---
license: cc-by-4.0
task_categories:
  - tabular-classification
  - tabular-regression
language:
  - en
tags:
  - healthcare
  - supply-chain
  - blood-bank
  - transfusion
  - blood-safety
  - TTI-screening
  - VNRD
  - maternal-mortality
  - sub-saharan-africa
  - lmic
pretty_name: "Blood Bank Supply Management (Collection, Screening, Shortage, Maternal Impact)"
size_categories:
  - 10K<n<100K
configs:
  - config_name: national_blood_centre
    data_files: data/blood_national_blood_centre.csv
  - config_name: district_hospital_bb
    data_files: data/blood_district_hospital_bb.csv
    default: true
  - config_name: rural_hospital_bb
    data_files: data/blood_rural_hospital_bb.csv
---

# Blood Bank Supply Management Dataset

## Abstract

This dataset provides **30,000 simulated facility-level observations** (10,000 per scenario) of blood product availability, collection, screening, and patient impact across three tiers of blood transfusion services in sub-Saharan Africa. Each record represents one blood product assessed at one facility during one monthly period. The dataset captures 40+ variables spanning blood product type, blood group distribution, collection volumes, TTI screening completeness, wastage, shortage dynamics, cold chain maintenance, and downstream clinical impact including maternal deaths and surgical cancellations. Three scenarios: national blood centre (76% availability), district hospital (44%), rural hospital (15%).

**This dataset is entirely simulated. It must not be used for clinical or transfusion decisions.**

## 1. Introduction

### 1.1 The Blood Supply Crisis

WHO (2022) reports that SSA collects approximately 5 units of blood per 1,000 population, compared to 30+ units in high-income countries. Only 58% of countries globally meet 100% voluntary non-remunerated blood donation (VNRD), with many SSA countries still relying heavily on family/replacement donors who carry higher TTI risk.

### 1.2 Blood Safety

Transfusion-transmissible infection (TTI) screening for HIV, HBV, HCV, and syphilis remains incomplete at many facilities, particularly at hospital-based collection sites without quality-assured testing. The Africa CDC has highlighted blood safety as a priority for health systems strengthening.

### 1.3 Clinical Impact

Blood shortages are a leading contributor to maternal mortality from postpartum haemorrhage, the single largest cause of maternal death in SSA. Surgical cancellations due to blood unavailability compromise emergency and elective surgical care.

## 2. Methodology

### 2.1 Parameterization

| Parameter | National Centre | District Hosp | Rural Hosp | Source |
| --- | --- | --- | --- | --- |
| Blood available | 75% | 45% | 15% | WHO blood safety |
| Shortage rate | 25% | 55% | 82% | Multi-country data |
| TTI screening | 95% | 70% | 30% | WHO/Africa CDC |
| Expiry wastage | 12% | 20% | 30% | Blood bank data |
| Collection/1000 pop | 5.0 | 2.5 | 0.5 | WHO AFRO |

### 2.2 Blood Product Selection

6 products: whole blood, packed RBC, fresh frozen plasma, platelet concentrate, cryoprecipitate, O-negative emergency blood. SSA blood group distribution modelled (O+ 47%, A+ 21%, B+ 19%).

## 3. Schema

| Column | Type | Description |
| --- | --- | --- |
| facility_level | categorical | Facility tier |
| product_name | categorical | 6 blood products |
| blood_group | categorical | 8 ABO/Rh groups |
| collection_type | categorical | VNRD / family_replacement / emergency_only |
| units_collected_month | int | Units collected |
| units_requested_month | int | Units requested |
| units_transfused | int | Units actually transfused |
| tti_screened_pct | float | TTI screening completeness |
| tti_positive_pct | float | TTI positivity rate |
| units_discarded_tti | int | Units discarded for TTI |
| units_expired | int | Units expired unused |
| available_on_survey_day | binary | Blood available |
| shortage_in_last_month | binary | Any shortage |
| shortage_days_last_month | int | Days of shortage |
| shortage_cause | categorical | 11 root cause categories |
| cold_chain_maintained | binary | Blood stored correctly |
| crossmatch_turnaround_hours | float | Time to crossmatch |
| maternal_deaths_no_blood | int | Maternal deaths from shortage |
| surgical_cancellations_no_blood | int | Surgeries cancelled |
| transfusion_reactions_month | int | Adverse transfusion reactions |

## 4. Validation

<p align="center">
  <img src="validation_report.png" alt="Validation Report" width="100%">
</p>

## 5. Usage

```python
from datasets import load_dataset

dataset = load_dataset(
    "electricsheepafrica/blood-bank-supply-management",
    "district_hospital_bb"
)
df = dataset["train"].to_pandas()

# Blood group availability analysis
print(df.groupby('blood_group')['available_on_survey_day'].mean().sort_values())
```

## 6. Limitations

- **Simulated**: Not from real blood bank information systems.
- **No seasonal dynamics**: Donation campaigns and seasonal variation not modelled.
- **Simplified TTI**: Binary screening rather than individual pathogen results.

## 7. References

1. WHO (2022). Blood safety and availability. 5 units/1000 in SSA.
2. Lancet (2019). Blood transfusion in sub-Saharan Africa.
3. Africa CDC. Blood safety programme documentation.
4. WHO AFRO. National Blood Transfusion Services assessments.

## Citation

```bibtex
@dataset{esa_blood_bank_2025,
  title   = {Blood Bank Supply Management Dataset},
  author  = {{Electric Sheep Africa}},
  year    = {2025},
  publisher = {Hugging Face},
  url     = {https://huggingface.co/datasets/electricsheepafrica/blood-bank-supply-management},
  note    = {Simulated dataset. Not for clinical or transfusion decisions.}
}
```

## License

[CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)