- I noticed a little glitch in your first example: in theory, Covid patients of December could explain additional deaths in January. So I checked and it turns out that you were right anyway, but there's still the matter of interpretation. If I see it correctly, the Promedica "Residents Total COVID-19 Deaths…
- I noticed a little glitch in your first example: in theory, Covid patients of December could explain additional deaths in January. So I checked and it turns out that you were right anyway, but there's still the matter of interpretation. If I see it correctly, the Promedica "Residents Total COVID-19 Deaths" count stood at 84 by the end of February. But the "Residents Total Confirmed COVID-19" count stood at only 69 at that time (both from the very beginning). In early March the Confirmed Covid number increased to 70 but the Covid death count reached 93 in April, therewith marking the end of a Covid wave. I guess that they didn't confirm all residents with suspected Covid by means of a test. In conclusion, those data may be quite correct on themselves but they are obviously unfit for usual statistical calculations such as taking their ratio.
- Out of curiosity I also looked at your third example, and yes that's really weird. Apparently, Iron River Care Center went from zero weekly deaths in May 2020 to 70 weekly deaths in October, peaking to 171 in November - but still with not even 10 weekly Covid deaths and with only around 45 occupied beds. In January 2021 their weekly deaths dropped back to mostly 0. Mind boggling!
Very interesting Steve, thanks!
- I noticed a little glitch in your first example: in theory, Covid patients of December could explain additional deaths in January. So I checked and it turns out that you were right anyway, but there's still the matter of interpretation. If I see it correctly, the Promedica "Residents Total COVID-19 Deaths" count stood at 84 by the end of February. But the "Residents Total Confirmed COVID-19" count stood at only 69 at that time (both from the very beginning). In early March the Confirmed Covid number increased to 70 but the Covid death count reached 93 in April, therewith marking the end of a Covid wave. I guess that they didn't confirm all residents with suspected Covid by means of a test. In conclusion, those data may be quite correct on themselves but they are obviously unfit for usual statistical calculations such as taking their ratio.
- Out of curiosity I also looked at your third example, and yes that's really weird. Apparently, Iron River Care Center went from zero weekly deaths in May 2020 to 70 weekly deaths in October, peaking to 171 in November - but still with not even 10 weekly Covid deaths and with only around 45 occupied beds. In January 2021 their weekly deaths dropped back to mostly 0. Mind boggling!