Can we identify and treat “schizophrenia light” to prevent true psychotic illness?

BMJ 2013; 346 doi: (Published 18 January 2013)
Cite this as: BMJ 2013;346:f304

Recent rapid responses

Rapid responses are electronic letters to the editor. They enable our users to debate issues raised in articles published on Although a selection of rapid responses will be included as edited readers' letters in the weekly print issue of the BMJ, their first appearance online means that they are published articles. If you need the url (web address) of an individual response, perhaps for citation purposes, simply click on the response headline and copy the url from the browser window.

Displaying 1-3 out of 3 published

Dear Editor

The editorial by van Os and Murray (1) recently published in the British Medical Journal correctly states that positive psychotic symptoms are an imperfect way of distinguishing between normal variants in the population, individuals with mood and anxiety disorders, young people at ‘ultra high risk’ (UHR) or in the prodrome of a psychotic disorder, and those who have already “transitioned” to a frank psychotic disorder. By presenting a straw man that there exists some ‘apparently strong ultra high risk [UHR] paradigm’, the article goes on to question the validity of the UHR concept and to suggest a focus on treating psychotic symptoms found in non-psychotic disorders.

The primary problem with the commentary by Van Os and Murray (1) is its attempt to present itself as a novel contribution to the literature. The authors appear to have gone out of their way to avoid citing authors (notably ourselves) who have made all their points and more over the past several years. For example, we have previously noted that a majority of individuals who meet UHR criteria do not develop psychosis (2), and also stated that schizophrenia is not the only psychotic outcome from this UHR population (for example, mood disorders with psychotic features are included in psychotic outcomes(2)) . Moreover, in 2010, we explicitly questioned the validity of the threshold at which frank psychotic disorder is deemed to have begun, and acknowledged the importance of poor functional outcome in the absence of transition to psychosis (3). We have stated that outcomes other than psychosis need be assessed (4), and that poor functioning may be a more valid method of detecting “true” (deficit syndrome) schizophrenic illness than the “transition threshold” (5).

Clearly it is not news that the UHR criteria, which are currently predominantly defined by positive symptoms, are still a work in progress and cannot be considered “schizophrenia light”. We have explicitly stated that the UHR state should not be considered the prodrome to schizophrenia and pointed out that research predicated on this belief is making an invalid assumption (3). As we note previously, “As representatives of the PACE Clinic, the original UHR (prodromal) service, we feel it is our responsibility to highlight the limitations of the concept, and to remind other researchers and clinicians that it was only ever an empirically defined cut-off in what are continuous symptoms. It has been a useful starting point, but ongoing thinking and research are needed.” (3).

A further problem with the editorial is the exclusive focus on positive symptoms. It ignores the fact that schizophrenia is not a constellation of severe positive psychotic symptoms. Even the admittedly imperfect DSM and ICD-10 diagnostic systems include the criteria of functional impairment and negative symptoms. When negative symptoms and poor psychosocial functioning are added to the UHR criteria as predictors, they have good validity for predicting DSM defined schizophrenia (6-9) . Furthermore, there is now clear evidence that structural brain imaging techniques are developing to the point that they may become useful additional tools in predicting who will or will not transition to psychosis (10).

In summary, while we agree with Van Os and Murray that not all psychosis is schizophrenia, we point out that not all schizophrenia is psychosis. And while we agree that the UHR criteria have a varied longitudinal course with outcomes including mood and other non-psychotic disorders, we highlight that the criteria still detect onset of schizophrenia in rates far higher than found in the general population, especially when psychosocial functioning and negative symptoms are used as predictors. As noted before, we suggest a way forwards is to broaden the outcome of interest to include psychosocial function negative symptoms, and cognition (11). Clearly non-psychotic outcomes of mood and anxiety disorders are important as well. We believe that the UHR criteria identify a highly symptomatic and distressed group and research in all these areas should be encouraged.

Alison R Yung, Professor, Institute of Brain, Behaviour and Mental Health, University of Manchester, UK and Orygen Youth Health Research Centre, University of Melbourne, Australia

Stephen J Wood, Professor, School of Psychology, University of Birmingham, UK and Melbourne Neuropsychiatric Centre, University of Melbourne, Australia

Ashleigh Lin, School of Psychology, University of Birmingham, UK


1. van Os J, Murray RM. Can we identify and treat “schizophrenia light” to prevent true psychotic illness? BMJ: British Medical Journal. 2013;346.
2. Yung AR, Phillips LJ, Yuen HP, Francey SM, McFarlane CA, Hallgren M, et al. Psychosis prediction: 12-month follow up of a high-risk. Schizophrenia Research. 2003;60(1):21-32.
3. Yung AR, Nelson B, Thompson A, Wood SJ. The psychosis threshold in Ultra High Risk (prodromal) research: is it valid? Schizophrenia Research. 2010;120(1-3):1-6.
4. Lin A, Nelson B, Yung AR. ‘At-risk’ for psychosis research: where are we heading? Epidemiology and Psychiatric Sciences. 2012;doi:10.1017/S2045796012000388.
5. Lin A, Wood SJ, Nelson B, Brewer WJ, Spiliotacopoulos D, Bruxner A, et al. Neurocognitive predictors of functional outcome two to 13 years after identification as ultra-high risk for psychosis. Schizophrenia Research. 2011;132(1):1-7.
6. Cannon TD, Cadenhead K, Cornblatt B, Woods SW, Addington J, Walker E, et al. Prediction of psychosis in youth at high clinical risk. Archives of General Psychiatry. 2008;65(1):28-37
7. Ruhrmann S, Schultze-Lutter F, Salokangas RKR, Heinimaa M, Linszen D, Dingemans P, et al. Prediction of psychosis in adolescents and young adults at high risk: results from the prospective European Prediction of Psychosis Study. Archives of General Psychiatry. 2010;67(3):241-51.
8. Nelson B, Yuen HP, Wood SJ, Lin A, Spiliotacopoulos D, Bruxner A, et al. Predictors of transition to psychosis in the ultra high risk (“prodromal”) population: A long term follow up study. JAMA Psychiatry. In press.
9. Fusar-Poli P, Bechdolf A, Taylor M, Bonoldi I, Carpenter W, Yung A, et al. At risk for schizophrenic or affective psychosis? A meta-analysis of DSM/ICD diagnostic outcomes in individuals at high clinical risk. . Schizophrenia Bulletin. 2012;doi:10.1093/schbul/sbs060.
10. Koutsouleris N, Borgwardt S, Meisenzahl EM, Bottlender R, Möller HJ, Riecher-Rössler A. Disease prediction in the at-risk mental state for psychosis using neuroanatomical biomarkers: results from the FePsy study. Schizophrenia Bulletin. 2011.
11. Lin A, Wood SJ, Yung AR. Measuring psychosocial outcome is good. Current Opinion in Psychiatry. 2013;26(2):138-43.

Competing interests: None declared

Alison R Yung, Professor of Psychiatry

Stephen J Wood, Ashleigh Lin

Institute of Brain, Behaviour and Mental Health, University of Manchester, UK, Oxford Rd, Manchester

Click to like:

Dear BMJ

I was interested to read my colleagues’ editorial addressing the controversial notion of the psychosis spectrum (1-3) or ‘schizophrenia-lite’ as they call it. The editorial was prompted by a meta-analysis by Stafford et al (2) on early interventions to prevent psychosis. Although they are critical of arbitrary cut-offs for psychosis diagnosis and other fuzzy concepts they are insufficiently critical of some implications of the psychosis continuum and the Stafford article.

Regarding the article, the authors’ classification of evidence was overly negative and possibly biased. It seems unfair that trials of novel agents such as omega fatty acids are marked down because their appearance in the literature may one day be seen to reflect publication bias. Of course a new treatment which seems to work, on the basis of a double blind randomised controlled trial (RCT), will tend to be published, and it may take some time for other studies to appear (positive and negative) to the point at which an estimate of publication bias can be made. This is always going to be the sequence. So to automatically impugn (‘downgrade’) a recent study because of the ‘risk of publication bias’ is unjustifiable. If that is allowed then all psychological treatment trials should automatically be downgraded too. It is undeniable that no psychological treatment can ever achieve the perfect balance and total blinding of a pharmacological RCT. Naturally there are many ways in which pharmacological trials may be flawed and similarly many ways in which psychological intervention trials can be rigorous but to be consistent Stafford et al., should ‘downgrade’ all psychosocial interventions on the basis that ‘blinding is impossible’.

As for van Os and Murray’s critique – they fail to note that reporting bias is an obvious explanation for the apparent ubiquity of psychotic symptoms, especially in those with anxiety and depression. If a person is feeling bad and wishes to convey this they may, when faced with a questionnaire full of serious sounding symptoms which, they are told, ‘many people experience’, endorse the items in a more or less unthinking way. If asked about, say, bowel symptoms they would no doubt endorse some of these too. Chest pain? Tick; burning eyes? Tick; Hearing voices? Tick. No wonder the ‘transition rates’ to frank psychosis are so low! It seems that the drive to diagnosis psychosis early so that it may be prevented has now waned and our new imperative is to sub-categorise common mental disorders as “with or without psychotic symptoms” – either way, the treatment is, of course, CBT.

I would suggest we step back and reconsider. Work on various ‘high risk’ patients around the world, as the editorialists note, shows that regardless of whether they present with medication-requiring ‘frank psychosis’, they are chronically unsatisfied, they lead unrewarding and unproductive lives, with enduring symptoms, social and cognitive impairments. Is it not possible that such people have always been with us? They perhaps existed unobtrusively in the preindustrial age and continue to do so in rural economies. They can lead fulfilling lives with support, for example from their parents. But they are unlikely to find life partners, leave home, and get jobs without some help (again maybe from their parents). Such people may benefit from psychological interventions, and as always, “long term follow up studies are needed”, but more likely they wont, because that is just the way they are. The important news for them and their families is that they don’t have schizophrenia. Diagnosis may be important after all.

Anthony S David
Professor of Cognitive Neuropsychiatry, Institute of Psychiatry, King’s College London, London UK.

1. BMJ 2013;346:f304
2. David AS. Why we need more debate on whether psychotic symptoms lie on a continuum with normality. Psychol Med 2010; 40: 1935-42.
3. David AS. Should the diagnostic boundaries of schizophrenia be expanded? Cogn Neuropsychiatry 2011; 16: 97-100.
4. BMJ 2013;346:f185

Competing interests: None declared

Anthony S David, Professor of psychiatry

Institute of Psychiatry, King's College London, PO Box 68, Institute of Psychiatry London SE5 8AF

Click to like:

Dear Editor

Interventions for drug use need to be considered in any plan to reduce transition and diagnostic delay in those with psychosis

I read with interest the editorial by van Os and Murray (BMJ 2013;346:f304)

I agree with the authors that longer term follow up studies are required to determine the effectiveness of early interventions in those at 'high risk' of psychosis. I also applaud their admission over the blurred diagnostic boundaries and difficulties in defining transition points. What I find odd is that no mention is made of the contribution of cannabis and other drug use to this early diagnostic and phenomenological blurring. With only vague temporal operational criteria in current diagnostic manuals I would suggest that the 'waste bin 'diagnosis of drug-induced psychosis also significantly contributes to the both the blurring of clinical responsibility between services and a delay in diagnosis. Concurrent substance use, so very common in first episodes, can, if not dealt with in a truly integrated fashion lead to delayed diagnosis and subsequently access to specialist services. As an addiction psychiatrist I see how in times of fiscal dieting, gatekeeping to specialist services sometimes appear focused on reducing access. Such practice is inconsistent with efforts to reduce transition in high-risk individuals and improve longer-term outcomes. Rather optimistically I would request for a wider adoption of clinical skills to address substance use within psychosis services and a willingness to follow those given a diagnosis of a drug induced psychosis long enough to decide what condition the person has and what treatment is appropriate. At present all too often concurrent drug use even in the most vulnerable of people seems to enough not only to be refused entry to research trials but also psychosis services.

Competing interests: None declared

Adam R Winstock, Consultant Addiction Psychiatrist

South London and Maudsley NHS Trust, 151 Blackfriars Road SE1 8EL

Click to like: