PSYCHLOPS (Psychological Outcome Profiles) has been designed as a mental health outcome measure. The group that produced PSYCHLOPS consists of a research group of NHS based health professionals working in primary care. Our research work was funded through an NHS body, STaRNet, the South Thames primary care Research Network. It is based at the Department of General Practice and Primary Care, King’s College London.
When we first started out in 1999, we wanted to come up with an outcome measure that was much more client-centred than existing measures. We started by obtaining a multitude of outcome measures derived from the mental health and quality of life literature. During the course of this search, we found a few examples of a different type of measure, the so-called ‘patient- generated measure’. This is the ultimate patient-centred instrument in that the items to be measured are not chosen by experts in the field; they are chosen by the patient themself. The patient thus decides what is important and it is this that forms the basis of the measure.
We used the simplest patient-generated instrument that we could find as a starting point. This was an instrument called MYMOP (Measure Your Medical Outcome Profile). Originally devised by Charlotte Paterson and published in the British Medical Journal in 1996, MYMOP was designed to elicit details about the change in physical wellbeing after treatment for physical conditions (conventional or alternative therapy) in primary care. Its attraction was its simplicity. It simply asked the patient about the problem that troubled them most, a second problem (if there was one), how this affected their functioning and their wellbeing. Each response was scored by the patient. Both the problem and the main activity affected by this problem were defined in free text by the patient.
We decided to adapt this questionnaire. It needed re-working so that it was suitable for use with clients with psychological problems. To adapt it, we had to work through four stages:
We consulted with the national user group, Depression Alliance.
We asked therapists within our own research group to pilot the questionnaire
We emphasised the importance of an eye-catching design which was also compatible with self-completion. The design was produced by Dr Henk Parmentier.
We submitted our final questionnaire to the Plain English Campaign. After several re-draftings, we were awarded the Crystal Mark for plain English. PSYCHLOPS is the only mental health outcome measure to have received this award.
Following the process above, we produced a first version of PSYCHLOPS. It consisted of three domains (just like MYMOP):
Problems (2 questions)
Function (1 question)
Wellbeing (1 question).
In order to gauge the usefulness of PSYCHLOPS in evaluating talking therapies in everyday primary care, we had to test it and compare it with an established instrument. We devised a protocol and submitted this for ethical approval. This was granted in 2002 (reference: MREC/01/2/84). We also submitted a grant application to employ a research fellow to lead this project and we were fortunate to obtain funding from the King’s Fund. In late 2002, we started the trial comparing PSYCHLOPS with CORE-OM, one of the most commonly used outcome measures in primary care. In most trials, about two-thirds of clients fail to complete post-therapy questionnaires and we aimed to generate 100 matched pairs of pre- and post-therapy PSYCHLOPS and CORE-OM questionnaires. We therefore needed to recruit some 300 clients before starting therapy. Four therapists who used CORE-OM in everyday practice agreed to participate in the trial.
We completed the trial in June 2004 and published the validation study in 2005 (Primary Care Mental Health 2005;3:261-70).
For further details of validation studies exploring validity and reliability parameters, see the PSYCHLOPS publication summary sheet on the website.
The numerical scores to questions on PSYCHLOPS enable a figure to be calculated by comparing pre- and post-therapy scores. The literature on other patient-generated instruments suggests that they tend to produce larger change scores than conventional, checklist type instruments. Perhaps that is because the item being measured relates to the very issue for which the client is seeking help. Certainly, in testing MYMOP (see above), the change scores after therapy were much greater than those obtained using the standard functional assessment scale, SF-36. Results from subsequent studies suggest that PSYCHLOPS compares favourably to more conventional measures in terms of change scores. In other words, it appears to be a more sensitive indicator of change after therapy.
PSYCHLOPS was originally designed as a quantitative measure. But analysis of the statements written by clients in the free text boxes has yielded some interesting results. These statements are responses to questions about problems and functioning. Qualitative analytic techniques have been used to demonstrate that many of the issues raised by clients do not appear on conventional outcome measures. We published our first analysis of the qualitative data in 2006 (Primary Mental Health Care 2006;4:165-173).
For details of further qualitative studies, see the PSYCHLOPS publication summary sheet on the website.
In 2009, we launched PSYCHLOPS Version 5. This version has arisen through a process of continual development since the original version, produced in 2004. However, Version 5 has been designed to be as future proof as possible and it is our intention that this version should remain unchanged for 10 years.
The process of piloting, responding to feedback and considering new designs has been overseen by a group consisting of: Dr Susan Robinson, Dr Mel Shepherd, Professor Chris Evans and Dr Mark Ashworth.
The main reason for producing Version 5 was the need to devise a ‘during therapy’ version of PSYCHLOPS. As an outcome measure, its principal usefulness was as a measure of change occurring between pre- and post-therapy. However, in practice, some two-thirds of clients do not continue therapy through to completion resulting in loss of change data. Far better, we thought, to capture change data throughout the course of therapy. By obtaining PSYCHLOPS scores at various points during therapy as well as at the end of therapy, we would generate more change scores and have data on a larger proportion of those clients who completed pre-therapy questionnaires.
Version 5 includes a ‘during-therapy’ questionnaire as well as the original pre- and post-therapy questionnaires of the earlier versions. Version 5 has been piloted in a trial based in Poland.
Version 5 also contains a simplified scoring system. The scoring of PSYCHLOPS Version 1 was based on the mean scores for each question. Later versions were based on the sum of scores for each domain, initially scored 1-6 but now scored 0–5. The original four domains remain. Thus, with scores of 0 - 5 for each of the four questions, the total score of Version 5 may range from zero to 20.
It is our hope that PSYCHLOPS will join other psychometric instruments in everyday use as outcome measures and complement them rather than replace them. It is now ‘approved’ by the Department of Health as an outcome measure and is incorporated in the Outcomes Compendium published by the National Institute for Mental Health in England, 2009.
Because it is patient generated, it will never have the diagnostic accuracy of conventional instruments. However, going back to its roots, its strength is that it enables the client’s voice to be heard and is the most client-centred way to evaluate the outcome of therapy. We hope also that it will prove a therapist-friendly instrument and one that therapists will feel comfortable with now that most talking therapists have to demonstrate outcomes to their Mental Health Trusts or Primary Care Trusts.
We will be regularly updating this sheet and hope to keep you informed of new developments.
| Total visits to Psychlops.org pages since 21.x.07 by free web page counter |
Return to PSYCHLOPS home page