HipAAsynth/sepsis-population-drift-benchmark-300

---
license: cc-by-nc-4.0
task_categories:
- tabular-classification
- tabular-regression
language:
- en
tags:
- healthcare
- ehr
- tabular
- evaluation
- deterministic
pretty_name: HipAAsynth Sepsis Multisite Distributional Shift Dataset
size_categories:
- n<1K
configs:
- config_name: default
data_files:
- split: train
path: "synthetic-sepsis-multisite-distributional-shift.csv"
---

# HipAAsynth Dataset

## Summary

This dataset is a validation artifact generated by HipAAsynth.

HipAAsynth is a deterministic testing and validation service that simulates real-world variability to evaluate how healthcare systems perform under deployment conditions.

---

## Description

This dataset represents a controlled cohort used for testing and benchmarking.

HipAAsynth generates cohorts to simulate how conditions present across:

* patient populations
* demographic distributions
* comorbidity patterns

Datasets are designed for repeatable evaluation and consistent comparison across systems.

---

## Schema

47-column structured CSV:

| Column | Description |
|---|---|
| site | Site identifier (minot_nd / fargo_nd / nd_tribal_standing_rock / etc.) |
| patient_id | Unique patient identifier |
| age | Age in years |
| sex | Biological sex |
| ethnicity | Patient ethnicity |
| height_cm | Height in centimeters |
| weight_kg | Weight in kilograms |
| bmi | Body mass index |
| bmi_category | BMI classification |
| conditions | Pipe-delimited list of active clinical conditions |
| num_visits | Total encounter visit count |
| num_labs | Total lab order count |
| engine_version | HipAAsynth engine version |
| schema_version | Schema version |
| synthetic | Always true — no real patient data |
| disclaimer | Data use disclaimer |
| anchor_hash | SHA-256 generation anchor for reproducibility |
| site_seed | Seed used for this site's generation |
| age_group | Age group classification (young / middle / elderly) |
| diabetes | Diabetes comorbidity flag |
| hypertension | Hypertension comorbidity flag |
| ckd | Chronic kidney disease flag |
| copd | COPD comorbidity flag |
| chf | Congestive heart failure flag |
| glucose_mgdl | Glucose (mg/dL) |
| creatinine_mgdl | Creatinine (mg/dL) |
| wbc_per_nl | White blood cell count (per nL) |
| ldl_mgdl | LDL cholesterol (mg/dL) |
| sepsis_flag | Sepsis-3 criteria met flag |
| infection_source | Infection source classification |
| temperature_c | Temperature in Celsius |
| heart_rate | Heart rate (bpm) |
| resp_rate | Respiratory rate (breaths/min) |
| sbp_mmhg | Systolic blood pressure (mmHg) |
| dbp_mmhg | Diastolic blood pressure (mmHg) |
| spo2_pct | Oxygen saturation (%) |
| lactate_mmol | Lactate (mmol/L) |
| wbc_sepsis | WBC meets sepsis threshold flag |
| creatinine_sepsis | Creatinine meets sepsis threshold flag |
| glucose_sepsis | Glucose meets sepsis threshold flag |
| afebrile | Afebrile presentation flag |
| ams_only | Altered mental status as sole presentation flag |
| delayed_hypotension | Delayed hypotension flag |
| dm_htn_ckd_stack | Combined DM + HTN + CKD comorbidity stack flag |
| hours_to_hypotension | Hours from presentation to hypotension |
| hours_to_icu | Hours from presentation to ICU admission |
| hours_to_vasopressors | Hours from presentation to vasopressor initiation |