Wednesday 4 December 2013

Could Patient Reported Outcome Measures (PROMs) and Patient Reported Experience Measures (PREMs) form a bridge crossing the public - professional divide?



PROMs event at King's Fund (in assoc w London School of Hygiene &Tropical Medicine)
3rd Dec 2013

Thanks to John Appleby and Nick Black for hosting a great, insightful, day covering all aspects of patient reported health status; from heart surgery to consultation with your family doctor, including outpatient services for women with heavy menstrual bleeding and mental health services. Topics varied from empowering patients to developing the statistics and economics underpinning this approach.

Placing PROMs in context
There is a growing focus on health outcomes for improving health services, especially relating to regulation and funding health services and hospitals. On the other hand, there is also a growing pressure to include patient’s opinions, both on moral grounds and in terms of accountability. This acknowledges that the clinician’s view of priorities for treatment for a patient are not always consistent with the patient's own view. One example may be where the clinician considers the benefit of a particular procedure on a particular aspect of health, whereas the patient may see this as only one part of their overall health and wellbeing, in the context of their social or family life. Therefore asking the patient their perceptions of their health status, in a systematic way, attempts to bridge the gap between a patient's lived experience, and the evidence-based outcomes focused approach of modern medicine and health systems.

Diversity of uses of PROMs
Different perspectives or motivations for using this approach were covered today. Robert Johnstone gave a patient's perspective and requested that PROMs should be used to inform and empower patients. Eugene Nelson and Tim Hughes mentioned how discussing PROMs within a patient consultation could improve communication between patient and clinician.  Using PROMs could improve clinicians understanding of patient's values as advocated by Ipek Gurol-Urganci, or improve services, as described by Philip van der Wees in an international study. Philip also discussed how PROMs could be used to make decisions about investment or commissioning of health services, whilst highlighting the conflicts of interest between using detailed information for organisations to improve services and transparency of sharing data to enable fair and transparent regulation and competition. Finally PROMs are used within clinical research, requiring assurance of robust validity and sensitivity; many issues around methods and analysis were debated today.

Patient Reported Experience Measure (PREM)
Improvement of quality of healthcare has emphasised different aspects over recent decades. These have included access to services, clinical effectiveness (frequently considered as 'evidence-based medicine/practice). Recently, patient experience and safety have stepped into the limelight. Institute of medicine describe four domains of quality of healthcare: effectiveness, cost, patient experience and safety. Patient Reported Outcome Measures can help judge effectiveness and safety, whilst the more recent Patient Reported Experience Measure addresses experience. Andrew Hutchings described his work developing this 'new kid on the block' using data collected by East Midlands Patient Experience Service. Andrew analysed how experience may relate to outcome and found there was a small but significant relationship between experience and outcome; i.e. patients who report good experience are likely to have better health outcomes, although being a statistician (!) he wouldn't be drawn on whether one caused the other. Similarly there is a relationship between medical complications following a procedure (eg surgery) and PREM score. Patients who reported better experience had lower probability of reporting a complication. Although some clinicians express concerns that PREMs might be influenced by the 'process' of delivering care, Andrew was confident (as much as a statistician can be!) that PREMs were actually measuring quality. Another issue was raised by Paul Sullivan who asked whether PREMs might be influenced by individual personality types, for example one person may respond in a 'negative' way, whereas another person receiving the same care might respond in a ' positive' way. If this were the case, PREM score may be predicted to follow PROM scores, as they are both subjective measures. Andrew felt this wasn't a strong factor, but hadn't included personality aspects into the analysis.

Use in clinic and aggregating data
Philip van der Wees discussed the tensions between using PROMs as part of the clinical consultation and as a way of reporting health organisation performance, which may relate to funding. Philip gave us two examples where these two aspects have been brought together; in Minnesota, clinical data is integrated with regional database and reporting. In the Netherlands mental health services routinely use PROMs as part of the patient consultation, while the provider company aggregates the data for quality improvement benchmarking.

An interesting point made by Jonathan Hill was that whilst 'before and after measures' were seen by physiotherapists as added bureaucracy, when they were asked to report the PROM at every session, they used this as a useful part of the consultation with the client. Eugene Nelson gave examples from US, Canada and Sweden where PROMs are used alongside e-health to fully engage patients with care planning.

Applying PROMs to long term conditions (LTC)
Adam Glaser, from National cancer survivorship programme gave a perspective from the growing need for support for people with long term conditions. Although PROMs were originally developed to capture a specific health intervention, they can be put to good use monitoring long term changes and unmet needs. Similarly Ben Ellis and Jonathan Hill are developing generic multidisciplinary PROMs for musculoskeletal conditions.

Challenges
Ongoing and future challenges with PROMs and PREMs were debated. Nancy Devlin described new measures of Quality of Life (EQ5D-5L); with increased sensitivity but potentially more difficult for patients to understand. James Coles and David Parkin debated case-mix adjustment which can be critical in comparing different providers or for research outcomes. Andy Street described how missing data can lead to misleading summary data. Eugene Nelson emphasised the need to communicate PROMs findings to patients as well as clinicians. Both of these groups could be put off by the more ‘academic’ interpretation of PROMs: reliability, validity, funnel plots etc...!

Overall I felt it was fascinating to hear so many diverse perspectives on capturing patient’s subjective assessments of their health and wellbeing. I was encouraged that these approaches are becoming mainstream and influencing clinical and health commissioning decisions. Hearing several comments about using PROMs/PREMs to enable patients to participate in decisions about their health is the epitome of ‘Nothing about me without me’, I think. However I was slightly disappointed that I hadn’t heard very much about patients and public being involved in the process of developing the measures and research questions. I’d like to think PROMs/PREMs may become a bridge between the public (patients and carers) and professionals (clinicians, managers and academics).