Job request: 23776

version: '3.0'

# Ignore this`expectation` block. It is required but not used, and will be removed in future versions.
expectations:
  population_size: 10000

actions:
  generate_dataset_prepandemic:
    run: ehrql:v1 generate-dataset analysis/dataset_definition_prepandemic.py --output output/dataset_prepandemic.csv
    outputs:
      highly_sensitive:
        dataset: output/dataset_prepandemic.csv

  generate_dataset_pandemic:
    run: ehrql:v1 generate-dataset analysis/dataset_definition_pandemic.py --output output/dataset_pandemic.csv
    outputs:
      highly_sensitive:
        dataset: output/dataset_pandemic.csv

  generate_dataset_postpandemic:
    run: ehrql:v1 generate-dataset analysis/dataset_definition_postpandemic.py --output output/dataset_postpandemic.csv
    outputs:
      highly_sensitive:
        dataset: output/dataset_postpandemic.csv

  # Generate datasets for analysis   001  
  generate_analysis_datasets:
    run: stata-mp:latest analysis/001_cr_define_covariates_cohorts.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic]
    outputs:
      highly_sensitive:
        log1: logs/001_cr_define_covariates_cohorts.log
        data1: output/prepandemic.dta 
        data2: output/pandemic.dta 
        data3: output/postpandemic.dta 

  # Drug prevalence dataset: 102
  generate_drugprevalence:
    run: stata-mp:latest analysis/102_cr_drug_prevalence_contraind.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets]
    outputs:
      highly_sensitive:
        log1: logs/102_cr_drug_prevalence.log
        data1: output/tabfig/prevalences_summary_*.dta
        data2: output/tabfig/prevalences_summary_redacted_rounded_overall.dta 

  # Drug prevalence dataset: 102A
  generate_drugprevalence_coms:
    run: stata-mp:latest analysis/102_A_cr_drug_prevalence_contraind_combinations.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets]
    outputs:
      highly_sensitive:
        log1: logs/102_A_cr_drug_prevalence_combinations.log
        data1: output/tabfig/drugcombinations.csv

  # Cohort rates: 103
  generate_rates:
    run: stata-mp:latest analysis/103_cr_cohort_rates_repeated.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets]
    outputs:
      highly_sensitive:
        log1: logs/103_cohort_rates_repeated.log
        data1: output/tabfig/rates_repeated_pandemic_redacted_rounded.dta 
        data2: output/tabfig/rates_repeated_prepandemic_redacted_rounded.dta 
        data3: output/tabfig/rates_repeated_postpandemic_redacted_rounded.dta 

  # Cohort rates: 103A
  generate_rates_A:
    run: stata-mp:latest analysis/103_A_cr_cohort_rates_repeated_diabetes.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets]
    outputs:
      highly_sensitive:
        log1: logs/103_A_cohort_rates_repeated_diabetes.log
        data1: output/tabfig/rates_repeated_pandemic_redacted_rounded_diabetes.dta 
        data2: output/tabfig/rates_repeated_prepandemic_redacted_rounded_diabetes.dta 
        data3: output/tabfig/rates_repeated_postpandemic_redacted_rounded_diabetes.dta 

  # Generate table 1 :104  
  generate_table1:
    run: stata-mp:latest analysis/104_cr_table1.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets]
    outputs:
      highly_sensitive:
        log1: logs/104_cr_table1.log
        data1: output/tabfig/table1_prepandemic_redacted_rounded.csv 
        data2: output/tabfig/table1_pandemic_redacted_rounded.csv 
        data3: output/tabfig/table1_postpandemic_redacted_rounded.csv 

  # Drug graphs: 105
  generate_druggraphs:
    run: stata-mp:latest analysis/105_cr_drug_prevalence_graphs.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets, generate_drugprevalence]
    outputs:
      highly_sensitive:
        log1: logs/105_cr_graphs.log
      moderately_sensitive:
        Figures1: output/tabfig/*_prevalences_by_drug_*.svg 
        Figures2: output/tabfig/prevalences_*.svg 

  # Drug graphs: 106
  generate_rategraphs:
    run: stata-mp:latest analysis/106_cr_graphs_rates.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets, generate_rates]
    outputs:
      highly_sensitive:
        log1: logs/106_cr_graphs_rates.log
      moderately_sensitive:
        Figures1: output/tabfig/rates_*.svg 

  # Drug graphs: 107
  generate_drugescalation:
    run: stata-mp:latest analysis/107_cr_cohorts_escalation.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets]
    outputs:
      highly_sensitive:
        log1: logs/107_cr_cohorts_escalation.log
      moderately_sensitive:
        Figures1: output/tabfig/escalation_*.svg 

  # Drug graphs: 108
  generate_30d_mort:
    run: stata-mp:latest analysis/108_cr_cohorts_posthosp_mortality.do
    needs: [generate_dataset_prepandemic, generate_dataset_pandemic, generate_dataset_postpandemic, generate_analysis_datasets]
    outputs:
      highly_sensitive:
        log1: logs/108_cr_cohorts_posthosp_mortality.log
      moderately_sensitive:
        Figures1: output/tabfig/30d_mort_*.svg