Specialist intellectual disability liaison nurses have been recommended to work across general hospital departments to improve the experience and outcomes of people with intellectual disability who are admitted to hospital. Our data show that only a minority of adults with intellectual disability who died in hospital in England between 2016 and 2021 received input from an intellectual disability liaison nurse prior to their death. This may be due to a person’s intellectual disability not being recognised or recorded when they are admitted to hospital meaning that they are not identified as being eligible for specialist input.25 Alternatively, it may be that not all hospitals employ intellectual disability liaison nurses or, where they do, coverage is limited by resource constraints or because the post is split across a number of sites. It is not mandatory for hospitals in England to employ an intellectual disability liaison nurse service and no national-level data exist that show the extent of intellectual disability liaison nurse provision in hospitals. A survey of children’s hospitals in England found that just over half had a dedicated intellectual disability nurse.26 A recent NHS benchmarking exercise found that 30% of staff in NHS secondary care services did not have access to specialist intellectual disability advice, although respondents included those working in community and mental healthcare settings, as well as those in hospitals.27

The regression analysis showed certain patient-level characteristics were significantly associated with the likelihood of receiving input from an intellectual disability liaison nurse. After adjusting for other factors, people who died in the two regions covering the north of England were significantly less likely to have seen an intellectual disability liaison nurse prior to their death. This may be due to difficulties in recruitment or retention of specialist staff in those areas. Intellectual disability nurses represent only a small proportion of all nurses in the UK and their numbers are falling.28

Cancer is the fourth most common cause of death in people with an intellectual disability29 and those who died of the condition were most likely to receive input from an intellectual disability liaison nurse. This might be because of relatively well-resourced and structured care pathways for people diagnosed with cancer and their integration with palliative care services for people with a terminal diagnosis that has a holistic and multidisciplinary focus. It is also possible that those with cancer tend to be admitted to larger hospitals which may be more likely to have intellectual disability liaison nurse input.

People with multiple long-term health conditions were more likely to receive input from an intellectual disability liaison nurse. It may be that the input of an intellectual disability liaison nurse was sought by the responsible care team in these cases due to the complexity of care being increased, or it may also be that intellectual disability liaison nurses prioritise these cases where their resource is limited. Although not statistically significant, we also observed that people living in supported or residential care were more likely than those living in their own home to receive the support of a specialist nurse; this may be because paid carers advocated for greater input in the hospital. It might also be due to intellectual disability not being recognised in people with less significant functional impairments (who live independently), although there was no observed association between degree of intellectual disability and liaison nurse input.

DNACPR orders provide guidance to medical staff about how to respond when someone has a cardiac arrest.30 DNACPR decisions are not legally binding and are not signed by the patient or a witness (in contrast to Advance Decisions to Refuse Care) but should be discussed with the person and their family or carers, where possible. Previous research has shown considerable variation in how DNACPR policies are written and enacted for the general population, with significant scope for improvement.31 32 In addition, there has been renewed focus on the appropriateness and rigour of the DNACPR process since the COVID-19 pandemic, with people with intellectual disability considered at particular risk of poor and discriminatory practice, including blanket policies to deny CPR based on a diagnosis of intellectual disability or specific genetic conditions.33 The finding that the involvement of an intellectual disability liaison nurse was associated with an increased likelihood of a DNACPR order being made perhaps suggests that the nurse can facilitate necessary end-of-life care discussions, including about DNACPR and avoiding inappropriate and futile resuscitation attempts. However, liaison nurse input was not associated with the quality of DNACPR documentation or whether the order was correctly followed; this may indicate that the liaison nurse’s influence is limited in a medically driven process or that they were not closely enough involved in the direct care of the person to provide advice and oversight about implementation when a decision has been made.

Under the UK Equality Act,34 hospitals and other providers of services have a legal duty to make reasonable adjustments to ensure that people are not disadvantaged by nature of any protected characteristic, including disability. In the context of intellectual disability, reasonable adjustments to hospital care may include environmental adaptations on the ward, providing information in accessible formats, or permitting carers to remain with a person outside usual visiting hours.35 A core component of the intellectual disability liaison nurse’s role is to ensure person-centred care when someone is admitted to hospital36; this was clearly evident in the findings which showed that those with input from a liaison nurse were almost twice as likely to receive any reasonable adjustments to care. Similarly, there was more likely to be evidence of best practice in the person’s care when their case was reviewed where an intellectual disability liaison nurse was involved. Best practice here is broadly defined and could include good practice when dealing with families and carers, following guidelines, or applying the principles of the Mental Capacity Act.37 One previous audit of hospital in-patient care of people with intellectual disability in England found that there were improvements in markers of high-quality care (including having appropriate assessments and care plans in place) in those hospitals where an intellectual disability liaison nurse was employed, but this study was underpowered and could not directly link care of individual patients to liaison nurse input.38

In contrast, there was no observed association between input of an intellectual disability liaison nurse and overall quality of care rating assigned by the reviewer. This question requires the reviewer to consider all aspects of care that the person received, not limited to healthcare or care received in hospital and includes a multitude of factors that an intellectual disability liaison nurse could not reasonably be expected to have any influence over, such as quality of support provided in the person’s home or the appropriateness of community activities. Nonetheless, our findings seem to indicate that the input of an intellectual disability liaison nurse is associated with tangible benefits for the individual. Qualitative studies that have included people with intellectual disability and carers have reported that intellectual disability liaison nurses are highly valued20 and improve hospital experiences for people with intellectual disability, including in communication, patient safety, holistic care and the provision of reasonable adjustments.15 36 39

There is little existing evidence of the impact of intellectual disability liaison nurses on objectively measured patient-level care outcomes. To our knowledge, the intellectual disability liaison nurse role has not been implemented beyond the UK and the Ireland,21 despite international evidence of poor care provision for this group.38

We obtained data from a very large national dataset of in-depth individual reviews of the care that people with intellectual disability received before their death. Reviews are completed using a standardised method by trained reviewers who can access a person’s complete medical record and who gather data from a range of sources, including by speaking to informants who knew the deceased person in either a personal or professional capacity. The reviews are of deaths that occur across all regions of England over a period of 5 years and findings can be considered representative on this basis. There are very low rates of missing data in each variable and the data gathered can be trusted as accurate and comprehensive. We were able to identify factors associated with the likelihood of receiving input from an intellectual disability liaison nurse, potentially highlighting disparities in access to care. The study adds to the limited empirical evidence of the impact of intellectual disability liaison nurses on individual outcomes which, given the largely positive findings, could be used to strengthen existing recommendations for acute hospitals to provide this service.

Although reporting deaths to the LeDeR programme is strongly encouraged, it is not mandatory and there may be a small number of people who died who are eligible for a LeDeR review who did not receive one; we do not have their details and are not able to characterise this group. The dataset includes only those people who have died, and therefore we are not able to draw inferences about the input and impact of intellectual disability liaison nurses for people who are admitted and later discharged from the hospital.

This study is a secondary analysis of data that were collected to improve services but not specifically to answer our research questions, and our analysis is limited by the data that are available. Intellectual disability liaison nurse input was identified using a single yes/no question, which cannot identify variation in the type or degree of involvement in individual cases. Liaison nurse models of working may differ between hospitals and across the country, but we do not have information available to define liaison models in use and to associate these with outcomes. As this is an observational study, we are not able to determine the direction of associations that were shown between intellectual disability liaison nurse input and other variables. We have no clear measure of patient or carer experience or reported outcomes and how these may have been influenced by input of an intellectual disability liaison nurse. Although there are safeguards to ensure LeDeR reviews are standardised, there may have been some unmeasured variability in how individual reviewers classified good practice or rated quality of care.

A national census of intellectual disability liaison nurse posts would provide more detailed information about coverage and access to specialist care and identify where gaps exist. This would also help to define different modes of liaison nurse work at organisational and individual levels which could be evaluated in a prospective study to determine the optimum model of liaison working. A health economic evaluation as part of this would demonstrate the financial implications of the role and is important where there are competing demands on the healthcare budget. It will be important that future work extends the current study by including people who did not die in hospital; this could be achieved through sampling those who have been identified as having intellectual disability on their discharge documentation or by using linked datasets. This could also include people who attend the hospital as outpatients as well as those who are admitted. It would also be possible to expand the evaluation to new settings, to examine the benefits of intellectual disability liaison nurses in primary care or acute mental health services.

The perspectives and experiences of people with intellectual disability and family carers should be sought so that practice advances can take these into account. Such studies may be conducted as in-depth qualitative evaluations, appropriately adapted to enable the participation of people with communication needs, and would complement the potential national-scale work described above.