# Antibiotic prescribing

> Systemic antibacterial items per 1,000 patients registered with London GP practices.

- Category: NHS
- Cadence: monthly
- Status: Published on a monthly cadence.
- Dataset key: `antibiotic-prescribing`
- Source: [NHS Business Services Authority](https://opendata.nhsbsa.net/dataset/english-prescribing-dataset-epd-with-snomed-code)
- Licence: Open Government Licence v3.0 — [terms](https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/)
- Canonical page: [https://londongraph.com/g/antibiotic-prescribing](https://londongraph.com/g/antibiotic-prescribing)
- Machine-readable record: [https://londongraph.com/data/antibiotic-prescribing](https://londongraph.com/data/antibiotic-prescribing)

## Methodology

BNF section 5.1 systemic antibacterial items in the monthly English Prescribing Dataset, divided by the same month's official registered-patient total for practices whose postcode resolves to each borough. Exact items, quantity, cost, registered patients and practice count remain visible. Registered practice location is not necessarily patient residence, and prescribing is not infection prevalence or prescribing quality on its own.

## Visualisations

- [Normalised map](https://londongraph.com/g/antibiotic-prescribing): Map systemic antibacterial items per 1,000 registered patients while retaining exact totals.
- [Distribution](https://londongraph.com/g/antibiotic-prescribing?view=distribution): Place borough prescribing rates around the London median and quartiles.
- [Items vs cost](https://londongraph.com/g/antibiotic-prescribing?view=cost-relationship): Compare antibacterial items per 1,000 patients with actual cost per registered patient.
- [Borough detail](https://londongraph.com/g/antibiotic-prescribing?view=borough-detail): Inspect exact items, quantity, cost, patient denominator and included practices.

## Citation guidance

Attribute the original publisher named above and identify LondonGraph as the visualisation and methodology layer. Do not describe live values as a historical census, and do not present placeholder values as observations.
