electricsheepafrica/ssa-reproductive-history-women

SSA Reproductive History Dataset (Women, Multi-ancestry, Synthetic)

Dataset summary

This dataset provides a synthetic reproductive history cohort of 10,000 women across multiple ancestry groups with a focus on sub-Saharan Africa (SSA). It includes:

• Age at menarche.
• Parity and parity category.
• Age at first birth (for parous women).
• Breastfeeding duration (aggregate months, for parous women).
• Menopausal age and postmenopausal status.

Distributions are informed by LMIC menarche/menopause literature, DHS fertility data, and SSA breastfeeding analyses, but all individuals are synthetic and non-identifiable.

Cohort design

Sample size and populations

• Total N: 10,000 synthetic women.

• Populations:

  • SSA_West: 2,000
  • SSA_East: 2,000
  • SSA_Central: 1,500
  • SSA_Southern: 1,500
  • AAW (African American women): 1,500
  • EUR (European reference): 1,000
  • EAS (East Asian reference): 500

• Age: 18–80 years, with population-specific means and SDs reflecting adult cohorts (e.g., slightly older ages in EUR/AAW).

Population labels align with other Electric Sheep Africa datasets, enabling integrated analyses (e.g., pairing reproductive history with body composition, CV metrics, or genomics).

Reproductive history variables

Age at menarche

• Variable: age_at_menarche (years).
• Modeled as a normal distribution with means by population:
  • SSA: ~13.2–13.5 years.
  • AAW: ~12.7 years.
  • EUR: ~12.5 years.
  • EAS: ~12.2 years.

These values are guided by:

• Systematic reviews of age at menarche in LMICs and global analyses, which report global means around 12–13 years with regional variation.
• SSA-focused studies suggesting slightly later menarche in some African populations compared with high-income settings.

Values are truncated between 9 and 18 years and constrained to fall before current age.

Parity and age at first birth

Variables:

• parity_category – one of:
  • 0 (nulliparous)
  • 1-2
  • 3-4
  • 5+
• parity_total – integer number of live births (0, or sampled within category range).
• age_at_first_birth – age at first live birth (years) for parous women only (null for parity 0).

Design anchors:

• SSA populations have higher fractions in 3-4 and 5+ parity categories and lower nulliparity, consistent with DHS patterns.
• AAW, EUR, EAS have higher 0 and 1-2 parity and lower high-parity fractions.
• Mean age_at_first_birth by population approximates:
  • SSA: ~19–20 years.
  • AAW: ~23 years.
  • EUR: ~28 years.
  • EAS: ~27 years.

These values reflect DHS fertility patterns in SSA and later childbearing in high-income settings. Ages at first birth are truncated to lie between 12 and 45 years and must precede current age.

Breastfeeding duration

Variables:

• breastfeeding_duration_months – approximate total months of any breastfeeding across all births, defined for parous women only.

Per-population means (months):

• SSA_West/East/Central/Southern: ~19–22 months.
• AAW: ~14 months.
• EUR: ~12 months.
• EAS: ~16 months.

These are guided by SSA breastfeeding meta-analyses showing median any-breastfeeding durations around 18–24 months in many SSA settings, with shorter durations in some high-income populations. Values are truncated between 0 and 48 months.

Menopausal age and status

Variables:

• menopause_age – modeled natural menopausal age (years), populated for women aged ≥40 years (null for younger women).
• is_postmenopausal – boolean flag indicating whether the woman is postmenopausal at the time of observation.

Per-population mean ages at menopause are set around:

• SSA: ~49.5–50.0 years.
• AAW/EUR/EAS: ~50.5–51.0 years.

These values are guided by:

• Global reviews indicating mean natural menopause around 50–52 years.
• High-income cohort data (e.g., Treloar et al.) targeting ~50–51 years.

Synthetic is_postmenopausal is derived consistently: women with age >= menopause_age are almost always marked True.

File and schema

reproductive_history_data.parquet / reproductive_history_data.csv

One row per synthetic woman with:

• Demographics / ancestry

  • sample_id
  • population – SSA_West, SSA_East, SSA_Central, SSA_Southern, AAW, EUR, EAS.
  • region – SSA subregion or Non_SSA.
  • is_SSA – boolean.
  • is_reference_panel – True for AAW/EUR/EAS.
  • sex – Female.
  • age – 18–80 years.

• Reproductive milestones

  • age_at_menarche – age at menarche (years).
  • parity_category – 0, 1-2, 3-4, 5+.
  • parity_total – integer total live births.
  • age_at_first_birth – age at first live birth (years) for parous women; null for parity_total == 0.
  • breastfeeding_duration_months – total months of any breastfeeding, defined for parous women; null for parity_total == 0.
  • menopause_age – modeled menopausal age (years) for women aged ≥40 at observation; null for younger women.
  • is_postmenopausal – boolean.

Generation

The dataset is generated with:

• reproductive_history/scripts/generate_reproductive_history.py

using configuration in:

• reproductive_history/configs/reproductive_history_config.yaml

and literature curated in:

• reproductive_history/docs/LITERATURE_INVENTORY.csv

Key modeling logic:

1. Sample table – ages per population drawn from truncated normal distributions.
2. Age at menarche – sampled by population with age-appropriate truncation and forced to precede current age.
3. Parity – parity category sampled from population-specific distributions, then integer parity derived from category.
4. Age at first birth – sampled only for parous women and truncated to be plausible and < current age.
5. Breastfeeding duration – sampled only for parous women, with per-population means.
6. Menopausal age and status – sampled for women age ≥40; is_postmenopausal derived based on a comparison of current age and menopause age.

Validation

Validation follows the GENOMICS Synthetic Data Playbook and is implemented in:

• reproductive_history/scripts/validate_reproductive_history.py

Checks include:

• C01–C02 – Sample size and population counts
  • Confirm N = 10,000 and per-pop counts match config within acceptable deviation.
• C03 – Mean age at menarche by population
  • Compare observed means to config targets.
• C04 – Parity category distributions by population
  • Compare observed parity category frequencies to config.
• C05 – Mean age at first birth (parous women)
  • Validate mean by population.
• C06 – Mean breastfeeding duration (parous women)
  • Validate mean by population.
• C07 – Menopausal age distribution and status consistency
  • Check mean menopause age by population and that women with age >= menopause_age are typically postmenopausal.
• C08 – Missingness in key variables (conditional)
  • Treat age_at_first_birth and breastfeeding_duration_months as required only for parous women; menopause_age as required only for women aged 40+.

The validator writes:

• reproductive_history/output/validation_report.md

For the released version, all checks completed with an overall status of PASS.

Intended use

This dataset is intended for:

• Methods development in reproductive epidemiology, e.g., modeling relationships between menarche, parity, breastfeeding, menopause, and disease risk.
• Teaching and demonstration of reproductive life-course variables in SSA and comparison populations.
• Benchmarking tools that require realistic but non-identifiable reproductive history data.

It is not suitable for:

• Clinical decision-making or counseling.
• Direct estimation of real-world prevalence or trends.
• Individual-level inference.

All records are synthetic.

Ethical considerations

• No real patient data are used.
• Population labels are for simulation and methodologic realism only.
• Analyses should be interpreted as methodological demonstrations rather than definitive statements about any specific country or group.

License

• License: CC BY-NC 4.0.
• Free to use for non-commercial research, education, and methods development with attribution.

Citation

If you use this dataset, please cite:

Electric Sheep Africa. "SSA Reproductive History Dataset (Women, Multi-ancestry, Synthetic)." Hugging Face Datasets.

and, as appropriate, cite the underlying literature used to inform the design (e.g., LMIC menarche timing reviews, DHS fertility analyses, SSA breastfeeding duration studies, and menopause age reviews).