Paediatric mental health liaison services in England: findings from a national survey

Abstract

Background

Paediatric Mental Health Liaison (PMHL) in England is poorly understood with limited data on services previously collected.

Aim

To describe the characteristics of PMHL services in England and create a service model typology.

Method

Data were collected using a cross-sectional online survey adapted from the questionnaire used by the existing adult-care-focussed Liaison Psychiatry Surveys of England (LPSE). The survey was distributed via email through professional networks and social media. Survey data were supplemented with direct correspondence with survey respondents. Responses were coded numerically and analysed descriptively using SPSS. Free-text responses were analysed using NVivo.

Results

Results were obtained from 28 services serving 47 hospitals. Three service models were identified; Type 1 services providing psychological care for children and young people (CYP) with long-term conditions (LTCs); Type 2 services providing care for CYP in crisis, and Type 3 services providing care for both groups. Few services provided 24/7 crisis care for all under-18s; many respondents reported this was provided by community child and adolescent mental health (CAMH) teams. Type 1 services were often constrained by condition-specific funding. Type 1 and Type 2 services lacked multidisciplinary staffing compared to Type 3 services.

Conclusions

This is the first survey of English PMHL service provision. Findings are exploratory, as they are based on a partial and non-representative sample of services. However, they suggest significant variation by hospital including a lack of dedicated 24/7 crisis care for all under-18s, and inequitable access to psychological support. Further research is required to replicate findings with a larger dataset, and to explore differences in service models. However, this preliminary data suggests that adequate PMHL provision would benefit from clearer national standards. Commissioning should ensure adequate funding and equity with adult provision. Terminology disagreements represent a research barrier.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-026-14206-w.

Introduction

The mental health of children and young people (CYP) is a global public health priority. The World Health Organisation estimates 10% of CYP worldwide experience a mental disorder, although most will not seek help or receive appropriate care [1]; furthermore, they cite suicide as the leading cause of death for 15–19 year olds.

CYP who attempt suicide, or experience conditions with both physical and psychological components, e.g. self-harm, eating disorders or medically unexplained/persistent physical symptoms (MUS/PPS) are frequently and increasingly seen in acute hospitals [2]. A further group of CYP frequently receiving mental health support in acute hospital settings are those with long term physical health conditions who are up to four times more likely to experience psychiatric symptoms than their peers [3].

Care for CYP in these groups requires an approach which addresses physical and psychological health simultaneously [4–7]. However, this population group is often treated in generalist settings such as emergency departments (EDs) or paediatric inpatient wards, where professionals may not have the training, skills, or knowledge to identify or appropriately manage their psychological needs [8]. Recent data suggest increasing use of generalist settings for this group of young people, partially due to the inability of community child and adolescent mental health services (CAMHS) to cope with increasing referrals over the past few years, upwards trends in ED presentations, and subsequent admissions of young people to general paediatric wards where their mental health is the primary concern [9–11]. Consequently, this group of CYP are at risk of inappropriate, inadequate, or unnecessary investigations and treatments, often with delayed referral to specialist CAMHS.

One potential solution to this problem is investment in service models that integrate mental and physical health services, which are usually referred to as mental health liaison (MHL) services or liaison psychiatry services [12]. In 2016, government policy identified the need to improve access to 24/7 emergency mental health care in England for adults and CYP, aiming for all acute hospitals to have ‘all-age’ liaison psychiatry services by 2020/21, with 50% of these services meeting a national standard known as the ‘Core 24’ standard [13]. The Core 24 standard requires a distinct, on-site 24/7 service, with specific staffing and skill-mix levels including 24/7 access to psychiatry, which is able to respond in 1 hour to emergency referrals and provide biopsychosocial assessment and psychological intervention for patients admitted to the hospital. However, Core 24 only relates to adult services; confusingly ‘all-age’ in this context means adults of all ages, rather than including under 18s [14]. Some services do provide a ‘cradle to grave’ MHL service, and these would be held to Core 24 because of their 18 + provision. However, it is not clear how appropriate this is and there are no such equivalent standards for standalone CYP services – including no suggested staffing numbers or ratios. Similarly, while the UK’s Psychiatric Liaison Accreditation Network (PLAN) has developed a set of one hundred and twenty five standards for MHL team accreditation, only eight of these currently relate to CYP [15]. There is, therefore, a significant inequity between standards and expectations for MHL provision for adults and for children.

The annual Liaison Psychiatry Surveys of England (LPSE) focus mainly on adult service provision; however they give some insights into English PMHL services in EDs. This data suggests that PMHL is less well-developed than adult provision [16] and often provided on an ad-hoc basis or by external providers, rather than by dedicated hospital-based teams. The most recent LPSE found that, out of 170 hospitals with EDs in England, just 24 (11%) currently provide an all-age liaison service that includes CYP provision; however, there is no information regarding the characteristics of these services [17]. Furthermore, research suggests that dedicated 24/7 services for CYP under 18 requiring mental health support in acute hospitals are limited [18]. In relation to mental health support for CYP with long-term physical health conditions, a study conducted in 2008 found that most paediatric psychology services operated as a standalone team within an acute hospital, with few operating as part of a MHL team in that setting [19]. However, there is limited detail about the differences between these two types of service provision and an overall lack of understanding about how these services are currently organised and provided and their integration with MHL teams [20].

Thus, there is a lack of current data about MHL services for CYP and consequently a limited understanding of services in England beyond the conclusion reached by Hines et al. [16] that they appear to be less reliably available than adult services, and possibly less likely to meet the Core 24 Standard. Indeed, the assumption that paediatric liaison services should be held to the same standards as adult services poses a problem given the specific needs of CYP at different developmental stages, and comparisons of paediatric services with Core 24 can only be understood as illustrative, rather than providing a clear evaluative benchmark. A recent systematic review concluded that further survey and cross-sectional work across different PMHL models is required to allow for comparisons between services and to set the scene for primary research [21]. Given the increase in CYP presenting to acute hospitals with mental and physical health needs, and the clear lack of equity between requirements and standards for adult MHL and paediatric MHL services, it is important to investigate current service provision for this group to support national evaluation, and development of appropriate standards.

Study aim

To describe the characteristics of PMHL Services in England and develop a typology of service models.

Research Questions

1. What are the main models of PMHL services currently being provided in England, and how do they differ from one another in terms of what is provided and for whom?

2. In the absence of clear requirements for PMHL services, how does PMHL service provision compare with key MHL service standards as set out in Core 24?

Methods

Survey development

A cross-sectional survey was developed based on the adult-focused LPSE questionnaire [17]. Changes were made to account for the differences between adult and paediatric services, to expand the scope to include psychological as well as psychiatric liaison, and to gain information about types of interventions offered and reasons for referral. Following discussions with leading experts in the field, we decided that the term ‘liaison psychiatry’ might constrain the understanding of services for CYP because such services are routinely provided by paediatric psychologists and mental health practitioners, with limited or no psychiatric input. Therefore, in this study we use the term ‘Paediatric Mental Health Liaison’ (PMHL). PMHL is defined as any service or intervention provided to CYP aged 0–25 by a single mental health professional or team of primarily mental health professionals of any discipline working in an acute hospital setting. ‘Mental health practitioner’ is understood to be a trans-professional role, usually referring to Mental Health Nurses, Social Workers, or Occupational Therapists.

Survey questions were reviewed by experts in the field, including the LPSE-7’s authors, and the survey was subsequently revised following feedback and piloting by professionals in the field. The finalised questionnaire was hosted using Qualtrics [22] and included questions about service operation (hours, age ranges served), scope of practice within the acute hospital, interventions offered, staffing, reasons for referral, and any changes in referral patterns observed since the start of the Covid-19 Pandemic (Appendix 1 contains full survey details). As the survey was not collecting personally identifiable or sensitive data, no Research Ethics permission was required or sought. This was clarified and confirmed by the University of Manchester, who provided an ethical exemption certificate.

Survey distribution

A link to the online survey and brief explanation of the study was distributed via email. All acute hospitals in England with ED departments were identified, and attempts were made to find contact details for a service lead or other suitable person at each hospital who might be able to complete the survey. Emails were sent to key professional networks, leads of services identified from publicly available details on National Health Service (NHS) Trust websites, and to lead practitioners at PMHL services identified in LPSE-7; they were asked to complete the survey if able, and disseminate it to others in their network known to provide PMHL services. A survey link was also shared via the lead author’s professional social media channels.

Inclusion and exclusion of responses

Respondents were asked to complete the survey if they felt their service fitted the following description: ‘By mental health liaison, we mean services provided by MH professionals, where a key part of the role is liaising with paediatric colleagues in acute hospital settings including EDs, paediatric wards, or outpatient clinics.’

Services were included if there were missing data, as long as the data provided were sufficient to determine the service location and that it met our definition of PMHL. Respondents were followed-up by email or telephone where there were missing data. Where appropriate, free text responses were also used to complete missing data, e.g. if the question about service hours was not complete but this information was provided in the free text responses.

Exclusion criteria are described in Table 1. Duplicate responses were merged.

Table 1.

Exclusion criteria

1.	Responses referring to a service with a primary safeguarding function (also known as ‘paediatric liaison’ in some hospitals).
2.	Responses referring to a paediatric department rather than a specific mental health, psychiatric or psychological service.
3.	Insufficient survey data to draw reliable conclusions about the scope or function of the service.
4.	Duplicate responses (i.e. already received sufficient data from the same hospital/service).

Analysis

Responses were downloaded from Qualtrics [22] into a Microsoft Excel [23] file. Individual responses were given a numerical code and reviewed for completeness and checked against the inclusion/exclusion criteria. Numerical data were then analysed using descriptive statistics in IBM SPSS Statistics [24]. This approach is in line with the survey’s aim to identify and describing PMHL services, and appropriate given that it is unknown to what extent to the responding services are a representative sample of the true population of PMHL services. Free-text responses about reasons for changes in referral were analysed using NVivo 12; as answers tended to be brief, and data limited, the software was simply used to group answers into categories. Vote counting was then used to report common trends [25]. If a service had data missing it was excluded from analysis of that question.

Through coding the data, a service model typology was developed. This was done by identifying the key characteristics of services that might differentiate them from one another e.g., whether the service provided emergency/crisis response, whether non-emergency intervention linked to the psychological dimensions of physical illness was offered, and which professional group led or dominated the service. Services were then coded against these characteristics and entered into a matrix which allowed for comparison of all services based on key characteristic codes. MA developed an initial typology consisting of five service types based on clusters of ‘similar’ services in an initial review of the coded data. MA then met with SK and SP to review the initial typology matrix. Each service was then reviewed against the initial typology and any discrepancies between service characteristics within the lead author’s groupings and proposed service models were identified and discussed. Once consensus on the final typology was reached between MA, SK, and SP, each service was categorised or excluded if the data available was insufficient to allow inclusion within the typology. Ultimately services were categorised into three service models.

Results

Overview and typology description

The survey opened in October 2023 and closed in February 2024. Responses were obtained from 28 services meeting the inclusion criteria and serving 47 hospitals. This represents 27% of hospitals in England with EDs, though this cannot be considered a response rate as we do not know how many hospitals provide a PMHL service. 31% of the 28 services were in London; 17% in other major cities; 20% in non-major cities; and 31% in towns, including those serving rural areas. Services located in London and the South-East of England accounted for almost 40% of responses, compared to a very limited representation of services in the West Midlands and the North-East with only one response received for each of these areas.

Table 2 presents the three service models identified following the coding and typology development process described previously, and the number of services categorised to each model. There was a fairly even split between services providing emergency liaison work (n = 15) and those only offering non-emergency liaison work (n = 13).

Table 2.

PMHL service models

	Type 1, Long-Term Conditions Focus: (n = 13)	Type 2, Crisis Focus: (n = 8)	Type 3, Integrated (both Type 1 and Type 2 work): (n = 7)
Usual Scope	Psychological impact of LTCs.	Physical impact of mental health conditions & mental health crisis, e.g. self-harm, eating disorders.	Physical impact of mental health conditions, mental health crisis and psychological impact of LTCs.
Hospital areas usually served	Paediatric wards, outpatient clinics	Mostly limited to ED and paediatric wards.	ED, paediatric wards, outpatient clinics
Usual staffing	Mostly clinical psychologists, some limited psychiatry input.	Mostly MHPs, some limited psychiatry input	Multidisciplinary (Psychologists, psychiatrists, MHPs)
Emergency/Crisis Response	No – usually provided by CAMHS in-reach or all-age liaison.	Yes – usually within 1 hour.	Yes – usually within 1 hour.

Key differences in service provision by service model

Age range

Type 1 services were most likely to offer services to the full range of ages typically seen as comprising CYP, i.e. 0–18 years, or 0–25 years where there is a learning disability, or the child/young person has been in care (Table 3). Free-text responses from Type 1 and Type 3 services indicated that they would only see CYP over eighteen if under the care of specific medical specialties, i.e. haematology/oncology (n = 3) and cleft (n = 2). None of the fifteen services offering crisis support reported offering this to CYP beyond the age of eighteen. Just two (25%) of the eight Type 2 crisis services saw CYP aged 0–18. Others reported supporting CYP 0–17 (n = 2), 0–16 (n = 2), 9 to 18 (n = 1) and 7 to 17 (n = 1).

Table 3.

Age range by service type

		Service Type
		Type 1 (Long-term conditions focus)	Type 2 (Crisis focus)	Type 3 (Integrated)
		n = 13	n = 8	n = 7
Age Range	0–25 (in at least one area)	4	0	1
	0–18, 19 or 20	7	2	4
	Less than 0–18	2	6	2

Operating hours

Operating hours varied widely (see Table 4). Most of the thirteen Type 1 services, none of which provided an emergency response, operated mainly within standard office hours (9am to 5pm, Mon-Fri), with two (15%) reporting reduced office hours, and one (8%) offering extended office hours. Of the fourteen services which reported covering the ED supporting CYP in crisis, only six (43%) reported having coverage at all times. A further six (43%) only operated within standard office hours, and two (14%) offered extended office hours 7 days per week.

Table 4.

Operating hours by service model

		Service Type
		Type 1 (Long-term conditions focus)	Type 2 (Crisis focus)	Type 3 (Integrated)
		n = 13	n = 8	n = 7
Operating Hours	24/7	0	3	0
	Office Hours + separate out of hours cover	0	0	3
	Extended Office Hours 5 or 7/7	1	1	1
	Office Hours only 5 days or less	12	4	3

Service location

In terms of service location, 29% (n = 8) of the 28 services reported being based either within the ED or paediatric ward that they served, with another 54% (n = 15) reporting that they were based on the hospital site and could reach the ED or paediatric ward within 10 minutes. Of the five services based off-site, all but one reported they could reach the hospital reliably in under one hour.

Service name

Three quarters (n = 21) of the 28 services represented in the survey data did not refer to themselves as a ‘liaison’ service, with those which did being mostly Type 2 and 3 services (Fig. 1). Examples of alternatives included ‘CAMHS Crisis and Urgent Support’ (Type 2 service) and ‘Paediatric Mental Health Team’ (Type 3 service). Type 1 services were usually named ‘paediatric psychology’, or something very similar.

Fig. 1.

Inclusion of ‘liaison’ in service name by service type. Data from all responding services (N = 28): Type 1 n = 13, Type 2 n = 8, Type 3 n = 7

Staffing

Services reported being staffed by a combination of psychologists, MHPs and psychiatrists. Relative proportions of staff reported by each service model are shown in Fig. 2; staffing data were missing from two Type 2 services, therefore only six Type 2 services are represented in the graph. Five (39%) of the thirteen Type 1 services reported only employing psychologists. Of the six Type 2 services reporting staffing details, 50% (n = 3) reported only employing MHPs. Of the seven Type 3 services, four (57%) reported employing at least one psychiatrist, one psychologist, and one MHP. Two (29%) of the seven Type 3 services were unidisciplinary – these were very small services employing in one case a lone consultant psychiatrist covering a hospital serving a largely rural area, and in the other, two clinical psychologists located in a hospital serving a London borough.

Fig. 2.

Staff groups by service type. Data from 26 responding services: Type 1 n = 13, Type 2 n = 6, Type 3 n = 7

Service leadership largely reflected the staff group in the majority within the team. Although two teams reported being psychiatrist-led, they were mainly staffed by either psychologists (Type 1 service) or MHPs (Type 2 service). Overall, twelve (45%) of the twenty-six teams who provided staffing data employed at least one psychiatrist; however, it was notable that only two Type 2 services reported doing so. One respondent described not a service, but operating as a lone consultant child psychiatrist covering a large teaching hospital in the south, as well as core CAMHS.

Service coverage within the hospital

Fig. 3 reports differences in hospital areas served depending on service type seen in the dataset. Just four (14%) of the twenty-eight services accepted referrals from all hospital areas; these were all Type 3 services.

Fig. 3.

Hospital areas served by each service model. Data from all responding services (N = 28): Type 1 n = 13, Type 2 n = 8, Type 3 n = 7. Abbreviations: ED = Emergency Department, PW = At least one Paediatric Ward, OP = At least one Outpatient Clinic

Service provision in the emergency department

Type 1 services, by definition, had no ED provision. Twelve (92%) of the thirteen Type 1 services provided information on who offered an emergency service in their hospital; these were mainly CAMHS in-reach (n = 5), a local CAMHS Crisis service (n = 2), an adult/all-age MHL (n = 3); or a separate PMHL team (n = 2) based in the hospital. Most Type 2 services (7 out of 8; 88%) reported covering the ED, the one outlier relying on CAMHS in-reach for ED assessments, and providing crisis support for CYP subsequently admitted to the paediatric ward. Six (86%) of the seven Type 3 services reported providing ED cover; the one which did not, reported that CAMHS in-reach saw CYP requiring MH support in this setting (Fig. 4). Of the thirteen services taking ED referrals eleven services (85%) responded with information about target response times. Four (36%) of these eleven services reported aiming to see CYP within one hour of presenting to ED, with another four (36%) reporting no target at all. Two (18%) had a target of four hours; the remaining service reported a 24-hour target.

Fig. 4.

Emergency department provision by service model. Data from 26 responding services: Type 1 n = 12, Type 2 n = 8, Type 3 n = 7

As well as offering psychosocial or psychiatric assessment for CYP presenting to the ED following a suicide attempt or self-harm, education and training was provided to ED paediatric staff by three (42%) of the seven Type 2 services providing data and six (86%) of the seven Type 3 services. A small number of Type 2 services (n = 2) reported also providing practical support, e.g., assisting with observations, or physical restraints.

Service provision on paediatric wards

Almost all responding services (97%, 27 out of 28) reported providing support for CYP admitted to paediatric wards However, of the ten Type 1 services who responded to the question about response times, 50% (n = 5) reported that they did not provide an emergency service, or that another team would be responsible for responding to emergency inpatient referrals. In contrast, just one of the seven Type 3 services stated they did not provide an emergency response on paediatric wards.

Referral reasons for CYP admitted to paediatric wards were more varied than for those seen in the ED (Fig. 5). As might be expected, Type 1 services reported receiving referrals primarily to support CYP in coping with physical illness or treatment. Type 3 reported a more evenly spread range of reasons, with MUS appearing to be most common. Type 2 services most commonly identified self-harm or crisis as a main reason for referral in this setting. Eating disorders were mentioned as the main reason for referral by only three services overall.

Fig. 5.

Reasons for referral from paediatric wards by service model. Data from 93% of responding services (N = 26): Type 1 n = 11, Type 2 n = 8, Type 3 n = 7

Services reported providing a range of interventions to CYP admitted to paediatric wards with 69% (n = 18/26 responding services) offering psychosocial assessment and/or brief psychological interventions. Of the ten Type 1 services operating on paediatric wards who provided data, 80% (n = 8) offered long-term psychological interventions, compared to 43% (n = 3) of the seven Type 3 services and 25% (n = 2) of the eight Type 2 services. All but one (n = 6) of the seven Type 3 services offered psychiatric consultation with patient contact, compared with 63% (n = 5) of the eight Type 2 services and half (n = 5) of the ten Type 1 services providing data. Education and training for paediatric staff was provided by 90% (n = 9) of the ten responding Type 1 services and six (86%) of the seven Type 3 services, compared to just 38% (n = 3) of the eight Type 2 services. Only 12% (n = 3) of the twenty-five services providing data reported that their service provided practical support to paediatric colleagues.

Service provision in outpatient clinics

Sixteen services (Type 1 n = 11; Type 3 n = 5) reported covering at least one medical outpatient clinic, representing 80% of the 20 services (Type 1 n = 13; Type 3 n = 7) providing non-emergency coverage. 94% [15] of the 16 services providing outpatient coverage named coping with medical treatment as a main reason for referral (Fig. 6). Of the five Type 3 services operating in this setting, 80% (n = 4) also named MUS as a main reason for referral; however, this was not reported by any Type 1 services. ‘Neuropsychological assessment’ was identified as a reason for referral only in the outpatient setting, although only a fifth of services (n = 3/15) named this within their top three.

Fig. 6.

Top Reasons for referral from outpatient clinics by service model. Data from 18 responding services: Type 1 n = 11, Type 3 n = 7

Psychosocial assessment and brief psychological interventions were provided by all sixteen teams working in the outpatient setting who responded to this question (Type 1 n = 11, Type 3 n = 5), and most also reported offering long-term psychological interventions (10 out of 11 Type 1 teams, 4 out of 5 Type 3 Teams). The main difference was that psychiatric consultation which included patient contact (rather than the psychiatrist providing consultation to paediatric staff without seeing the CYP) was offered by all five Type 3 services providing data, compared with less than a third (3 out of 11) Type 1 services providing data.

Provision across physical health specialties

Respondents were asked to specify which medical specialties they accepted referrals from (Fig. 7). Type 1 and Type 3 services tended to accept referrals from a variety of specialties in both inpatient and outpatient settings. Support for CYP with diabetes was most widely available, offered by 56% of the 25 teams responding to this question in an inpatient setting (n = 15) and 64% (n = 16) in an outpatient setting. Other specialties frequently mentioned were Gastroenterology and Rheumatology although services less commonly accepted inpatient referrals from them than for outpatients. Few services (n = 2 inpatients, n = 1 outpatients) described accepting any referrals from the specialties of Cardiology, Dental, Dermatology, Differences in Sex Development, Ear Nose Throat, Genetics, Gynaecology, Immunology, and Hepatology. All Type 2 teams represented here reported accepting General Paediatrics referrals from paediatric wards, with virtually none reporting accepting referrals from any other inpatient specialty in the hospital, and none offering any kind of service in an outpatient setting.

Fig. 7.

Medical specialities served in inpatient and outpatient setting. Data from 25 responding services: Type 1 n = 11, Type 2 n = 7, Type 3 n = 7. Key for Fig. 7: Blue column = inpatient referrals, purple dot = outpatient referrals

Changes in service provision since the Covid-19 pandemic

Twenty of the 28 services responded to a question about changes in referral patterns post Covid-19. Seventeen (85% of those responding) reported increased referrals since March 2020. Free-text responses to this question were received from eleven services. These responses indicated that more than 50% (6 out of 11) of services have additionally encountered a general increase in perceived severity or complexity of presentations, with 45% (5 out of 11) of services reporting increased presentations of neurodivergence. Increased referrals for CYP experiencing disordered eating (36%, 4 out of 11), or problems with school attendance (27%, 3 out 11) were also noted.

Comparison of service characteristics with the Core 24 standard

Information provided by each respondent was compared with key elements of the Core 24 standard, namely: whether the hospital(s) served by them had a 24/7 on-site paediatric or truly all-age liaison service with the ability to respond within one hour for emergency CYP referrals. Whether or not hospitals represented in this dataset were likely to meet these criteria is presented in Fig. 8. Just 36% (n = 10) of the 28 services indicated that their hospital(s) were likely to meet these Core 24 requirements. Of the three service models, respondents from Type 3 services were most likely to meet these criteria (71%, 5 out of 7 services).

Fig. 8.

Hospital likely to meet key core 24 standards. Data from all responding services (N = 28): Type 1 n = 11, Type 2 n = 8, Type 3 n = 7

Discussion

This exploratory study has contributed to knowledge by demonstrating previously undocumented variations in PMHL work happening in a subset of hospitals with EDs in England. A service typology was developed encompassing services focused on non-emergency psychological work for CYP with LTCs (Type 1); services focused solely on crisis support (Type 2); and services providing both types of support for CYP, including those experiencing MUS/PPS (Type 3). The variation in team names reported by study respondents suggests a lack of agreement over how ‘paediatric liaison’ is defined. This discrepancy in parlance and understanding has been reported elsewhere, e.g. [26], and represents a potential barrier to accurately mapping research on the topic.

However, given the survey collected data about emergency response within the hospital, regardless of team type, this dataset suggests that responsive on-site round the clock services for CYP aged 0–18 presenting to ED or admitted to paediatric wards in crisis or following self-harm may be lacking. Hospitals with comprehensive, dedicated Type 3 PMHL services likely to meet the Core 24 standard for emergency care and providing psychological work for inpatients were represented in just 18% of responses, indicating that in some places CYP may not be able to access the type of PMHL input they require or a standard of care equitable with adult provision – where the latest survey found 100% of adult MHL services are working to a one hour emergency response target [17]. In contrast, almost a third of services responding to this survey reported that they relied on Community CAMHS in-reach in their hospital for an emergency response, making a one-hour response very unlikely. If replicated nationally, this would indicate a major discrepancy between MHL provision for adults and CYP. This situation may be linked to historical guidance that CYP should be admitted to a paediatric ward following any act of self-harm, whilst they wait for assessment from a community CAMHS team [4], which has essentially negated the need for an on-site PMHL team, but leaves CYP in crisis open to long waits to see an appropriate team when presenting to the ED or leads to unnecessary paediatric admissions. Limited UK research on what crisis care should look like for CYP [27] or what CYP views are of receiving mental health care in this setting [21] make it difficult to establish how service standards should differ from those applied to adult provision and to some extent limit the utility of evaluating PMHL services against the Core 24 Standard. However, given the potential negative impact of hospital admission for CYP who self-harm [9], and pressure on both paediatric and CAMHS inpatient beds [28], there is a clear need for 24/7 MHL teams who are able to provide timely CYP and family focused-assessment for CYP of all ages presenting in crisis to the general hospital. Investment in teams able to provide this are likely to improve both patient experience and reduce health service costs; indeed, available research indicates that on-site PMHL teams can reduce both length of stay and hospital costs [21].

In terms of the MH consequences of LTCs, survey responses indicated that where provision is available, this is largely depending on medical speciality or diagnosis. One Type 1 service, for example, was commissioned exclusively for CYP with diabetes, another only for diabetes and cardiology. This situation reflects the specific requirement within NICE guidelines for psychological support for CYP with diabetes, not replicated across guidance for most other LTCs [19], despite emerging evidence indicating that cognitive behavioural therapy is both effective and acceptable for CYP with range of LTCs [29]. If replicated nationally, this suggests there may be gaps in service provision for those CYP with LTCs requiring psychological support where support is not commissioned and who do not meet the risk threshold for input from local community CAMHS.

Survey responses indicated a sharp differentiation between distress classed as being linked primarily with an LTC and distress where the primary cause would likely be deemed to be mental health-related, e.g. following an overdose. Typically, Type 1 teams see the former, Type 2 the latter, with Type 3 supporting both groups of CYP. Is worth noting that Type 3 teams, who commonly reported seeing CYP with MUS, appeared to be more likely to be multidisciplinary in nature, to include more psychiatry input, and to provide a cross-hospital service emergency and non-emergency service including education and teaching to paediatric staff. This model perhaps creates a culture more aware of MH risks associated with LTCs, including MUS, echoing research findings that expanding or formalising PMHL teams results in increased referrals, presumably at least in part due to an increased awareness and recognition of mental health needs in this setting by paediatric staff [30]. It is also worth considering if there is something about the nature of a multidisciplinary team which provides both crisis and long-term psychological input which allows more room for the consideration of physical and mental distress as causally bidirectional (Fig. 9).

Fig. 9.

PMHL in acute hospitals typology diagram

Interestingly, Type 1 teams represented in the survey data appear to be dominated by clinical psychologists, who are largely absent from Type 2 teams, despite a clear need for crisis work to be more psychologically informed [31]. Conversely, MHPs appear largely absent in Type 1 teams and dominant within this crisis model. If this reflects the national picture, it potentially represents missed opportunities for more expansive and multidimensional thinking and intervention both in the realm of crisis intervention and in the longer-term mental health support of CYP with LTCs. For example, mental health nurses with a grounding in anatomy and physiology are likely to have much to add when it comes to considering coping with illness and medical intervention. Across all services the lack of access to child psychiatry echoes an ongoing national concern [32], and represents an element of CYP mental health care which is simply not equitable with adult provision.

Surprisingly just three teams reported eating disorders as a top reason for referral from paediatric wards. This may be down to fact that, over the last decade, funding has shifted to community ED services [33] and while these services can theoretically offer in-reach services, they may struggle to do so in the face of the unprecedented increase in demand since 2020 [34]. This issue warrants further investigation, given the challenges paediatric teams face in providing appropriate care to this group of CYP [11]. Perhaps linked with this, only a small number of teams reported providing practical support to paediatric clinicians such as help with carrying out observations or assisting where restraint is required to facilitate care; this echoes the findings of a recent systematic review of PMHL work, whereby paediatric staff reported a disconnect between support they required from PMHL teams, and what was actually on offer [21]. In England at least this is potentially due to the challenges associated with the provision of MH care in a hospital setting geared towards physical healthcare, or where MH professionals are employed by an MH Trust work on an acute Trust site, thus limiting the duties for which they are insured, e.g. prescribing, restraint. It would be useful to know to what extent this is reflected in other hospitals offering emergency and inpatient care to CYP, and to understand further the views of health professionals on how best paediatric liaison teams can support them to care for CYP needing mental health care.

Conclusion

This exploratory survey provides novel insights into the provision of PMHL services at a subset of acute hospitals across England. Despite policy mandates and targets to improve PMHL provision within acute hospitals, the data suggest that – within the partial sample of hospitals represented here at least, there remains a long way to go before CYP requiring both physical and psychological care achieve parity of service provision with adults, especially in the realm of crisis-care. It is apparent – even from a small dataset such as this - that not all hospitals offer even a basic 24/7 crisis service for all CYP under-18 or have facilities to provide specialist psychological support for CYP with LTCs. Furthermore, the data suggests that there may be a lack of multidisciplinary integrated MH teams operating in this setting; this makes it likely that CYP experiencing MUS are likely to be particularly underserved. The lack of specific requirements and standards for PMHL services in this setting have likely contributed to this situation; indeed, where these are in place e.g. for diabetes care, it does seem that a service is more likely to be provided. A push for all-age liaison teams has likely increased access for CYP in crisis over the past few years; however, the lack of data collected detailing exactly what all-age teams can or do offer to CYP means that we are still in the dark about how many adult teams have simply ‘expanded’ to provide CYP crisis assessment in addition to their usual adult work, or whether there are those which include professionals with specific CYP expertise. Although looking to the Core 24 Standard is useful to highlight the disparities in adult and paediatric service provision, and evaluate some basic service requirements, it lacks the nuance needed to support development of CYP services that are fit for purpose, and cannot be used as a benchmark for them. Given that it is not known how representative this sample of PMHL services is, further work is required to test whether the findings reported here are replicated across all acute hospitals, and to establish stakeholder views on key elements of PMHL service provision. These should be used to develop national guidance to ensure a consistent and appropriate minimum service offer both for crisis and non-crisis support for CYP requiring mental health input in this setting.

Strengths

There has been no previous national survey specifically of PMHL services, consequently our data provides a much-needed starting point for future research and service development. It has provided some clarity around service delivery models and their differences, as well as highlighting likely gaps in service provision for the hospitals represented in the data. When combined with data from LPSE-7, this survey suggests it is likely to be very much a lottery as to what services are available in an area. These findings will be useful to direct further research and to inform the development of CYP-specific MHL standards.

Limitations

A complete enumeration (census) of PMHL services is difficult for the same reasons as any other census: the resource and time costs in identifying and accessing the services, respondent burden or refusal, and diverse or elusive services. As such, data were not collected from all hospitals in England, and a partial snowball strategy to sampling may have introduced bias. As the total number of PMHL services is unknown, this is a limited dataset which may not adequately represent the national picture. Therefore, there may be variations in service provision elsewhere which do not fit within the proposed typology; as such the typology requires validation with a larger sample. Furthermore, small numbers of each service type meant that sub-group analyses were somewhat limited, e.g. it was not possibly to contrast by hospital size. Not knowing the size or nature of the true population of PMHL services meant a response rate could not be calculated, nor could it be determined whether the services responding were representative of this population. This meant analysis was limited to descriptive statistics: counts, frequencies, and crosstabulations based on service model as determined by our typology. It is also possible that some services were not aware fully of the roles or existence of other teams providing liaison work at their hospital (e.g. crisis-care focused teams with no awareness of paediatric psychology services for CYP in wards or outpatient clinics). Errors in the data may therefore be more likely for respondents in larger Trusts with multiple hospitals/services.

Recommendations for Practice

• Standards for PMHL are needed to ensure a consistent national minimum service offer, across both acute and non-acute domains of liaison work, with funding available to support delivery.

• At minimum, there should be CYP focused crisis provision available on site 24/7 with a 1-hour target response time, working across both emergency and inpatient settings, and a robust mechanism for ensuring CYP with LTCs are able to access specialist psychological support when needed.

• All forms of PMHL work should be at least 0–18, extending to 25-year-olds where there is a learning disability or ongoing care need.

Recommendations for Research

• Further survey work to establish the replicability of these findings across all acute hospitals and establish the role that expanded ‘all-age’ liaison services play in PMHL work.

• In-depth qualitative research exploring stakeholder experiences of the different service models to support the development of appropriate national standards for PMHL.

• Regular monitoring to assess how acute hospitals are meeting the MH needs of CYP, particularly with regards to equity of access to appropriate and timely crisis assessment, and across medical specialties for psychological work.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

PMHL

Paediatric Mental Health Liaison

LPSE

Liaison Psychiatry Surveys of England

CYP

Children and Young People

NHS

National Health Service

LTC

Long Term Condition

CAMH(S)

Child and Adolescent Mental Health (Services)

MH

Mental Health

MUS/PPS

Medically Unexplained Symptoms/Persistent Physical Symptoms

ED

Emergency Department

MHL

Mental Health Liaison

PLAN

Psychiatric Liaison Accreditation Network

NICE

National Institute for Health and Care Research

MDT

Multidisciplinary Team

MHP

Mental Health Practitioner

Author contributions

MA – conceptualisation of the research, study design, survey development, data collection, data analysis, draft and revision of final manuscript. SK – conceptualisation of research, study design, survey development, data analysis, revisions to and approval of final draft for publication. SP – conceptualisation of research, study design, survey development, data analysis, revisions to and approval of final draft for publication. WL – survey development, support with survey distribution and data collection, revisions to and approval of final draft for publication.KW – survey development, support with survey distribution and data collection, revisions to and approval of final draft for publication.

Funding

This work has been completed using a doctoral fellowship funding award from the National Institute of Health and Care Research to the corresponding author, Miriam Avery. Reference NIHR302168.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study conforms to principles laid out in the Declaration of Helsinki. As the survey was not collecting personally identifiable or sensitive data, no formal Research Ethics permission was required or sought, however. This was clarified and confirmed by the University of Manchester, who provided an ethical exemption certificate.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.