There are several intriguing data points in tonight’s evening update from the Florida Department of Health (FDOH). Before we jump into the numbers, a word of caution: these datasets are small, so identifying trends is an endeavor fraught with error. Just because we get a spike or decline doesn’t mean we have a trend. Even so, we can look at one set of data, such as testing or new cases, and compare it with what other data points are telling us, to see if there’s any consistency or patterns that might indicate a trend.
We should also note that we have identified some relatively minor inconsistencies in data provided by FDOH, particularly as it relates to the difference between Florida residents versus non-Florida residents testing positive for the virus. To keep our data as consistent as possible, we pull it from the evening report provided every day at 7pm by FDOH, and compare it with the previous evening report to determine any changes in the data. One known difference between our data analysis and some FDOH data is that FDOH case numbers reported in certain sections of the report are totaled for the 24 hour period between 12:00am and 11:59pm, while our data typically runs roughly from 7:01pm to 7:00pm each day, accounting for some slight differences in the tally between days. However, all of the total case and testing data matches closely with that reported by FDOH.
With all that out of the way, let’s jump into the report.
Monday night, Florida’s total confirmed coronavirus cases climbed to 5,704, but the numbers are consistent with expanded testing, or in today’s specific case, a slight decline in testing levels versus the day before. Whether that’s due to fewer people seeking tests, or fewer tests available is not known. Still, more than 13,700 new tests were completed in the last 48 hours, while the number of new cases, just 715, fell 158 short of Sunday’s record of 912 new cases – again, that slight decline is commensurate with the slight decline in testing.
If there’s bad news to be found, it’s in the positive test rate, which is among the highest we’ve seen so far: 12.42% of all statewide tests were positive on Monday (FDOH says the rate was actually 13%, due to the fact that the FDOH reporting “day” is different from the numbers provided on their evening report, which is posted 5 hours before the day actually ends). Here’s a look at daily testing and daily new cases.
The more important and significantly more reliable statistic, hospitalizations, jumped to 715 after 82 more patients were admitted today (versus an average of 65 the previous two days). But despite the slight increase, even hospitalizations are trending below the level that steady exponential growth suggests it should be, an indication that Florida’s hospitalization growth rate may be slowing down. Don’t misunderstand, though. A slowing growth rate is still growing. It’s just growing less fast over time:
As we mentioned at the top of this post, this is a very small dataset, so taken alone, may not mean much, except that daily hospitalizations are not obviously growing at a steady exponential rate. It’s possible we could get a spike in cases that would put the growth back on track, but for now at least, the numbers are coming in lower than expected, and that’s a very good thing.
There’s another bit of data out there worth looking at, which reinforces our analysis that we are trending below the expected level for hospitalizations. A widely cited study by the University of Washington’s Institute for Health Metrics and Evaluation (IHME) projects that Florida will peak at around 16,000 hospitalizations. That’s a big, scary number. The good news is that we’re currently far below where IMHE thinks we’re supposed to be already. Remember, we only have 715 hospitalized for coronavirus in Florida right now. But IHME projected we would already have more than twice that many hospitalizations (1,695) by now:
Keep an eye out tomorrow for the hospitalization data. A number of around 85 new admissions means growth is remaining flat over yesterday. Anything above that figure indicates some level of exponential growth, while a number lower than 85 means we’ve trended downward for three straight days – a welcome slowdown.
Brian, thanks for the data analysis on the COVID-19 outbreak in Florida. It is the only report I have seen that makes sense of what is happening here.
142 new hospitalizations today. Not slowing down.