Continuing analysis - Githubissues

trangdata commented 3 years ago

Re: demographics

[x] more detailed age and race breakdown for examining these potential risk factors for the short-term outcomes. (No worries if this is not available yet.)
[x] When we have results from all sites, we’ll add geography as a potential risk factor

Re: readmission / death

[x] In addition to comparing the difference between patients with or without any of the neurological conditions, examine the breakdown among the individual neuro ICD codes ( similar to hospitalization duration) as well as categories of ICD codes (per neuro paper 1)

Re: hospitalization duration (and readmission / death)

[x] Add the patients without any neurological condition in the figure containing individual neuro ICD codes

Re: comorbidity

[x] To discuss on Wednesday: Meghan Hutch's analysis on NWU data. This will be an important risk factor along with age, sex, race/ethnicity and geography.

trangdata commented 3 years ago

When we have results from all sites, we’ll add geography as a potential risk factor.

I'm currently not sure we can do this. What we will receive from each site is summary statistics (within site analysis). Perhaps we can examine the significance level ~ country? 🤔

trangdata commented 3 years ago

Discussed on 2020-11-18. Continuing discussion in #2.

trangdata / neuro-penn

Continuing analysis #1