key: cord-0849894-dl01vbml
authors: Yuan, H.-Y.; Wu, L.; Wang, D.-P.
title: Surges in COVID-19 are led by lax government interventions in initial outbreaks
date: 2020-07-23
journal: nan
DOI: 10.1101/2020.07.17.20156604
sha: 904d8a23b7bacbcaf741cb92410ab8f4538acc15
doc_id: 849894
cord_uid: dl01vbml

Sharp increases in COVID-19 cases occurred after reopening in the United States. We show that the post-intervention effective reproduction number is a strong predictor of the surge in late June. Lax interventions in the early stages coupled with elevated virus spread are primarily responsible for surges in most affected states.

We obtained the daily new cases from "The COVID Tracking Project", https://covidtracking.com/, for the fourteen states of among the highest total cases in the United States. The study period is from the date of the first reported cases in each state up to July 6. The total cases () Ct tend to increase exponentially with a growth rate such that the number of daily new cases is CC  = . The growth rate is a measure of the percentage increase of daily new cases, commonly used to describe the virus spread (1) . To capture the current outbursts, we calculated the two-week averaged growth rates for each state using data from the last 14 days of records (Fig. 1A) . Arizona, Florida, and Texas stood out with the highest growth rates > 0.04 days -1 . California, North Carolina, Georgia, and Louisiana also had fast growth rates. In contrast, New York, New Jersey, and Massachusetts, had the lowest growth rates < 0.002 days -1 .

To explain the pattern of observed growth rates, a susceptible-infectious-recovered (SIR) model is In the United States, each state has adopted a different policy towards lockdowns and reopening. In many states, the surges have often been attributed to early reopening. Arizona, Florida, Georgia, Louisiana, North Carolina, and Texas have all opened up (long) before the late May. The e R above or near 1 in early June could result from premature relaxation of the lockdowns. Most of these states moreover have experienced 2-3 folds increases in growth rates in the month of June alone; the crowded bars and beaches may have been a key factor. The current crisis is thus a result of preexisting conditions due to relaxed initial interventions intensified by elevated virus spread rates since June.

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 23, 2020. . https://doi.org/10.1101/2020.07.17.20156604 doi: medRxiv preprint California presents a seemingly contradictory case. The state didn't open up till late May; yet, e R was already above 1 in early June. After the reopening, the growth rate in California had only increased slightly. The surge thus is primarily associated with a persistent upward trend. At a growth rate of  =0.025 days -1 in early June (Fig. 1A) , the daily counts would double in about a month. Indeed, the weekly averages had increased from about 2,600 in late May to an alarming 5,700 in late June. The current spike reflects a failure of initial interventions in fully containing the outbreak. North Carolina has a similar situation of an upward trend and no increased growth rate. Illinois and Maryland where e R in early June were close to 1, offer another interesting case (Fig. 1B) . The growth rates decreased considerably after the reopening, and both states have (so far) avoided a surge.

The post-intervention effective reproduction number provides a general measure of intervention outcomes, and is a robust indicator of the current surges. While the early reopening intensified the current crisis, it may not be the critical driver. To prevent resurgence, it is important to understand why in certain states, lockdowns did not lead to the anticipated effects. We suggest that behavioral factors related to compliance with public health measures, such as social distancing and face coverings, must be carefully considered (5, 6) .

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 23, 2020. . . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 23, 2020. . https://doi.org/10.1101/2020.07.17.20156604 doi: medRxiv preprint

The basic reproductive number of Ebola and the effects of public health measures: the cases of Congo and Uganda

Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus

Forecasting seasonal outbreaks of influenza

The impact of COVID-19 and strategies for mitigation and suppression in low-and middleincome countries

The effect of control strategies to reduce social mixing on outcomes of the COVID-19 epidemic in Wuhan, China: a modelling study

Identifying airborne transmission as the dominant route for the spread of COVID-19