As a user, I want to know the probability that the recent hospitalization numbers reported by OC HCA are accurate.

[klenwell]

Trello Card: https://trello.com/c/bDfRT5qP

Ended up not doing as deep a statistical analysis as I planned. Really just put together a table and eye-balled it. Find the table here:

• docs.google.com/spreadsheets

A couple interesting things turned up:

• Administered Test-Positive Rate (TPR) has been much lower (~2%) than Reported Test-Positive Rate suggesting that reported test results are skewing older and more positive. Are delays in test reporting getting even longer?
• There has been a marked increase in Skilled Nursing Facility (SNF) cases in last week as hospitalizations and administered TPR have declined, including the largest single one-day rise ever (. I would expect a proportionate decrease
• The decline in hospitalizations does track with the decline in cases, I think. Is OC hiding hospital cases as SNF? Maybe. But not enough to really affect the narrative of the epidemic in OC right now, which is: curve flattened at an elevated plateau after June surge.

For more information, see these Reddit comments:

• https://www.reddit.com/r/CoronavirusOC/comments/i18lwu/new_case_reports_are_down_but_testpositive_rate/g00ssqe/
• https://www.reddit.com/r/CoronavirusOC/comments/i18lwu/new_case_reports_are_down_but_testpositive_rate/g024e2s/

[klenwell]

Note: Extra data files. I believe because I forgot to push develop here to origin after merging master into it.