The ethos of PCMIS and the research team is to get everyone involved so many of our customers actively participate in the research which informs our software development.
Embedded research tools pave the way for PCMIS
The Outcome Feedback (OF) tool helps psychological therapists and other mental health professionals to track progress and to identify cases at risk of poor response to treatment in order to intervene with solutions as early as possible.
Studies in USA and Europe indicate that using OF can help improve clinical outcomes for patients at risk of poor progress.
Our research aims to introduce OF into IAPT services.
The goals of the programme include:
November 2015 update
An NHS Research Capability Funding (RCF) grant was obtained in May 2015 to conduct a feasibility study to explore factors that prevent and enable the utilisation of OF in routine practice
NHS ethical approval and research governance permissions for the feasibility study were obtained in September 2015
The feasibility study started 1 October 2015, with a training day, including 15 therapists working in Leeds IAPT. The group of participants includes Psychological Wellbeing Practitioners, CBT and IPT therapists
IAPT services in London, North West and East of England have joined the collaboration and agreed in principle to support a multi-site randomised controlled trail. Its aim will be to test the effectiveness of using OF compared to a 'usual IAPT care' control group
Tools created through research
In 2014 our research collaborators* developed state-of-the-art outcome prediction tools using a large IAPT dataset. There are two tools integrated within PCMIS that can be used by clinicians to identify cases at high risk of poor outcomes.
Figure 1. ETR model using PHQ9 depression scores. This model shows a case that is 'on track' (progressing as expected) Expected Treatment Response (ETR) models
ETR models are used to alert clinicians to cases that are not progressing well in therapy. These models are based on two visual markers: an upper ETR boundary and a lower boundary. ETR boundaries, also referred to as 'curves', are superimposed onto routine symptom tracking charts available on PCMIS. These curves are derived from large datasets of psychotherapy cases that had similar baseline scores on symptom measures (eg PHQ9, GAD7). Approximately 80% of patients will show session-to-session symptom scores contained within the upper and lower ETR curves. The remaining cases show atypical scores, with 10% of cases falling above the upper ETR curve (high risk cases) and 10% falling below the lower boundary. In routine practice, if your patient shows a symptom score that is above the upper ETR curve, this indicates that the patient is not progressing as expected and may be at risk of poor outcomes.
(Figure 2. ETR model using GAD7 anxiety scores. This model shows a case that is 'not on track' (at risk of poor outcomes) Research in action (Leeds Risk Index)
The Leeds Risk Index (LRI) is a numerical scale ranging between 0 and 21. Each patient can be assigned an LRI score based on their clinical and demographic characteristics. A higher score has been found to predict greater likelihood of dropout, poor response to therapy, and non-recovery after treatment. This tool was developed using patient profiling methods, and can be used as a measure of case complexity and risk.
Together, ETR and LRI tools can be used by clinicians to detect high risk cases before they drop out, and early enough to review and modify treatment to improve clinical outcomes.
(Figure 3. Associations between LRI scores and recovery rates. This graph shows that cases with the highest LRI scores had the lowest % of reliable and clinically significant improvement (RCSI). *Collaborators:
Dr Jaime Delgadillo (University of York, UK),
Mr Omar Moreea (Leeds Community Healthcare NHS Trust, UK),
Professor Wolfgang Lutz (University of Trier, Germany)
For further information about our research: http://www.pcmis.com/research.html
For further information about our research group: http://www.york.ac.uk/healthsciences/research/mental-health/