Long delays in seeking treatment for schizophrenia

Abstract

Sir—Beng-Choon Ho and Nancy Andreasen, in their March 24 commentary, offer a timely and balanced commentary on the lengthy and damaging duration of untreated psychosis in first-episode schizophrenia. We agree that the importance of shortening this time is self-evident. We also agree that the secondary preventive potential of this action is by no means self-evident but remains to be shown, especially for early intervention in the prodromal or preonset phase of schizophrenia. We do not differ with Ho and Andreasen on these points but take issue with one interpretation and one polarisation. The interpretation involves their classification of Linszen and colleagues’ 1998 study among studies in which no association between long duration of untreated psychois and poor outcome was noted. Linszen and colleagues showed the lack of association was probably secondary to an overall absence of variation in duration of untreated psychosis in their sample— ie, the probability of a false-negative association was high. The polarisation in Ho and Andreasen’s commentary involves separating workers in the field of early detection and intervention into optimists and the understandably cautious. As such workers, we regard ourselves as embodying both attributes—optimism because we are there and caution because we are there as researchers. Overall, we wish Ho and Andreasen had made a clear distinction in their commentary between practice and research, because appropriate caution about early intervention as a practice is too frequently applied inappropriately and destructively at research efforts trying to show whether or not early intervention has risks or benefits. Ho and Andreasen end by calling for scepticism about the feasibility of early intervention. Research as hypothesis testing is formalised doubting, or scepticism par excellence. They, in essence, call for research, but their failure to distinguish between practice and research undermines the message.