Top news, reports and insights for today:
- Daily headline summaries for Saturday:
- Alabama sees new cases, hospitalizations and test positivity rates on the rise amid reopening (PBS Newshour)
- Good news on the treatment front: a peer-reviewed study from a randomized clinical trial of remdesivir shows that while survival was not improved, COVID-19 patients in the treatment group had a shorter recovery time compared to placebo, making this the first drug to be shown to be effective as a treatment (New England Journal of Medicine)
- A new poll from AP/NORC shows only half of Americans are ready to resume pre-epidemic activities like going to the movies or a sporting event. This helps explain why allowing businesses to reopen has not translated into widespread resumption of business activity. From the article: “Americans fear spreading or contracting infection, so much so that they’ve overwhelmingly participated in social distancing measures. They tell pollsters by wide margins that they fear lifting those restrictions too soon much more so than too late. ” (Vox)
- Leading experts believe the CDC’s new lowered projections for hospitalizations and deaths are too optimistic, signaling that politics may be driving the science, rather than the reverse (NPR)
- U.S. Cases remain flat, continuing an overall whack-a-mole pattern
On Thursday and Friday, May 20-21, the U.S. reported about 49,000 new COVID-19 cases. One month ago, on April 21-22, there were 55,000 new cases. On the same 2 days in March there were 14,000 new cases. While cases may have peaked in the U.S. for now, the epidemic is still generating approximately the same number of new daily cases as it did last month. The graph below shows a generally flat trend in new cases since May 11. There were 155,836 new cases reported in the last week, rise in cumulative cases of 10.9%. If the U.S. were to continue to increase cases by 10% a week for the next month, there would be 2.5 million cases by June 21.
The lower graph shows how cases are growing by state over the last week. Twelve states saw cases grow by 10% or less last week. However, 38 states had more than 10% growth and 4 saw growth of greater than 25%, equivalent to a doubling of cases in 4 weeks (Minnesota, North Dakota, Arkansas and North Carolina).
Bottom line: These graphs basically say that while the news continues to be good in the states that suffered the worst initial waves (New York, New Jersey, Louisiana and Michigan), the situation is best characterized as continued whack-a-mole, not a general waning of the epidemic at a national level.
- Modeling update: The spread of coronavirus has not slowed in 24 states
The modelers have been taking it on the chin in recent weeks, with numerous headlines detailing their ups and downs. As today’s Top Pick of the Day reports, modelers from the Imperial College of London are now projecting that COVID-19 may be spreading at epidemic rates in 24 states, particularly in the South and Midwest. This is based on the latest version of their model, described in Report 23, which now includes anonymous cell phone data capturing changes in people’s movement patterns. Their model shows an effective reproduction number (Rt) of greater than 1 in Texas, Arizona, Illinois, Colorado, Ohio, Minnesota, Indiana, Iowa, Alabama, Wisconsin, Mississippi, Tennessee, Florida, Virginia, New Mexico, Missouri, Delaware, South Carolina, Massachusetts, North Carolina, California, Pennsylvania, Louisiana and Maryland. Nationally, they estimate that about 4.1% of American’s have so far been infected; it’s about 17% in New York. That means even New York is far from herd immunity. Their report says “…the epidemic is not under control in much of the US”. The figure below was taken from this report; it shows the probability that each state has an Rt of less than 1, meaning the epidemic is controlled. Green states have a high likelihood, purple states are low. These data don’t support the idea that the epidemic is going away with warm weather. The hot spots (with the possible exception of Minnesota and Iowa), are, well hot.
What does this mean? The debate continues to rage about the utility of these models. The Imperial College model has proven to be among the best performing models we have. The addition of real-time travel data has the potential to enhance it’s performance. If correct, their model predicts that if travel continues to increase in some states at the current rate, cumulative deaths may double over the next two months.