Datasetpublic

AfiadataKe/ppb-kenya-recalls-dataset

PPB Kenya Recalls Structured data on medicine and medical device recalls and rapid alerts issued by the Pharmacy and Poisons Board (PPB) of Kenya, covering 2016 to 2025. Load the dataset from datasets import load_dataset # Recall notices (default) recalls = load_dataset("afiadata/ppb-kenya-recalls", "recalls") # Rapid alerts alerts = load_dataset("afiadata/ppb-kenya-recalls", "rapid_alerts") Or with pandas directly: import pandas as pd recalls =… See the full description on the dataset page: https://huggingface.co/datasets/AfiadataKe/ppb-kenya-recalls-dataset.

sourceHugging Facecc-by-4.0updated 6mo agoView on Hugging Face

1likes62downloads

Dataset Card

PPB Kenya Recalls

Structured data on medicine and medical device recalls and rapid alerts issued by the Pharmacy and Poisons Board (PPB) of Kenya, covering 2016 to 2025.

Load the dataset

python

from datasets import load_dataset

# Recall notices (default)
recalls = load_dataset("afiadata/ppb-kenya-recalls", "recalls")

# Rapid alerts
alerts = load_dataset("afiadata/ppb-kenya-recalls", "rapid_alerts")

Or with pandas directly:

python

import pandas as pd

recalls = pd.read_csv("https://huggingface.co/datasets/afiadata/ppb-kenya-recalls/resolve/main/recalls_combined.csv")
alerts  = pd.read_csv("https://huggingface.co/datasets/afiadata/ppb-kenya-recalls/resolve/main/rapid_alerts.csv")

Dataset structure

recalls (183 records, 2016–2025)

Column	Description
`year`	Year the recall was issued
`date`	Date of the recall notice
`recall_ref`	Official PPB reference number (`REC/YYYY/NNN`). Empty for pre-2022 records.
`product_name`	Brand name of the recalled product
`inn_name`	International Non-proprietary Name (generic name) and strength
`batch_no`	Affected batch number(s)
`manufacturer`	Manufacturer name and country
`reason`	Reason for recall as stated in the PPB notice

rapid_alerts (62 records, 2018–2025)

Column	Description
`year`	Year the alert was issued
`date`	Date of the alert
`title`	Full title of the alert
`pdf_url`	URL to the PDF on the PPB website, where available
`pdf_filename`	Corresponding local PDF filename, if available
`product_type`	Product category (Medicine, Medical device, Excipient)
`source`	Issuing body (PPB or WHO)
`manufacturer`	Stated or confirmed manufacturer

Additional files

The full dataset repository also includes:

—recalls_full_text.jsonl — extracted text from recall PDFs (digital + OCR)
—rapid_alerts_full_text.jsonl — extracted text from rapid alert PDFs
—manifest_recalls.csv — PDF-to-record linkage for recalls
—manifest_rapid_alerts.csv — PDF-to-record linkage for rapid alerts
—curation_log.csv — documents manual corrections and records needing review
—recall_pdfs/ — source PDF documents
—rapid_alerts_pdfs/ — source PDF documents

See data_dictionary.md for full column definitions.

Source

All records originate from the PPB website (pharmacyboardkenya.org). Rapid alerts include both PPB-issued notices and WHO alerts circulated by PPB.

Citation

bibtex

@dataset{afiadata_ppb_kenya_recalls_2026,
  author    = {Afiadata},
  title     = {PPB Kenya Recalls Dataset},
  year      = {2026},
  publisher = {Hugging Face},
  url       = {https://huggingface.co/datasets/afiadata/ppb-kenya-recalls}
}

License

CC BY 4.0 — free to use, share, and adapt with attribution.

Attribution: PPB Kenya Recalls Dataset, compiled by Afiadata. Original source: Pharmacy and Poisons Board of Kenya (pharmacyboardkenya.org).