The NHS long-term plan focuses on the improvement of Child and Adolescent Mental Health (CAMHS) community services including the roll out of 24/7 Crisis teams universally across the country. Crisis and Liaison teams form an important alternative to inpatient admission, offering intense, short-term support to young people in mental health crisis and often high levels of risk. The number of referrals to Crisis and Liaison services are rising. In order to maintain patient flow and meet demand, these teams also need safe, evidence-based protocols for efficient discharge, transition and handover of young people to community teams and services. The SAFER care bundle was designed to facilitate discharges from hospital, and this has been adapted to the SAFER-YMH bundle for discharges from adolescent mental health wards. A similar care bundle for discharge from teams offering alternatives to inpatient care has not yet been developed.

This study aimed to investigate the acceptability and necessary adaptations required for the use of the SAFER-YMH care bundle to facilitate transitions out of CAMHS Crisis and Liaison teams.

This study used stakeholder feedback from multiple sources through focus groups to adapt the SAFER-YMH care bundle for use in young people in transitions out of CAMHS Crisis and Liaison teams. Normalisation process theory was utilised as the theoretical foundation upon which the development of the adapted care bundle, and its potential implementation in the complex multifaceted healthcare landscape was based.

Initial focus groups were held with young people, parents/carers, healthcare professionals from CAMHS crisis and liaison teams, CAMHS NHS management, NHS IT services, community CAMHS teams and NHS commissioners in two trusts in England. Following each focus group adaptations were made to the care bundle in an iterative manner. In the second round of focus groups, the adapted care bundle was presented to a mixed group of participants and agreed to be acceptable.

Through stakeholder feedback this study has adapted the SAFER-YMH to create the SAFER-YCL care bundle; an acceptable version for use in discharges from CAMHS crisis and liaison services. End-user design and involvement is vital in the development of clinical applications and pathways which are user-friendly and time-saving for healthcare professionals and also helpful for young people and their families.

The Central Mental Hospital is the Republic of Ireland's only secure forensic hospital and the seat of its National Forensic Mental Health Service (NFMHS). We scrutinised admission patterns in the NFMHS during the period 01/01/2018–01/10/2023; before and after relocating from the historic 1850 site in Dundrum to a modern facility in Portrane on 13/11/2022.

This prospective longitudinal cohort study included all patients admitted during the above period. The study initially commenced in Dundrum and continued afterwards in Portrane. Data gathered included demographics, diagnoses, capacity to consent to treatment, and the need for intramuscular medication (IM) after admission. Therapeutic security needs and urgency of need for admission were collated from DUNDRUM-1 and DUNDRUM-2 scores rated pre-admission. Hours spent in seclusion during the first day, week, and month after admission were calculated. Data were collected as part of the Dundrum Forensic Redevelopment Evaluation Study (D-FOREST).

There were 117 admissions during the 69-month period. The majority were male (n = 98). Most were admitted from prisons (87%). Schizophrenia was the most common diagnosis (55.8%). Mean DUNDRUM-1 triage security scores were in the medium-security range (2.84–3.15) during this period. At the time of admission, 53.8% required seclusion, 25.6% required IM medication, and 79.5% lacked capacity to consent to treatment. Those who required seclusion on admission had worse scores on the DUNDRUM-2 triage urgency scale (F = 20.9, p < 0.001). On linear logistic regression, the most parsimonious model resolved with five predictors of hours in seclusion during the first day and week, which were: D1 item 8 – Victim sensitivity/public confidence issues, D1 item 10 – Institutional behaviour, D2 item 2 – Mental health, D2 item 4 – Humanitarian, and D2 item 6 – Legal urgency. 50% required IM medication during their first week of admission and these patients had significantly worse scores on: D1 item 8 – Victim sensitivity/public confidence issues, D1 item 10 – Institutional behaviour, D2 item 2 – Mental health, and D2 item 4 – Humanitarian (all p < 0.05).

There was an increase in the frequency of admissions since relocating to Portrane. The results suggest that there was no change in overall triage security and urgency needs during the time period in question. Major mental illness related factors impacted the need for seclusion early in the admission, whereas factors linked to prison behaviour or personality-related factors were more associated with an ongoing need for seclusion at month one.

Side-effects are a recognised burden of all medications and are linked to poor compliance. In psychiatry, poor compliance can result in a relapse and significant deterioration in mental health. This has an impact on both the patient and the wider healthcare system. It has been speculated that if patients had more control/recognition of side-effects, compliance would increase.

GASS is a self-rating scale for side-effects of antipsychotic medication. It has the added effect of being able to stratify side-effects by their severity and biological system involved (Central Nervous System (Sedating) effects, Neurological (Movement) disorders, Anticholinergic, Gastrointestinal and Endocrine). The form consists of 22 questions with a scoring sheet attached to the reverse. Symptoms are graded by frequency and patient's perceived burden.

The British National Formulary has ‘minimum standards’ expected. These are designed to create a standardised approach to side-effect reviewing, encouraging a proactive reviewing process. These are meant to take place: After initiation and dose titration, at 3 months and annually thereafter. The National Institute for Clinical Excellence Guideline CG178 and the Scottish Intercollegiate Guidelines Network (Guideline-131) both advocate this standardised approach with the gold standard adding a review at 1 month.

The aim for this project was to audit the current completion rate of GASS forms in inpatient wards. The secondary aim was to improve completion rates after intervention.

1. 1. Search case notes and extrapolate data to Microsoft excel.

2. 2. Review data and identify challenges perceived from staff.

3. 3. Implement changes targeting highlighted challenges.

   1. 1. Present at ward QI meetings.

   2. 2. Create & discuss Infographic for staff.

   3. 3. Highlight role/importance of forms and usefulness to clinicians.

4. 4. Re-audit after 2 months.

Initial results found a completion rate of 7% across both wards reviewed (n = 41). Within this, 1 form was actually valid. One of the wards had no completed forms. The post-intervention group had fewer patients involved (n = 35), but an increased number of completed forms. Completion rate had risen from 7% to 26% (3–9 patients). Within this, the valid forms had increased from 1 to 4.

There was a clear impact on completion rate after initial interventions. The sub-optimal increase in completion highlighted the ongoing need for further input to improve completion rates.

This was a small, local audit of patients in an acute inpatient psychiatric ward. There was a recognised limitation on the number of patients in the study and acuity of some patient's illness, preventing completion.

Patients with severe mental illness are at a significantly higher risk of poor physical health outcomes than the general population and die on average 15–20 years earlier. The Royal College of Psychiatrists has published extensively on how to improve routine physical health monitoring in this cohort. Despite this, there is little data or guidance on improving emergency medical care for this cohort. We aimed to analyse and optimise the process of patients being sent to general hospital on an emergency basis.

A review was undertaken of the clinical notes for medical emergencies over a 12 month period by two core psychiatry trainees. Site visits and interviews with local A&E department clinicians, the covering General Practitioner and pharmacy were completed. A questionnaire was distributed to all nursing staff to gather their perspective on the considerations for emergency medical transfers out of The State Hospital.

On review of the case notes, 44/44 emergency outings were deemed to be clinically necessary for investigations/interventions that would not have been possible on-site. Qualitative methodology highlighted a disconnect amongst staff groups and stake holders regarding thresholds for transfers relating to medical emergencies leading to a high level of staff dissatisfaction.

The interface between psychiatric and medical services is an area of risk to patients. Levels of staff confidence, knowledge and available resources all contribute to the risk of transfer. Further work is required to explore other aspects of patient care and treatment which can be impacted as a consequence of emergency transfers.

UCL 5th year medical students undertake 3-week North London Mental Health Partnership inpatient psychiatric wards placements.

Before this project the management of these placements was at the discretion of individual ward teams. A varied, and potentially unsatisfactory, medical student experience resulted.

This project sought to implement a structured approach to placements.

Ward teaching best practice was, for the purposes of this project, considered to be (i) sending students a welcome email prior to placement, (ii) issuing placement timetable, (iii) using tutorial materials for onward tuition.

Project aims: 75% of wards sending welcome email, 75% issuing timetables, 75% using tutorial materials, 75% of students stating placement exceeded expectations.

This project consists of 1 PDSA cycle.

Prior to project baseline measures of ward teaching best practice were collected. The project started at the commencement of 2022/2023 academic year; duration: 12 weeks. The intervention was that inpatient medical teams were supported to send an introductory email to each student cohort, provide a placement timetable, and use supplied tutorial materials.

Questionnaires were emailed to inpatient medical teams at 6 and 12 weeks and to medical students at the end of placements. Medical team questionnaire covered engagement with best practice teaching. The student questionnaire addressed placement experience.

Outcomes at project conclusion:

• • 33.3% of wards sent introductory email.

• • 66.7% of wards issued a placement timetable.

• • 16.7% of wards used tutorial materials.

• • Less than 75% of student reported that the placement exceeded expectations.

The medical inpatient teams did not engage with this project as hoped. Feedback suggested reasons:

• • Lack of knowledge about the project.

• • Time pressures.

• • Perceived lack of medical student engagement.

• • Team had preferred teaching practices.

Despite this intervention, student inpatient placement experience remains varied.

It may have been optimistic to expect medical teams to change their established practice regarding medical students with only very modest additional support.

Some teams are enthusiastic and thoughtful about student teaching. Other are less so; this may be associated with temporary staff.

Following PDSA cycle 1 no further cycles were attempted as outcome suggested an alternative approach is required.

Possibilities for further PSDA cycles include:

• • Supporting placements via regular teaching-focused ward team meetings where expertise can be shared.

• • Appointing ward teaching fellows.

• • Explicitly rewarding inpatient teams displaying teaching excellence.

To confirm that 100% of patients treated by Electroconvulsive Therapy (ECT) have weekly assessment of mental state via Montgomery–Åsberg Depression Scale (MADRS).

To confirm that 100% of patients treated with Electroconvulsive Therapy (ECT) have regular assessment of their cognition before treatment and every 4 treatment sessions via the Montreal Cognitive Assessment Scale (MOCA).

This Audit included all service users attended ECT suite regularly at Worcestershire Specialist Mental health services over a period of 12 months between April 2022 and April 2023.

Twenty patients were included in this audit for whom data was collected from both electronic and paper records to analyse the percentage of compliance with the Electroconvulsive Therapy Accreditation service (ECTAS) standards with regards to the recommended weekly assessment of mental state via MADRS and the recommended regular assessment of cognition before treatment and every 4 treatment sessions via the MOCA Scale.

The Audit included 20 patients having ECT treatment done regularly over a year.

Overall, 87.32% of the patients were found to have MOCA assessment done before their first ECT session and every 4 treatment sessions as per guidelines. While 96.29% of the patients had MADRAS assessment done weekly or every two treatment sessions as per guidelines.

Regarding MOCA assessment, it has been found that 80% of the patients had MOCA done before their first treatment session. 94.73% of the patients had MOCA done after their 4th treatment session. 89.47% of the patients had MOCA done after their 8th treatment session. And 84.61% of the patients had MOCA done after their 12th treatment session.

With regards to MADRAS, 100% of the patients had MADRAS done before the start of the treatment. 90% of the patients had MADRAS done after the second treatment (1st week). 100% of the patients had MADRAS done after 4th treatment (second week). 100% of the patients had MADRAS done after 6th treatment (third week). 93.33% of the patients had MADRAS done after 8th treatment (4th week). 92.85% of the patients had MADRAS done after the 10th treatment (5th week).

Overall, ECT practice at Worcestershire Specialist Mental health services has been found to be in compliance with the ECTAS guidelines.

The majority of patients had MOCA assessments done regularly every 4 weeks with the highest compliance found to be after the first 4 treatment sessions and the lowest compliance was for the MOCA assessment done before the start of the ECT treatment.

In terms of MADRAS assessment, there was an overall adherence with the guidelines with very few patients missing MADRAS assessment only once over their course of treatment.

Catatonia is a neuropsychiatric syndrome that affects motor, speech, and behavioural functions. The link between catatonia and cannabis use is complex and poorly understood, with limited evidence from case reports about the neuropsychiatric manifestations. This paper aims to describe an unusual presentation of catatonia precipitated by a drink made from cannabis leaves (a.k.a. ‘Bhang’).

Mr JP, a 22-year-old college student, was admitted to an acute medical ward in North India. The medical team sought psychiatry opinion following the unusual presentation: sudden onset of mutism, staring, and rigidity. Physical examination revealed tachycardia and redness of the eyes. Routine blood investigations, EEG and MRI-Brain were unremarkable. Urine drug screen was positive for cannabis. Initially reluctant due to fear of legal troubles, the accompanying friends later revealed a history of ingestion of cannabis leaves (Bhang) for recreational purposes twelve hours ago. Following the clinical diagnosis of catatonia, the lorazepam challenge test led to improvement in rigidity and verbal responsiveness. No overt psychotic symptoms, such as delusions or hallucinations, were noted at the time or during follow-up. JP had previously experimented with smoked cannabis without any diagnosed psychiatric or medical complications requiring inpatient management. He was abstinent from all forms of cannabis use over the past three months due to college exams and denied any illicit substance use. Over the next two days, as the effects of the ingested cannabis wore off and oral lorazepam (6 mg/day) was continued, JP was back to his previous self with stable vital signs. He was discharged from the hospital with a plan to taper and stop lorazepam on an outpatient basis.

‘Bhang' has been a culturally acceptable cannabis form in the Indian subcontinent for centuries, providing an interesting cultural aspect to the case. This case highlights an unusual clinical instance of cannabis use; Oral ingestion led to a drastic presentation requiring hospitalisation, while the smoked form did not result in any similar sequelae. The study's limitations include the inability to test for synthetic cannabinoids and the lack of objective catatonia scoring scale(s).

With the surging popularity of cannabis use in recent years, it is essential to be aware of its various forms and exercise a high degree of suspicion towards unusual presentations. Further research is needed to understand the link between cannabis use and catatonia at the neurotransmitter level, mediated by individual risk factors.

Childhood maltreatment (CM) and peer victimisation (PV) are common sources of early-life interpersonal stress. CM is associated with atypical fronto-limbic emotion processing and regulation, and increased vulnerability for self-harm/suicide. However, few studies have compared the neurofunctional correlates between caregiver-inflicted versus peer-inflicted mistreatment. We compared the alterations of neurofunctional correlates of facial emotion processing in young people exposed to CM or PV and explored their associations with self-harm.

fMRI data were collected from 114 age- and gender-matched youths (39 CM, 37 PV and 38 controls) during an emotion discrimination task. Region-of-interest (amygdala, insula) and whole-brain analyses were conducted.

Groups differed significantly during processing of disgust only. Both CM and PV groups had lower activation in right amygdala and bilateral posterior insula than controls, where the left insular underactivation was furthermore related to increased self-harm in maltreated youths. At the whole-brain level, both CM and PV groups also had underactivation compared with controls in a cluster of bilateral limbic-thalamic-striatal, precuneus/posterior cingulate, temporal, fusiform/lingual and cerebellar regions, which was negatively associated with emotional problems in controls, as well as a cluster of somatosensory regions associated with increased self-harm in maltreated youths.

Early-life interpersonal stress from caregivers or peers is associated with common underactivation of limbic-thalamic-striatal, precuneus/posterior cingulate and somatosensory regions during disgust processing. The hypoactivation of key emotion and sensory processing and self-referential brain regions could be a potential suppressive mechanism to cope with the aversive emotion; however, it may also entail increased risk of affective psychopathology in seemingly healthy youths.

Persons with schizophrenia typically have a 20% shorter lifespan and mortality rates two times higher than the general population. More than 2/3 of this is due to different forms of physical diseases, like cardiovascular and metabolic syndrome. Systematic meta-analyses and various studies in schizophrenic patients revealed the prevalence of metabolic syndrome to range from 11 to 69%, poor sleep quality 30% to 80%, and impaired lung function ~30%. Both in the general population and in persons with schizophrenia, poor sleep quality and impaired lung function are associated with a heightened risk of metabolic and cardiovascular diseases. Hence, this study aimed to look for the magnitude of metabolic syndrome, poor sleep quality, and impaired lung function, and any association among them, if proven, may be helpful in better management.

We included sixty cooperative patients through purposive sampling with an age range of 18 to 65 years, meeting the DSM–5 criteria for schizophrenia, and excluded patients with co-morbid substance use disorder except for smokeless tobacco and caffeine. Harmonized criteria were used to diagnose metabolic syndrome; the Pittsburgh Sleep Quality Index (PSQI) for sleep quality and lung function was interpreted as per the Spirometry for Health Care Providers, Global Initiative for Chronic Obstructive Lung Disease.

55% were found to have metabolic syndrome. Poor sleep quality (PSQI > 5) was found in 60% of cases, with the most common sleep abnormality being increased sleep latency (95%). Restrictive Lung Dysfunction (RLD) was found in 46.7% of cases. 66.7% of the participants with metabolic syndrome had RLD, whereas only 22.2% without metabolic syndrome had RLD. The difference was statistically significant. No statistically significant difference was found between metabolic syndrome and sleep quality or sleep quality and RLD.

From the results obtained, it is clear that the prevalence of metabolic syndrome in people with schizophrenia is twice that of the general population, which also contributes to their increased mortality. Thereby, early identification of metabolic disturbances and correcting poor sleep quality and impaired lung function that are associated with an increased risk of metabolic syndrome will lead to increased life expectancy and a decrease in the mortality rate. Since lung function is studied in only a very few studies all over the world and ours being a novel approach in India showing significant association, it needs to be replicated in a larger sample size.

To investigate risk assessment and management processes across a health board in the context of the implementation of a new risk screening tool and risk policy using comparison of DATIX incidents before and after implementation of the CRAFT tool.

In mental health services, risk assessment and management are key responsibilities for clinical staff. A risk management tool that is structured and evidence-based aims to assist staff in managing risks including violence, self-harm, suicide and self-neglect.

It is not clear whether risk tools have clinical utility in influencing risk-related decision making and previous reviews within the health board indicated that risk policy was not being adhered to, prompting a review of the policy. Furthermore, policy recommends service user and carer collaboration with staff in all areas of mental health in Scotland but despite these recommendations there is little evidence to suggest they are routinely involved in risk assessment and management processes.

The present study is an opportunity to explore how teams think about and discuss risk management.

We looked at data on patient incidents that occurred over 30 months from 1/1/19 to 30/09/21. The Datix data were subdivided into five main categories: Violence & Aggression, Challenging behaviour, self-harm, absconding/missing and Suicide.

Throughout the study period the category of Violence & Aggression was the most frequently reported Datix category for 28 out of 30 months, followed by Challenging Behaviour which was the second most frequent category for 22 out of 30 months and in the last year reports in this category have increased by 39.35%. The third most frequently reported category was self-harm and the fourth most reported category was Abscondment/Missing. The frequency of reports in this category increased over the study period.

The rate of suicide was consistently the lowest reported category and remained stable throughout the study period. With the exception of Violence and Aggression, all categories showed a general upwards trend in Datix report numbers.

We have seen an increase in significant incidents in all categories reported using the DATIX system with the exception of suicide and violence and aggression during the study period. This suggests that further work is required to ascertain the reasons for this and what impact, if any, the change in CRAFT risk assessment tool has had.

People who have Alzheimer's disease (AD) often experience sleep disturbances due to the nature of the illness. Melatonin has been prescribed for sleep disturbance in individuals with AD, although there is a lack of national guidelines for pharmacological care for this presentation. Prolonged sleep disturbances for individuals with AD tend to lead to poor quality of life for the individual, behavioural challenges, carers' exhaustion and potential placement breakdowns.

The objective of this literature review is to determine whether the available evidence supports recommending melatonin to patients with AD for sleep, along with other benefits and adverse effects.

The hypothesis for this review is that melatonin is beneficial for sleep disturbances and has neuroprotection for individuals with AD.

Literature search on the online electronic database from 2010 to November 2023, using the title of “Melatonin's effectiveness and the side effects on Alzheimer's Disease''. This literature review was done by screening the 125 searched titles. The inclusion criteria included systematic review (SR), meta-analysis, randomised controlled trial (RCT), animals and cell studies. Exclusion criteria included case studies, literature and peer reviews. A total of 12 papers are included in this review.

The three SRs, two meta-analyses and one RCT showed the potential effect of melatonin on ameriolating cognitive decline, improving cognition, quality of life and sleep qualities, with the conclusion that further studies are required. One combined meta-analysis and SR showed melatonin might be an effective treatment for mild AD. One Cochrane review showed melatonin has no evidence of improving sleep for moderate-to-severe AD.

One animal study and two cellular studies showed a melatonin effect in the control progression of AD. One animal study and one cellular research study concluded that melatonin has potential treatment effects.

Adverse effects were mentioned at the higher dose (10mg) with negative reaction times, sedation and confusion.

There is a potential favourable effect of prescribing melatonin for mild to moderate AD, but there is limited evidence for prescribing it for moderate to severe AD. Furthermore, there is emerging evidence on melatonin's neuroprotective effect and potential treatment options for mild to moderate AD; further research is required for both sleep and neuroprotection in AD.

Lithium is a mood-stabiliser with a narrow therapeutic index. Patients are known to be at risk of lithium toxicity if they are unaware of how to recognise its signs. NICE guidelines outline the information that must be relayed to these patients. Furthermore, GMC Good Medical Practice highlights the importance of clear and contemporaneous patient records that contain relevant clinical information.

The aim of this two-cycle audit was to assess the quality of documentation for patients reviewed in lithium clinic and to analyse the consistency of the notes recorded to ensure high quality care provision and communication within the department, in line with the NICE guidelines.

The inclusion criteria were patients over the age of 65, prescribed lithium and were actively reviewed in the monthly lithium clinic at the Older Person's Mental Health Service (OPMHS) at Princess of Wales Hospital in Wales.

A data collection form was created to ensure all the relevant data in line with NICE guidelines was captured including serum lithium level, lithium dose, other psychotropics, side effects, renal function, patient mood, safety netting advice provided including signs of toxicity and awareness of lithium card. A standard of 100% was set for this data to be captured for each patient.

Cycle 1 was completed in November 2023 where a total of 18 patient records were selected (N = 18). Lithium dose, lithium level and renal function were recorded in over 83.3% (n = 15) of the files audited. Details on psychotropics were recorded in 61.1% (n = 11), side effects in 50% (n = 9) and patient mood in 77.8% (n = 14). Safety netting advice was recorded in 11.1% (n = 2). Furthermore, it was noted data recorded varied between clinicians.

The results of this audit were disseminated to OPMHS team. A proforma was introduced to encourage capture of all relevant information and to ensure consistency. Feedback was collected from clinicians using the proformas and relevant changes were made.

A second cycle of this audit was carried out after the proforma was introduced to the subsequent clinic (N = 12). This showed an improvement in record-keeping including lithium dose, lithium levels, psychotropics and side effects of 100% (n = 12). Renal function and mood were recorded in 91.7% (n = 11) of files and safety netting advice provision in 75% (n = 9) of files audited.

Introduction of a proforma is a simple and effective way to ensure relevant and important details are documented. This is not only for good clinical practice, but for medico-legal reasons also.

There is a significant mortality gap between the general population and people with SMI. This is especially prominent in those with psychotic disorders, underpinned by an increased risk of cardiometabolic disease. Identifying patients at risk early in their psychotic disorder is of key importance to reduce this mortality gap. Despite the recognised importance of regular physical health assessments in this group, completion rates are suboptimal. Point-of-care testing (POCT) to screen for diabetes and hyperlipidaemia, providing a result from a fingerprick sample in under 10 minutes presents a potential solution to enhance delivery of physical health checks and improve health outcomes in a proactive manner.

We introduced POCT across EIP teams in Southeast of England and evaluated the impact on physical health check completion rates and the quality of clinician-patient interactions in EIP teams.

A stepped wedge study was performed, introducing Abbot Afinion-2 machines across 30 EIP teams in all eight Mental Health Trusts in South East England (2021–2022). Numbers of completed physical health checks, and HBA1C and lipids blood tests completed in six months before and six months after introduction of POCT were collected from individual patients. Data were compared with those from the South West, which acted as a control region. Data were analysed from National Clinical Audit of Psychosis (NCAP) over comparable date range (2021–2022) to corroborate the findings. Clinician questionnaires were administered at three timepoints (after training, two-months, and eight-months), capturing training experiences, device usability and impacts on patient interactions.

In Southeast England, the rate and quality of physical health checks increased after introduction of POCT HbA1c testing OR 2.02 (95% CI 1.17 to 3.49), lipids 2.38 (1.43 to 3.97), and total completed health checks 3.61 (1.94 to 7.94). These increases were not seen in the Southwest region that did not introduce the machines. A post-hoc review of national audit data also showed a greater improvement of health checks in the intervention group compared with the comparator group over an overlapping timescale. Findings from the questionnaires evidenced improved patient engagement, clinician empowerment and the preference of POCT over traditional blood tests in this setting.

POCT is associated with improvements in the rate and quality of physical health checks, and this study emphasizes the potential of POCT in reducing health inequalities and enhancing holistic care for individuals living with severe mental illness.