Datasets:

electricsheepafrica
/

ssa-risk-factor-prevalence

sample_id stringlengths 15 15	population stringclasses 7 values	region stringclasses 5 values	is_SSA bool 2 classes	is_reference_panel bool 2 classes	sex stringclasses 2 values	age int64 18 85	bmi_category stringclasses 4 values	physical_activity_level stringclasses 2 values	alcohol_use_pattern stringclasses 3 values	smoking_status stringclasses 3 values	dietary_pattern stringclasses 3 values
RF_SAMPLE_00001	SSA_West	West	true	false	Female	49	Obese	Sedentary	None	Never	Mixed
RF_SAMPLE_00002	SSA_West	West	true	false	Female	31	Underweight	Sedentary	None	Never	Traditional
RF_SAMPLE_00003	SSA_West	West	true	false	Female	55	Overweight	Active	None	Never	Traditional
RF_SAMPLE_00004	SSA_West	West	true	false	Female	57	Normal	Active	None	Never	Western
RF_SAMPLE_00005	SSA_West	West	true	false	Female	20	Obese	Sedentary	None	Never	Mixed
RF_SAMPLE_00006	SSA_West	West	true	false	Male	28	Obese	Sedentary	Low	Never	Traditional
RF_SAMPLE_00007	SSA_West	West	true	false	Female	47	Obese	Active	None	Never	Mixed
RF_SAMPLE_00008	SSA_West	West	true	false	Female	41	Normal	Active	None	Never	Mixed
RF_SAMPLE_00009	SSA_West	West	true	false	Male	45	Underweight	Active	Low	Never	Mixed
RF_SAMPLE_00010	SSA_West	West	true	false	Female	34	Normal	Active	Low	Never	Traditional
RF_SAMPLE_00011	SSA_West	West	true	false	Female	56	Obese	Active	None	Never	Traditional
RF_SAMPLE_00012	SSA_West	West	true	false	Male	55	Normal	Active	Harmful	Current	Traditional
RF_SAMPLE_00013	SSA_West	West	true	false	Female	46	Obese	Active	None	Former	Mixed
RF_SAMPLE_00014	SSA_West	West	true	false	Female	60	Overweight	Sedentary	None	Never	Traditional
RF_SAMPLE_00015	SSA_West	West	true	false	Female	51	Underweight	Sedentary	None	Never	Western
RF_SAMPLE_00016	SSA_West	West	true	false	Male	34	Normal	Active	Low	Never	Traditional
RF_SAMPLE_00017	SSA_West	West	true	false	Female	50	Normal	Active	Low	Never	Western
RF_SAMPLE_00018	SSA_West	West	true	false	Female	33	Obese	Sedentary	Low	Never	Mixed
RF_SAMPLE_00019	SSA_West	West	true	false	Female	56	Normal	Sedentary	None	Never	Traditional
RF_SAMPLE_00020	SSA_West	West	true	false	Female	44	Overweight	Active	None	Never	Traditional
RF_SAMPLE_00021	SSA_West	West	true	false	Male	43	Normal	Active	Harmful	Never	Western
RF_SAMPLE_00022	SSA_West	West	true	false	Female	36	Obese	Active	Harmful	Never	Traditional
RF_SAMPLE_00023	SSA_West	West	true	false	Male	61	Obese	Active	None	Never	Traditional
RF_SAMPLE_00024	SSA_West	West	true	false	Female	43	Normal	Sedentary	Harmful	Never	Mixed
RF_SAMPLE_00025	SSA_West	West	true	false	Female	39	Underweight	Active	Harmful	Never	Traditional
RF_SAMPLE_00026	SSA_West	West	true	false	Male	40	Overweight	Sedentary	None	Current	Traditional
RF_SAMPLE_00027	SSA_West	West	true	false	Female	52	Overweight	Sedentary	None	Never	Mixed
RF_SAMPLE_00028	SSA_West	West	true	false	Female	50	Normal	Active	Low	Never	Traditional
RF_SAMPLE_00029	SSA_West	West	true	false	Male	50	Overweight	Active	None	Current	Traditional
RF_SAMPLE_00030	SSA_West	West	true	false	Female	51	Obese	Active	None	Never	Mixed
RF_SAMPLE_00031	SSA_West	West	true	false	Female	73	Overweight	Active	None	Never	Western
RF_SAMPLE_00032	SSA_West	West	true	false	Male	40	Obese	Sedentary	Harmful	Never	Mixed
RF_SAMPLE_00033	SSA_West	West	true	false	Female	38	Normal	Active	None	Never	Traditional
RF_SAMPLE_00034	SSA_West	West	true	false	Female	34	Obese	Active	Low	Never	Traditional
RF_SAMPLE_00035	SSA_West	West	true	false	Female	53	Underweight	Sedentary	None	Former	Traditional
RF_SAMPLE_00036	SSA_West	West	true	false	Female	60	Overweight	Active	Harmful	Never	Western
RF_SAMPLE_00037	SSA_West	West	true	false	Female	44	Overweight	Active	Low	Never	Mixed
RF_SAMPLE_00038	SSA_West	West	true	false	Male	34	Overweight	Active	Harmful	Never	Traditional
RF_SAMPLE_00039	SSA_West	West	true	false	Female	34	Normal	Active	Low	Never	Traditional
RF_SAMPLE_00040	SSA_West	West	true	false	Male	53	Normal	Active	Low	Never	Traditional
RF_SAMPLE_00041	SSA_West	West	true	false	Female	55	Obese	Active	None	Never	Traditional
RF_SAMPLE_00042	SSA_West	West	true	false	Female	52	Normal	Sedentary	None	Never	Traditional
RF_SAMPLE_00043	SSA_West	West	true	false	Female	36	Obese	Active	None	Never	Traditional
RF_SAMPLE_00044	SSA_West	West	true	false	Male	48	Overweight	Active	Low	Former	Western
RF_SAMPLE_00045	SSA_West	West	true	false	Female	47	Obese	Active	Low	Never	Traditional
RF_SAMPLE_00046	SSA_West	West	true	false	Female	48	Overweight	Active	None	Former	Traditional
RF_SAMPLE_00047	SSA_West	West	true	false	Male	56	Overweight	Active	None	Current	Mixed
RF_SAMPLE_00048	SSA_West	West	true	false	Male	48	Overweight	Active	Low	Never	Mixed
RF_SAMPLE_00049	SSA_West	West	true	false	Female	54	Normal	Active	None	Never	Mixed
RF_SAMPLE_00050	SSA_West	West	true	false	Male	46	Obese	Active	None	Never	Mixed
RF_SAMPLE_00051	SSA_West	West	true	false	Female	49	Normal	Active	None	Former	Traditional
RF_SAMPLE_00052	SSA_West	West	true	false	Female	53	Obese	Active	Low	Never	Traditional
RF_SAMPLE_00053	SSA_West	West	true	false	Female	26	Obese	Sedentary	None	Current	Mixed
RF_SAMPLE_00054	SSA_West	West	true	false	Female	41	Obese	Sedentary	None	Never	Traditional
RF_SAMPLE_00055	SSA_West	West	true	false	Female	39	Normal	Sedentary	Low	Never	Western
RF_SAMPLE_00056	SSA_West	West	true	false	Male	37	Normal	Sedentary	Harmful	Current	Mixed
RF_SAMPLE_00057	SSA_West	West	true	false	Female	41	Obese	Sedentary	None	Never	Traditional
RF_SAMPLE_00058	SSA_West	West	true	false	Male	64	Overweight	Active	None	Never	Western
RF_SAMPLE_00059	SSA_West	West	true	false	Male	34	Overweight	Active	Low	Never	Traditional
RF_SAMPLE_00060	SSA_West	West	true	false	Male	58	Normal	Active	Low	Never	Traditional
RF_SAMPLE_00061	SSA_West	West	true	false	Male	23	Underweight	Active	Harmful	Current	Traditional
RF_SAMPLE_00062	SSA_West	West	true	false	Male	41	Normal	Active	Low	Never	Western
RF_SAMPLE_00063	SSA_West	West	true	false	Male	47	Normal	Active	Low	Never	Traditional
RF_SAMPLE_00064	SSA_West	West	true	false	Female	53	Overweight	Sedentary	Low	Never	Traditional
RF_SAMPLE_00065	SSA_West	West	true	false	Female	54	Obese	Active	None	Current	Mixed
RF_SAMPLE_00066	SSA_West	West	true	false	Female	55	Normal	Active	Low	Never	Western
RF_SAMPLE_00067	SSA_West	West	true	false	Female	40	Overweight	Sedentary	None	Never	Traditional
RF_SAMPLE_00068	SSA_West	West	true	false	Female	39	Underweight	Active	Low	Never	Traditional
RF_SAMPLE_00069	SSA_West	West	true	false	Male	56	Normal	Active	None	Never	Mixed
RF_SAMPLE_00070	SSA_West	West	true	false	Female	43	Normal	Sedentary	Low	Never	Traditional
RF_SAMPLE_00071	SSA_West	West	true	false	Male	28	Normal	Active	Low	Current	Traditional
RF_SAMPLE_00072	SSA_West	West	true	false	Female	30	Overweight	Active	None	Never	Mixed
RF_SAMPLE_00073	SSA_West	West	true	false	Female	33	Overweight	Active	None	Never	Traditional
RF_SAMPLE_00074	SSA_West	West	true	false	Female	51	Overweight	Active	None	Never	Western
RF_SAMPLE_00075	SSA_West	West	true	false	Female	47	Overweight	Active	None	Never	Traditional
RF_SAMPLE_00076	SSA_West	West	true	false	Female	54	Normal	Active	Harmful	Never	Traditional
RF_SAMPLE_00077	SSA_West	West	true	false	Female	39	Normal	Active	Low	Never	Mixed
RF_SAMPLE_00078	SSA_West	West	true	false	Female	47	Underweight	Sedentary	Harmful	Never	Western
RF_SAMPLE_00079	SSA_West	West	true	false	Female	53	Normal	Active	None	Never	Traditional
RF_SAMPLE_00080	SSA_West	West	true	false	Female	41	Overweight	Sedentary	None	Never	Mixed
RF_SAMPLE_00081	SSA_West	West	true	false	Female	51	Obese	Active	Low	Never	Mixed
RF_SAMPLE_00082	SSA_West	West	true	false	Female	36	Normal	Active	Low	Never	Traditional
RF_SAMPLE_00083	SSA_West	West	true	false	Female	40	Normal	Active	Harmful	Never	Mixed
RF_SAMPLE_00084	SSA_West	West	true	false	Female	40	Overweight	Active	None	Never	Mixed
RF_SAMPLE_00085	SSA_West	West	true	false	Female	29	Normal	Active	None	Never	Western
RF_SAMPLE_00086	SSA_West	West	true	false	Female	51	Obese	Active	Harmful	Never	Traditional
RF_SAMPLE_00087	SSA_West	West	true	false	Male	39	Overweight	Sedentary	Harmful	Never	Mixed
RF_SAMPLE_00088	SSA_West	West	true	false	Male	45	Overweight	Active	Low	Never	Traditional
RF_SAMPLE_00089	SSA_West	West	true	false	Male	51	Overweight	Active	Low	Former	Traditional
RF_SAMPLE_00090	SSA_West	West	true	false	Female	51	Normal	Active	None	Never	Mixed
RF_SAMPLE_00091	SSA_West	West	true	false	Female	54	Obese	Sedentary	None	Never	Mixed
RF_SAMPLE_00092	SSA_West	West	true	false	Female	44	Overweight	Active	None	Never	Mixed
RF_SAMPLE_00093	SSA_West	West	true	false	Male	39	Obese	Active	Low	Never	Mixed
RF_SAMPLE_00094	SSA_West	West	true	false	Male	44	Obese	Active	None	Current	Mixed
RF_SAMPLE_00095	SSA_West	West	true	false	Male	23	Obese	Active	Low	Never	Mixed
RF_SAMPLE_00096	SSA_West	West	true	false	Male	26	Underweight	Active	Low	Current	Traditional
RF_SAMPLE_00097	SSA_West	West	true	false	Female	28	Normal	Active	None	Never	Mixed
RF_SAMPLE_00098	SSA_West	West	true	false	Male	32	Overweight	Active	Low	Current	Traditional
RF_SAMPLE_00099	SSA_West	West	true	false	Male	50	Overweight	Active	Harmful	Current	Traditional
RF_SAMPLE_00100	SSA_West	West	true	false	Female	33	Overweight	Active	None	Never	Traditional

End of preview. Expand in Data Studio

SSA Risk Factor Prevalence Dataset (Adults, Multi-ancestry, Synthetic)

Dataset summary

This dataset provides a synthetic adult cohort (age 18–85 years) with key modifiable non-communicable disease (NCD) risk factors, designed around sub-Saharan African (SSA) populations with comparative reference groups.

Included risk factors:

Obesity prevalence using WHO BMI categories.
Physical activity: sedentary vs active.
Alcohol use patterns: none, low, harmful.
Smoking status: never, former, current.
Dietary patterns: traditional vs mixed vs Westernized diet.

Prevalence targets are qualitatively anchored to WHO AFRO factsheets, DHS-based tobacco analyses, global physical activity surveillance, and nutrition-transition work in SSA. All individuals are fully synthetic.

Cohort design

Sample size and populations

Total N: 10,000 synthetic adults.
Populations:
- SSA_West: 2,000
- SSA_East: 2,000
- SSA_Central: 1,500
- SSA_Southern: 1,500
- AAW (African American reference): 1,500
- EUR (European reference): 1,000
- EAS (East Asian reference): 500
Sex:
- Female, Male (approximately 55% vs 45% overall).
Age:
- Range 18–85 years.
- Modestly older mean ages in AAW/EUR/EAS vs SSA clusters.

Population labels are harmonized with other Electric Sheep Africa synthetic datasets for multi-dataset integration.

Risk factors and variables

BMI (WHO categories)

Variable:

bmi_category – one of:
- Underweight (<18.5 kg/m²)
- Normal (18.5–24.9 kg/m²)
- Overweight (25.0–29.9 kg/m²)
- Obese (≥30.0 kg/m²)

Configuration reflects:

High overweight/obesity in urban and southern African clusters, especially among women, informed by:
- WHO AFRO obesity factsheet.
- Multi-country surveys showing combined overweight+obesity >60–70% among South African and urban professional women.
Lower, but rising, obesity in other SSA clusters, with sex differences.
High obesity prevalence in AAW and EUR groups and more moderate prevalence in EAS.

Physical activity

Variable:

physical_activity_level – one of:
- Sedentary – insufficient moderate/vigorous activity.
- Active – meets or exceeds guideline-consistent activity.

Patterning reflects:

Global estimates of insufficient activity (~25–30%) from pooled surveillance analyses, with:
- Lower sedentariness in many SSA rural/traditional contexts.
- Higher sedentariness among urban women, AAW, and EUR.

Alcohol use

Variable:

alcohol_use_pattern – one of:
- None – lifetime abstainers or very rare use.
- Low – low-to-moderate consumption without harmful pattern.
- Harmful – binge/heavy episodic or sustained heavy use.

Distributions are informed by WHO AFRO alcohol factsheets and regional reviews of harmful alcohol use:

Higher harmful use fractions in some southern and western SSA clusters and among men.
Substantial abstention in many SSA populations, especially among women.

Smoking

Variable:

smoking_status – one of:
- Never
- Former
- Current

Calibrated using DHS-based tobacco analyses across 30 SSA countries:

Current smoking in men often 10–30%, depending on country.
Current smoking in women generally <5% in most SSA settings.
Higher prevalence in some southern African contexts and lower in many West/East African settings.

Dietary pattern

Variable:

dietary_pattern – one of:
- Traditional – cereal/legume/root-based diets, relatively low animal fat/sugar.
- Mixed – transition diets with both traditional staples and processed foods.
- Western – higher sugar, animal fats, refined grains, processed foods, and alcohol.

Patterns are anchored to nutrition transition analyses in SSA:

Southern African/island cluster shows more Westernized diets.
Other clusters remain more traditional or mixed, depending on geography and urbanisation.

File and schema

`risk_factor_prevalence_data.parquet` / `risk_factor_prevalence_data.csv`

Each row represents one adult:

Demographics
- sample_id
- population
- region – SSA subregion or Non_SSA.
- is_SSA – boolean.
- is_reference_panel – boolean for AAW, EUR, EAS.
- sex – Female or Male.
- age – integer years.
Risk factors
- bmi_category
- physical_activity_level
- alcohol_use_pattern
- smoking_status
- dietary_pattern

Generation

The dataset is generated using:

risk_factor_prevalence/scripts/generate_risk_factor_prevalence.py

with configuration in:

risk_factor_prevalence/configs/risk_factor_prevalence_config.yaml

and literature inventory in:

risk_factor_prevalence/docs/LITERATURE_INVENTORY.csv

Key steps:

Sample generation – multi-ancestry cohort with age and sex distributions by population.
BMI assignment – sample bmi_category by population and sex using distributions inspired by SSA obesity studies and WHO AFRO.
Physical activity assignment – sample physical_activity_level using global insufficient-activity ranges, with lower sedentariness in many SSA clusters.
Alcohol and smoking – sample alcohol_use_pattern and smoking_status by population and sex, reflecting higher male prevalence and regional differences.
Dietary pattern – sample dietary_pattern by population, reflecting traditional vs Westernised nutrition clusters.

Validation

Validation is performed with:

risk_factor_prevalence/scripts/validate_risk_factor_prevalence.py

and summarized in:

risk_factor_prevalence/output/validation_report.md

Checks include:

C01–C02 – Sample size and population counts vs config.
C03 – BMI category distributions by population and sex.
C04 – Physical activity distributions by population and sex.
C05 – Alcohol use patterns by population and sex.
C06 – Smoking status distributions by population and sex.
C07 – Dietary pattern distributions by population.
C08 – Missingness in key variables.

The released version passes all checks within the configured tolerance, yielding an overall validation status of PASS.

Intended use

This dataset is intended for:

Risk factor clustering and joint modeling of modifiable NCD risks.
Simulation studies of intervention scenarios (e.g., reducing sedentariness or harmful alcohol use).
Educational use for teaching epidemiologic methods with realistic SSA-focused patterns.

It is not suitable for:

Estimating true prevalence in any specific country or health system.
Direct clinical or policy decision-making without reference to real surveillance data.

All individuals and outcomes are synthetic.

Ethical considerations

No real persons are included or re-identifiable.
Population labels and risk factor prevalences are constructed for methodological realism and should not be interpreted as exact country-level estimates.
Users should avoid stigmatizing language or interpretations when working with population-stratified risk-factor data.

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 Risk Factor Prevalence Dataset (Adults, Multi-ancestry, Synthetic)." Hugging Face Datasets.

and, where appropriate, key background sources such as WHO AFRO obesity and alcohol factsheets, global physical activity surveillance (e.g., Lancet Global Health pooled analyses), DHS-based tobacco prevalence analyses, and nutrition transition studies in sub-Saharan Africa.