search

oxford-pharmacoepi / MegaStudy

4 stars 2 forks source link

Incidence Attrition results #57

Closed Dorian-APHM closed 4 months ago

Dorian-APHM commented 6 months ago

Hello,

I went through the feasability steps for APHM. Now we're trying to complete Incidence Prevalence steps, the code ran without problem but the results seems to be suspicious.

Below a screenshot from the shiny application :

Incidence attrition APHM

The starting population is good, but I don't understand why the variable Missing year of birth includes all the starting population while this data seems to be present in our cdm. The same goes for Missing Sex.

Does anyone have an explanation ? What did I do wrong ?

Best regards,

Dorian

tiozab commented 6 months ago

hi @Dorian-APHM the numbers are the number of people retained, not the number dropped. So, I think your population looks good.

Dorian-APHM commented 6 months ago

Hi @tiozab, thanks for the reply, that's what we were thinking. However, I can't see the results in ‘study results’. For example for amoxicilin which is mapped, and in the base we have this :

Screenshot_20240530_102428

There must be a variable, an indicator or something else that is not in the base and that makes it impossible to display the results, but we can't find out what it is. Or maybe some filters to modify.

I've uploaded our results in the new results section under this name: Results_CDM_APHM_Incidence_Prevalence if that helps, as we can't find the solution.

tiozab commented 5 months ago

@Dorian-APHM sorry about that This should fix it, please amend the databases.csv according to your own information (cdm_name as seen in the snapshot) https://github.com/oxford-pharmacoepi/MegaStudy/issues/62

tiozab commented 4 months ago

@Dorian-APHM did it work?

Dorian-APHM commented 4 months ago

Yes it works !

Thanks for the reply and sorry for the delay in replying.

Best regards,

Dorian

tiozab commented 4 months ago

Great! thanks!