Stage_1_data_cleaning issues

[ZoeMZou]

Decisions from the Meeting on 28th Jan 2025

1. Removal of Vaccination Product Variables

Missing Variables: No vax_cat_product_* variables were found in either the old or new repositories.
Action: The following code can be removed, as it is no longer needed:
https://github.com/opensafely/post-covid-mentalhealth/blob/ddb50bd8fc9e4e1a9855dc9f9a03991487c908aa/analysis/stage1_data_cleaning.R#L136-L142

2. New Data Cleaning Structure
To improve modularity and maintainability, we will split the current stage_1_data_cleaning.R into four separate scripts:

• Setting Reference Levels – Defines reference levels for categorical variables.

• Quality Assurance – Applies data validation checks and flags potential issues.

• Inclusion Criteria – Implements study-specific inclusion and exclusion rules.

• Main Data Cleaning Script – Calls the above functions and outputs the final cleaned datasets.

No intermediate outputs (e.g., consort, consort_midpoint6, output.rds) will be generated after the Quality Assurance step.
Final cleaned datasets will be produced for each cohort (prevax, vax, unvax).