Some aspects of crisis care appear effective according to a broad NIHR overview of mental health crisis services. Crisis resolution teams that offer home treatment and support early discharge from hospital, is one successful example – although UK implementation is very variable. Early intervention service models are helpful for people with psychosis, as is collaborative care for those with depression. Crisis houses and acute day hospitals, known as “safe havens”, are popular with service users, and supported employment may help prevent future crises. Finally, liaison psychiatry teams might improve waiting times and reduce re-admission. Many of these interventions are recommended in NICE guidance. Overall, there was sparse evidence on cost-effectiveness.
This was a very broad and complex review, comparing very different kinds of services from police street triage to telephone helplines. The quality of evidence was generally low and high-quality UK evaluations are needed. The authors identified particular gaps in evidence around support before crisis and access to emergency services during crisis. But this review for the first time gives a useful map of current service models and existing evidence along the mental health crisis pathway.
Why was this study needed?
Spending on mental health is the largest single category of NHS expenditure. In England, in 2014/15, 3617 people per 100,000 of the population in England accessed mental health and learning disability services. The impact in economic and social costs of mental health problems in 2009/10 was £105 billion.
It has been remarked that services for people having a mental health crisis are not always fit for purpose. A CQC survey in 2015 found that only 14% of people who experience a mental health crisis felt that the care they received provided the right response and helped to resolve their situation.
Recognising this, the 2014 Crisis Concordat – an agreement between services involved in the care and support of people in crisis in England – set out how organisations should work together to make sure that people get the help they need. It focused on four main areas of crisis response: support before crisis point; urgent and emergency access to crisis care; quality treatment and care when in crisis; and promoting recovery.
The NIHR funded this review to examine the current state of guidance and evidence across these four stages of care.
What did this study do?
This was a pragmatic review that included evidence-based guidelines, reviews and individual studies. Guidelines, by NICE or NICE accredited bodies, were prioritised. These were usually for specific conditions, such as psychosis or depression. Where these weren’t available, systematic reviews and “reviews of reviews” were consulted. In cases where neither guidelines nor systematic reviews were available, primary studies were reviewed.
Nine NICE guidelines were included, one review of reviews, six systematic reviews and 15 primary studies. All articles focused on developed country healthcare systems, though only three of the 15 primary studies were in the UK. Primary studies were evaluation studies, such as before and after studies, and qualitative research.
Study quality was assessed with a variety of tools according to the study type being appraised. The evidence was generally of low quality. The authors noted a general lack of rigorous randomised trials or appropriate controlled evaluations.
What did it find?
- Access to support before crisis point: NICE guidelines recommend that people at risk of mental health crisis should receive care and referral quickly. This might include making self-referral easy, telephone support and triage, early detection and training of general health staff. However, there is little reliable evidence as to how this should be done.
- Urgent and emergency access to crisis care: There was some evidence that liaison psychiatry teams may improve waiting times and reduce readmission (five studies, two of which were in the UK), and that police officers with mental health training were less likely to arrest people with mental health problems but rather take them to a health-care setting (six studies, one in the UK). However, studies in both cases were mostly small and uncontrolled.
- Quality treatment and care in crisis: Crisis resolution teams were found to be both clinically and cost effective compared to inpatient care (from six randomised controlled trials and two UK economic analyses). However, implementation is variable, with few teams meeting all evidence-based criteria for good practice. Crisis houses and acute day hospitals, which are recommended by NICE, were as effective as inpatient care (one RCT in the USA), but associated with greater service user satisfaction. No cost-effectiveness studies were found. Again, however, the quality of the evidence was low.
- Promoting recovery/preventing future crises: This was a very broad category of services. Early intervention service models were found to be effective (and probably cost-effective) for people with psychosis and other serious mental illnesses, as was collaborative care for people with depression, particularly those with chronic physical health problems. Both are recommended by NICE. There was some evidence for supported employment and for peer support. Joint crisis plans are recommended by NICE, but more recent trials have found that implementing crisis plans in routine practice is challenging.
What does current guidance say on this issue?
The nine NICE guidelines included in this review and their publication dates were: Psychosis and schizophrenia in adults 2014; Psychosis and schizophrenia in children and young people 2013; Service user experience in adult mental health 2011; Self harm 2011; Borderline personality disorder 2009; Psychosis with coexisting substance misuse 2011; Bipolar disorder 2014; Depression in adults 2009; and Depression in adults with a chronic physical health condition 2009.
NICE guideline recommendations were reported as part of the results, as described above.
What are the implications?
The scope of this review was extremely broad and it is clear that much evidence on mental health services for crisis care is both patchy and of low quality. There are however some interventions that appear promising. Crisis resolution teams appear effective, and crisis houses and acute day hospitals are popular with service users. Early intervention service models are effective for people with psychosis, and collaborative care for those with depression. Finally, there is some evidence for supported employment, although this covers a range of initiatives. Most studies did not measure cost-effectiveness compared with usual care.
There is also slightly weaker evidence that liaison psychiatry teams may improve waiting times and reduce readmission, and that training police officers may avoid run-ins with the justice system. Peer support interventions may be effective too.
The low quality of the evidence is at least partly a reflection of the challenges of conducting and evaluating complex interventions in this area. However, recent methodological advances in the evaluation and synthesis of complex interventions should offer future results that will help commissioners. For instance, an ambitious NIHR evaluation is underway of liaison psychiatry services in hospitals. This uses a sophisticated study design, including linked data and careful matching of cohorts of patients using liaison psychiatry services in different hospitals with those not using these services. There are also a number of ongoing trials, such as the NIHR funded Crisis resolution team Optimisation and RElapse prevention (CORE) study, which will report soon.
This is an area of research that is likely to deliver better and more reliable evidence during the next few years.
Citation and Funding
Paton F, Wright K, Ayre N, et al. Improving outcomes for people in mental health crisis: a rapid synthesis of the evidence for available models of care. Health Technol Assess. 2016;20(3):1-162.
This project was funded by the National Institute for Health Research Health Technology Assessment (HTA) (project number 14/51/01).
CQC. Right here, right now – help, care and support during a mental health crisis. London: Care Quality Commission; 2015.
DH. Mental Health Crisis Care Concordat Improving outcomes for people experiencing mental health crisis. London: Department of Health; 2014.
HSCIC. Mental health bulletin. Annual report from MHMDS Returns 2013-14. Leeds: Health and Social Care Information Centre; 2015.
NHS England. Valuing mental health equally with physical health or “Parity of Esteem”. London: NHS England; 2015.
Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre