Authors' objectives: There are multiple reasons for involving patients and families in incident investigations. Fiscally, costs due to clinical negligence claims approximate £4 billion annually. Logically, patients and families provide important information about patient safety incidents. Morally, involving harmed patients and families helps address their concerns. However, little United Kingdom-based evidence was available to support systematic involvement. To co-design processes and resources to guide the involvement of patients and families in incident investigations at a national and local level, and to test these processes to understand their impact upon experience, learning and likelihood of litigation. The burden of patient safety incidents is significant. Fiscally, 10,000 are estimated to occur annually within the UK, and costs associated with clinical negligence claims and their administration approach £4 billion per year. Some of these claims are thought to result from the experience of incident investigations themselves, emphasising the need to improve how patients and families are involved. Involving patients and families is also important logically, due to their well-established role in patient safety. Evidence from the USA has shown that patients and families can identify contributory factors to patient safety incidents, which if gathered, may support better organisational learning. However, at the time of the award, evidence from the UK was limited, with no known studies from the UK exploring the process of involving patients and families in incident investigations, no evidence that this leads to improved experience for those involved, improved learning or reduction in the likelihood of litigation, and no UK-relevant structured processes for how to systematically undertake successful involvement. The proposed research programme aimed to address these gaps, with the following overall aim: To co-design processes and resources to guide the role of patients/families in incident investigations at a national and local level, and to test these processes to understand their impact upon experience, learning and likelihood of seeking legal recourse. Further funds were awarded during the programme to explore the perceived utility of investigations following suicide. Within mental health care, incident investigations are the primary source of exploring risk, as well as understanding factors contributing to or surrounding suicide, and how to prevent recurrence. However, there are currently no known UK-based studies exploring local investigations following suicide, and how they contribute to organisational strategies to prevent incidence. Further, given the established links between familial isolation and suicide, what meaningful involvement means within such investigations is complex, with no known UK-based studies exploring who should be involved. RQ1: What is the current involvement of patients and families in incident investigations? RQ2: What is the experience of patients and families who have been involved in an incident or incident investigation, and what might have influenced decisions to litigate? RQ3: What is the experience of front-line healthcare staff and investigators who have been involved in an incident investigation, and what might have influenced decisions to litigate? RQ4: What are the views of front-line healthcare staff and investigators on the potential involvement of patients and families in incident investigations? RQ5: What are the common principles necessary for involving patients and families in incident investigations? RQ6: How might these common principles be reflected in local and national processes for involving patients and families in incident investigations? RQ7: Are co-designed processes for involving patients and families in incident investigations feasible and acceptable to patients, families, healthcare staff and investigators? RQ8: How do co-designed processes influence incident investigations in terms of depth of learning, recommendations, action plans and decisions. RQ9: Do stakeholders involved in incident investigations following death by suicide believe they contribute to organisational learning, and risk management and suicide prevention strategies? RQ10: How do we define meaningful involvement in investigations to prompt learning following death by suicide?

Authors' results and conclusions: Investigations are complex, relational processes. Our guidance was found to be feasible, with stakeholders being positive about involvement and the impact on organisational learning. It may help to reduce the significant and long-lasting experience of compounded harm for patients and families. However, involvement may always be challenging due to the divergent needs of patients/families and organisations. Incident investigations are complex, relational processes, which have the potential to either repair or compound the harms from patient safety incidents. Our co-designed processes are feasible and acceptable, and support more comprehensive and systematic involvement of patients and families in investigations, and the learning that arises from them. Importantly, they might support the reduction in the significant and long-lasting effects of compounded harm for patients and families, although this study design cannot establish this. Investigations are complicated by divergences in the needs of different stakeholders involved in investigations, often relating to their purpose and focus. Navigating this complexity requires skilled, well-supported investigators, who work with an organisational infrastructure flexible enough to allow them to individualise their approach.

Authors' recommendations: Policy and procedures need to formally recognise that there are multiple purposes for responding to a safety incident and that organisational learning is only one of these purposes. The relational work of involving patients, families and staff is important, but complex, and needs to be resourced, valued and recognised within policy and processes. When embedding processes for involving and engaging patients and families in incident investigations and responses, organisations need to first seek to understand how this is currently done and seek to adapt current organisational infrastructure to support them. Organisations should undertake ongoing monitoring of processes for involving and engaging patients and families, which centre their experiences, as well as objective outcomes, such as subsequent complaints and litigation. Research should explore how current approaches to involving and engaging patients and families in incident responses might be different for people from minoritised and underserved groups, and what adaptations might be required. Research should explore the possibilities for ‘harm-centred’ rather than ‘incident-centred’ responses to safety, and how this might address the seemingly intractable divergences in the needs of patients, families and organisations.

Authors' methods: A mixed-methods programme of research was undertaken. Stage 1 comprised a scoping review of evidence for the experience of patients/families in incident investigations, and a documentary analysis of 43 National Health Service Trust incident investigation policies. Stage 2A extended this with 41 qualitative interviews with patients/families, healthcare staff and investigators. Stage 2B synthesised previous data to develop common principles and programme theory. Stage 3 involved a 6-month co-design phase with a ‘co-design community’ of > 50 stakeholders. In stages 4 and 5, co-designed guidance was evaluated in a 15-month ethnography, within four National Health Service Trusts and the national independent investigatory body. Twenty-nine investigations were followed in real time, including 127 interviews and 45 hours of observation. Four final co-design workshops supported iterations to the final guidance and website. A substudy explored meaningful involvement in, and learning from, investigations following suicide via interviews and a qualitative survey involving 32 people (healthcare staff, policy-makers and managers; people bereaved by suicide). Undertaking research during the pandemic may have impacted sample representativeness in stage 2A. Ethnically minoritised and lower socioeconomic groups were under-represented across the programme. This stage comprised two studies. First, a scoping review of the qualitative literature explored the experiences of patients and families within incident investigations, and what prompts decisions to litigate. Three databases were searched, followed by independent screening of title and abstracts (20% double-screened). Data were extracted from included studies, before undergoing a narrative synthesis. Second, we conducted a documentary analysis of policy documents referring to the involvement of patients/families in incident investigations. A random sample of 103 NHS Trusts (50% of all trusts in England) were approached, supplemented with searching trust websites. A total of 43 documents were sourced and submitted to a qualitative documentary analysis, with particular attention to how involvement of patients, families and staff was described, and how this was presented in the context of the whole document. A qualitative semistructured interview study was conducted with patients/families, healthcare staff, legal staff and investigators. We invited participants via (1) communication from partner sites, (2) charitable organisations, (3) social media, (4) snowball sampling. One hundred and seventeen people registered interest and 42 participated (18 patients/families, 7 staff, 16 investigators, 1 legal representative). Data were analysed using an inductive reflexive thematic approach. A three-phase analysis and synthesis was conducted on data from stages 1 and 2A. The inductive analysis phase involved creating short descriptive reports for each of the previous three studies. An abductive analysis phase based on three foundational theories and approaches created a new analytical framework to support the development of the draft common principles and narrative programme theory. A synthesis phase ran parallel with the first two, bringing them together through a series of analytical workshops. We conducted a longitudinal, largely remote co-design process, informed by the UK Design Council Double Diamond for Innovation. Co-design activities comprised two large stakeholder events bookended by a series of three co-design workshops that ran in three parallel workstreams, reflecting (1) acute settings, (2) mental healthcare settings and (3) national independent investigatory body setting. A ‘co-design community’ of > 50 stakeholders was formed, with members invited to participate in all activities. A range of innovative co-design activities were used to build relationships and trust and support generation of ideas. A 15-month focused ethnography was conducted across four NHS Trusts, and within maternity investigations undertaken by the national investigatory body. We trained 49 investigators in the new guidance, before working with 16 investigators within 29 real-time investigations. We undertook 127 remote qualitative interviews (92 semistructured, recorded interviews, and 35 ethnographic interviews) at different investigation touchpoints, with investigators, patients/families and staff. A duration of 44.5 hours of non-participant observation was undertaken, including investigator training and site-based meetings. Analysis proceeded using an inductive approach. First, an adapted version of pen portrait methodology brought together all data into individual cases at the level of the investigation, along with contextual organisational summaries. Following this, we conducted a multicase analysis using a reflexive thematic approach, with the investigation case as the unit of analysis, in conjunction with the organisational summaries. Following the ethnography, a series of four final workshops were convened with co-design partners and other stakeholders to discuss findings and agree to the required iterations to the final guidance. We conducted a qualitative study combining interviews and a qualitative free-text survey. Semistructured interviews were conducted with staff, managers and policy-makers with experience of investigations following suicide. Purposive sampling was employed, with recruitment undertaken via social media and approaching policy-makers directly. A fully qualitative survey was developed to explore the experience of those bereaved by suicide, of the investigation and other processes that follow. The survey was based on the stage 2 findings and further developed with a family representative. The survey was distributed via social media, charities relating to suicide, and the Care Quality Commission (CQC). Data were combined and analysed using reflexive thematic analysis. This programme produced two important contributions to co-design methodology. First, we expanded the understanding of how co-design might be undertaken with policy-makers, and the benefits of this approach. Second, we have outlined an innovative approach to co-designing policy, where evidence was moved into policy and practice at the point of need – in this case, during a significant national policy shift. There are two principal limitations to the work presented here. First, the programme commenced just ahead of the pandemic, which impacted the stage 2 interview study particularly in terms of the potential representativeness of the sample. However, given the similarity in experiences of investigations between stage 2 (interview study) and stage 4 (ethnography), we believe the sample was unlikely to be unusual. Second, and more significant, is our difficulty achieving a sample across studies that is representative of healthcare populations. In particular, we struggled to connect with or hear from ethnically minoritised people or those from lower socioeconomic groups.

Authors' identified further research: Research should explore how people from minoritised groups experience investigations and any required adaptations to the approach. Research should also explore the possibilities for ‘harm-centred’ rather than ‘incident-centred’ responses to safety.

Project Status: Completed

URL for project: https://www.journalslibrary.nihr.ac.uk/programmes/hsdr/18/10/02

Year Published: 2025

URL for published report: https://www.journalslibrary.nihr.ac.uk/hsdr/KJHT3375

URL for additional information: English

English language abstract: An English language summary is available

Publication Type: Full HTA

Country: England, United Kingdom

DOI: 10.3310/KJHT3375

Organisation Name: NIHR Health Services and Delivery Research programme

Contact Address: NIHR Journals Library, National Institute for Health and Care Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK

Contact Name: journals.library@nihr.ac.uk

Contact Email: journals.library@nihr.ac.uk