Has the stay-at-home order affected the COVID-19 growth rate?
BY LEN AND JENNIFER CABRERA
When Alachua County issued a stay-at-home order on March 23, the County Commission feared that coronavirus was an existential threat, a once-a-century pandemic that would see exponential rates of infection and death.
It’s been two weeks since the stay-at-home order, and we haven’t seen either exponential growth or any discernible difference in the rate of new cases since before the order was issued. The general understanding of the progression of COVID-19 is that new cases take up to 14 days to show up, so results of any change should show up around 14 days later—but we haven’t seen any change at all.
Looking at data from 3/19 to 4/7 shows a linear relationship in the total number of COVID-19 cases in Alachua County.
The chart starts on 3/19 because that was the first day the Florida Department of Health reported the number of positive, negative, and pending tests by county per day. For the analysis, we only considered the number of tests each day that had a positive or negative result, assuming pending results would be included in future days.
The graph for the total number of cases looks very similar to the graph for the total number of tests. There is no exponential growth in total tests, which could be because there are only a limited number of tests, but the results of those tests do not suggest exponential growth in COVID-19.
Presumably COVID-19 tests are only given to those mostly likely to have the disease. (To our knowledge, Alachua County has not instituted randomized testing to identify the rate of asymptomatic infections.) If COVID-19 were in fact growing exponentially, not only should we see an accelerating number of cases, we should also see a higher percentage of positive results from the limited testing resources. That’s not what the data show.
The graph below shows that new cases by day are not increasing at an increasing rate as they would if there was exponential growth. Despite the slightly positive trend line, the regression model slope is not statistically different from zero (p-value is 0.36). There is also no discontinuity in the graph, suggesting the stay-at-home order had no effect on the number of cases.
The graph below showing the percentage of positive tests per day also has no significant trend over time. Taken day-by-day, the average positive test rate is 5.5%. This is about the same as the overall rate computed by dividing total cases by total positive tests (5.3%).
During the March 24 online County Commission discussion, Commissioner Ken Cornell said the cases were doubling every two days, which would lead to 8000 cases after 20 days. He said he hoped that the stay-at-home restrictions would lead to a doubling only every 4 days, or 2000 cases after 20 days. We don’t know when his 20 days started, but 28 days after the first case was reported, there are 139 cases in Alachua County, and our rate of new cases has been consistently linear, not exponential.
The bottom line is that there has been no statistically-significant change in the number of positive tests per day or the percentage of positive tests per day since the first testing data became available from the state on March 19.
Excellent analysis – especially love the graphs and showing that it is not growing exponentially. Keep up the GREAT work. Sue
This clear evidence that the State, BOCC and City have taken down the local jobs and economy for no reason.
To the contrary. Compare Alachua county growth rates with other counties with similar exposure and no lock down. They’re growth rates are increasing at larger rates. That is a the counterfactual. That fact that growth rates are stable is nothing but great news!
What is your example? Miami is Florida’s worst case, but their shelter-in-place came 24 hours after Alachua County’s.
Your analysis doesn’t tell us whether the measures are changing the natural growth rate of the disease or not. First of all you need to have a counterfactual or comparison point. Just by looking daily at the numbers in the map of the Fl dept of health one can see that Alachua county is now at par with the least infected counties in the whole state. Starting this month Alachua was more comparable to Southern Florida in terms of per capita cases. (I don’t have the data or the time right now to do the full analysis, but you could try it yourself). Second, as everyone has shown, the confirmed cases don’t tell us much about the real spread of the disease. More accurate measures, at least about the severity of it, are related to hospitalizations and deaths. Third, it would be important to control for number of tests performed.
And threaten to tighten even more, now requiring the wearing of masks and threatening a curfew and closure of boat ramps and golf courses. Does the Chinese virus become more virulent at a certain hour? Do you go boating or golfing with strangers?
In a Feb 28 article in the NEJM, Anthony Fauci back-tracked on the Covid-19 pandemic to claim it was no worse than a seasonal Flu. Mainstream Media has not reported this story, yet the State of Florida remains in lock-down and small businesses and their employees are shut-down as if they were in Communist China.
” If one assumes that the number of asymptomatic or minimally symptomatic cases is several times as high as the number of reported cases, the case fatality rate may be considerably less than 1%. This suggests that the overall clinical consequences of Covid-19 may ultimately be more akin to those of a severe seasonal influenza (which has a case fatality rate of approximately 0.1%) or a pandemic influenza (similar to those in 1957 and 1968) rather than a disease similar to SARS or MERS, which have had case fatality rates of 9 to 10% and 36%, respectively.”
The case of lockdown vs no lockdown is a false dichotomy. While facing a disease with unknown mortality rates, we are seeing convincing evidence that it can be devastating, which merits a precautionary principle. However, as more evidence is collected and our ability to monitor the infection increases, measures could be gradually loosened. this requires having the means to confidently evaluate infection growth rates and be ready to go back to stronger measures of the rates start growing again.