Abstract
Nash and Gorrell and their colleagues are rightly cautious when discussing the outcome study of their early psychosis intervention program, not least because of the high proportion of patients who declined to participate in the research and frequent non-completion of intake and follow-up assessments[1]. The headline finding that the program may have led to greater improvement in negative symptoms at 12 months is based on a dubious comparison of 14 versus 9 subjects from 215 potential patients. They conclude that this study and their companion case note audit [2] support the introduction of specialized teams for initial treatment of psychotic illnesses. Could I suggest that they have, in fact, provided evidence against the overall usefulnessofthis service innovation?
The authors emphasize the importance of continuity of care – one of the specifically audited indicators from the Australian Guidelines for Early Psychosis. Their examination of continuity of care is only based on visits by the community case manager during inpatient stays. However, by their very nature, subspecialist teams for early psychosis fracture continuity of care. There would have to be a massive demonstrable superiority in other areas of treatment to justify transfer ofpatients with severe and enduring mental disorders who are then returned to ordinary services after only 18 months. For 14 out of 24 indicators from the Australian Guidelines there were no clinically or statistically significant differences before and after introduction of the specialized service. Modest improvements were found for some relatively trivial indicators (patient offered group therapy 17% versus 37%; prescription of new rather than older antipsychotic medicines 62% versus 87%; as required benzodiazepines instead of antipsychotic medicines 17% versus 46%; family offered group work 6% versus 29%). There were advantages in regard to more important aspects of care but take-up was disappointing with both types of service (psychoeducation received 36% versus 56%; attendance at group therapy 8% versus 27%; relapse prevention plan prepared 11% versus 26%; attendance at family groups 2% versus 13%). Ordinary multidisciplinary teams that persevere in attempts to help mentally ill people and their families will have many more opportunities to provide these interventions. Similar unexciting advantages for the subspecialist service at the end of the study period may be counterbalanced by lower admission rates for patients under the care of generic teams.
Gorrell et al. cite a paper by Yung and colleagues published in this journal claiming to show clear superiority for an elaborate early intervention program compared with a neighbouring generic service [3]. This study has been used to bolster the unashamed political lobbying of the early intervention movement – even though Fitzgerald pointed out a year ago that there were problems due to possible confounding factors [4]. The results in regard to duration of untreated psychosis strongly suggest that the generic service provided better care and it is possible they used less seclusion, fewer community treatment orders and lower doses of neuroleptic agents. It is important that further analyses of these data should be published after a simple controlling for age.
Iam concerned that the recent papers by Nash et al. and Gorrell et al. will also be used inappropriately to influence mental health policy. In the UK, resources are being diverted from ordinary psychiatric services to highly protected early intervention teams. This has caused havoc in some areas for the ongoing treatment of people with serious mental illnesses. Further comparisons of models of care are urgently required [4], but with careful presentation and balanced discussion of the results.