New services and systems of mental health crisis care: what works for whom and how?

Demand on mental health services has increased at a time when both inpatient and community mental health services are facing financial constraint, with many people accessing care at a time when they are experiencing an acute mental health crisis. The 2014 Crisis Care Concordat tasked service providers to improve access to appropriate care before and during mental health crisis, to improve the quality of crisis care, and to prevent future crisis by ensuring people are referred to appropriate community services. A number of new service models have emerged as a result, with the menu of crisis services on offer and the nature of the crisis care pathway differing considerably from one locality to the next.

The evidence-base for many of these new models of care is limited while system-level evaluations of the impact of changes to care pathways are lacking. It is not clear who has benefited from changes in crisis care provision, or whether persistent health inequalities have been ameliorated or perpetuated. This session will seek to address these issues. We will begin with a presentation at the system level, exploring service mix and care pathways, and then provide an in-depth focus on an innovative emerging response to the challenge; the Psychiatric Decision Unit.

A realist synthesis to explain how, for whom and in what circumstances different community mental health crisis services work (MH-CREST)

Dr Nicola Clibbens, University of Leeds

A reduction in psychiatric hospital beds alongside variation in service thresholds for crisis care has resulted in a variety of postcode-dependent service designs that can be difficult for people to access and navigate, meaning urgent needs are not always met. This study is carrying out a realist evidence synthesis with the collection and analysis of primary (stakeholder engagement) and secondary (realist synthesis) data. Using stakeholder expertise, evidence from research and current practice we will develop and test programme theories to identify mechanisms to explain how, for whom and in what circumstances community crisis services for adults work to resolve crises. By taking a theory-driven approach, our findings will be sensitive to context across diverse service designs and providers. This research will inform ongoing development and evaluation of existing mental health crisis services as well as future intervention development.

Psychiatric Decision Units - Mapping and Systematic Review

Jo Lomani, Dr Katie Anderson, St George’s, University of London

Psychiatric Decision Units (PDUs) have emerged as an approach to providing more appropriate assessment and care during a time of acute crisis, reducing unhelpful visits to Emergency Departments (ED) and unnecessary inpatient psychiatric admissions, providing a more therapeutic response to crisis and signposting people to community-based resources. We mapped the prevalence and operational structure of PDUs in England, receiving responses from 50 out of 53 mental health NHS trusts in England. Sixteen percent of mental health trusts had or were soon to open a PDU, the maximum unit length of stay was 48 hours, almost all units were co-located with a Place of Safety and they had a range of staffing patterns. We undertook the first systematic review of the PDU literature. International counterparts are often known as Psychiatric Emergency Services and have become increasingly critical to the delivery of mental health crisis care, particularly in the US. We evaluated the impact of PDUs on the frequency and length of inpatient admissions, and presentations and wait at EDs.

Psychiatric Decision Units – Pilot Interrupted Time Series Evaluation

Dr Jared Smith, St George’s, University of London

We piloted an interrupted time series (ITS) approach to assess the impact of introducing a PDU in a single Mental Health Trust site with a view to informing a large multisite ITS study. Routinely collected service data for a single Mental Health Trust were collected in the periods 24 months pre- and 24 months post-PDU implementation. Outcomes focussed on changes in the number and pattern of informal admissions to mental health Trust acute wards and psychiatric presentations in ED, including for those people with high service use. Changes in outcomes were explored using segmented time series regression models, allowing for serial correlation and underlying trends independent of the intervention, according to time aggregation units (weekly, monthly) that provided proportionately better model fits.