search

episphere / epitracker

epidemiology data tracker with real-time interactive visual analytics
0 stars 0 forks source link

Handle Connecticut's new counties #305

Closed LKMason closed 4 days ago

LKMason commented 5 months ago

Connecticut!!!! 😡🤌

Connecticut has changed its counties to county-equivalents in the census data, but the mortality data hasn't caught up yet. More details here. So we need to treat Connecticut as a special case in the data processing.

The new counties do NOT map cleanly to the old counties. I am not quite sure what to do about this. CDC Wonder seems to provide the 2021 population counts for these counties when you aggregate over 2021-2022 (for example), and provides no population counts or rates when you select only 2021. Perhaps we should consider just using the old population counts? Or we could try the approximate mapping from here? Or some combination of both?

Right now the data is missing for 2020-2022 due to the updated dataset I used. But we should in practice have counts for the old counties up to and including 2021.

Perhaps NCHS will update their county definitions next year? We need a long term plan if not.

LKMason commented 4 days ago

For Connecticut, we now just use the most recent available population data.