updates

[hjforbes]

• Study def changes: asthma now categorical with on/off ocs, cancer_haem now spelt correctly, renamed TPP covid case var names as they were too long for STATA.
• Interaction analyses: added sex and time period, and made do file to output data into table+figure

[CarolineMorton]

Hi @hjforbes This is currently failing tests because the code is looking for a file called 'codelists/opensafely-asthma-diagnosis.csv' which does not exist. This is why you are getting the error message:

 No such file or directory: 'codelists/opensafely-asthma-diagnosis.csv'

I have had a brief look at your code and you have a file called opensafely-current-asthma.csv. This is because there are 2 asthma lists knocking about:

• The one called opensafely-asthma-diagnosis includes all asthma diagnoses and includes things like childhood asthma. This was used in the RF study and combined with medication to determine people who have actual asthma. See study definition here.
• The ICS study used a smaller list called opensafely-current-asthma which removed more historical codes and then looked for these codes in the last 3 years. This is because it was trying to define a cohort of only asthma patients. See study definition here.

I think you just need to decide which one you want to use and how you want to define asthma. I.e does it need to have medication in there as well. Hope that helps.