New York City: have we hit the peak?

This guest blog is thanks to Salt Hill colleague Olivia Hebner of Summit Consulting.  I am going to add a caveat -- the NYC Dept of Public Health data, shown below, is very different from the NY Times data, for which the leveling off is unclear (though the increase has at least slowed).  This caveat is in no way reflective of Olivia's work -- she was very careful to use what we both agreed was the best data at the moment.  However, all the data right now comes with a lot of questions as to reliability.

Today we add another important data stream to the coronavirus analysis – daily hospitalizations.  One of the biggest fears is that hospitals will be quickly inundated as cases rise.  New York City hospitals are already at or near capacity in terms of both beds and ventilators.

Have we hit the peak?  The good news is that, if measured by new hospitalizations, we appear to have hit the peak.  This means new hospitalizations do not appear to be increasing exponentially anymore. The bad news is that as long as more people are getting hospitalized than are leaving the hospital, we will not hit the peak of total people hospitalized and therefore, we will still need more beds and ventilators each day. 

In the graph below, we track New York City daily deaths and new hospitalizations, and, as before, we put it on a log scale.  That means a straight upward sloping line is an exponential increase.  Since these are new deaths and hospitalizations rather than totals, a flattening of the line does not mean more people are not dying or being hospitalized, it only means those numbers are no longer increasing. 

 The data comes from the NYC Department of Health and Mental Hygiene (DOHMH) and is available here.  The graph shows a flattening of new hospitalizations as of a few days ago.  We seem to be near the peak in terms of deaths as well, although if we are, this is a more recent development. Unfortunately, from our understanding of the data in Italy and China, we will likely stay at the peak level of daily deaths for around 2 weeks, as a high percentage of long-term hospitalized patients do not recover.  We know less about hospitalizations, but will continue to research them and report as data becomes available.