Skip to the content.

Data Sources — abortion-cause-of-death

All data used in this project is from authoritative government or research institute sources. No crowd-edited references (Wikipedia, etc.) are used.


1. CDC WONDER — Underlying Cause of Death (Expanded), 2024

WONDER Queries Performed

File Group By Year
wonder_mortality_2024.tsv Single Race 6 × Sex × ICD-10 113 Cause 2024
wonder_mortality_ages_2024.tsv Single Race 6 × Five-Year Age Groups × ICD-10 113 Cause 2024
wonder_mortality_gender_2024.tsv Sex × Five-Year Age Groups × ICD-10 113 Cause 2024
wonder_mortality_gender_2018_2024.tsv Sex × Single Race 6 × ICD-10 113 Cause 2018–2024 (aggregated)

All queries: entire U.S., all ages, all races, both sexes (unless grouped by). Zero values shown. Export format: tab-delimited with notes footer.


2. Guttmacher Institute — Monthly Abortion Provision Study, 2024


3. CDC Abortion Surveillance, 2022 (for age and gestational age distribution)


4. U.S. Census Bureau — Population Estimates (via WONDER)


Methodology Notes

Rate calculation

For the “with abortion” comparison, crude rates are calculated as:

crude_rate = count / (population + abortions) × 100,000

Rationale: Aborted lives are never counted in the living population. To be consistent with how other crude death rates work (deaths drawn from a population that includes those who died), we add abortions to the denominator.

Year alignment

The 2-year lag on age and gestational proportions is acceptable because these distributions change slowly over time. CDC 2020 and 2021 surveillance show nearly identical distributions year-over-year.

Source hierarchy

  1. Guttmacher is the primary source for abortion counts (more complete than CDC)
  2. CDC Abortion Surveillance provides validated age-group proportions
  3. CDC WONDER provides mortality and population data
  4. All sources are cross-referenced where possible

Series breaks & comparability (read before any cross-year comparison)

The point of this section is to stop apples-to-oranges comparisons. Two known breaks touch this project’s sources:

Everything charted in this project is single-year 2024, so no series is drawn across either break; these notes exist to protect any future extension.


Source Provenance in DuckDB

Every table in data/project.duckdb has a corresponding entry in the _sources metadata table:

SELECT duckdb_table, source_name, methodology, series_breaks FROM _sources;

The _sources table carries provenance alongside the data itself. Columns: duckdb_table, source_name, url, license, notes, retrieved, and — added so methodology travels with the data — methodology (how the source collects and defines the data) and series_breaks (dates/boundaries across which the numbers are NOT comparable). Keep these in sync with the per-source sections above.

Two tables in the DB extend beyond the four sources documented above and carry their own break notes in _sources:


Discrepancies Between Sources

Issue Detail Resolution
Guttmacher vs CDC abortion totals Guttmacher ~1.12M vs CDC ~613K (2022) CDC excludes CA, MD, NH, NJ (~20% of abortions). Use Guttmacher.
Age proportions 2022 vs 2024 Slight shift possible Historical data shows <1pp/year change. Acceptable.
WONDER 2024 provisional vs final 2024 data may be provisional Noted; unlikely to affect rankings materially.