September 3 COVID Update: Testing Rates Continue Falling and Midwest Sees Cases Grow

The following summary and its attached report are the twentieth in the ongoing series by the W. Capra Data Science team on the impact of the Covid-19 outbreak on the industries we service over time.  The previous reports can be found in the following links: April 14, April 23, April 30, May 6, May 13, May 20, May 28, June 4, June 11, June 18, June 26, July 2, July 9, July 16, July 23, July 31, August 7, August 14, and August 21.


Using data current as of September 2nd, this report compiles updated results for testing, cases, and deaths, and categorizations of each state’s current circumstances. This summary will highlight some of the findings and conclusions for the past week. Additionally, the summary will examine business impacts, both domestically and internationally, that can inform the US of its possible outcomes and timescales, considering the trends seen today.

A note before we begin: This outbreak and the data surrounding it changes daily. This report was created when looking at the outbreak as a data problem that might benefit from data-driven solutions and insights. It is not intended to be a substitute for medical or safety advice, nor is it a recommendation on outbreak response currently in place in various locations around the country. Individual assessment of local laws and current official government and health guidance should be reviewed before making any decisions.


24.19% of the entire US population has been tested. This number only represents the total number of tests administered, so the actual testing rate is lower as some individuals have been tested more than once. The distribution of these tests is also unevenly distributed – states around the US, and especially in the Northeast, are leading the country with testing rates.  Expanding testing nationally is important as more testing results in more information about the impact and dangers of COVID.  More tests mean the sampling bias of the death rate is lowered, which is skewed by symptomatic patients.  Case rates around the country also vary widely, with a few states reaching greater than a 15% case rate over a two-week rolling period. The case rates in states like Mississippi – which have a very high positivity rate but declining testing rate – shows the potential of sampling bias as a much higher percentage of tests are positive which indicates testing going to symptomatic persons while access has been reduced for those potentially asymptomatic. Looking at the most recent two weeks allows for a more accurate view of how states are doing now, compared to over the course of the pandemic.  Case Fatality rates, or the rate of positive cases that result in morbidity, are now being reported with a two-week lag on the moving average of new cases.  The lag-adjusted Case Fatality rate has remained at 1.8% for the US since the last report. This case-fatality rate is for all discovered cases and does not include an estimate for cases undiscovered, so the true fatality rate is likely lower. The US appeared to have peaked with new cases and deaths sometime in April followed by a period of decline in both statistics. However, in late May and early June, cases began increasing rapidly again across the country, with the number of daily new cases in the US passing the previous peak in new cases from April.  Now, the case curve has begun to decrease again as more states regain control of the virus.  The death curve, after the typical lag period, began increasing as well – albeit at a slower rate than the initial peak and the second wave of cases – but appears to have flattened and begun declining over the past week.  The decrease in the case curve and the death curve will be updated as the weeks progress.  There had also been a continual increase in testing throughout the pandemic until mid-July.  Now, testing has decreased slightly over the past few weeks.  It was hypothesized that the decrease in testing may be attributed to high testing states like FL closing some testing sites in response to Hurricane Isaias, but the testing decreases have continued.  Decreasing the amount of testing has a negative effect as test growth will approach true population rates.

The acceleration or deceleration of cases is used to classify where each state is regarding the outbreak situation.  The situation looked dire for many states in June and early July but has seen some improvement in the past few weeks.  Many states have shown that they are moving away from experiencing a rapid spread of the virus and have flattened their growth curves.  At the beginning of reopening, only a few states were in the worst of four classifications – Exponential Growth – which is a positive new case velocity with a positive new case acceleration.  In June and July, nearly the entirety of the South and the West was in the exponential growth stage with many in the Midwest experiencing similar situations.  In the last month, most of the states have moved to the Linear growth stage – positive new case velocity but near-zero case acceleration – and many of those have continued into the Improvement stage. In the last two weeks, most of the South and West have made the transition to Improvement, but the Midwest has again mostly reverted to Exponential growth, with only IL, WI, and MI remaining in the Linear stage – a worrying trend. Overall, there are now 15 total states in the exponential stage, up from 8 in the last report.  Linear growth now accounts for 10 states, down from 15 in the last report.  Overall, that means that 25 states are not improving their outbreak situations, a number increasing from 24 states in the previous report which is a worrying trend.  The other two classifications, Improving and Contained, account for 19 and 6 states, respectively. This means that the Improving stage still has the most states, a very positive trend. The Northeast was the only region where these two stages were despite being the original hotspot of the virus, but states across the country including hotspots like FL and LA have entered the improvement stage as well. To read an unabridged version of the results, please see the attached report for a complete view of specific states.


The United States’ death curve appears to have begun declining and the acceleration rate of new cases has continued to decrease.  At the beginning of this new period of case increases, southern states were struggling with significant case growth while many northern states continued to improve, reaching a point where every state across the US besides those in the Northeast were seeing negative trends.  Most states in the past two weeks have begun to decrease their number of cases.  Many states, including hotspots CA and TX, have also begun to improve, with some hard-hit states like FL and LA reaching the improvement stage in the past week as well. However, a large portion of the central US – from MT to OK to VA – have begun seeing negative trends indicating a further spread of the virus.   Lockdowns being lifted and a decreasing adherence to health guidelines have contributed to the struggles of containing the virus, causing some states to remain in a stage of growth or return to a stage of growth of the outbreak.  There had been an increase in deaths in late June and July, which follows the lag from increasing cases, but deaths have flattened as well.  Indications from new case velocities and accelerations point to longer recovery periods than those of rapid acceleration experienced in March and April.  This trend is in line with what other countries across the world experienced with COVID-19. Overall, the effects of reduced mobility from lockdowns and social distancing measures continue to be strongly correlated with the deceleration of new cases.

Many states are rapidly decelerating their new daily cases, giving relief to the hardest-hit states in the South and West US.  With the deceleration in cases, however, testing has also decelerated which may explain some of the decreases in new case discovery. Additionally, despite the huge increase in peak cases when compared to the first peak in cases from April, it does not appear that deaths will peak anywhere near the previous percentage.  The number of deaths is near 50 percent of the peak in April, while cases have more than doubled, with the death rate near two percent over the recent wave compared to seven percent in April.  While many states are beginning to improve, some states continue to struggle.  HI has become a hotspot with its magnitude above that from April with no signs of slowing.  Monitoring these trends will be important as all states have reopened their economies to varying degrees with increasing consumer activity outside of the home. 

The Business Environment

Globally, the business environment remains in a precarious position.  As of this past week, all G7 countries have officially entered a recession.  The seven countries – Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States have reported economic contractions of varying degrees over the course of the first half of 2020.  The pandemic has affected each country differently – with differing degrees of decline and combatted with differing strategies.  The United Kingdom suffered the worst, with an economic decline of over 20% in the second quarter or 2020. Countries with harder lockdowns and restrictions saw sharper declines in the second quarter of 2020, but as seen in the United States the softer restrictions may lead to a weaker third quarter as the pandemic continues and some regions need to reimpose restrictions due to spikes.  Currently, the US economy is operating at 78% capacity compared to early March and the continuation of the pandemic is causing some to expect the economy to recover by late 2022.  The years following the latest recession post-2008 saw half-speed growth but had strong stock-market rebounds.  The COVID recession is closely tied to the pandemic’s trajectory with a strong stock-market rebound.  However, differences such as the massive decline in output and human capital coupled with the loss of small businesses and permanent changes such as the digital transformation of many economic sectors make the new recession difficult to forecast. The recovery is dependent on consumer confidence and the development of therapies to combat the virus, but also on the policies pursued in response.  

To restart these habits and help the economy, more than just lifting lockdowns is necessary.  There is no significant difference in employment and consumer spending between US states that maintained longer lockdowns versus those that reopened early according to McKinsey.  The difference in lockdown restrictions only accounted for seven percentage points of the average 60 percent drop in consumer traffic.  Most of the drop in spending is due to uncertainty – including from job prospects – even for items in regular, everyday life.  To turn around and restart the economy, certainty in virus control and confidence in discretionary mobility shows how consumers have reengaged in commercial activity.

For further discussion of data modeling or anticipated COVID-19 business impacts, contact the W. Capra Data Science team:

Nate at [email protected]

Stu at [email protected]


Data from The New York Times, based on reports from state and local health agencies. (2020, April 6). Retrieved April 3, 2020, from

The COVID Tracking Project. (2020, April 6). Retrieved April 3, 2020, from

COVID-19 Community Mobility Reports, Google,

Annual Estimates of the Resident Population for Counties in the United States, U.S. Census Bureau,

Boskin, Michael J. “How Does the COVID Recession Compare?” World Economic Forum, August 28, 2020.

Charumilind, Sarun, Ezra Greenberg, Jessica Lamb, and Shubham Singhal. “COVID-19: Saving Thousands of Lives and Trillions in Livelihoods.” McKinsey & Company. McKinsey & Company, August 21, 2020.

Horowitz, Julia. “Covid-19 Dealt a Shock to the World’s Top Economies. Here’s Who Has Fared the Worst.” CNN. Cable News Network, August 28, 2020.

Laeven, Luc. “COVID-19 and the Effects of Social Distancing on the Economy.” VOX, CEPR Policy Portal, August 31, 2020.

Verified by MonsterInsights