Ask the GP Access Research Data

~6,000 English GP practices, in two tables:

Monthly trends, March 2023 – May 2026 — appointment volumes, same-day vs booked-ahead mix, waiting bands, and online consultation use, for every practice, every month.

A practice profile snapshot — staffing, population, patient experience, quality and prescribing measures. These are anchored to the year ending March 2025, because the patient survey they centre on was fielded in early 2025; the matching appointment measures cover the same year so that like is compared with like. The July 2026 patient-survey wave (fieldwork Jan–Mar 2026) is included as _2026 columns — including, for the first time, the share of patients told to contact their practice again another day. Survey results back to 2012 and registered list sizes back to 2013 are included for every practice — including around 2,300 that have since closed.

Built entirely from open data:

• Appointments in General Practice — volumes, booking-to-appointment waits, mode, staff type
• Online Consultation Submissions — OC use, suppliers, and time-of-day of submissions
• Cloud Based Telephony — call volumes and the 8–10am phone rush
• GP Patient Survey 2025 — satisfaction, continuity, experienced access, respondent demographics
• Friends & Family Test (GP) — monthly % would-recommend per practice, Jul 2022 onward (timeliness signal between annual surveys)
• General Practice Workforce — GP/nurse/other FTEs, registered population age
• NHS Payments to General Practice — rurality, dispensing status
• OHID Fingertips — IMD deprivation, QOF points and diabetes prevalence, cancer detection/referral measures (NDRS)
• NHSBSA English Prescribing Data — prescription items, antibiotics
• NHS ODS — practice postcodes, closures

What the survey measures actually ask (GP Patient Survey 2025; full questionnaires at gp-patient.co.uk):

• Satisfaction — "Overall, how would you describe your experience of your GP practice?" (% very/fairly good)
• Continuity — "How often do you get to see or speak to your preferred healthcare professional when you ask to?" (% always/almost always or a lot of the time; asked only of patients with a preferred professional)
• Experienced same-day access — "How long after you first contacted your GP practice did the appointment take place?" (% "on the same day")
• Waited too long — "How do you feel about how long you waited for your appointment?" (% "it took too long")
• Failed contact — a composite: "told to contact my practice again another day, as they couldn't help that day" + "I couldn't contact my practice" (from "did you know what the next step would be?")
• A&E after failed contact — "What did you do when you couldn't contact your GP practice…?" (% "I went to A&E", among those whose contact failed)

Full build pipeline, data dictionary and derived files: GitHub.

Good questions look like: a place ("access in Brighton", "compare Cardiff Road Surgery with its PCN" — practices are named), a comparison ("deprived vs affluent", "rural vs urban", "big vs small"), a trend ("how have long waits changed since 2023?"), a relationship ("does phone consulting relate to antibiotic prescribing?"), or a findings question ("what predicts continuity?") which is answered from the research notes.

It can't see: individual patients or GPs (everything is practice-level), hospital data beyond the cancer measures, anything outside England, or events after May 2026.