Croydon University Hospital

On this page

• Overview
• Learning culture
• Safe systems, pathways and transitions
• Safeguarding
• Involving people to manage risks
• Safe environments
• Safe and effective staffing
• Infection prevention and control
• Medicines optimisation

This was a focused follow‑up assessment for this service. At our last inspection, we rated this key question as requires improvement. During this assessment, the rating stayed the same. We looked for evidence that the breach relating to safety and risk had been addressed. The service had made some improvements. However, some aspects of the service were not always safe, and there remained an increased risk that people could be harmed.

We assessed elements of 4 quality statements for the safe domain. We looked at the service’s learning culture and how the service set up and maintained safe systems of care. We looked at how the service safeguarded people using the service and ensured the environment was safe and met people’s needs. Patients’ pathway of care sometimes meant they were cared for in areas which were not designed for patient care. This included patients requiring support with their mental health.

We have not awarded this service a score for Safe.

Find out about when we will not publish a key question score and what we look at when we assess Safe.

The service did not always have an active and positive culture of safety based on openness and honesty. Learning opportunities were not always taken to continually identify and embed good practice. However, the service investigated and reported incidents.

Staff reported that the service did not have effective systems to share learning from incidents, complaints, or compliments. There was no regular forum for nursing staff to review incidents, themes, or trends, and nurses did not routinely attend the doctors’ monthly incident meetings. The service did not routinely share learning through newsletters, emails, or other structured communications. Feedback following incident reporting was inconsistent; some staff received no feedback, while others received only basic email responses, without shared learning about themes, trends, or service improvements. Staff could recall serious incidents but could not describe the learning outcomes that followed. Similarly, staff were aware of complaints and compliments but could not describe any learning arising from them. These gaps limited opportunities for collective learning and service-wide improvement, particularly for nursing staff.

Staff understood what an incident was and how to raise one using the incident reporting system. Leaders took responsibility for investigating incidents and managing complaints. Staff understood the duty of candour and demonstrated openness and transparency with patients and families, providing full explanations when things went wrong. Staff described a positive reporting culture in which leaders encouraged incident reporting, and staff felt comfortable raising concerns with managers. Staff found the incident reporting system easy to access and use, and they used it routinely, including for pressure ulcers. Staff reported an active attitude toward managing and solving risks, and information on raising concerns, and providing feedback was clearly displayed for patients, with multiple feedback channels available.

Data provided by the trust showed the service received 114 complaints from March 2025 to February 2026, mainly relating to the Emergency Department (ED). These highlighted constant pressure and recurring issues which the service did not always resolve through effective learning. Most complaints (109) related to the ED overall, with further concerns identified in majors and sub-wait areas.

The most common complaints involved admissions, discharge and transfer processes (27), patient care (22), communication breakdowns (20), and waiting times (15). Patients and relatives reported delays in care, poor flow through the department, and a lack of timely or clear information. These repeated themes showed that the service did not always embed learning from earlier complaints to prevent recurrence.

Despite these concerns, the service showed some learning and improvement. The service clearly recorded, themed, and categorised complaints, which enabled trend identification and provided a foundation for service improvement.

From September 2025 to February 2026 the service reported 80 incidents of mental health patients absconding, which indicated a significant and ongoing risk to patient safety. In 79 of these incidents, staff contacted the police, showing a reliance on external agencies to manage absconding events.

The repeated need for police involvement suggested that the service did not always address the underlying causes of absconding through learning and system improvement. This highlighted gaps in environmental safety, patient engagement, observation practices, and individualised risk management. The data showed that learning from previous incidents was not always embedded into practice to reduce recurrence or improve patient experience.

The service did not always manage or monitor people’s safety. They did not always make sure there was continuity of care, including when people moved between different services. There were delays in ensuring patients were cared for in the correct area once a decision to admit had been made.

The department did not consistently ensure that patients were cared for in the correct clinical area once a decision to admit had been made. Ongoing patient flow pressures, including 40–60 bed requests at any one-time, limited discharges, delays in social care provision, and challenges with care homes accepting patients over weekends, contributed to this. This resulted in patients remaining in areas not ideal for their needs, often for long periods of time. During periods of crowding, staff used escalation areas which were not designed for patient care. This increased safety risks. However, the trust had daily consultant led rounds for patients waiting more than 24 hours for an inpatient bed. This was delivered via a shared acute medicine rota with dedicated registrar and junior doctor support. A consultant of the week model was used across inpatient wards to strengthen senior clinical decision-making and improve responsiveness to referrals from the Emergency Department.

The use of the corridor care remained a concern. Corridor care refers to the practice of treating patients in non-designated areas, such as hallways, trolleys, or waiting rooms, due to high hospital demand and capacity constraints. Although the service introduced additional staffing and safety controls, the physical layout restricted space, and beds could not pass through the middle of the corridor, with only trolleys able to fit through. This continued to pose a risk if patients deteriorated and required urgent care or evacuation. Leaders ensured that clear exclusion criteria for corridor care were in place and were followed. The Nurse in Charge (NIC) of the Emergency Department (ED) worked in collaboration with the Rapid Assessment and Triage (RAT) team and the Physician in Charge to individually assess each patient’s suitability for corridor care. Staff applied professional judgement consistently to determine whether patients could be safely managed in this environment.

Staff excluded specific patient groups from corridor care to protect safety, dignity and clinical outcomes. These included children; patients presenting with mental health needs; and patients with learning disabilities, neurodivergence or autism. Staff also excluded patients with physical disabilities and those living with dementia, confusion, delirium, or reduced levels of consciousness (low GCS).

Patients with confirmed or suspected infectious conditions were not managed in corridor areas. Staff did not place patients in corridor care if they had a National Early Warning Score (NEWS2) greater than 5, recognising the increased risk of clinical deterioration. Additional exclusions included patients who were pregnant or breastfeeding, severely frail, approaching end of life, or immunocompromised. Staff also ensured that patients experiencing severe pain, including chest pain and sickle cell crises, were not cared for in corridors. Patients with suspected sepsis or those meeting trauma call criteria were consistently excluded from corridor care to ensure timely access to appropriate clinical environments and interventions.

The service recorded corridor care on their risk register and reviewed it regularly. Staff reported that duty staff from other clinical specialties did not always respond promptly when contacted. This caused some of the delays in patient movement and decision-making. The department also experienced pressure from a high volume of mental health attendances within a building that did not have sufficient space to meet demand. There was an increase in complaints from patients with physical health needs about proximity to patients with mental health needs.

However, the service took action to limit risks and strengthen patient safety. Following the warning notice issued by the CQC, leaders formalised staffing in the cohort corridor, allocating 2 nurses and 1 healthcare assistant specifically to that area to improve observation, visibility, and risk management. On the day of inspection, although no patients were being cared for in the corridor, leaders moved staff into other areas of the Emergency Department to maintain safe staffing overall. Staff only cared for lower-acuity patients in this area and moved them promptly to majors if they deteriorated.

The data provided by the trust demonstrated constant pressure within the Emergency Department. The trust recorded high numbers of ambulance arrivals, alongside a notable level of delayed handovers exceeding 45 minutes. This indicated ongoing weakness in patient transfer processes and increased the risk of harm associated with long waits.

The trust’s data also showed that a significant number of patients experienced extended stays in the department, with patients remaining on trolleys for more than 4 to 6 hours while awaiting admission. These figures showed that escalation processes and flow management systems were not always working to manage capacity pressures.

Data relating to patients presenting with mental health needs further highlighted system challenges. The trust recorded a high number of mental health attendances within the department over the reporting period, alongside long average waiting times for assessment and onward care. This showed that pathways for mental health patients did not always support timely access to the right services.

However, leaders completed fire risk assessments for escalation areas and introduced additional controls when more than 6 patients, and fewer than 12, were cared for in the corridor. They ensured additional staff were trained to support evacuation and enhanced visibility in these areas.

Staff maintained effective triage processes and demonstrated skilled and timely assessment. They used standardised triage systems, completed regular National Early Warning Score NEWS and risk assessments, and ensured early identification and escalation of clinical deterioration.

During our inspection we saw the department worked well together with internal teams and external partners to improve patient flow and continuity of care. The transfer of care team supported discharge processes, and leaders had introduced additional support services. This included a borough-funded homelessness team and drug and alcohol practitioners who actively supported patients within the Emergency Department. Staff worked closely with mental health partners and described mental health provision as good. There was a ligature-free environment and staff completed risk assessments for possible self-harm. Standard operating procedures were up to date, and staff showed a good approach to reviewing risks, supporting a culture which prioritised patient safety and responsiveness under pressure.

The service did not always work well with people and healthcare partners to understand what being safe meant to them and how to achieve that.

Staff did not consistently understand adult safeguarding arrangements, which created uncertainty and risk. Staff gave different accounts about whether the trust had a safeguarding team, who the safeguarding lead was, and how adult safeguarding referrals should be made. There was no shared or well‑understood process, particularly when patients were from outside the local area. Staff described out‑of‑area safeguarding as harder to manage. Some staff were unsure how safeguarding worked and said they would rely on escalation during handover or seek advice from the nurse in charge rather than follow a clear process. This lack of shared understanding limited confidence that adult safeguarding concerns would always be identified and managed consistently.

Staff we spoke with told us they had completed safeguarding training and understood how to recognise and respond to abuse. However, data provided by the trust showed mandatory training compliance for safeguarding adults (51.2%) and children (60.4%) were below the trust target. The trust target for compliance was 90%, and at the time of inspection, overall mandatory training compliance in the Emergency Department was 82.6%. We were not provided with information about planned training sessions or whether staff who were not up to date had been scheduled to attend, so we could not be assured that all staff would complete training within a reasonable timeframe.

The service did not achieve full compliance with mandatory Mental Capacity Act (MCA) training at the time of review. Leaders were able to see that 29 staff had not completed this training, which created a risk that some patients may not have had their capacity assessed or best‑interest decisions documented in line with legislation. However, leaders took steps to reduce risk and improve staff competence. This included delivery of a mental health key message presentations in June 2025, covering all Emergency Department handovers. These sessions reached medical, nursing, and healthcare support staff and focused on the recognition and risk assessment of patients presenting with mental health needs. Managers supported learning retention by circulating information via email and displaying information on key message boards for staff reference.

Staff demonstrated strong awareness and confidence in children’s safeguarding processes and early identification of vulnerability. During initial assessments, staff clearly identified potential safeguarding concerns and described appropriate actions to address immediate risks. Systems alerted staff to frequent attenders among children and young people through alerts. Staff had access to 24/7 safeguarding advice from a paediatrician with safeguarding expertise, and a paediatric safeguarding team was based within the department during core hours (9am – 5pm).

Staff escalated concerns appropriately, contacted local authorities when concerns arose, and accessed child protection information as needed. Nursing staff completed referrals directly to local authorities, including Multi-Agency Safeguarding Hub (MASH) referrals (This is a formal report made by a healthcare professional to a centralised team when they have serious concerns about the safety or welfare of a child or vulnerable adult), and senior staff supported them to do so, even when patients were out of area. Staff gave clear examples of safeguarding concerns they had identified and acted upon, such as injuries inconsistent with explanations and parental mental health concerns. Safeguarding team members attended daily handovers, matrons provided visible support, and staff described paediatric safeguarding arrangements as effective, well embedded, and responsive.

Safeguarding polices were available to staff, and these were supported by relevant legislation.

We did not look at Involving people to manage risks during this assessment. The score for this quality statement is based on the previous rating for Safe.

The service did not always detect and control potential risks in the care environment. They did not always make sure equipment, facilities and technology supported the delivery of safe care.

Patient privacy and dignity was not always maintained, mainly because of corridor care and prolonged waits for admission. Staff cared for patients in areas that were not designed for clinical use, including the corridor linking the majors sub‑waiting area, resuscitation area, and ambulance entrance. The corridor was continuously brightly lit and offered no privacy. Whilst staff were aware and managed some risks of corridor care, the environment in these areas was not right for patient care.

Patients often waited for long periods in chairs in the escalation area known as the sub wait area, which limited privacy and dignity. Staff reported delays in providing personal care and undertaking examinations because there were no side rooms available. Doctors confirmed they could not clinically examine patients fully in these areas. Although staff moved personal care activities to the resuscitation area to preserve dignity where possible, this was not always possible. The central area of majors was not designed to deliver patient care, and despite the use of privacy screens, it remained unsuitable. Leaders identified concerns about operational drift in normalising care delivery in inappropriate environments. This caused a high volume of complaints about patient experience in corridor areas.

Although ED teams generally made timely referrals, limited capacity within mental health services delayed onward care and discharge. As a result, patients remained in ED for extended periods in an environment not designed for long-term stays. However, the service completed a self‑assessment against NHSE corridor care guidance and took action to address identified gaps.

Environmental limitations further affected patient experience and safety. The corridor escalation area had access to one toilet, and although leaders explored options to add another, they found this was not possible. The majors sub‑waiting area also had limited toilet facilities, which added pressure during busy periods. During our inspection, staff told us that delays and overcrowding, particularly in the sub-waiting area, increased frustration among patients, which they felt contributed to incidents of aggression and violence.

We reviewed the temporary escalation spaces in the Emergency Department standard operating procedure (SOP), which was in date and version controlled. Overall, the SOP showed that the trust had set up a structured and risk-aware framework to manage corridor care, with defined governance, escalation processes, and safety controls. However, despite these measures, concerns remained as corridor care continued to be used during peak periods, indicating that the practice persisted beyond exceptional circumstances and continued to pose risks to patient safety, privacy, and patient experience.

Despite these challenges, leaders had made efforts to manage risks to safety and the environment. The service used CCTV in response to incidents of violence and aggression. While steps had been taken to minimise intrusion, such as adjusting camera angles, we could not fully assess whether patients were clearly informed about its use or how privacy was managed.

The service employed security staff who were familiar with the department layout and patient groups. Security staff were stationed at the front entrance and undertook regular walk‑arounds, focusing on high‑risk areas such as the sub‑waiting area. They were trained to manage aggression, support patients with mental health needs, and assist with restraint and rapid tranquillisation when required.

We did not look at Safe and effective staffing during this assessment. The score for this quality statement is based on the previous rating for Safe.

We did not look at Infection prevention and control during this assessment. The score for this quality statement is based on the previous rating for Safe.

We did not look at Medicines optimisation during this assessment. The score for this quality statement is based on the previous rating for Safe.