📊 covid: deaths by vax status (#3297)

* 📊 covid: deaths by vax status

* snapshot wip

* meadow

* garden

* change entity->country colname

* add title_public

* grapher wip

* infections: add title_public

* add country name to indicator title

* remove import

dag/covid.yml

@@ -268,3 +268,14 @@ steps:
     - data://meadow/covid/latest/infections_model
   data://grapher/covid/latest/infections_model:
     - data://garden/covid/latest/infections_model
+
+  # Deaths by vaccination status
+  data://meadow/covid/latest/deaths_vax_status:
+    - snapshot://covid/latest/deaths_vax_status_england.csv
+    - snapshot://covid/latest/deaths_vax_status_us.csv
+    - snapshot://covid/latest/deaths_vax_status_chile.csv
+    - snapshot://covid/latest/deaths_vax_status_switzerland.csv
+  data://garden/covid/latest/deaths_vax_status:
+    - data://meadow/covid/latest/deaths_vax_status
+  data://grapher/covid/latest/deaths_vax_status:
+    - data://garden/covid/latest/deaths_vax_status

etl/steps/data/garden/covid/latest/deaths_vax_status.meta.yml

@@ -0,0 +1,120 @@
+# NOTE: To learn more about the fields, hover over their names.
+definitions:
+  common:
+    description_short: |-
+      Death rates are calculated as the number of deaths in each group, divided by the total number of people in this group. This is given per 100,000 people.
+    unit: doses
+    presentation:
+      topic_tags:
+        - COVID-19
+
+
+# Learn more about the available fields:
+# http://docs.owid.io/projects/etl/architecture/metadata/reference/
+dataset:
+  update_period_days: 0
+  title: COVID-19, deaths by vaccination status
+
+
+tables:
+  us:
+    common:
+      description_key:
+        - The mortality rate for the 'All ages' group is age-standardized to account for the different vaccination rates of older and younger people.
+    variables:
+      us_unvaccinated:
+        title: Death rate (weekly) of unvaccinated people - United States, by age
+        presentation:
+          title_public: Death rate (weekly) of unvaccinated people - United States, by age
+        display:
+          name: Unvaccinated
+      us_vaccinated_no_biv_booster:
+        title: Death rate (weekly) of fully vaccinated people (without bivalent booster) - United States, by age
+        presentation:
+          title_public: Death rate (weekly) of fully vaccinated people (without bivalent booster) - United States, by age
+        display:
+          name: Vaccinated without bivalent booster
+      us_vaccinated_with_biv_booster:
+        title: Death rate (weekly) of fully vaccinated people (with bivalent booster) - United States, by age
+        presentation:
+          title_public: Death rate (weekly) of fully vaccinated people (with bivalent booster) - United States, by age
+        display:
+          name: Vaccinated with bivalent booster
+
+  chile:
+    common:
+      description_key:
+        - The mortality rate for the 'All ages' group is age-standardized to account for the different vaccination rates of older and younger people.
+    variables:
+      chile_0_1_dose:
+        title: Death rate (weekly) of people with 0 or 1 dose - Chile, by age
+        presentation:
+          title_public: Death rate (weekly) of people with 0 or 1 dose - Chile, by age
+        display:
+          name: 0 or 1 dose
+      chile_2_doses:
+        title: Death rate (weekly) of people with 2 doses - Chile, by age
+        presentation:
+          title_public: Death rate (weekly) of people with 2 doses - Chile, by age
+        display:
+          name: 2 doses
+      chile_3_doses:
+        title: Death rate (weekly) of people with 3 doses - Chile, by age
+        presentation:
+          title_public: Death rate (weekly) of people with 3 doses - Chile, by age
+        display:
+          name: 3 doses
+      chile_4_doses:
+        title: Death rate (weekly) of people with 4 doses - Chile, by age
+        presentation:
+          title_public: Death rate (weekly) of people with 4 doses - Chile, by age
+        display:
+          name: 4 doses
+
+  england:
+    common:
+      description_key:
+        - Unvaccinated people have not received any dose.
+        - Partially-vaccinated people are excluded.
+        - Fully-vaccinated people have received all doses prescribed by the initial vaccination protocol.
+        - The mortality rate is age-standardized to account for the different vaccination rates of older and younger people.
+    variables:
+      england_unvaccinated:
+        title: Death rate (monthly) of unvaccinated people - England, by age
+        presentation:
+          title_public: Death rate (monthly) of unvaccinated people - England, by age
+        display:
+          name: Unvaccinated
+      england_fully_vaccinated:
+        title: Death rate (monthly) of fully vaccinated people - England, by age
+        presentation:
+          title_public: Death rate (monthly) of fully vaccinated people - England, by age
+        display:
+          name: Fully vaccinated
+
+  switzerland:
+    common:
+      description_key:
+        - Data coverage includes both Switzerland and Liechtenstein. Unvaccinated people have not received any dose. Partially-vaccinated people are excluded.
+        - Fully-vaccinated people have received all doses prescribed by the initial vaccination protocol.
+        - The mortality rate for the 'All ages' group is age-standardized to account for the different vaccination rates of older and younger people.
+    variables:
+      swi_unvaccinated:
+        title: Death rate (weekly) of unvaccinated people - Switzerland, by age
+        presentation:
+          title_public: Death rate (weekly) of unvaccinated people - Switzerland, by age
+        display:
+          name: Unvaccinated
+      swi_vaccinated_no_booster:
+        title: Death rate (weekly) of fully vaccinated people (without booster) - Switzerland, by age
+        presentation:
+          title_public: Death rate (weekly) of fully vaccinated people (without booster) - Switzerland, by age
+        display:
+          name: Fully vaccinated, no booster
+      swi_vaccinated_with_booster:
+        title: Death rate (weekly) of fully vaccinated people (with booster) - Switzerland, by age
+        presentation:
+          title_public: Death rate (weekly) of fully vaccinated people (with booster) - Switzerland, by age
+        display:
+          name: Fully vaccinated, with booster
+

etl/steps/data/garden/covid/latest/deaths_vax_status.py

@@ -0,0 +1,28 @@
+"""Load a meadow dataset and create a garden dataset."""
+
+from etl.helpers import PathFinder, create_dataset
+
+# Get paths and naming conventions for current step.
+paths = PathFinder(__file__)
+
+
+def run(dest_dir: str) -> None:
+    #
+    # Load inputs.
+    #
+    # Load meadow dataset.
+    ds_meadow = paths.load_dataset("deaths_vax_status")
+
+    # Read table from meadow dataset.
+    tables = list(ds_meadow)
+
+    #
+    # Save outputs.
+    #
+    # Create a new garden dataset with the same metadata as the meadow dataset.
+    ds_garden = create_dataset(
+        dest_dir, tables=tables, check_variables_metadata=True, default_metadata=ds_meadow.metadata
+    )
+
+    # Save changes in the new garden dataset.
+    ds_garden.save()

etl/steps/data/garden/covid/latest/infections_model.meta.yml

@@ -41,23 +41,31 @@ tables:
     variables:
       icl_infections:
         title: Daily new estimated COVID-19 infections (ICL, <<estimate.title()>> estimate)
+        presentation:
+          title_public: Daily new estimated COVID-19 infections (ICL, <<estimate.title()>> estimate)
         description: |-
           <% set model_name = "ICL" %>
           {definitions.others.description}
 
       ihme_infections:
         title: Daily new estimated COVID-19 infections (IHME, <<estimate.title()>> estimate)
+        presentation:
+          title_public: Daily new estimated COVID-19 infections (IHME, <<estimate.title()>> estimate)
         description: |-
           <% set model_name = "IHME" %>
           {definitions.others.description}
 
       lshtm_infections:
         title: Daily new estimated COVID-19 infections (LSHTM, <<estimate.title()>> estimate)
+        presentation:
+          title_public: Daily new estimated COVID-19 infections (LSHTM, <<estimate.title()>> estimate)
         description: |-
           <% set model_name = "LSHTM" %>
           {definitions.others.description}
       yyg_infections:
         title: Daily new estimated COVID-19 infections (Youyang Gu, <<estimate.title()>> estimate)
+        presentation:
+          title_public: Daily new estimated COVID-19 infections (Youyang Gu, <<estimate.title()>> estimate)
         description: |-
           <% set model_name = "Youyang Gu" %>
           {definitions.others.description}

etl/steps/data/grapher/covid/latest/deaths_vax_status.py

@@ -0,0 +1,32 @@
+"""Load a garden dataset and create a grapher dataset."""
+
+from etl.helpers import PathFinder, create_dataset
+
+# Get paths and naming conventions for current step.
+paths = PathFinder(__file__)
+
+
+def run(dest_dir: str) -> None:
+    #
+    # Load inputs.
+    #
+    # Load garden dataset.
+    ds_garden = paths.load_dataset("deaths_vax_status")
+
+    # Read table from garden dataset.
+    tables = list(ds_garden)
+
+    #
+    # Process data.
+    #
+
+    #
+    # Save outputs.
+    #
+    # Create a new grapher dataset with the same metadata as the garden dataset.
+    ds_grapher = create_dataset(
+        dest_dir, tables=tables, check_variables_metadata=True, default_metadata=ds_garden.metadata
+    )
+
+    # Save changes in the new grapher dataset.
+    ds_grapher.save()

etl/steps/data/meadow/covid/latest/deaths_vax_status.py

@@ -0,0 +1,85 @@
+"""Load a snapshot and create a meadow dataset."""
+
+from etl.helpers import PathFinder, create_dataset
+
+# Get paths and naming conventions for current step.
+paths = PathFinder(__file__)
+
+
+def run(dest_dir: str) -> None:
+    #
+    # Load inputs.
+    #
+    # Retrieve tables from snapshots
+    tb_en = paths.read_snap_table("deaths_vax_status_england.csv")
+    tb_us = paths.read_snap_table("deaths_vax_status_us.csv")
+    tb_swi = paths.read_snap_table("deaths_vax_status_switzerland.csv")
+    tb_cl = paths.read_snap_table("deaths_vax_status_chile.csv")
+
+    #
+    # Process data.
+    #
+    # US
+    rename_cols = {
+        "Entity": "country",
+        "Day": "date",
+        "unvaccinated": "us_unvaccinated",
+        "vaccinated_without": "us_vaccinated_no_biv_booster",
+        "vaccinated_with": "us_vaccinated_with_biv_booster",
+    }
+    tb_us = tb_us.rename(columns=rename_cols)[rename_cols.values()]
+    tb_us = tb_us.format(["country", "date"], short_name="us")
+
+    # England
+    rename_cols = {
+        "Entity": "country",
+        "Day": "date",
+        "Unvaccinated": "england_unvaccinated",
+        "Fully vaccinated": "england_fully_vaccinated",
+    }
+    tb_en = tb_en.rename(columns=rename_cols)[rename_cols.values()]
+    tb_en = tb_en.format(["country", "date"], short_name="england")
+
+    # Switzerland
+    rename_cols = {
+        "Entity": "country",
+        "Day": "date",
+        "Unvaccinated": "swi_unvaccinated",
+        "Fully vaccinated, no booster": "swi_vaccinated_no_booster",
+        "Fully vaccinated + booster": "swi_vaccinated_with_booster",
+    }
+    tb_swi = tb_swi.rename(columns=rename_cols)[rename_cols.values()]
+    tb_swi = tb_swi.format(["country", "date"], short_name="switzerland")
+
+    # Chile
+    rename_cols = {
+        "Entity": "country",
+        "Day": "date",
+        "0 or 1 dose": "chile_0_1_dose",
+        "2 doses": "chile_2_doses",
+        "3 doses": "chile_3_doses",
+        "4 doses": "chile_4_doses",
+    }
+    tb_cl = tb_cl.rename(columns=rename_cols)[rename_cols.values()]
+    tb_cl = tb_cl.format(["country", "date"], short_name="chile")
+
+    # Table list
+    tables = [
+        tb_us,
+        tb_en,
+        tb_cl,
+        tb_swi,
+    ]
+
+    #
+    # Save outputs.
+    #
+    # Create a new meadow dataset with the same metadata as the snapshot.
+    ds_meadow = create_dataset(
+        dest_dir,
+        tables=tables,
+        check_variables_metadata=True,
+    )
+
+    # Save changes in the new meadow dataset.
+    ds_meadow.save()

snapshots/covid/latest/deaths_vax_status.py

@@ -0,0 +1,41 @@
+"""Script to create a snapshot of dataset.
+
+This data was downloaded from Grapher. It had been imported to Grapher before covid-19-data repository was created.
+"""
+
+from pathlib import Path
+
+import click
+
+from etl.snapshot import Snapshot
+
+# Version for current snapshot dataset.
+SNAPSHOT_VERSION = Path(__file__).parent.name
+
+
+@click.command()
+@click.option("--upload/--skip-upload", default=True, type=bool, help="Upload dataset to Snapshot")
+@click.option("--england", default=None, type=str, help="Path to ICL local data file.")
+@click.option("--us", default=None, type=str, help="Path to IHME local data file.")
+@click.option("--switzerland", default=None, type=str, help="Path to LSHTM local data file.")
+@click.option("--chile", default=None, type=str, help="Path to Youyang Gu local data file.")
+def main(england: str, us: str, switzerland: str, chile: str, upload: bool) -> None:
+    estimates = [
+        ("england", england),
+        ("us", us),
+        ("switzerland", switzerland),
+        ("chile", chile),
+    ]
+    # Create a new snapshots.
+    for estimate in estimates:
+        name = estimate[0]
+        filename = estimate[1]
+
+        if filename is not None:
+            snap = Snapshot(f"covid/{SNAPSHOT_VERSION}/deaths_vax_status_{name}.csv")
+            # Copy local data file to snapshots data folder, add file to DVC and upload to S3.
+            snap.create_snapshot(filename=filename, upload=upload)
+
+
+if __name__ == "__main__":
+    main()

snapshots/covid/latest/deaths_vax_status_chile.csv.dvc

@@ -0,0 +1,33 @@
+# Learn more at:
+# http://docs.owid.io/projects/etl/architecture/metadata/reference/
+meta:
+  origin:
+    # Data product / Snapshot
+    title: COVID-19, Incidencia de casos según estado de vacunación, grupo de edad, y semana epidemiológica (Chile)
+    description: |-
+      Incidence of deaths according to vaccination status, age group, and epidemiological week.
+
+      Vaccination status is classified as "Fully vaccinated" for those people who have received two doses and more than 14 days have passed since their second dose, or have received a vaccine from a vaccination protocol that includes only a single dose and more than 28 days have elapsed since inoculation. This variable takes the value "Unvaccinated or not fully vaccinated" if people do not have a complete vaccination schedule.
+
+      The mortality rate corresponds to the incidence rate of deaths per 100,000 inhabitants for the age group, corresponding vaccination status and corresponding epidemiological week.
+
+      The mortality rate for the "All ages" group is age-standardized by Our World in Data, using single-year age estimates from the 2022 revision of the United Nations World Population Prospects for Chile. Rates for specific age groups are calculated as crude incidence rates.
+    date_published: "2023"
+
+    # Citation
+    producer: Departamento de Epidemiología, Ministerio de Salud de Chile.
+    citation_full: |-
+      Departamento de Epidemiología, Ministerio de Salud de Chile. Accessed via GitHub (https://github.com/MinCiencia/Datos-COVID19). 2023.
+
+    # Files
+    url_main: https://web.archive.org/web/20230408120752/https://github.com/MinCiencia/Datos-COVID19/tree/master/output/producto89
+    date_accessed: 2024-09-16
+
+    # License
+    license:
+      name: CC BY 4.0
+
+outs:
+  - md5: 42f4ead672fe4284bc5d8a59ecfac666
+    size: 51346
+    path: deaths_vax_status_chile.csv