Nine public services, tracked from official government statistics against a pre-pandemic (2019) baseline. Each figure shows where it is now and which way it is heading.
6
▼ Worsening
2
▬ Broadly flat
2
▲ Improving
General practice
Fully qualified GPs (full-time equivalent)
28,956▼ Worsening
-1% since 2015-09
As of 2026-05 · reviewed quarterly
Fully qualified FTE GPs (excludes trainees and wider practice staff). Down ~400 on September 2015 while registered patients have risen, so patients per GP have increased. NHS Digital, monthly.
Source: NHS England Digital, General Practice Workforce
Hospitals
A&E: seen within 4 hours (all types)
77.1%▼ Worsening
-6.9 pts since 2019
As of 2026-03 · auto-updated
All A&E types, England. Parsed from the NHS monthly A&E CSV (sum of over-4hr / attendances across all provider types). Interim 82% target (Mar 2026); 95% constitutional standard. 2019 baseline approximate.
Source: NHS England, A&E Attendances and Emergency Admissions
Source: NHS England, Consultant-led Referral to Treatment (RTT)
Adult social care
Adult social care vacancy rate
6.2%▲ Improving
-1.6 pts since 2019-20
As of 2025-26 · reviewed quarterly
Lowest vacancy rate in a decade (~96,000 vacant posts) and a fourth year of workforce growth, but still around three times the wider-economy rate, with home-care vacancies above 10%. Skills for Care, annual.
Source: Skills for Care, State of the adult social care sector and workforce · by council
Children’s social care
Children’s homes rated good or outstanding
82%▲ Improving
+6.0 pts since 2019-20
As of 2024-25 · reviewed quarterly
Home quality has improved (82% of 3,633 homes good or outstanding). The sector’s crisis is cost, not inspection: residential placement spend has risen sharply and some placements exceed £60,000 a week, see our investigation, Who Profits When Councils Have No Choice.
Source: Ofsted, Children’s social care in England · by council
Homelessness
Households in temporary accommodation
134,210▼ Worsening
+56% since 2019
As of 2025-Q4 · auto-updated
Record high, up ~56% on 2019. Around 176,000 children were in temporary accommodation by December 2025, a twelfth consecutive record; two-thirds of families in TA are in London. MHCLG, quarterly. England.
Source: MHCLG, Statutory homelessness live tables · by council
Schools
Pupils persistently absent
18.7%▼ Worsening
+7.8 pts since 2018-19
As of 2024-25 · reviewed quarterly
Persistent absence = missing 10% or more of sessions. 18.7% in 2024/25, easing slightly year on year but far above the 10.9% pre-pandemic (2018/19) rate; severe absence still rising. DfE, termly. Attainment measures available as a second indicator.
Source: DfE, Pupil absence in schools in England
Police
Crimes resulting in a charge or summons
7.3%▼ Worsening
-8.2 pts since 2015
As of 2024-25 · reviewed quarterly
Charge/summons rate 7.3% in the year to March 2025, a slight improvement on recent years but less than half the 15.5% of 2015. England and Wales. HMICFRS PEEL force gradings tracked separately by quarterly review.
Source: Home Office, Crime outcomes in England and Wales
Criminal courts
Crown Court outstanding caseload
80,061▬ Broadly flat
+111% since 2019-Q4
As of 2026-Q1 · auto-updated
Open cases at quarter end. Q1 2026 down 37 quarter on quarter but +5% year on year; cases open a year or more at a record 22,124. More than double the Dec 2019 pre-Covid level (~38,000).
Source: Ministry of Justice, Criminal Court Statistics Quarterly
Prisons
Prison occupancy (population vs usable operational capacity)
96.2%▬ Broadly flat
+0.2 pts since 2019
As of 2026-06 · auto-updated
Population 85,761 against usable operational capacity 89,120 (29 June 2026). Occupancy is held below the roughly 99% ceiling only by the SDS40 early release scheme running since 2024, so the headline percentage understates the strain; 72% of prisons were overcrowded in 2024/25.
Source: Ministry of Justice / HMPPS, Prison population weekly estate figures
Figures are for England, except the justice indicators (police, criminal courts and prisons), which cover England and Wales. Automated indicators are parsed directly from the published data file; the rest are reviewed each quarter. Where a definition changes, the series is versioned rather than spliced.