project.yaml

version: '3.0'

expectations:
  population_size: 1000

actions:
  
  generate_ethnicity:
    run: cohortextractor:latest generate_cohort --study-definition study_definition_ethnicity --output-format=feather
    outputs:
      highly_sensitive:
        cohort: output/input_ethnicity.feather

  generate_study_population:
    run: cohortextractor:latest generate_cohort --study-definition study_definition
    outputs:
      highly_sensitive:
        cohort: output/input.csv

  generate_rates_1:    
    run: cohortextractor:latest generate_cohort --study-definition study_definition_rates --index-date-range "2015-01-01 to 2018-12-01 by month" --skip-existing --output-dir=output --output-format=feather
    outputs:      
      highly_sensitive:
        cohort: output/measures/inpu*.feather
  
  generate_rates_2:    
    run: cohortextractor:latest generate_cohort --study-definition study_definition_rates --index-date-range "2019-01-01 to 2022-12-01 by month" --skip-existing --output-dir=output --output-format=feather
    outputs:      
      highly_sensitive:
        cohort: output/measures/inp*.feather

  join_ethnicity:
    run: python:latest python analysis/join_ethnicity.py
    needs:
      [
        generate_rates_1,
        generate_rates_2,
        generate_ethnicity,
      ]
    outputs:
      highly_sensitive:
        cohort: output/measures/in*.feather
  
  generate_measures:
    run: cohortextractor:latest generate_measures --study-definition study_definition_rates --skip-existing --output-dir=output/measures
    needs: 
      [
        generate_rates_1,
        generate_rates_2,
        join_ethnicity,
      ]
    outputs:
      moderately_sensitive:
        measure_csv1: output/measures/measure_self_harmAE*_rate.csv
        measure_csv2: output/measures/measure_self_harmAEbyRegion*_rate.csv
        measure_csv3: output/measures/measure_self_harmAEbyIMD*_rate.csv
        measure_csv4: output/measures/measure_self_harmAEbyEthnicity*_rate.csv
        measure_csv5: output/measures/measure_self_harmAEbyAge*_rate.csv

  describe_trends:
    run: r:latest analysis/Descriptive_trends.R
    needs: 
      [
        generate_measures,
      ]
    outputs:
      moderately_sensitive:
        fig1: output/self_harmAE.png
        fig2: output/self_harmAEbyReg.png
        fig3: output/self_harmAEbyIMD.png
        fig4: output/self_harmAEbyEth.png
        fig5: output/self_harmAEbyAge.png   
        csv1: output/self_harmAE_rounded.csv
        csv2: output/self_harmAEbyReg_rounded.csv
        csv3: output/self_harmAEbyIMD_rounded.csv
        csv4: output/self_harmAEbyEth_rounded.csv
        csv5: output/self_harmAEbyAge_rounded.csv