The sad truth is that most clinical research has absolutely no effect on clinical practice—rarely does a study impact how we diagnose and treat patients. A paper by Morrison and colleagues1 in the British Medical Journal is that rare exception. For the pleasure of reading it, see http://www.bmj.com/content/344/bmj.e2233.
In this one beautiful study, the authors have succeeded in definitively answering two of the most timely and consequential questions facing psychiatry today: (1) Is it a good idea to include a “psychosis risk syndrome” in DSM 5? (2) Can early treatment programs prevent psychosis? The answers to both questions are conclusively and emphatically—No.
First off, the study proves that we don't yet have the basic diagnostic tools needed to build a preventive psychiatry. Fundamental to any meaningful prevention program is the accurate identification of those who are really at risk. This study clearly demonstrates the impossibility of predicting who is (and who is not) likely to become psychotic. It is striking that only 8% of so called “ultra high risk” patients actually go on to have a psychotic episode? The unacceptably high false-positive rate means that more than 9 of 10 people entered into a psychosis prevention program will be misidentified and told they are at risk (or even worse “ultra high risk”) for eventually becoming psychotic—when quite clearly they are not. There can be no justification for burdening young people with such unnecessary stigma and worries, reduced ambitions, and the risk of receiving potentially dangerous antipsychotic drugs. If they have problems that need addressing, these should be addressed, but not under the misleading and pejorative auspice of being at “ultra high risk.”
The authors follow their findings to the by now obvious conclusion that “psychosis risk syndrome” is still in its early research phase of development and makes no sense as an official category for DSM-5. Their powerful data support the almost unanimous opposition of the leading researchers on psychosis risk—including most notably the pioneers in the field, Patrick McGorry and Alison Yung. It is puzzling that DSM-5 persists in offering a suggestion so far out of touch with current clinical reality.
The treatment results are even more important and should discourage all unrealistic expectations that early and intense intervention can prevent psychosis. Those who received the preventive intervention were just as likely to become psychotic as those who did not. The conclusion is inescapable—prevention of psychosis is a wonderful but elusive goal, one that we cannot now achieve. We are still in the early stages of prevention research and the findings so far are not encouraging. With such low rates of conversion to psychosis, it seems unlikely that a treatment effect will ever be demonstrated.
This one excellent study has killed two birds with one stone. It is a clear caution against the DSM-5 proposal for a psychosis risk syndrome and it should temper enthusiasm for rushing ahead with “ultra high risk” prevention programs. It does the wonderful cause of prevention no good to prematurely set impossible tasks that it cannot yet accomplish. The research foundation must be much more developed before psychosis prevention will be ready for the real world.
Reference
1. Morrison AP, French P, Stewart SL, et al. Early detection and intervention evaluation for people at risk of psychosis: multisite randomised controlled trial. BMJ. 2012;344:e2233. doi: 10.1136/bmj.e2233.
