A simulation of general & acute (G&A) hospital beds in England. Change any setting and the model re-runs a year of simulated patients in your browser. Results are compared with a 2025/26 baseline calibrated to published NHS data. Every assumption is explained on the Methods page, and every input is listed with its source on Data & sources.
Bed occupancy across the year
Share of available G&A beds occupied at midnight (as counted by NHS England's KH03 return). Year runs April to March.
Show as table (monthly averages)
How many beds? The same scenario at other bed numbers
Updates automatically after each change. Vertical line: the beds in this scenario. Two charts rather than one with two scales.
Length of stay: history and this scenario
Published figures for England, with this scenario's simulated value added at the end.
Nights per overnight emergency admission
Mean nights per emergency admission among patients who stay at least one night (one per stay, however many consultants), as defined by the Health Foundation; HES, England, general and acute care, measured on diagnoses where at least 85% of admissions are emergencies. The scenario point is the simulator's own count for the same group.
Beds held by patients in hospital over 21 days
Share of occupied G&A beds, December to February (winter daily sitreps; not collected in 2020/21 and 2021/22). The scenario value is the simulated year-round share at 08:00, the same measure the model is fitted to.
Show as table
What the model is running
The inputs your settings produce, next to the baseline. This is exactly what is fed to the simulation.