📊 Update public expenditure on health OMM

[paarriagadap]

Recreate the public expenditure on health OMM that combines three datasets to get the indicator as share of GDP in the long run.

Main issue here https://github.com/owid/owid-issues/issues/1594

[owidbot]

Quick links (staging server):

Site Dev	Site Preview	Admin	Wizard	Docs

Login: ssh owid@staging-site-data-public-expenditure-heal

chart-diff: ❌

• 2/4 reviewed charts
• Modified: 2/2
• New: 0/2
• Rejected: 0
• Data changes: 0
• Metadata changes: 0

data-diff: ❌ Found differences

= Dataset garden/artificial_intelligence/2025-02-04/epoch_aggregates_affiliation
  = Table epoch_aggregates_affiliation
= Dataset garden/artificial_intelligence/2025-02-04/epoch_aggregates_domain
  = Table epoch_aggregates_domain
= Dataset garden/artificial_intelligence/2025-02-04/epoch_compute_intensive_countries
  = Table epoch_compute_intensive_countries
= Dataset garden/artificial_intelligence/2025-02-04/epoch_compute_intensive_domain
  = Table epoch_compute_intensive_domain
= Dataset garden/cdc/latest/measles_cases
  = Table measles_cases
    ~ Column cases (changed metadata)
-       -     date_accessed: '2025-03-05'
        ?                              ^
+       +     date_accessed: '2025-03-04'
        ?                              ^
    ~ Column outbreaks_n (changed metadata)
-       -     date_accessed: '2025-03-05'
        ?                              ^
+       +     date_accessed: '2025-03-04'
        ?                              ^
    ~ Column states_with_cases (changed metadata)
-       -     date_accessed: '2025-03-05'
        ?                              ^
+       +     date_accessed: '2025-03-04'
        ?                              ^
= Dataset garden/emissions/2025-02-12/ceds_air_pollutants
  = Table ceds_air_pollutants
= Dataset garden/health/latest/measles_long_run
  = Table measles_long_run
    ~ Column case_rate (changed metadata)
-       -     date_accessed: '2025-03-05'
        ?                              ^
+       +     date_accessed: '2025-03-04'
        ?                              ^
    ~ Column case_rate_static (changed metadata)
-       -     date_accessed: '2025-03-05'
        ?                              ^
+       +     date_accessed: '2025-03-04'
        ?                              ^
    ~ Column cases (changed metadata)
-       -     date_accessed: '2025-03-05'
        ?                              ^
+       +     date_accessed: '2025-03-04'
        ?                              ^
    ~ Column cases_static (changed metadata)
-       -     date_accessed: '2025-03-05'
        ?                              ^
+       +     date_accessed: '2025-03-04'
        ?                              ^
+ Dataset garden/health_expenditure/2025-03-04/health_expenditure_omm
+ + Table health_expenditure_omm
+   + Column share_gdp
+ Dataset garden/health_expenditure/2025-03-04/lindert
+ + Table lindert
+   + Column share_gdp
= Dataset garden/iea/2024-11-20/fossil_fuel_subsidies
  = Table fossil_fuel_subsidies
= Dataset garden/minerals/2024-07-15/minerals
  = Table minerals
+ Dataset garden/oecd/2025-03-04/health_expenditure_1993
+ + Table health_expenditure_1993
+   + Column share_gdp


Legend: +New  ~Modified  -Removed  =Identical  Details
Hint: Run this locally with etl diff REMOTE data/ --include yourdataset --verbose --snippet

Automatically updated datasets matching weekly_wildfires|excess_mortality|covid|fluid|flunet|country_profile|garden/ihme_gbd/2019/gbd_risk are not included

Edited: 2025-03-05 19:20:32 UTC
Execution time: 5.46 seconds

[paarriagadap]

Hi @spoonerf! I have recreated your work in the public expenditure as a share of GDP OMM, so we can maintain it up-to-date. Please take a look if everything makes sense. We now have the OECD Health Expenditure and Financing Database, please take a look if you still think that picking Government/compulsory schemes is the best option. It looks like it, but we also have Government schemes available. You can even play with the categories by changing CATEGORY_OECD in the garden step of health_expenditure_omm

[paarriagadap]

@spoonerf If helpful, you can also take a look at this comparison between the current OECD data, your version of the OMM and this new version: http://staging-site-data-public-expenditure-heal/admin/charts/8532/edit

[spoonerf]

Hey @paarriagadap!

The code all looks good to me, much nicer than my previous version!

As far as whether we should use Government/Compulsory or Government, I'm happy for you to make the call, I think you have a better understanding of the implications than me. For some countries, like Germany (below), it seems like Government/Compulsory is clearly more aligned with the earlier time-series.

I added just Government schemes to your comparison chart and made a scatter plot directly comparing them here, just for interest/exploration, but feel free to delete it!

[spoonerf]

small typo "bt"

[spoonerf]

Perhaps should be 0 as I don't expect we'll update this