneity can include differences in local population density, movement patterns, suitability of environmental conditions for transmission, among other factors [13–15]. These heterogeneities can impact the spread of any infectious disease. For example, Keeling et al. [16] showed how spatial distribution and size of farms affected the 2001 UK Foot and Mouth Epidemic, Apolloni et al. studied the role of age structure and travel patterns for 2009 H1N1 pandemic, and the GLEAM framework has been used to model diseases from Zika to COVID-19 while including population density and mobility data [17–19].

Here we provide a descriptive analysis of the reported progression of COVID-19 at the state level within the United States. We examine how commonly-used metrics, focusing on doubling time, can vary by state. Clearly, controlled and randomized experiments of COVID-19 spread are not possible. Therefore, although states are not independent units, we can use state-level data to understand the progression of the outbreak as detected by heterogeneous testing systems across different replicates within a country [20]. There are three periods of interest during an epidemic: the initial exponential phase, the slowing of the curve before before the peak, and the decrease in new cases after the peak. We expect all three of these phases to look different for each state depending on both demographics and interventions. Specifically, we hypothesize that the doubling time of the outbreak in each state will be correlated with population density and metrics of overall population health. In the middle of the outbreak, after government interventions have been enacted, we hypothesize that doubling times will be correlated with a combination of demographics and these government interventions. We also assess two correlates not traditionally tied to work on infectious disease dynamics: volunteerism rates and community tightness. We examined volunteerism rates as a measure of how willing individuals may be willing to adjust their behavior to help reduce overall disease spread. Community tightness is a measure indicating adherence to rules and strong cultural norms [21]. Recently, Gelfand et al. [22], found that countries with both efficient governments and high tightness scores were more effective in reducing the spread of COVID-19 and reducing mortality rates. In this paper, we focus on the early dynamics of the pandemic, where inadequate and heterogeneous testing capacities are as, if not more, important in driving epidemic dynamics than geospatial differences [23]. This allows us to test for a wide-range of potential correlates—testing, intervention, population density, behavior, etc.—all potentially driving early reported data. That being said, some of the results highlighted here can also be useful in understanding the varied decay patterns of COVID-19 [24] as well as its resurgence in different places [25]. Indeed, testing improved in all states but human behavior and control policies remain heterogeneous and inconsistent [26].

Materials and methods

We used data on the daily number of cases compiled by Johns Hopkins University Center for Systems Science and Engineering [2]. The United States experienced an initial exponential growth in the number of cases, especially since February 29th (S1 Fig) and slowed starting in mid-March before rising again in summer. In addition, outbreaks have been reported as schools, including college campuses, have reopened [2]. Country-level results, however, hide underlying dynamics within each state [8, 9].

Therefore, we examined how the number of cases changed over time within each state. To properly compare the progression of the epidemic across states, we looked at the log number of cases since the first day a state reported 25 (after which exponential curves were more reliable) cases (Fig 2). Although no starting value is perfect, we chose 25 cases as the modeling results were qualitatively more similar after 25 cases. In other words, models with a starting value of less than 25 cases were not very consistent, likely due to early differences in reporting and case definition [27]. On a log scale, a straight line of the cases over time indicates exponential growth where the slope of the line is the exponential growth parameter. This slope informs us about the growth of the epidemic, and is related to a useful quantity known as doubling time: the expected length of time required to see the number of cases double. A longer doubling time in reported cases is therefore preferred as it suggests a slower spread of the disease.

Fig 2.

Fig 2

Open in a new tab

(a) The log number of cases over time for each individual state for the 10 days since their first day of 25 total cases. (b) The log number of cases over time for each individual state for the next 10 days. The light grey diagonal lines represent the growth trajectory for doubling times of 2, 4, and 10 days. The log number of the starting value (initial number of cases on first day when at least 25 cases were recorded) had to be subtracted on the y-axis to standardize the graph across states. (c) Rolling doubling times calculated over 10-day windows for each individual states. (d) Distributions of state-level doubling times early and more recent in the course of the outbreak.

We calculated the doubling times throughout the course of the epidemic within each state. We then looked at the early (first 7 days since 25 cases) and overall doubling times (first 21 days since 25 cases) as our two response variables. We did not examine the doubling time after 21 days as some states were already recorded a decrease in the daily number of cases per day at that point (S2 Fig). The early doubling time should not be strongly correlated with government interventions as these were often implemented late and there is a delay in which effects from government actions are detectable [28].

We used a simple linear regression to evaluate possible predictors of early and overall doubling times. We evaluated all possible combinations of explanatory variables using stepwise model selection. We compared models using Akaike Information Criterion (AIC) to determine the best fitting model while accounting for model complexity. We did not examine interactions between predictor variables given the limited number of data points. All analyses were performed in R [29]. To check model assumptions, we examined the Pearson residuals plotted against model predictions and time (S4 Fig). In line with past work, we also did not include variables that were strongly correlated (greater than 0.7) in the same model [30, 31]. Specifically, several demographic variables (e.g. population percent in rural areas versus population percent in urban areas) were strongly correlated. In those cases, we arbitrarily only included one variable in the model.

Demographic correlates

Following Chin et al. [8], we collected demographic, health, education, and income variables from a variety of sources:

State government interventions

Much of the response to COVID-19 in the United States has been done at the state, as opposed to federal, government level [20, 32]. Several policies started at city or county levels before being implemented across an entire state [33, 34]. Although interventions, guidelines and policies have varied, some slowly became almost ubiquitous across all states. We focused on five such state-wide mandates: declaring a state of emergency, limiting gatherings (usually to 10 people or less), closing schools, restricting restaurants, restricting businesses generally, and stay-at-home mandates. This is also the general sequence in which state-wide mandates were implemented (see Fig 1). To quantify the timing of the intervention relative to COVID-19, we used information collected by Adolph et al. [20] on whether or not a state had mandated a specific action by the first day they had 25 or more cases. We adjusted this number to 150 cases for more severe restrictions like closing all non-essential business or stay at home mandates). We follow Adolph et al. [20] and use the first date of gathering restrictions announced, regardless of the size of the gatherings restricted.

Fig 1. Rank distribution of different interventions.

Fig 1

Open in a new tab

Per state, every intervention is given a rank from 1 to 6 depending on when it was implemented (1 being the first put into place) and ties are given an average rank (e.g. 2.5 for tied 2nd and 3rd rank).

Results

We found considerable differences between states in how the outbreak initially progressed (Fig 2). These doubling times are, of course, changing over time. We found that doubling times for all states did increase with time and that the heterogeneity between states was reduced (see the change in spread of lines between Fig 2a and 2b). We mapped doubling time across the US and found regional differences where the West and Northeast have seen large doubling times, i.e. slower outbreak dynamics (S3 Fig).

Early dynamics

Given the large heterogeneity between states early on (Fig 2), we examined correlates of doubling time for only the first seven days after a state reached 25 total cases. Comparing the doubling times across states, we found that lower population density along with higher flu vaccination rates and wealth were associated with slower epidemics. (Fig 3, Table 1).

Fig 3.

Fig 3

Open in a new tab

Doubling time (in number of days) versus (a) log (population density), (b) population density in rural areas, (c) life expectancy (years), (d) percent of population above age 65, (e) gross income per capita (in 1000s USD), (f) expected years of schooling, (g) yearly flu vaccination rate, (h) volunteer rate, and (i) tightness score.

Table 1. Best fitting linear models (according to AIC) and corresponding parameter estimates for the doubling time both early (first 7 days since 25 cases) and for the first three weeks.

A parameter was not included in the table if it was not selected in the best fitting linear model. The overall model included the following parameters (see Methods for more detail): log(Population density), population percent in rural areas, percent of population over age 65, influenza vaccination rate, gross income per capita, expected number of years of schooling, community tightness score, tests per capita (at the end of 7 days or 21 days, respectively), and binary variables for the presence or absence of each social distancing restriction (see Fig 4).

Dependent variable:
Early doubling time Overall doubling time
Restrict restaurants 0.560**
(0.150, 0.969)
log (Population density) −0.342*** −0.293***
(−0.467, −0.218) (−0.421, −0.166)
Vaccination rate 11.308***
(6.914, 15.702)
GNI per capita 0.037***
(0.021, 0.054)
Population percent in rural areas 0.014*
(−0.0004, 0.028)
Tightness score −0.018**
(−0.032, −0.004)
Constant −12.041*** 0.901
(−16.384, −7.697) (−0.521, 2.324)
Observations 50 50
R2 0.554 0.589
Adjusted R2 0.525 0.553
Open in a new tab

Note:

*p<0.1;

**p<0.05;

***p<0.01

Overall dynamics

We then examined the overall (first 21 days since 25 total cases) doubling times at the state level. Except for population density and percent of population living in rural areas, we found that demography, education, and wealth were poor predictors of the state-level overall doubling times (Fig 3, Table 1). Therefore, we also examined the correlation between doubling time and state government interventions. Of state-wide mandates, only restricting restaurant operations was a significant (at α <0.05) predictor of doubling time (Fig 4, Table 1). However, these restrictions were also additive, as states that implemented more actions early had higher doubling times (Fig 4f). The ordering of these restrictions was also fairly consistent between states (Fig 1). While declaring a state of emergency is an obvious first intervention, closing public schools tend to be implemented at different times across states. More importantly, by the time local governments restrict businesses and declare a stay-at-home order, all other interventions tend to have already been implemented (Fig 1, [20]). Lastly, a state’s tightness score, a measure indicating adherence to rules and strong cultural norms [21], was also negatively correlated with doubling time with tighter states having lower doubling times (Fig 4, Table 1). The overall doubling time was not strongly correlated with overall tests rates for each state (Table 1, Fig 5).

Fig 4.

Fig 4

Open in a new tab

Doubling time (in number of days) across the US states for five different statewide government restrictions: (a) limit gatherings (usually to less than 10 people, but see Methods) by first day of 25 cases, (b) close public schools by first day of 25 cases, (c) restrict restaurants by first day of 25 cases, (d) restrict non-essential businesses by first day of 150 cases, (e) stay at home order by first day of 150 cases, and (f) total number of restrictions before number of cases threshold.

Fig 5.

Fig 5

Open in a new tab

Doubling time (in number of days) for each state according to the total tests conducted per capita (a) early in the outbreak (within the first week since 25 cases) and (b) for within the first three weeks.

Discussion

A lot of work on the COVID-19 pandemic has drawn on comparisons between countries to evaluate the effectiveness of different interventions [6, 35, 36]. However, this hides underlying variation within each country. We investigated the variation in how the epidemic unfolded across states in the United States. We found substantial variation between states, particularly in the early phase of the epidemic (Fig 2). We found that the early dynamics (first seven days) were most strongly correlated with population density, percent of population living in rural areas, income, and yearly flu vaccination rate (Fig 3). This is in line with other work on COVID-19 that found population demography important at the US county level [37] and in other countries. Interestingly, while the correlation between the intensity of a contagious respiratory infection and population density is intuitive, our findings are actually contrary to patterns observed during Influenza seasons [38] and for COVID-19 in China [39]. The different behavior observed here could be due to COVID-19 being less reliant on environmental factors and mostly driven by social behaviours. Similarly, it also suggests that patterns observed in China might be driven by the timing and scale of government interventions and not by density itself given the opposite behavior observed in the United States.

In addition to population density, the early doubling time was also correlated with flu vaccination rate and average income (“GNI per capita” in Table 1). For each percentage increase in vaccination rate, there was a 0.11 day difference in doubling time. For example, a states with at least 0.85 vaccination coverage had an average early doubling time of 3.5 days compared to 2.5 days for states with vaccination rates under 0.85. We hypothesize this has to do with general health and access to healthcare for people in more vaccinated states. In addition, the early doubling time increased with average income per capita (Table 1). We would also hypothesize that average income also related to access to healthcare and baseline health. In addition, other work has shown that those from higher income brackets were able to socially distance earlier and to a larger degree [25].

The overall doubling time (first 21 days of the epidemic) in each state was also correlated with demographic parameters. However, in line with past work [28, 40, 41], we show early doubling time was also associated with government actions, most notably restrictions on restaurant operations (Fig 4, Table 1). For example, a parameter estimate of 0.560 for restaurant closures implies that states with restaurant closures experienced a half-day increase in their doubling time. The specific timing of restrictions was not as important as simply a mandate to limit operations early in the epidemic (before 25 cases were recorded). This does not provide causal evidence for government actions as a key factor in controlling the disease spread, only an association. Instead, business restrictions and school closures could simply be coarse metrics of actual social distancing.

Lastly, after accounting for other demographic variables, a state’s “tightness” score was also correlated with overall doubling time (Table 1). A state with a high tightness score has “many strongly enforced rules and little tolerance for deviance” [21]. Recently, Gelfand et al. [22], showed that countries with efficient governments and high tightness scores in reducing the spread of COVID-19 and its mortality rate. Thus, we expected that states with highly enforced rules should have higher doubling times compared to “loose” states. Instead, we found the opposite where tight states had low doubling times and consequently faster disease spread. We hypothesize this may be the result of people in tight cultures finding it more difficult to adjust their behavior when new rules are imposed. This could lead individuals to protest or avoid closures. Althouse et al. (2020) showed that states with higher tightness scores had decreased number of visits to non-essential business, but also increased distance traveled for any visits that did take place. Using a simple mathematical model, they were able to show that this type of behavior leads people to cluster and can cause outbreaks to occur [26]. Future work should focus on better understanding this relationship through mobility data and more detailed mechanistic modeling.

There are several limitations to this work that require further investigation. In many places, government restrictions went into effect at a smaller scale, often in cities or counties, before implemented at the state level [34]. Thus, county-level analyses are a natural extension of this work [33, 37]. In addition, government restrictions are a crude measure that does not capture actual social distancing. For instance, work has shown that individuals reduced their movement often before certain government actions were implemented. Similarly, public buy-in to various restrictions has varied across the US [42]. Testing rates were not a significant predictor of doubling time (Table 1). However, each state implemented their own policies for who could receive a test and these policies changed throughout the course of the epidemic [43]. Thus, this heterogeneity in testing may influence doubling time (Fig 5) and should be evaluated further [43].

Conclusion

We found a large degree of heterogeneity in the reported number of COVID-19 cases over time across the United States. After state-level government actions were implemented, doubling time was most strongly correlated to social distancing restrictions, in particular restaurant operations (Fig 4). More detailed work will be needed to understand how these dynamics differ within each state, especially as many government actions started on more local scales [33, 34]. Similar to the implementation of restrictions, states have loosened, and sometimes reimposed, restrictions, but this varies across the country. This will present another natural experiment on the effect of government actions on the course of the epidemic.

Supporting information

S1 Fig. COVID-19 cases over time for the US as a whole.

(TIF)

Click here for additional data file. (90.8KB, tif)
S2 Fig. COVID-19 cases over time for each US state.

(TIF)

Click here for additional data file. (1.1MB, tif)
S3 Fig. Map of COVID-19 case doubling time (first three weeks since 25 cases) for each US state.

(TIF)

Click here for additional data file. (323KB, tif)
S4 Fig. Residual diagnostic plots for the best fitting models for both the early (first 7 days) and overall (first 21 days) doubling times.

(TIFF)

Click here for additional data file. (568.7KB, tiff)

Acknowledgments

We thank three anonymous reviewers for feedback on an earlier version of the manuscript.

Data Availability

All code and corresponding data is freely available at \url{https://github.com/eastonwhite/COVID19_US_States}. The original raw data has been compiled by the Johns Hopkins University Center for Systems Science and Engineering at (\url{https://github.com/CSSEGISandData/COVID-19}).

Funding Statement

L.H.-D. acknowledges support from the National Institutes of Health 1P20 GM125498-01 Centers of Biomedical Research Excellence Award. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

  • 1.WHO. Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19); 2020.
  • 2. Dong E, Du H, Gardner L. An Interactive Web-Based Dashboard to Track COVID-19 in Real Time. The Lancet Infectious Diseases. 2020; p. S1473309920301201. 10.1016/S1473-3099(20)30120-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Xu B, Gutierrez B, Mekaru S, Sewalk K, Goodwin L, Loskill A, et al. Epidemiological data from the COVID-19 outbreak, real-time case information. Scientific Data. 2020;7(1):1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Anderson RM, Heesterbeek H, Klinkenberg D, Hollingsworth TD. How Will Country-Based Mitigation Measures Influence the Course of the COVID-19 Epidemic? The Lancet. 2020;395(10228):931–934. 10.1016/S0140-6736(20)30567-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Jombart T, van Zandvoort K, Russell T, Jarvis C, Gimma A, Abbott S, et al. Inferring the Number of COVID-19 Cases from Recently Reported Deaths. medRxiv. 2020. 10.1101/2020.03.10.20033761 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Meunier TAJ. Full Lockdown Policies in Western Europe Countries Have No Evident Impacts on the COVID-19 Epidemic. medRxiv. 2020. 10.1101/2020.04.24.20078717 [DOI] [Google Scholar]
  • 7. Utsunomiya YT, Utsunomiya ATH, Torrecilha RBP, Paulan SC, Milanesi M, Garcia JF. Growth Rate and Acceleration Analysis of the COVID-19 Pandemic Reveals the Effect of Public Health Measures in Real Time. medRxiv. 2020. 10.1101/2020.03.30.20047688 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Chin T, Kahn R, Li R, Chen JT, Krieger N, Buckee CO, et al. U.S. County-Level Characteristics to Inform Equitable COVID-19 Response. medRxiv. 2020. 10.1101/2020.04.08.20058248 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Lin S, Huang J, He Z, Zhan D. Which Measures Are Effective in Containing COVID-19? Empirical Research Based on Prevention and Control Cases in China. medRxiv. 2020. 10.1101/2020.03.28.20046110 [DOI] [Google Scholar]
  • 10. Levin SA. The Problem of Pattern and Scale in Ecology. Ecology. 1992;73(6):1943–1967. 10.2307/1941447 [DOI] [Google Scholar]
  • 11. Hanski I. Spatially Realistic Theory of Metapopulation Ecology. Naturwissenschaften. 2001;88(9):372–381. 10.1007/s001140100246 [DOI] [PubMed] [Google Scholar]
  • 12. Schreiber SJ. Interactive Effects of Temporal Correlations, Spatial Heterogeneity and Dispersal on Population Persistence. Proceedings of the Royal Society B: Biological Sciences. 2010;277(NA):1907–1914. 10.1098/rspb.2009.2006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Grenfell BT, Bolker BM, Kleczkowski a. Seasonality and Extinction in Chaotic Metapopulations. Proceedings of the Royal Society B: Biological Sciences. 1995;259(1354):97–103. 10.1098/rspb.1995.0015 [DOI] [Google Scholar]
  • 14. Park AW. Infectious Disease in Animal Metapopulations: The Importance of Environmental Transmission. Ecology and Evolution. 2012;2(7):1398–1407. 10.1002/ece3.257 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.St-Onge G, Thibeault V, Allard A, Dubé LJ, Hébert-Dufresne L. School Closures, Event Cancellations, and the Mesoscopic Localization of Epidemics in Networks with Higher-Order Structure. arXiv:200305924 [nlin, physics:physics]. 2020.
  • 16. Keeling MJ, Woolhouse MEJ, Shaw DJ, Matthews L, Chase-Topping M, Haydon DT, et al. Dynamics of the 2001 UK Foot and Mouth Epidemic: Stochastic Dispersal in a Heterogeneous Landscape. Science. 2001;294:813–817. 10.1126/science.1065973 [DOI] [PubMed] [Google Scholar]
  • 17. Apolloni A, Poletto C, Ramasco JJ, Jensen P, Colizza V. Metapopulation Epidemic Models with Heterogeneous Mixing and Travel Behaviour. Theoretical Biology and Medical Modelling. 2014;11(1):3 10.1186/1742-4682-11-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Zhang Q, Sun K, Chinazzi M, Pastore y Piontti A, Dean NE, Rojas DP, et al. Spread of Zika Virus in the Americas. Proceedings of the National Academy of Sciences. 2017;114(22):E4334–E4343. 10.1073/pnas.1620161114 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Chinazzi M, Davis JT, Ajelli M, Gioannini C, Litvinova M, Merler S, et al. The Effect of Travel Restrictions on the Spread of the 2019 Novel Coronavirus (COVID-19) Outbreak. Science. 2020; p. eaba9757. 10.1126/science.aba9757 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Adolph C, Amano K, Bang-Jensen B, Fullman N, Wilkerson J. Pandemic Politics: Timing State-Level Social Distancing Responses to COVID-19. medRxiv. 2020. 10.1101/2020.03.30.20046326 [DOI] [PubMed] [Google Scholar]
  • 21. Harrington JR, Gelfand MJ. Tightness-Looseness across the 50 United States. Proceedings of the National Academy of Sciences. 2014;111(22):7990–7995. 10.1073/pnas.1317937111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Gelfand M, Jackson JC, Pan X, Nau D, Dagher MM, Chiu Cy. Cultural and Institutional Factors Predicting the Infection Rate and Mortality Likelihood of the COVID-19 Pandemic. PsyArXiv. 2020. 10.31234/osf.io/m7f8a [DOI] [Google Scholar]
  • 23. Lopes-Júnior LC, Bomfim E, da Silveira DSC, Pessanha RM, Schuab SIPC, Lima RAG. Effectiveness of mass testing for control of COVID-19: a systematic review protocol. BMJ open. 2020;10(8):e040413 10.1136/bmjopen-2020-040413 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Ruktanonchai NW, Floyd J, Lai S, Ruktanonchai CW, Sadilek A, Rente-Lourenco P, et al. Assessing the impact of coordinated COVID-19 exit strategies across Europe. Science. 2020;369(6510):1465–1470. 10.1126/science.abc5096 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Solis J, Franco-Paredes C, Henao-Martínez AF, Krsak M, Zimmer SM. Structural Vulnerability in the US Revealed in Three Waves of COVID-19. The American journal of tropical medicine and hygiene. 2020;103(1):25 10.4269/ajtmh.20-0391 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Althouse BM, Wallace B, Case B, Scarpino SV, Berdahl AM, White ER, et al. The unintended consequences of inconsistent pandemic control policies. arXiv preprint arXiv:200809629. 2020.
  • 27. Tsang TK, Wu P, Lin Y, Lau EH, Leung GM, Cowling BJ. Effect of changing case definitions for COVID-19 on the epidemic curve and transmission parameters in mainland China: a modelling study. The Lancet Public Health. 2020. 10.1016/S2468-2667(20)30089-X [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Auger KA, Shah SS, Richardson T, Hartley D, Hall M, Warniment A, et al. Association between statewide school closure and COVID-19 incidence and mortality in the US. JAMA. 2020. 10.1001/jama.2020.14348 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.R Core Team. R: A Language and Environment for Statistical Computing; 2020. R Foundation for Statistical Computing.
  • 30. Dormann CF, Elith J, Bacher S, Buchmann C, Carl G, Carré G, et al. Collinearity: A Review of Methods to Deal with It and a Simulation Study Evaluating Their Performance. Ecography. 2013;36(1):27–46. 10.1111/j.1600-0587.2012.07348.x [DOI] [Google Scholar]
  • 31. Vatcheva K, Lee M, McCormick Joseph B, Rahbar Mohammad H. Multicollinearity in Regression Analyses Conducted in Epidemiologic Studies. Epidemiology: Open Access. 2016;06(02). 10.4172/2161-1165.1000227 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Gostin LO, Hodge JG, Wiley LF. Presidential Powers and Response to COVID-19. Jama. 2020. 10.1001/jama.2020.4335 [DOI] [PubMed] [Google Scholar]
  • 33.Gupta S, Nguyen T, Rojas FL, Raman S, Lee B, Bento A, et al. Tracking Public and Private Responses to the COVID-19 Epidemic: Evidence from State and Local Government Actions. Cambridge, MA: National Bureau of Economic Research; 2020. w27027.
  • 34.Mervosh S, Lu D, Swales V. See Which States and Cities Have Told Residents to Stay at Home; 2020.
  • 35. Dowd JB, Rotondi V, Adriano L, Brazel DM, Block P, Ding X, et al. Demographic Science Aids in Understanding the Spread and Fatality Rates of COVID-19. medRxiv. 2020. 10.1101/2020.03.15.20036293 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Lonergan M, Chalmers J. Estimates of the Ongoing Need for Social Distancing and Control Measures Post-“lockdown” from Trajectories of COVID-19 Cases and Mortality. medRxiv. 2020. 10.1101/2020.04.26.20080994 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Stier AJ, Berman MG, Bettencourt LMA. COVID-19 Attack Rate Increases with City Size. arXiv:200310376 [physics, q-bio]. 2020.
  • 38. Dalziel BD, Kissler S, Gog JR, Viboud C, Bjørnstad ON, Metcalf CJE, et al. Urbanization and Humidity Shape the Intensity of Influenza Epidemics in U.S. Cities. Science. 2018;362(6410):75–79. 10.1126/science.aat6030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Rader B, Scarpino S, Nande A, Hill A, Dalziel B, Reiner R, et al. Crowding and the Epidemic Intensity of COVID-19 Transmission. medRxiv. 2020. 10.1101/2020.04.15.20064980 [DOI] [Google Scholar]
  • 40. Courtemanche C, Garuccio J, Le A, Pinkston J, Yelowitz A. Strong Social Distancing Measures In The United States Reduced The COVID-19 Growth Rate: Study Evaluates the Impact of Social Distancing Measures on the Growth Rate of Confirmed COVID-19 Cases across the United States. Health Affairs. 2020;39(7):1237–1246. 10.1377/hlthaff.2020.00608 [DOI] [PubMed] [Google Scholar]
  • 41. Du Z, Xu X, Wang L, Fox SJ, Cowling BJ, Galvani AP, et al. Effects of Proactive Social Distancing on COVID-19 Outbreaks in 58 Cities, China. Emerging Infectious Diseases. 2020;26(9):2267–2269. 10.3201/eid2609.201932 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Kantor BN, Kantor J. Nonpharmaceutical Interventions for Pandemic COVID-19: A Cross-Sectional Investigation of US General Public Beliefs, Attitudes, and Actions. medRxiv. 2020. 10.1101/2020.04.26.20078618 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Kaashoek J, Santillana M. COVID-19 Positive Cases, Evidence on the Time Evolution of the Epidemic or an Indicator of Local Testing Capabilities? A Case Study in the United States. arXiv. 2020; p. 14. [Google Scholar]

Decision Letter 0

Jeffrey Shaman

13 Jul 2020

PONE-D-20-13138

State-level variation of initial COVID-19 dynamics in the United States

PLOS ONE

Dear Dr. White,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we have decided that your manuscript does not meet our criteria for publication and must therefore be rejected.

Specifically: the reviewer notes a number of methodological flaws that suggest the findings are biased.

I am sorry that we cannot be more positive on this occasion, but hope that you appreciate the reasons for this decision.

Yours sincerely,

Jeffrey Shaman

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this paper, the authors explore differences in state-level COVID outcomes and policy responses. There have already been a number of papers exploring exactly the same topic (e.g. Courtemanche et al. 2020) and with much better modeling. I know PLOS ONE has a policy that novelty is not a criterion, but what these other papers highlight is that the models in this paper are very likely misspecified.

The authors note the independent variables are highly correlated with one another and only remove them if they are correlated at 0.7 or more. This is still very high collinearity so I'm not sure we can trust the model results. The fact that the authors' note that their results are contrary to those for influenza and for analyses of COVID in other countries, makes me worry that the model is misspecified.

The policy variable is very crude, equal to 1 if policy interventions happened before 25 cases and 0 if they happened after. Why not use the number of days prior to or before some arbitrary threshold? That's the way the IHME model works (https://covid19.healthdata.org).

I'd also like the authors to justify the arbitrary threshold of 25 cases. And since population varies so much across states, why use number of cases rather than percent of population?

The models very likely suffer from omitted variable bias. A better analysis would forego the state level independent variables and instead use variation over time and fixed effects to account for differences between states.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

- - - - -

For journal use only: PONEDEC3

Decision Letter 1

Jeffrey Shaman

21 Aug 2020

PONE-D-20-13138R1

State-level variation of initial COVID-19 dynamics in the United States

PLOS ONE

Dear Dr. White,

Thank you for re-submitting your manuscript to PLOS ONE. We have obtained comments from two reviewers and  feel, based on their reviews, that the manuscript has merit but does yet not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please respond to the reviewer comments on a point-by-point basis and revise the manuscript accordingly.

Please submit your revised manuscript by Oct 05 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Jeffrey Shaman

Academic Editor

PLOS ONE

Journal Requirements:

1) Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2) We note that your manuscript is not formatted using one of PLOS ONE’s accepted file types. Please reattach your manuscript as one of the following file types: .doc, .docx, .rtf, or .tex (accompanied by a .pdf).

If your submission was prepared in LaTex, please submit your manuscript file in PDF format and attach your .tex file as “other.”

3) We note that Figure S3 in your submission contains map images which may be copyrighted.

All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (a) present written permission from the copyright holder to publish this figure specifically under the CC BY 4.0 license, or (b) remove the figure from your submission:

a. You may seek permission from the original copyright holder of Figure S3 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Partly

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: I Don't Know

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: While there seem to be interesting and overall likely well done analyses, there are places where the methods are underspecified to the point where it is unclear if the methods are adequate.

Major points

• While I understand the need for stepwise regression and concerns about correlation. It is unclear how the authors selected variables when there were highly correlated variables. If two variables are highly correlated, it would be likely (although not absolutely true) that they would either both meet stepwise inclusion or neither.

• Please clearly list all candidate predictors (state variables as well as policies) that were considered as part of the stepwise regression.

• Please provide more context for the inclusion of the volunteer rate in the methods. This inclusion seems interesting but the reader doesn’t have context for this variable. Similarly, the tightness score (this comes through a little more in the discussion, but explaining in methods would be helpful)

• Please give more information regarding the gathering limitations. You say “usually to 10 people” but the reality is that almost all states started at 250 or 100, then decreased to 50, then 10, some went to smaller numbers too. So what date are you using? The first date or the date the states went to 10 people?

• The references (and at times the text) need to updated. I believe publishing on the beginning of the pandemic is useful but putting these findings into context of other publications around the same time period is useful. A prior reviewer mentioned the Courtemanche manuscript, there is also the Hsiang manuscript in Nature. This submitted manuscript confirms many of the findings of Auger re schools in JAMA. Thus, I think the concerns raised by prior reviewers that the model is not correctly specified are not founded.

In terms of updating the manuscript, the second sentence discusses the case counts as of April 29th. The reader will be confused by this statement given how different we are currently. I see why you are presenting this data as this is at the end of the study period but you need to first introduce the study period and then give context of the numbers.

Similarly, you need to at least acknowledge the latest phase of the pandemic (lines 40-45) there is another rise in cases/deaths after reopening. I understand that you are not focused on that time period but it is strange to not acknowledge it. In fact you state the numbers have “slowed since mid-March” (line 50) which is simply not true.

• Results—Table 1line 123. You state that demography, education, etc are poor predictors and cite table 1; yet none of these variables are in table1.

• Please put in context the parameter estimates from Table 1 into the text. What does a doubling time parameter estimate of 0.492 for schools mean??? This would be helpful for the entire table.

• It is kind of a big deal that testing wasn’t accounted and is only listed as a limitation. The statement “overall doubling time was not strongly correlated with overall tests rates for each state” line 135 is woefully underspecified. (the text says figure 4, but It appears to refer to figure 5). The reader has no idea what is meant by test kits per capita. Is this the number of tests per people over some time? What time period? Why not just include the daily testing rate as a time varying covariate?

• Figures 3 and 4 would be improved with some sort of correlation coefficient in each panel. It is unclear if all of the panels in figure 4 were considered as candidate covariates. Why aren’t business restriction, stay at home mandates, and number of state actions in the final model? Were they not retained in stepwise? Were they too highly correlated?

Specific points

Abstract

• The abstract states “we examine how commonly used epidemiological metrics differ for each individual state within the US during the initial COVID-19 outbreak. Yet, the manuscript does not report out individual state level results so this sentence should be revised.

Methods

• “The early doubling time should not be severely affected by government interventions s these were rarely implemented that early in the course of the epidemic” (line 65-66) I think the sentiment of this statement is true… but it is likely more that the effects of the policies wouldn’t take place that fast. Several states closed schools and implemented gathering bans when they had very very few cases (e.g. West Virginia closed schools with 0 cases). Given that it would take 10+ days for policy change to impact coronavirus spread (see lag in model derived by Auger, JAMA), I believe this is a reasonable approach; however, you may want to tweak the language because it isn’t that the policies weren’t implemented it is more they wouldn’t have had time to take effect.

Results

• “We found that doubling times for all states increased with time and that heterogeneity between states was reduced” line 109-110. This seems very subjective. It is strange not to quantify this in some way. How much did they change?

Discussion

• The first paragraph of the discussion is a bit long and feels tangential to the main point of the manuscript. Perhaps this could be largely reduced and additional references regarding emerging COVID studies (listed above) could be incorporated. There are some nice additions (or considerations) in this manuscript which haven’t been in other models (tightness, volunteer rate, etc). So highlighting these is appropriate (as you do in lines 170-180).

Supplemental figure 3

This one is a bit confusing to me. You talk about the changes in the doubling time throughout the study period; yet, this is just a single number per state... so when is this from? Early? The average of the study period???

Conclusions

• appropriate

Reviewer #3: Summary

The reviewed paper presents an interesting analysis on how state-level variation can explain the early pandemic dynamics of COVID-19. The analysis appears to be logical and statistically sound. I don’t have any major comments, but have a few minor comments that I think could enhance the manuscript.

Major comments

N/A

Minor comments

The authors note: “We show that early non-pharmaceutical government actions were the most important determinant of epidemic dynamics.” I think this statement can be interpreted in two ways: (1) government actions need to happen early to impact the epidemic dynamics, or (2) the government actions that happened early had the largest impact on subsequent epidemic dynamics. These have two different interpretations in my mind, that the specific timing in the early phase matters or that implementing them early at all matters. I believe the authors mean the latter, but I think it would be useful to clarify, maybe something like this?: “We show that non-pharmaceutical government actions during the early phase of the epidemic were the most important determinant of epidemic dynamics.” One could reference previous research from China that agrees with this conclusion as well (I know it likely wasn’t published when this was written): https://wwwnc.cdc.gov/eid/article/26/9/20-1932_article

Figure 5: The authors note: “The overall doubling time was not strongly correlated with overall tests rates for each state (Fig. 4).”

Did the authors compare with testing rates as a raw number rather than per capita? I wonder if there would be a stronger correlation, as the early pandemic dynamics with small numbers might not have any per capita impact.

In general, I found it unintuitive to follow the doubling time results. E.g. on line 120: “We found that population density, flu vaccination rates, and wealth were all positively correlated with doubling times” I would suggest rephrasing for clarity for these types of results throughout to something like: “Comparing the doubling times across states, we found that increasing population density, flu vaccination rates, and wealth were all associated with slower epidemics” -- I’ll leave it to the authors to decide if they agree

Line 120 indicates that all three variables are positively correlated with doubling times, but the results in Table 1 have different signs for population density from the other two variables, so I think there might be a typo?

Line 173, I’d add that tightness was negatively correlated with doubling time.

Typo in line 180: “We hypothesis this may be the result of people in tight cultures finding it more difficult to adjust their behavior when new rules are imposed.”

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Decision Letter 2

Jeffrey Shaman

1 Oct 2020

State-level variation of initial COVID-19 dynamics in the United States

PONE-D-20-13138R2

Dear Dr. White,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jeffrey Shaman

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Acceptance letter

Jeffrey Shaman

5 Oct 2020

PONE-D-20-13138R2

State-level variation of initial COVID-19 dynamics in the United States

Dear Dr. White:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Jeffrey Shaman

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Fig. COVID-19 cases over time for the US as a whole.

    (TIF)

    Click here for additional data file. (90.8KB, tif)
    S2 Fig. COVID-19 cases over time for each US state.

    (TIF)

    Click here for additional data file. (1.1MB, tif)
    S3 Fig. Map of COVID-19 case doubling time (first three weeks since 25 cases) for each US state.

    (TIF)

    Click here for additional data file. (323KB, tif)
    S4 Fig. Residual diagnostic plots for the best fitting models for both the early (first 7 days) and overall (first 21 days) doubling times.

    (TIFF)

    Click here for additional data file. (568.7KB, tiff)
    Attachment

    Submitted filename: White2020 Response_to_PLoSONE_review.pdf

    Click here for additional data file. (251.7KB, pdf)
    Attachment

    Submitted filename: Response To Reviewers.pdf

    Click here for additional data file. (85.6KB, pdf)

    Data Availability Statement

    All code and corresponding data is freely available at \url{https://github.com/eastonwhite/COVID19_US_States}. The original raw data has been compiled by the Johns Hopkins University Center for Systems Science and Engineering at (\url{https://github.com/CSSEGISandData/COVID-19}).

    Articles from PLoS ONE are provided here courtesy of PLOS

    RESOURCES