This project examined NHS England A&E attendance and emergency admission data across the full 2025/26 financial year covering April 2025 to March 2026. The analysis covers all NHS trusts and providers submitting monthly A&E returns to NHS England, producing 15 analytical outputs that track performance trends, identify providers with the greatest performance challenges, quantify the impact of winter pressure and assess performance consistency across the system.
- Quantify monthly A&E attendance volumes and identify seasonal patterns.
- Track 4-hour performance against the NHS England standard across the year.
- Measure the scale and trend of patients waiting 12 or more hours before admission.
- Identify highest volume providers and those with the greatest performance challenges.
- Identify best performing providers and assess performance consistency across the year.
- Analyse emergency admission trends across the financial year.
- Compare regional attendance volumes and breach rates across NHS England.
- Assess performance variation across NHS England regions.
- Analyse month on month changes in attendance volumes and breach rates.
- Assess the impact of winter pressure from December 2025 to February 2026.
- Present findings in a summary performance heatmap across the top 20 providers.
- Execute SQL analytical queries against a PostgreSQL database.
- Present key findings in an interactive Power BI dashboard.
| Publisher | NHS England |
| Dataset | A&E Attendances and Emergency Admissions 2025/26 |
| Coverage | Approximately 200 NHS trusts and providers per month |
| Frequency | Monthly |
| Access | NHS England Statistics |
| Tool | Purpose |
|---|---|
| Python 3.12 | Core programming language |
| pandas | Data loading, cleaning and transformation |
| matplotlib | Chart production and formatting |
| seaborn | Chart styling and colour palettes |
| PostgreSQL 16 | Database storage and SQL analysis |
| SQLAlchemy | Python to PostgreSQL connection |
| psycopg2 | PostgreSQL database adapter |
| Power BI Desktop | Interactive dashboard |
| Jupyter Notebook | Interactive analysis environment |
26,965,026 A&E attendances were recorded across England in 2025/26. The overall 4-hour breach rate was 39.4%, nearly double the NHS target ceiling of 24%. The gap to target was 15.4 percentage points. The NHS missed the 24% target in all 12 reporting periods. A breach rate consistently operating at nearly double the target ceiling across every single month of the financial year indicates a structural performance problem, not a seasonal or marginal shortfall.
March 2026 was the busiest month with 2,351,317 attendances. February 2026 was the quietest month with 2,043,572 attendances. Monthly volumes were broadly stable ranging from 2.04 million to 2.35 million throughout the year, with no evidence of a sustained demand spike driving the system-wide performance challenge.
January 2026 recorded the highest breach rate at 44.8%. March 2026 recorded the lowest at 37.9%. The breach rate exceeded 40% in 10 out of 12 reporting periods. March 2026 was simultaneously the busiest month by attendances and the best month by breach rate. Higher demand in March did not produce worse performance, a finding that warrants closer investigation at trust level to understand what operational conditions drove the improvement.
570,931 patients waited 12 or more hours before being admitted to hospital across the full year. January 2026 recorded the highest monthly volume at 71,517 patients, nearly double the July 2025 low of 35,467. The concentration of ultra-long waits in winter months, combined with lower overall attendance volumes in that period, points to bed capacity and patient flow constraints rather than demand as the primary driver.
6,451,848 emergency admissions were recorded across 2025/26, an emergency admission rate of 23.9%, approximately 1 in 4 attendances resulted in an emergency admission. July 2025 was the highest month at 559,392 admissions and February 2026 the lowest at 493,015. Monthly volumes were broadly stable throughout the year with no significant seasonal variation in admission rates.
The year followed a clear seasonal pattern. Performance improved in spring from May through July, deteriorated continuously through autumn and into winter from August through January, then recovered sharply in February and March. January 2026 recorded the worst single month deterioration at 2.2 percentage points. Attendance in January was nearly flat at 0.2% below December, indicating that January's deterioration was driven by capacity and staffing constraints rather than a demand spike. March 2026 recorded the largest single month improvement at 4.7 percentage points despite also recording the largest attendance surge of the year at 15.1%, reinforcing that demand volume alone does not determine performance outcomes.
Winter from December 2025 to February 2026 averaged 2,173,285 attendances per month, lower than the rest of the year at 2,271,686. Despite seeing fewer patients per month, winter produced a breach rate 2.7 percentage points higher at 41.5% versus 38.8% and generated 1.35 times more 12 hour or longer waits at an average of 58,980 per month compared to 43,777 for the rest of the year. Winter's performance challenge is a capacity and flow problem, not an attendance volume problem. The system sees fewer patients in winter but manages them less effectively, a finding that has direct implications for winter planning and resource allocation.
Mid Cheshire Hospitals NHS Foundation Trust recorded the highest breach rate at 56.4%. Every trust in the worst 10 recorded a mean breach rate above 52%, more than double the 24% NHS target ceiling. These figures represent sustained year-round performance challenges that persist across these organisations throughout the year rather than isolated seasonal pressures. University Hospitals Plymouth NHS Trust appeared in both the worst performers list at 55.0% and the most variable performers list with a standard deviation of 6.8, a combination that is consistent with performance that lacks both a reliable baseline and a stable improvement trajectory.
Sheffield Children's NHS Foundation Trust recorded the lowest breach rate at 7.6%. The top 5 best performing providers were all children's hospitals or specialist trusts. Specialist trusts treating a narrower patient population with more predictable care pathways consistently outperformed large general acute trusts against the 4-hour standard. Only 5 of the best 10 providers actually met the 24% NHS target, reinforcing that the performance challenge extends across the system and is not limited to the worst performers.
Princess Alexandra Hospital NHS Trust recorded the highest standard deviation of monthly breach rates at 7.5, indicating performance that swings by approximately 7.5 percentage points around its mean from month to month. Performance variability of this scale is consistent with a trust that lacks a stable improvement trajectory. Chesterfield Royal Hospital NHS Foundation Trust appeared in both the worst performers list at 53.1% and the most variable list at 7.3. The combination of persistently elevated breach rates and high monthly variability represents the most challenging performance profile in the dataset. Mid and South Essex NHS Foundation Trust showed the most dramatic trajectory, improving to 26.3% by July 2025 before deteriorating to 43.5% by January 2026, an early improvement that was not sustained across the year.
NHS England Midlands was the busiest region with 4.90 million attendances. NHS England South West was the quietest with 2.38 million. NHS England North West recorded the highest breach rate at 42.6% and NHS England South East the lowest at 36.9%. The spread between the worst and best regions was just 5.7 percentage points, indicating that the A&E performance challenge is broadly uniform across England rather than concentrated in specific geographies. There is no meaningful relationship between attendance volume and breach rate at regional level. London handled 4.84 million attendances with a 37.5% breach rate while the North West handled fewer at 3.79 million yet recorded the worst breach rate at 42.6%, demonstrating that demand volume alone does not explain regional performance variation.
Among the top 20 providers by Type 1 attendance volume, Lewisham and Greenwich NHS Trust recorded the highest mean breach rate at 49.4%, never falling below 45% in any month of the year. Calderdale and Huddersfield NHS Foundation Trust recorded the lowest mean at 16.6%, the only trust in this group consistently operating below the 24% target ceiling throughout the year. The 32.8 percentage point gap between the best and worst performers in this high volume group represents one of the most striking performance differentials in the dataset and demonstrates that trust level operational conditions drive outcomes alongside system-wide pressures.
| Analysis | Description |
|---|---|
| 1 | Monthly attendance volume: total A&E attendances by month |
| 2 | Monthly 4-hour breach rate trend compared to NHS target |
| 3 | Monthly 12 hour or longer wait trend |
| 4 | Top 10 providers by total Type 1 attendance volume |
| 5 | Worst 10 providers by 4-hour breach rate |
| 6 | Monthly emergency admissions trend |
| 7 | Regional attendance breakdown by NHS England region |
| 8 | Winter pressure: December 2025 to February 2026 vs rest of year |
| 9 | Full year summary scorecard |
| 10 | Month on month change in attendance volumes and breach rates |
| 11 | Best 10 providers by 4-hour breach rate |
| 12 | Regional breach rate comparison across NHS England regions |
| 13 | Performance variation across NHS England regions |
| 14 | Provider performance consistency — mean and standard deviation |
| 15 | Summary heatmap: monthly breach rates across top 20 providers |
git clone https://github.com/Kingsley-Eboh/nhs-trend-analysis.git
cd nhs-trend-analysispip install pandas matplotlib seaborn sqlalchemy psycopg2-binary jupyterDownload the monthly CSV files from NHS England for each month from April 2025 to March 2026 and place them in the data folder:
england.nhs.uk/statistics/statistical-work-areas/ae-waiting-times-and-activity
Look for A&E Attendances and Emergency Admissions 2025/26 and download the full provider level extract for each month.
Launch Jupyter Notebook and open nhs_trend_analysis.ipynb:
jupyter notebookSelect Kernel → Restart and Run All Cells to execute all cells in sequence.
Create a local PostgreSQL database and user. Update the connection string in the notebook SQL section with your credentials before running.
Once the database is populated execute the analytical queries:
psql -U your_user -d your_database -h localhost -f sql/nhs_ae_queries.sqlOpen powerbi/nhs_ae_dashboard.pbix in Power BI Desktop and reconnect to your local PostgreSQL instance using your configured credentials.
nhs-trend-analysis/
├── nhs_trend_analysis.ipynb
├── figures/
│ ├── analysis1_monthly_attendances.png
│ ├── analysis2_4hr_breach_rate.png
│ ├── analysis3_long_waits.png
│ ├── analysis4_top10_providers.png
│ ├── analysis5_worst10_breach.png
│ ├── analysis6_emergency_admissions.png
│ ├── analysis7_regional_breakdown.png
│ ├── analysis8_winter_pressure.png
│ ├── analysis10_month_on_month_change.png
│ ├── analysis11_best10_providers.png
│ ├── analysis12_regional_breach.png
│ ├── analysis13_icb_performance.png
│ ├── analysis14_provider_consistency.png
│ └── analysis15_summary_heatmap.png
├── sql/
│ └── nhs_ae_queries.sql
├── powerbi/
│ └── nhs_ae_dashboard.pbix
├── data/
├── .gitignore
└── README.md
Kingsley Eboh GitHub
Data sourced from NHS England. This project is intended for portfolio and educational purposes.