For the second consecutive Saturday, King County Public Health has not updated its COVID-19 “daily summary” dashboard, so we have no roundup tonight. As we did last week, we’ll include the West Seattle trends – which we’ve been updating on Saturdays – in the next roundup.
The data quality generated by Washington State has been so poor compared to other states that I’ve tracked. You would think it would get better by now.
Yeah, no kidding. DOH has been so far behind in the reporting of negative test results that the last “complete” date is Feb 1. In addition to that, every time I view the state dashboard, there is always a notice about that there may be 500, or 700 as it was today, duplicates. If our state reports around 1000 positives per day, an estimate of 700 possible duplicate makes the data utterly and completely useless. So are we talking about about 1000 new cases, or 500 or 300? That is quite the range, especially since public policy is based on this very data that DOH admits is flawed. This ineptitude in data processing is not normal and shouldn’t be acceptable.
I don’t believe the 700 is a daily thing. That alert was posted some days back and hasn’t been changed/updated. The KC dashboard lets old “alerts” linger too.
Oh, I know it’s not a daily thing. It seems more like a weekly thing. DOH does update the number of possible duplicates quite frequently. It started out at 200 at the beginning of the week if I remember correctly and has been going up since then. What is more infuriating is the continued backlog in reporting negative test results, which has been an ongoing problem since Thanksgiving. The positivity rate is used as a key metric to gauge the trajectory of the epidemic, and I don’t know of another state that has such problems providing this critical data. The last “complete” date for negative test results is three weeks ago. It doesn’t take a data scientist to realize that this is a a serious problem.
I agree. Last week I asked the question of what the positive rate number is based on. I’m beginning to really question any number coming from the state. Scary policy is based on these numbers.
Every state has been doing data adjustments. We in West Seattle are fortunate to have a news source that gives us daily updates on COVID numbers so we notice the adjustments every time they happen. Texas’s testing numbers dropped 85,000 from previous counts yesterday. A big surge because of reporting delays after the storm would make sense, but the correction went the other way so who knows. Wyoming had a drop of 100,000 tests from their count one day a few weeks ago, when duplicates were removed. Alaska, Idaho, and Wyoming all had days last week where they didn’t update their state dashboards. Ohio added ~1100 extra deaths to their count at one point and ~2500 at another because of data reconciliation. California has done a number of data corrections changing their positive cases by thousands. It’s happening in every state, and it’s probably normal to correct data when duplicates are found, etc. for other records (flu deaths, people affected by wind storms, etc.), but we just don’t have a reason to look at those numbers all the time and don’t see the corrections when we’re looking at snapshots.
A note about the vaccine: the Senior Center of West Seattle was very helpful in finding me an appointment for the vaccine in West Seattle. If you are a senior struggling to get a vaccination, or know a senior who needs assistance, you might want to register with them. They found me a spot within a couple of days, for which I am super grateful.
Sorry, comment time is over.
| 7 COMMENTS