Read on Twitter
1/ I was going to wait until Friday to post updated data, but since @Allegheny_Co (PA) officials will be speaking later today about the county's situation, I thought I'd post now in hopes of providing context for the alarmist reports of increased covid19 cases (thread).
2/ Context is *critical* to understand for reporting media, for the public, and for policy-makers. To be blunt (and obvious) the media needs to report it; the public needs to understand it; and policy-makers need to use it.
3/ These data are drawn from the @HealthAllegheny public website and are current as of this morning 7/7/20. Again, the numbers move around quite a bit vs daily press releases; and we know that's from data cleaning that allows a firmer date to be assigned to + tests, deaths, etc.
4/ The data represent 7-day averages and dates range from 3/11/20-6/30/20 (hospitalization data go back only as far as 4/9). Data from the past week are preliminary, according to @HealthAllegheny and will likely change substantially - that's why these chart data only go to 6/30.
5/ # Tests administered and # confirmed positive covid19 tests. Testing is still way, way up (e.g., 6/29 recorded over 2,000 tests; 6/30 recorded nearly 2,000 tests). Positive cases increased with increased testing but not nearly at the same rate.
6/ It is very important to ask: Is the county chasing "hotspots" where positive tests are more likely to be returned? Is contact tracing responsible? What is the % of symptomatic vs asymptomatic cases? These are additional, critical pieces of the puzzle.
7/ # confirmed cases and # hospitalized. Hospitalizations are up slightly, an average of 1-2 new admits/day in the past 2 weeks (vs prior 2 weeks). Hospitals confirm ample capacity even though there is no evidence, so far, that increased # cases mean crazy high hospital admits
8/ # confirmed cases and # deaths. Despite the fact that confirmed cases have "surged", deaths have remained low (almost non-existent) and flat for some time. Deaths peaked in mid-April with an average of ~4/day for the week of 4/15 and have decreased with each passing week.
9/ # hospitalizations and # deaths. Numbers are very low (note y-axis). Hospitalizations are often seen as a more proximal indicator of impending death than case counts. But there is no indication that the county's slight hospitalization bump is a harbinger of a bump in death.
10/ So, despite the alarm that reports of increased cases seem to generate, there is no reason for these adverse reactions if those case numbers are placed in the proper context. @Allegheny_Co is healthy and there continues to be *significant* reason for optimism.
11/ I hope that policy-makers can share in this optimism and act accordingly and with these contextualized data in mind.
