Appendix 2. Summary of Nuffield report on Inquiry questionnaire

Introduction

The Inquiry commissioned the Nuffield Trust (Nuffield) to administer and analyse the results of a questionnaire for NHS Trusts in England providing neonatal care.¹
The questionnaire, drafted by the Inquiry, was sent to 121 Trusts with neonatal units and completed by 120 Trusts.^* Trusts were asked to arrange for the questionnaire to be completed by the Medical Director and a non-clinical director with responsibility for neonatal care. These instructions were followed by 91 of the Trusts. Disappointingly, the remaining 29 Trusts either provided a single response or a joint response signed by more than one person. These responses were accepted by Nuffield.
The report summarised the responses received under a number of headings, as summarised below.

Governance and accountability

Trusts were asked for a general description of the way in which they were governed and to whom they were accountable. Many responses identified the board as responsible for the Trust’s strategy, implementation of objectives for setting the culture of the organisation, and ensuring that the Trust is providing safe, high-quality care.² Some added that the board’s role included other aspects of governance.
There was very little uniformity, even between similar Trusts, in the manner in which Trusts organised their governance.
Committees were described as playing a central part in governance structures. Some committees are required by statute, such as audit, appointment and remuneration committees. Other committees have specific remits to lead other areas of the Trust’s work. These include quality and safety; people; research; safeguarding; equality, diversity and inclusion; and patient experience.³
There was also little uniformity in how Trusts organised and managed their clinical functions. These were variously organised into ‘divisions’, ‘care groups’ or ‘clinical service units’ and could have different titles and configurations. For example, neonatal care was reported to sit within ‘Family Care’, ‘Women’s and Children’s’, ‘Paediatrics and Neonates’ or ‘Family Health’.⁴
When asked about their accountability, some Trusts gave details of the organisations that hold them to account and to which they need to provide reports. Where Trusts gave details of accountability, they pointed to the fact that they are accountable to NHS England, their own governance structures, ICBs, CQC and government. Others referred to regulators, either in general or by way of a list of the specific regulators they report to.⁵ Others were more general and so rather uninformative, referring to, for example, “[t]he population we serve … NHSE and our regulators (CQC and others)”.⁶
The questionnaire revealed the complexity in both the accountability relationships in the NHS, and the wide range of ways in which Trusts organise their own governance within the boundaries of their statutory responsibilities.

Trust policies

Trusts were asked to submit copies of policies which dealt with the following:
1. safeguarding babies
2. investigating a neonatal death
3. Freedom to Speak Up Guardians
4. any other policies about escalating and raising concerns
5. whistleblowing
6. complaints.
Many of the policies were lengthy, with multiple sub-sections. From my own observation, this is typical of policies within the NHS. They were usually catch-all policies, which covered all ground relevant to the policy’s purpose. For instance, safeguarding policies covered adult, young person, child and neonatal safeguarding. The policies were described as “generic but comprehensive”.⁷
For many (but not all) Trusts, Freedom to Speak Up and whistleblowing policies were one and the same thing.⁸
There was very little useful information provided by Trusts about how often the policies relevant to reporting of incidents were used, with many reporting that either data was not available, or that the policies were always used when appropriate. This suggests that this information is not routinely collected, and Trusts may not have a clear picture from staff on how useful individual policies are in directing responses to incidents. As I have observed elsewhere in this Report, lengthy policies are unlikely to be read because there is no time. As a result they are not of any real use in directing responses to incidents. The nature of the questionnaire responses supports that view.
Approximately one-third of those surveyed (39 Trusts) had reviewed their policies as a direct result of Letby’s conviction. A further 22 had reviewed their policies for some other reason since Letby’s conviction, either as part of a regular review process or triggered by a change in national policy. 55 Trusts had not recently reviewed their policies. Of those who had not reviewed their policies, there was a suggestion that the review, when it happened, would have regard to the Letby case and the findings of this Inquiry, when published.⁹
19 of the Trusts that did review their policies made no changes. This appears to have been a deliberate and informed decision that had regard to the circumstances of the Letby conviction. There was often an explicit acknowledgement that there would be further review as this Inquiry published findings or national policies were changed.¹⁰
In total, 25 Trusts gave details of the changes made to their policies as a result of Letby’s conviction. Details of those changes can be found in the Nuffield report.¹¹ There was no general or common response from Trusts, which have responded to the findings in different ways.
Some Trusts pointed out further changes they had implemented that went beyond policy review. These included psychological support for staff, family support and information sharing, board briefing, training, and improvements to relevant processes (i.e. monitoring and governance). While there are no clear trends in this information, it does make clear that some Trusts have clearly tried to be proactive in supporting staff and parents at a difficult time for neonatal care.¹²

Staffing

Trusts measure their staffing levels against guidelines produced by BAPM. These guidelines include the levels of experience required for medical staff and nurses. Trusts appeared to have more access to data about nursing vacancies than medical vacancies, with many Trusts being ambiguous about the number of medical vacancies they had. However, even with this in mind, 68 of the Trusts surveyed (i.e. significantly over half) were not compliant with nursing levels under the guidelines, while 56 reported not being compliant with medical staffing requirements.¹³
Nursing levels were universally difficult to maintain, whereas medical staffing was more of an issue for units that treated sicker neonates (Level 3). This is a result of a greater level of specialism being required to fill those roles.
There was also evidence that staffing levels would vary depending on the shifts required to be filled. This was borne out in the data provided about staff cover, where evenings and weekends were, unsurprisingly, harder to cover than day shifts.¹⁴
There was also a reported problem in half of surveyed Trusts in recruiting or training nurses with a QIS in neonatal care.¹⁵
Many Trusts made the very good point that having a full rota and complement of staff did not mean that the guidelines were met. Sickness, parental leave and other absences all contributed to the continued need for temporary staff.¹⁶
Trusts had a number of different methods for analysing staffing needs. In general, they looked at occupancy rates and at acuity to consider the staffing needs of a unit. Where more staff are needed, employment of bank and temporary staff, cancellation of non-clinical activity, shift swaps and redeployment of staff from other units are all ways managers try to ensure that staffing levels are compliant.¹⁷ If necessary, cots can also be closed to ensure that the units remain safe. Given that staff shortages exist even where there is a full complement of staff on the books, thought should be given to assessing staff requirements in a way that takes account of predicted sickness levels, maternity leave and so on. This must be something that software can calculate, given the huge amount of staff data across the NHS.
It is clear from the evidence provided in the questionnaires that staffing remains a considerable area of difficulty for many Trusts. This means that staff are overworked. They do not have time to read, still less absorb, the mass of digital paperwork that comes in their direction.

Culture

Trusts’ responses to questions on culture varied significantly. This neatly makes the point that culture means different things to different people. It also supports my own view that there is no single NHS culture, no matter which adjective or descriptor is placed before the word ‘culture’. This is consistent with the responses of those who felt they should not pass comment as they were not well placed to describe the culture of a large organisation.¹⁸ They suggested that a Trust would have many different cultures within it. This is unsurprising and important, given the reliance placed on culture as an effective tool to improve patient safety.
Trusts often reported that their culture was good, but referred to a range of different facets in their answers. These reflect the diversity of ways in which culture has been referred to in other evidence before the Inquiry, and include ‘safety’, ‘justice’, ‘openness’, ‘transparency’, ‘dedication’ and ‘cooperation’.¹⁹
Broadly, where specifics were given about the nature of culture on neonatal units, they aligned with the wider Trust’s. There was also a view that neonatal units suffered from a greater level of scrutiny than other units. This can be an added tension to the culture on those units, on top of other pressures that included staffing levels, industrial action, Covid-19, waiting lists and burnout.²⁰
The increased scrutiny of outcomes can leave staff feeling that they are constantly under assessment.²¹ This is likely to be the result of events at the Countess. It is also likely that maternity units feel under intense scrutiny now given the widespread and long-standing criticisms and concerns about maternity units with the consequent inquiries, local and national. This year, two inquiries have reported and there is a further one ongoing.^†
That there are so many different factors listed by Trusts as evidence of both positive and negative aspects of culture shows how broad a concept culture has become. A positive culture encompassed initiatives such as ‘board breakfasts’ and celebrating International Nurses’ Day, while a negative culture was evidenced by bullying, harassment and discrimination, and failures to comply with Trust policies.²²
There was a near universal acknowledgement from Trusts that, even where culture was reported to be positive, there was still work to do and improvements could still be made.²³

Working relationships

In general, the respondents reported that working relationships were good. Some general challenges were noted, which involved the context of the Trusts in question. In particular, mergers and investigations were seen as acute stressors to working relationships between the different professions.²⁴
Working relationships between doctors and managers were reported as largely positive. This was especially true in Trusts that had a medical-led or multidisciplinary model, where clinicians were responsible for decisions.²⁵
Relationships between nurses, midwives and managers were not seen as positively. There was greater emphasis placed on the need for these relationships to be nurtured with regular, effective communication.²⁶ This echoes the findings of the Picker survey (see Appendix 3).
The relationships between different groups of clinical professionals were also broadly positive. However, it is worth noting that this was the view of senior managers, a group that the Picker survey suggested routinely overestimated the quality of their relationships with other occupations. Further, despite broadly positive views, disagreements could occur that required specific management and redress. Specific examples were given by different Trusts, but the suggestion is that different professionals may be more likely to disagree about care decisions, and there can be an element of tribalism involved that requires specific management.²⁷

Reporting and managing concerns and complaints

Nuffield recorded the very many ways in which patients and staff can raise concerns and complaints about NHS services. Patients wishing to raise a concern are often directed to their local PALS, while staff reporting routes will likely vary depending on internal relationships.²⁸ The most common starting points for staff were the Freedom to Speak Up Guardian and the staff member’s direct line manager.²⁹
Nine Trusts described specialised anonymous reporting routes for staff, while others made clear that the Freedom to Speak Up Guardians could be reached anonymously over staff intranet or other internal incident-reporting software. A ‘National Whistleblowing hotline’ (presumably NHS England’s Speaking up to NHS England initiative) was mentioned by only one Trust as a potential mechanism for reporting.³⁰ This speaks for itself. The hotline is either not known about or considered irrelevant. DHSC will have the figures for use of the hotline, but the responses to the survey suggest this hotline is not being used.
When a concern was raised, the most common response was for the incident to be logged on the Datix risk management system or some other similar system. This would then trigger a number of actions, including an investigation.³¹
Informal concerns or complaints would often be treated informally, with conversations between staff and signposting to formal complaint procedures if this was insufficient. These investigations or reviews were often not recorded, as they were dealt with at a ward level.³²
More serious concerns would involve investigation by staff at different levels of the Trust. This group was variously referred to as the ‘patient safety team’, ‘executive review group’ or ‘decision-making group’. The membership of such a group varied between Trusts but would often be multi-professional. The group would then often provide an incident report to the board for consideration, though this may not be the process for less serious incidents, which would be reviewed by an intermediate body. Sign-off of the final written responses to the complaint or concern would often be the responsibility of a board member, usually the Chair or Chief Executive.³³
108 Trusts reported at least one complaint or concern raised about neonatal care in the year from 19 October 2022 to 19 October 2023. Of those, 97 Trusts had carried out at least one investigation in the same timeline. Respondents did not give much consistent information about the nature of the concern or complaint, the severity or complexity of the issues raised, or the details of the investigation carried out.³⁴
Trusts were clear that there were several external bodies that they were required to share information with, in specific circumstances. These include MBRRACE-UK, StEIS, Maternity and Newborn Safety Investigations, the Disclosure and Barring Service and CDOPs. Some Trusts considered it appropriate to refer issues to professional regulators for further investigation and action.³⁵
In cases where disciplinary processes also needed to be followed, the Trust’s HR department would be involved, alongside local employer liaison advisers from the GMC or the NMC. This process is guided by a framework called Maintaining High Professional Standards.³⁶
Trusts referred to regulators, the Disclosure and Barring Service and requesting a Healthcare Professional Alert Notice from NHS Resolution as mechanisms that could be used to manage a situation where there were concerns regarding the fitness to practise of an employee who had left the Trust.³⁷
Serious Incidents were dealt with according to the national Serious Incident Framework. They were discussed at regular meetings with ICBs. The reporting of incidents to the Learn from patient safety events service has replaced the NRLS. This data is reviewed by CQC and NHS England.³⁸
If the incident involved an early neonatal death, a report would be made to the coroner and reviewed using the national Perinatal Mortality Review Tool. A referral would also be made to Maternity and Newborn Safety Investigations, which investigates early neonatal deaths.^‡ Where the incident was categorised as a safeguarding issue, the LADO policy would be followed for escalation.³⁹
50 Trusts escalated a concern to their own board in the year from 19 October 2022 to 19 October 2023.⁴⁰ 56 Trusts reported that they had referred to external agencies in the same timeframe. The most common external referral pathways were MNSI and StEIS.⁴¹
24 Trusts had made a report to a professional regulator or CQC, with the NMC being the most common regulator referred to.⁴²
Trusts were asked to give indications of factors that either inhibited or encouraged staff to raise concerns. The factors listed included:⁴³
1. A culture of not reporting: They mentioned in particular competitive cultures between professions, and a fear of raising incidents being seen as punitive or risking punishment.
2. A culture of reporting: Positive reporting and learning cultures were seen as encouraging reporting, as were initiatives such as active bystander programmes.
3. Staff voices: Some voices would be prioritised, and some did not speak out in circumstances where they thought that their voice would not be listened to.
4. Complex reporting structures: These were often seen as inhibiting, but some Trusts saw the many routes through which concerns could be raised as a positive, as there were many opportunities for staff to raise issues.
5. Public scrutiny: There was a suggestion by some Trusts that the level of public scrutiny over maternity and neonatal services inhibited reporting of concerns.
The most common factors inhibiting managers from acting on concerns were financial strain and understaffing. Another concern that was repeatedly referred to was the capacity and capability of managers who handled complaints. This included the fact that neonatal care is so specialist that it can be hard for managers to deal effectively with complaints in this area.⁴⁴

Reviewing evidence after a death

Trusts provided information in different ways about the number of requests for perinatal pathology they had submitted. Some gave a full account of every request, while others reported having no information available. Further, the information provided about the length of time requests took to be processed varied significantly across Trusts. There were clear regional disparities in waiting times, from six to eight weeks to over nine months.⁴⁵
There was no consistency in the way in which unexplained or unexpected deaths were recorded by Trusts.⁴⁶ This is striking and echoes the findings I have made about the position ten years ago. This important matter needs to be addressed by DHSC/NHS England and as a serious training issue. I acknowledge and hope that the introduction of the medical examiner system in September 2024 has begun to improve consistency in this regard, given the responses to the questions about medical examiners in Chapter 32.
Nuffield conducted a thematic analysis of the reasons given for the review process after a death. 119 Trusts referred to ‘learning’ as the key reason for any such review process. 31 Trusts spoke of the need for a robust and thorough investigation. Most Trusts reported learning, explanations and transparency/accountability as the purpose of a post-death review, rather than a robust and thorough investigation.⁴⁷ It seems obvious that after a death the circumstances should be investigated thoroughly, the outcome should be shared and responsibility identified. A review after a death should be transparent and should look at all four of those aspects.
An important finding is that every Trust that responded had access to a medical examiner. 82 Trusts had used a medical examiner to review a neonatal death. Most importantly, 112 Trusts had processes in place for medical examiners to identify trends or patterns in deaths across the Trust.⁴⁸ This is encouraging, not least because the questionnaire predates the full statutory system now in place in England and Wales.
Trusts reported attending between zero and ten CDOPs in the year from October 2022 to October 2023. CDOPs were convened at various intervals, depending on the way in which they were organised in the local area, with some Trusts attending every panel and some sharing attendance with other local Trusts. As CDOPs review deaths based on the area in which the patient was resident and not by Trust, this could mean that some deaths would not be reviewed by the relevant Trust at CDOPs. While some Trusts were able to contribute to panels convened out of their area, others found they were unable to.⁴⁹ I have dealt with this issue in Chapter 12.
Nuffield completed a thematic review of the Trusts’ responses to a question about the effectiveness of CDOPs. The primary benefit of CDOPs was seen by Trusts as their contribution to local learning, with 64 Trusts referring to this benefit. The independence of the scrutiny and the panels’ expertise across multiple disciplines were also consistently seen as a benefit (57 Trusts). The most common limitation of CDOPs’ effectiveness was the time delay between deaths and the panel being convened, reported by 14 Trusts.⁵⁰ This requires the attention of DHSC. I acknowledge that the CDOP process comes after all other investigations are completed, but since that is identified as hampering its effectiveness, a review is needed to establish whether the order of events could be changed or, more radically, to consider the purpose of CDOPs when an inbuilt lengthy delay is apparently unavoidable.

What safety nets exist?

Trusts reported that parental engagement in neonatal care is prioritised because of the benefits of parents being involved in both decisions relating to care and the care process itself. 118 Trusts reported that parents were involved in ward rounds, with many Trusts clearly adapting their practice to ensure that parents could attend. This included changing the timing of ward rounds, offering virtual attendance, or alternatives such as telephone consultations. It was also clear from responses that parental attendance at ward rounds was monitored by a number of Trusts, emphasising the point that this was taken seriously.⁵¹ These were encouraging responses.
116 of the 120 Trusts surveyed routinely gave discharge summaries to parents, with some providing these before discharge to allow parents to review and discuss the summary with clinicians as appropriate.⁵²
It was not routine for parents to have access to medical records. While some Trusts did provide access to some medical records through online portals, it was more common for these to be available on request, either through PALS or subject access requests.⁵³

CCTV

CCTV was reported at either the entrances and exits or in communal areas of the neonatal unit by 99 Trusts. No Trust reported that CCTV was present in clinical areas. 57 Trusts reported that they provided some form of access to webcam footage or video calling for parents. This was either live, ad hoc or pre-recorded.⁵⁴

Medicine management

Policies for the management of medicines varied across Trusts but broadly followed the same principles. No Trust reported a specific neonatal or maternity drug storage and administration policy. Medicines were kept locked away and required either a key or a swipe card to access them. Often, this key would be kept by a senior member of staff on duty. That member of staff had ultimate responsibility for the medication used on the ward. 19 Trusts had a system whereby the swipe card used to access medication was logged, allowing for a register to be created showing who had accessed medicine at what time. For those Trusts without electronic registers, physical logbooks were used. Where information was included, insulin was reported to be kept in locked fridges or, in one case, was now treated in the same way as a controlled drug as a result of the Letby case.⁵⁵
Where drugs were being administered to neonates, most Trusts reported that their preparation and administration had to be checked by two authorised members of staff. Both members of staff then signed the relevant record, either on paper or electronically.⁵⁶

Support for bereavement

The primary mechanism through which families are supported when they suffer a bereavement after their child’s treatment on a neonatal unit is PALS. The service was either signposted when neonates were admitted to the unit, or after death. Bereavement midwives were also trained and present in some Trusts, and would refer families to PALS as appropriate.⁵⁷
A further layer of support referred to by Trusts is the continued contact between the neonatal service and bereaved families that happens as part of review processes. This can either be as part of the duty of candour, where meetings are arranged and information sent out, or part of the work of medical examiners or clinicians completing the Perinatal Mortality Review Tool. Where appropriate, Trusts can also refer family members on to therapy services (such as those offered by the charity Petals) and other specialist support.⁵⁸
Trusts were less clear about the process through which support would be offered to parents who had previously suffered a neonatal bereavement when there is a new pregnancy. Where available, Trusts mentioned that they would refer to a specialist clinic (a so-called rainbow clinic). However, in areas where this was not available, Trusts seemed to rely on good communication between neonatal and maternity units to ensure that families got the care they needed.⁵⁹
There was no consistent process for recording on medical records that families had been bereaved. 37 Trusts used a sticker system (either in physical form or on electronic notes) to denote bereavement. To work effectively, these required the community midwife or other clinician to look at medical notes before the appointment. Another common process was for discharge letters sent to local health networks and to GP practices to highlight the need for future support.⁶⁰

Learning and making improvements

Trusts were asked to suggest structural improvements that could be made to the management and governance of neonatal services. The clearest and most frequently made suggestion was that there needed to be simpler and more standardised processes, especially for external reporting and regulation. It was felt there was duplication of work in this area, and that the level of scrutiny placed too high a burden on Trusts.⁶¹ I agree.
Some suggestions were more local and Trust specific, while others were broader. For instance, some Trusts suggested that what was needed was better leadership, or an improved culture.⁶²
There was disagreement between Trusts about the appropriate way of organising neonatal units within wider maternity or obstetric services. Some preferred a ‘perinatal service’ to ensure continuity of care. Others thought this may risk a lack of focus on neonatology, and could reduce visibility for neonatal services at board level as they are taken together with other services.⁶³ This disagreement is reflected in the manner in which Trusts do, in fact, organise their services. Neonatology often sits in different places, which suggests that there is no one agreed answer to the question of how to structure the provision of neonatal care.
When asked what specific lessons Trusts had learnt in the period since October 2022, many Trusts took the opportunity to reassure the Inquiry that they are learning organisations that are continually looking to develop and improve.⁶⁴ When looking at specific answers to the question, the most common theme of Trusts’ answers was that there needed to be an improved culture and a greater focus on reporting concerns.

Regulation of managers

Respondents gave a variety of answers to the question of whether managers in the NHS should be regulated. Four themes were identified by Nuffield that were relevant to the question:
1. the purpose of regulation
2. other existing processes
3. whether regulation is an appropriate response
4. how regulation would work in practice.⁶⁵
Managers said that regulation could be beneficial for providing accountability and credibility to senior managers. There was some concern that regulation would be used as a method to identify scapegoats. To help prevent this, regulation was also identified as a way in which professionalism, training and development of senior managers could be properly supported. Respondents felt that better training could help to develop a career pathway for middle managers to take on higher-level roles. This would help with recruitment and retention, and could help senior managers to feel properly valued.⁶⁶
Some respondents pointed out that there are already a number of formal accountability processes in place for senior managers, either with their own professional regulators or via other routes such as CQC, employment contracts, the fit and proper person test and codes of conduct (including the Nolan principles).⁶⁷
There was concern (even amongst those respondents who felt that regulation was, in principle, a good thing) that it was not the proper response to the issues that the case of Letby highlights. There is a clear fear that regulation is a byword for blame, and that a move towards regulation in response to Letby is actually a thinly veiled attempt to scapegoat senior managers. Some respondents feel strongly that regulation (or what they describe as over-regulation) would not improve patient safety, and would in fact create more bureaucracy that would not be beneficial (see Chapter 37).⁶⁸

Alternative to regulation

Instead of regulation, it was suggested that strengthening existing processes would better support patient safety and prevent a repeat of the events seen at the Countess. Three Trusts mentioned improving oversight and governance, and six Trusts mentioned improving the process for reviewing neonatal deaths.⁶⁹

Summary

It is plain that there is wide variation amongst Trusts in the way they manage the provision of neonatal services. What is striking is the clear evidence that managers see culture as a catch-all solution to improve patient safety and care outcomes and also a reason, if not the reason, that structural improvements are needed in the first place. Culture was mentioned as a cause and a solution. When asked what was learnt in the aftermath of Letby’s conviction, culture was mentioned as central. When asked to evaluate factors that improve or inhibit the reporting of concerns, culture was mentioned as both a help and a hindrance. Predictably, given this context, the responses to questions specifically about culture make clear that it is not a term that is well understood or consistently applied across the NHS. I deal further with the question of culture in Chapter 39.

Footnotes

* This did not include the Countess.
† These are: the National Maternity and Neonatal Investigation led by Baroness Amos, which covers 12 local investigations; the Independent Review of Maternity Services at Nottingham University Hospitals NHS Trust; and the Independent Review into Maternity and Neonatal Services at Leeds Teaching Hospitals. There are also calls for individual inquiries into Oxford and Sussex Trusts.
‡ These are deaths of babies who are born alive and die in the first week (zero to six days) of life.