Viv3ckj/extend-validation-and-study-periods

[viv3ckj]

The major change in this PR is the updating of the get_nhsbsa_data function.
When running the code in get_pf_medication_validation_data.R, I ran into a lexical error, which was a critical issue that prevented the code from proceeding when the API returned invalid responses. Therefore the script seemed to fail immediately whenever the server responded with anything other than valid JSON. By introducing retries, the script could recover from temporary issues like rate limiting or server overload.

The validation datasets now include September (for consultation validation data), and October (for medication validation data). The study monthly intervals have all been increased by one month.

Closes #92

[milanwiedemann]

Thanks, looks great! Just one small thing before we rerun, should we make all measures intervals end in Nov 2024 or was there a reason for different end dates?

[milanwiedemann]

Just checking if there's a reason for the differnt end date that we're getting here? I remember that we decided to have different start dates depending on what we're looking at, but cant remember why we're having different end dates. Should all intervals end in November 2024?

  measure            start_date add_months end_date  
1 pf_breakdown       2023-11-01         12 2024-11-01
2 condition_provider 2023-01-01         22 2024-11-01
3 descriptive_stats  2024-02-01          9 2024-11-01
4 medications        2023-11-01          9 2024-08-01
5 med_counts         2024-02-01          7 2024-09-01

[viv3ckj]

@milanwiedemann Yup I can change the monthy intervals to make it more consistent, do we want the end date to be November 2024 or December 2024?