Article Text

Improving referrals to the Liaison service at the Royal United Hospital in Bath
  1. Alister Gomes-Pinto,
  2. Ruta Kuzminskyte,
  3. Katie Wooding,
  4. Katherine Asplin,
  5. Liz Ewins
  1. Avon and Wiltshire Mental Health Partnership NHS Trust, UK
  1. Correspondence to
    Alister Gomes-Pinto alister.gomespinto{at}


Psychiatry liaison services provide the interface between mental and physical health in the acute medical hospital, however there can be logistical and operational difficulties to overcome.

This quality improvement project aimed to improve the timeliness of referrals to a liaison service from an acute hospital through simple interventions of a newsletter, email to staff, and a pilot including attending post-take ward rounds on the Medical Assessment Unit (MAU) of the hospital. This resulted in a faster referral process to liaison as well as improved staff satisfaction with the liaison service, both of which will have a positive benefit on the clinical management of patients and the patients experience in hospital.

There was a significant improvement in overall staff satisfaction with the referral pathway, appropriateness of referrals and working hours of the Mental Health Liaison Team - increasing from 14% at baseline to 100% at the end of the study. Referral outcomes also showed a considerable improvement, with the percentage of junior doctors successfully able to locate the referral form increasing from 60% at baseline to 100%.

This is an open-access article distributed under the terms of the Creative Commons Attribution Non-commercial License, which permits use, distribution, and reproduction in any medium, provided the original work is properly cited, the use is non commercial and is otherwise in compliance with the license. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


The Mental Health Liaison Team (MHLT), based at the Royal United Hospital in Bath and provided by Avon and Wiltshire Partnership NHS Trust, is expanding in response to the needs of the service.

The MHLT aims to provide a flexible, adaptive, and needs-led service to adults aged 18 and over, who present to the RUH with a range of mental health related problems. The team also provides a dedicated service to the Older People’s Unit, and works collaboratively with maternity services to offer a monthly perinatal clinic.

Last year (2014) the MHLT undertook a survey across the hospital to capture the views of staff about the accessibility of mental health services for the patients admitted to the RUH, and their satisfaction with the service. A total of 129 surveys were completed and 84% of those surveyed were staff who were likely to refer to the liaison team, or had done so in the past.

Staff who were surveyed expressed a lack of confidence about the types of presentations that warranted a referral to the liaison service. They were dissatisfied with the referral process, and largely unaware of the working hours of the team including availability out-of-hours. Dissatisfaction extended to communication provided by the liaison team between initial referral and assessment, but communication post-assessment was rated more favourably.


More than one quarter of general hospital patients have a coexisting mental disorder. 60% of people over the age of 65 who are admitted to a general hospital have or will develop a mental disorder during their admission [1]. Mental illness can frequently cause or aggravate physical disorders, which are seen and treated in acute hospital settings.

Most acute liaison services could provide the following:

- brief interventions, advice, and signposting to services in a range of agencies for patients in acute hospital settings

- rapid response to requests for assessment on acute hospital wards

The service could bring the following benefits:

- improved wellbeing of staff in acute hospital settings, by relieving the anxiety these staff sometimes feel when dealing with patients with complex needs

- reduced stigma associated with mental health care

There is currently no single, uniform model for liaison services across the country. Where such services exist, they are often provided by the local mental health trust within the estate of the acute hospital trust, which may present logistical and operational challenges. [2]

A search of related projects on databases such as Cinahl, Medline, PsycInfo, and BMJ Quality Improvement Reports yielded no such work which has been completed with regards to liaison services attached to acute hospitals, to date.

Baseline measurement

The diagram below identifies the current process for referral and subsequent review by the MHLT at the RUH, Bath. In the original 2014 survey undertaken by the MHLT across the RUH, staff were asked to indicate via anonymized commentary, their specific areas of concern in relation to the referral system. The most common complaints within the captured data set were as follows:

- Junior doctors and nursing staff had difficulty locating the referral form, as it was not stored in the same location as the mainstream Medical/Surgical referral forms. It was actually stored with 'other associated Healthcare Professionals', such as Occupational Therapy and Tissue Viability. This meant that Mental Health was the only medically affiliated specialty referral to be stored with other non-medical specialties.

- Referrers were unclear on appropriateness of mental health referrals to the Liaison Team, and what referral information exactly was required of them

- They complained as well about the limited space on the form for them to provide more detailed information

Areas of concern with regard to the referral process would be addressed as follows :

- To provide further information and training on form completion

- Training on appropriateness of patient referral to the MHLT

- Details of how to contact the MHLT both during and out of hours

See supplementary file: ds6062.ppt - “RUH QIP driver diagram”


Looking at the issues outlined above the initial aim was to change both the location of the referral forms, and the layout of the referral form itself to make it more user friendly. However upon discussion with the MHLT Manager this apparently had been attempted a few years ago, and proved to be unsuccessful due to the then ongoing political nature of the relationship between the acute hospital and mental health trust. The manager was not convinced that much had changed in the way of making swift amendments without enduring a process that might ultimately prove not to be productive.

The next step was to target information streams that were readily accessible to most if not all staff at the RUH. The RUH Communications Team, for instance, notify their staff of general information updates on a weekly and monthly basis, through the RUH newsletter and via email.

Not only was this felt to be an efficient way of contacting the staff of the RUH, but tapping into this resource meant not having to alter systems as had been attempted earlier with the referral form. The Team Manager was happy with these outcomes, as was the Liaison Consultant Psychiatrist.

3 PDSA cycles were to be utilized to achieve the following outcomes:

  1. An article would be published in the RUH internal newsletter, by text and picture, to identify the members of the MHLT, highlight the situations warranting referral and outline referral process.

  2. Next, an e-mail would be sent to all of the RUH staff, with the same information.

  3. Lastly, it was agreed that the Medical Assessment Unit (MAU) within the RUH would be targeted as a pilot ward for the third stage of this study. The reason behind this decision involved the MAU having the least favourable feedback when results from the 2014 study were collated.

The pilot scheme would involve doctors from the MHLT (AGP, RK, KA) attending the morning round of the MAU (either the Working Age Adult team - 18 to 65 year olds; or the Older Adult team - the over 65s) and offering psychiatric advice to the responsible Medical Consultant for the day, or signposting to affiliated services as appropriate.

A quantitative study was conducted to measure time taken to locate the referral forms to the MHLT. This consisted of measuring time and number of mouse clicks taken to locate the relevant forms by referring junior doctors, which was tested before and after each PDSA cycle. A different group of junior doctors was used each time.

A survey consisting of 4 questions was conducted before and after each PDSA cycle as well, to gauge MAU staff understanding of the referral processes and staff understanding of the referral system. These questions were a simplified form of the previous longer staff survey.


PDSA cycle 1 - Publication. An article was published in the RUH Newsletter - detailing the contacts, hours of working, and appropriateness of referrals to the MHLT.

PDSA cycle 2 - E-mail version of PDSA cycle 1. This was distributed to all RUH staff through their NHSMail inboxes.

PDSA cycle 3 - A pilot study was conducted on the Medical Assessment Unit of the RUH, attending the post-take round with either the Adults of Working Age or Older Adult Consultant of the day for a whole week. During this time laminated pocket cue cards which identified the online referral process, working hours and contact details of the MHLT were handed out.

For each of the three PDSA cycles, quantitative data was collected on speed of referral completion and number of clicks used to locate the referral form on the hospital computer system, and also measured qualitative data before and after each cycle.

See supplementary file: ds5922.docx - “quantitative data analysis”


Between 8 -14 junior doctors were used to complete the survey questionnaire for each PDSA cycle. The Foundation Year 1 doctor (KW) would randomly allocate questionnaires to junior doctors on duty in MAU in any given working week, in order to reduce bias. The response rate was 100% as KW would wait in person for the surveys to be completed in each cycle. If KW was unavailable during the week the GP SHO (KA) would assist with data collection from MAU instead.

Those indicating 'yes' in response to the following questions were compared between baseline and results from the 3rd (and final) PDSA cycle:

- feeling confident about the referral pathway. Baseline - 28%; Cycle 3 - 90%

- being certain of the working hours of the MHLT. Baseline - 0%; Cycle 3 - 60%

- knowing which mental health presentations warranted a referral. Baseline - 28%; Cycle 3 - 90%

- overall satisfaction with the process. Baseline - 14%; Cycle 3 - 100%

There was a demonstrable improvement in the awareness of referral pathway and appropriateness, working hours of the MHLT and overall satisfaction as seen in our qualitative data analysis bar charts (attached).

The reason for the sudden spike in positive outcomes by the 3rd PDSA cycle may be due to the fact that AGP, RK and KA had a physical presence during the pilot phase of this study on MAU, which translated into a 'real time' engagement with staff as opposed to accessing their time and attention via newsletter and email. As well the presence of the MHLT medics and respective roles was reinforced with the distribution of the cue cards mentioned earlier during the rounds of the MAU.

Data (both qualitative and quantitative) was usually collected a week after each intervention was implemented. The quantitative data was also collected by KW, who approached junior doctors in MAU and counted the number of mouse clicks used by these doctors to locate the referral form - including timing the doctors as they did so (and again, the response rate was 100%). In some cases the same junior doctor who completed the qualitative survey also completed the quantitative segment of the study as they were on a rotational shift work timetable for the duration of the QIP.

Quantitative data interpretation:

Number of doctors used - 10 at baseline, 5 thereafter for each cycle. It was difficult to raise numbers due to the unpredictable nature of shift work on MAU

- number of clicks from start to locating referral form on a RUH computer. Baseline - 13 mouse clicks (this is an average of the total number measured); cycle 3 - 10 clicks

- time taken to find and complete mental health referral form. Baseline - 1 minute 44 seconds (average time); cycle 3 - 2 minutes 1 second

- referral outcome (i.e. was the referral successfully sent to the MHLT?). Baseline - 60% success rate; cycle 3 - a 100% success rate

Although there was a reduction in the number of clicks used to locate the referral and an increase in successful referrals submitted; a slight increase in time taken to complete the referral was noted from baseline. The increase in time taken to complete the referral may be a byproduct of increasing success rate as it was observed that junior doctors were more likely to spend time searching for the form rather than simply giving up, as the project progressed.

See supplementary file: ds5924.docx - “qualitative data analysis”

Lessons and limitations

The Team attempted to change the layout and format of the Referral form. The Team also attempted to incorporate the MHLT Referral form into the Hospital's internal referral system, used for other specialty referrals.

The problems that were identified in Option 1 were discussed with the Team Manager, and felt to not be viable at this stage. Option 2 was also explored with the team manager, who felt that previous discussions had been held at Trust level, and not been tenable or welcomed.

Other issues that were raised in the process of developing the QIP included some initial resistance from the MAU Lead Consultant, who was unconvinced about the potential merits of the project. His reservations were however allayed with the timely intervention of the MHLT Consultant Psychiatrist, who envisaged a need for a MHLT presence there.

The key lesson learned from producing this QIP, is that the nature of the relationship between physical and mental health within an acute hospital setting is very much a political one - which can impact on funding habits. And in an age of recession, funding has implications for any potential changes or conclusions that may be gleaned from quality improvement projects.


This study did have a positive impact both in terms of quantitative gains and qualitative improvements. From a quantitative perspective, the study identified improvements in ease of referral location and an improvement in timeliness of referral transfer to the MHLT. However, this study also identified qualitative improvements such as overall staff satisfaction with the referral system and MHLT contacts.

(Please note: the following additional improvements were based on subjective observations, including informal feedback provided by RUH staff and have not been objectively measured).

Attendance at the MAU post-take round appeared to provide additional benefits such as: identifying patients already known to psychiatric services, and expediting referral to the MHLT and subsequent discharge. In one instance, simply explaining to a treating team how difficult a mental health problem was for one MAU patient appeared to improve staff awareness and patient management, which was later on felt to be more empathetically delivered.

Other subjective (not objectively tested) quality improvements were observed too. Inappropriate drug interactions were reduced when medication reviews were conducted jointly by the post-take Consultant and MHLT medics on MAU during the pilot week. Out-of-area patients who did not necessarily warrant a local service for their mental health problems were appropriately signposted to the relevant teams. On another occasion a timely review facilitated the intervention of the Community Mental Health Team, in the case of a patient who was admitted - with the sudden and unexpected death of her spouse who was visiting her in hospital at the time.

In terms of sustainability: the interventions of putting an article in the newsletter and sending an email were one off measures. However, although the MHLT presence on post take ward rounds was a pilot study, thanks to the liaison presence being a positive intervention overall these findings are due to be presented at the next Joint Operations Committee meeting (which is regularly attended by the RUH and AWP as a forum for service development and discussion).

This forum will be used to put forward a recommendation for a new MHLT role to be created within the MAU setting of the RUH on a permanent basis. This would not have been possible without the outcomes generated by this Quality Improvement Project.


  1. No Health Without Mental Health. Royal College of Psychiatrists, The ALERT Summary Report, July 2009:

  1. Guidance for commissioners of liaison mental health services to acute hospitals, Joint Commissioning Panel for Mental Health:

Declaration of interests

No Declaration of Interests noted.


Thank you to Dr Matt Jelley, Consultant Psychiatrist, for assisting in the development of this QIP.

Thank you to the Communications Team, RUH Bath for assisting in the publication of liaison related information in their Newsletter and online.

A final thank you to Rachael Redman, Manager of the Mental Health Liaison Team at the RUH Bath, for her support and input in the development of this QIP.

Ethical approval

This project was discussed at the Joint Operations Meeting which is attended by Avon and Wiltshire Partnership Mental Health Trust (AWP), and The Royal United Hospital in Bath (RUH). It was felt that ethical approval would not be needed for this study.

View Abstract

Supplementary materials