Ipswich Hospital — The Patient Flow Bundle That NHS England Mandated Nationally
Ipswich Hospital (East Suffolk and North Essex NHS Foundation Trust) identified that corridor care was driven by patients staying in beds beyond clinical need — not by A&E demand. They implemented a strict six-point patient flow bundle: Red/Green Day tracking, Criteria-Led Discharge, and 10:30am board rounds. NHS England studied the results and mandated the SAFER bundle nationally. The approach is one of the most widely replicated QI interventions in NHS corridor care improvement.
- Understand the SAFER patient flow bundle and how each element addresses a specific flow failure.
- See why Criteria-Led Discharge is a Level 2–3 intervention and why it is more sustainable than most flow interventions.
- Assess the bundle effect problem — why six simultaneous interventions make it difficult to know which element worked.
- Know what pre-committed Bootstrap CUSUM evaluation would add to the policy confidence NHS England expressed through national mandate.
☰ Contents
The key insight — the problem is upstairs, not at the front door
Ipswich’s starting point was a diagnostic insight that most corridor care improvement programmes miss: corridor care is not primarily caused by too many patients arriving at A&E. It is primarily caused by patients in the wards staying beyond their clinical discharge date, preventing bed availability for patients waiting in the corridor below.
This is the constraint identification step applied correctly. The pile of work (patients in corridors) is downstream of the constraint (delayed discharge from wards). Adding resource at the pile — more A&E capacity, more corridor nursing staff — addresses the symptom. Addressing delayed discharge addresses the constraint.
Before the Red/Green Day system, wards had no real-time visibility of which patients were ready for discharge and which had something clinically blocking their discharge. Ward rounds happened, decisions were made, but the information about discharge status was not visible in a form that enabled immediate management action.
Red/Green Day tracking made the delay visible at the point of the constraint: a red day meant a patient was occupying a bed while clinically fit to leave, with an administrative or social care delay preventing discharge. Once visible, the delay could be acted on immediately rather than appearing in a weekly report.
The SAFER patient flow bundle — six elements
| Element | What it does | Visibility instrument |
|---|---|---|
| S — Senior review | All patients reviewed by a senior clinician before midday | Going to the Gemba — senior clinical presence at the point where discharge decisions are made |
| A — All patients have an Estimated Discharge Date | Every patient has a named date when they are expected to be discharged, set at admission | Makes the expected flow visible before the actual flow fails — the pre-committed prediction applied to individual patient journeys |
| F — Flow of patients before midday | Target: 33% of daily discharges before noon | Process measure that makes morning coordination visible and creates pressure to discharge earlier in the day |
| E — Engagement of patients and families | Patients and families involved in discharge planning from admission | Reduces the information gap between clinical team and patient/family that produces last-minute discharge delays |
| R — Review of all patients with length of stay >7 days | Weekly review of long-stay patients with active discharge planning | Makes the tail of the length-of-stay distribution visible — the patients most likely to be occupying beds beyond clinical need |
| Red/Green Day tracking | Every patient’s day flagged as green (progressing toward discharge) or red (delayed for any reason) | Visual management at the ward level — makes administrative and social care delays visible in real time |
Joiner level analysis
| Intervention | Joiner level | Sustainability |
|---|---|---|
| Red/Green Day tracking | Level 2 — changes the process by which delay visibility is created and acted on | High. Once embedded in ward culture, does not require external support to maintain. Makes invisible delays visible permanently. |
| Criteria-Led Discharge | Level 2–3 — changes accountability structure for discharge; nurses empowered to discharge without waiting for a doctor’s signature | High. Capability change — nurses empowered to discharge retain that capability. Requires ongoing clinical governance to maintain criteria quality but does not reverse when external teams leave. |
| 10:30am board rounds | Level 2 — changes the process by which discharge decisions are coordinated | Medium. Depends on sustained clinical leadership commitment. Can drift back to later or less structured rounds under pressure. |
| Estimated Discharge Date from admission | Level 2 — changes information availability throughout the patient journey | High once embedded in admission processes and electronic records systems. |
Why NHS England mandated it nationally
NHS England packaged the Ipswich methodology as the SAFER patient flow bundle and mandated it across NHS trusts. This is strong evidence of policy confidence — NHS England does not mandate approaches without substantial evidence of effectiveness. However, policy confidence is a different standard from Bootstrap CUSUM pre-committed prediction evidence.
The national mandate means the methodology has been evaluated by NHS improvement analysts and found to be worth spreading. It does not mean:
- The results at Ipswich have been formally replicated under controlled conditions elsewhere
- The specific elements of the bundle that drove improvement have been isolated from the bundle effect
- A pre-committed prediction was made before Ipswich implemented the bundle that was subsequently confirmed
It does mean: the results at Ipswich were sufficiently convincing to NHS improvement leaders that they chose to spread the approach nationally rather than continue testing it locally. For practitioners, this is useful evidence of plausible effectiveness.
The bundle effect problem
Ipswich implemented six elements simultaneously as a bundle. This is clinically sensible — the elements are interdependent and work better together than separately. But it creates a statistical problem: if outcomes improve, which element caused the improvement?
This matters for replication. If a trust implementing only three of the six elements sees no improvement, is that because the three missing elements are critical, or because of different local conditions, or because the bundle effect is not linear? Without isolating the mechanism, the learning from Ipswich is “implement the whole bundle” — useful guidance but not causal understanding.
Bootstrap CUSUM applied to each element sequentially — implementing one at a time with a pre-committed prediction for each — would resolve the bundle effect problem. It would take longer but produce transferable causal knowledge rather than policy-based correlation.
Sustainability assessment
The SAFER bundle is the most likely of the five Bright Spots to be structurally sustained, for one specific reason: Criteria-Led Discharge is a capability change rather than a process change. Once nurses are trained and empowered to discharge against pre-agreed criteria, that capability does not disappear when an improvement team leaves. The knowledge is in the workforce, not in the programme.
The risk factors are:
- Clinical governance burden — Criteria-Led Discharge requires regular review of the criteria to remain safe and current. If clinical governance capacity is insufficient, criteria drift and discharge quality falls.
- Consultant buy-in — Criteria-Led Discharge requires consultants to trust nurses to make discharge decisions. In hospitals where consultant authority is deeply embedded, this trust is not automatic and requires sustained leadership to maintain.
- Red/Green drift — Red/Green Day tracking can become a paperwork exercise rather than an action trigger if ward managers are not empowered to act immediately on red days. Visibility without authority produces frustration rather than improvement.
Bootstrap CUSUM — the honest position
Ipswich has run charts showing discharge before noon rates and length of stay trends. These are consistent with improvement. The bundle was mandated nationally on the basis of these and similar results across early adopter trusts. No pre-committed Bootstrap CUSUM evaluation has been published.
What would complete the evidence:
- A pre-committed prediction for each element of the bundle, implemented sequentially, with Bootstrap CUSUM applied to each
- A sustained data series (12 months minimum) from trusts that implemented the full bundle, showing whether the improvement held after NHS England monitoring reduced
- Balancing measures: 30-day readmission rates (did faster discharge produce unsafe discharge?), staff experience scores (did the bundle add unsustainable pressure?)