UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

The lonely soul wanders: on the role of impaired social functioning in the prediction of a firstpsychosis

Velthorst, E.

Link to publication

Citation for published version (APA):Velthorst, E. (2011). The lonely soul wanders: on the role of impaired social functioning in the prediction of a firstpsychosis.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 08 Sep 2020

On the role of  impaired social functioning in the prediction of  a first psychosis

On the role of  impaired social functioning in the prediction of  

a first psychosis

E V A V E L T H O R S T

1

The Lonely Soul WandersOn the role of impaired social functioning in the prediction of a first psychosis

Eva Velthorst

2

THE PUBLICATION OF THIS THESIS WAS FINANCIALLY SUPPORTED BY: ASTRAZENICA, LUNDBECK B.V.,

ACADEMIC MEDICAL RESEARCH B.V., GRADUATE SCHOOL ADADEMIC MEDICAL CENTER.

Print: Koninklijke drukkerij c.c. callenbach b.v. , te Nijkerk.ISBN: 978-90-9026308-3

Illustration Cover: SKYBOX CONCEPTUAL DESIGNDesign & Layout: SKYBOX CONCEPTUAL DESIGN

Copyright: Eva Velthorst

3

The Lonely Soul WandersOn the role of impaired social functioning in the prediction of a first psychosis

Academisch proefschrift

ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam

op gezag van de Rector MagnificusProf. Dr. D.C. van den Boom

ten overstaan van een door het college van promoties ingestelde commissie,

in het openbaar te verdedigen in de Aula der Universiteitop woensdag 5 oktober 2011, te 11:00 uur

door

Eva Velthorst

Geboren te Hoorn

4

Promotiecommissie

Promotores: Prof. dr. D.H. Linszen Prof. dr. L. de HaanCo-promotor: Dr. D.H. Nieman

Overige leden: Prof. dr. M. van der Gaag Dr. M.W.J. Koeter Dr. C. Morgan Prof. dr. J. van Os Prof. dr. A.H. Schene Prof. dr. G.M. Schippers

Paranimfen: Mr. N. Velthorst en drs. F.L. Schneiders

Faculteit der Geneeskunde

5

CONTENTS

Preface 7

Chapter 1. General Introduction 9 Part IChapter 2. Ethnicity and baseline symptomatology in patients with an 21 “At Risk Mental State” for psychosis

Chapter 3. Factor analysis of the scale of prodromal symptoms: 41 differentiating between negative and depression symptoms Part IIChapter 4. Baseline differences in clinical symptomatology between ultra high risk 59 subjects with and without a transition to psychosis of psychosis

Chapter 5. Disability in people clinically at high risk 75

Chapter 6. The Strauss and Carpenter Prognostic Scale predicts transition 95 to psychosis in subjects clinically at high risk for psychosis

Chapter 7. Social disability at admission for a first psychosis does not predict 111 clinical outcome at 5-year follow-up Part IIIChapter 8. Three-year course of clinical symptomatology in young people at 125 ultra high risk for transition to psychosis

Chapter 9. Ultra high-risk state for psychosis and non-transition: 139 a systematic review

Chapter 10. Summary, Conclusions and Discussion 167

Nederlandse samenvatting en conclusie 181

Dankwoord. List of Publications, Curriculum Vitae 187

6

7

The Soul’s Superior instants

Occur to Her – alone –

When friend – and Earth’s occasion

Have infinite withdrawn –

Emily Dickinson The complete poems of Emily Dickinson, Edited by Thomas H. Johnson, 1961

It was in the first half of the 1860’s, after she had already largely withdrawn from social contact, that Emily Dickinson (1830-1886) wrote the bulk of her outstanding poems. In her case, abandoning social contact did not only lead to nearly 1800 verses, she also believed that only after having severed all ties that bind to the daily world (friend and earth’s occasion) one could come to spiritual heights, or ‘superior instants of the soul’ (for an elaboration of this and many other poems, see The Emily Dickinson Journal, 2010).

Notwithstanding the positive experiences that loneliness may bring about, a loss of social contact, either through social dysfunction, anhedonia, or adverse social experiences, can have tremendous negative effects. It is thought to play an important role in - and might even contribute to- the develop-ment of ‘wandering’ thoughts (or delusional ideas) in one of the most disabling diseases to date: schizophrenia (Üstun et al., 1999).

8

9

Chapter 1General Introduction

SCHIZOPHRENIA

Since the early twentieth century, schizophrenia has been described as a severe mental disorder which can manifest itself through a wide variety of symptoms and deficits (Kraepelin, 1919; Bleuler, 1950; Mueser & McGurk, 2004). Despite the diversity in course, impairment, symptoms and patho-genesis, the syndrome is invariably characterized by recurrent distortions in reality testing during one or more psychotic episodes. These distortions are generally referred to as ‘positive symptoms’ and relate to experiences such as hallucinations, delusions, bizarre behaviour and formal thought disor-ders (American Psychiatric Associations, 1994). The positive symptom dimension can be distin-guished from the negative symptom dimension which refers to the absence of functions normally present in the healthy population (i.e. alterations in drive and volition such as affective flattening, social withdrawal, alogia and avolition: Mueser & McGurk, 2004; van Os & Kapur, 2009). In addition to the well-known positive and negative symptoms, also described as the ‘char-acteristic symptoms’ in the DSM-IV (American Psychiatric Associations, 1994), a third broad cluster of symptoms comprises of cognitive impairments (Mueser & McGurk, 2004; van Os & Kapur, 2009). Cognitive impairments are thought to be a direct expression of underlying brain pathology, and start off before the first psychotic episode (Cornblatt et al., 2003; Harvey, 2009; Becker et al., 2010). They have been related to both negative symptoms as well as social impairments (Mueser & McGurk, 2004; Weinberg et al., 2009; Dominguez et al., 2009; Insel, 2010); impairments which Bleuler already recognized as one of the four main disturbances (or ‘A’s) in schizophrenia (‘Autistic Isolation’; Bleul-er, 1950, cited in Insel, 2010). As to date, schizophrenia is still among the world’s top ten causes of long-term disability (Mueser & McGurk, 2004). In fact, estimates in Europe and the United States in-dicate employment rates of less than 20% (Marwaha, 2007; McGurk et al., 2009). A first psychosis in the schizophrenia-spectrum generally emerges in late adolescence or young adulthood, although in particular cognitive and social impairments are evident much earlier (Häfner et al., 1995; Cornblatt et al., 2003; Cannon et al., 2003; Shapiro et al., 2009; Harvey, 2009). Schizophrenia, only diagnosed when illness duration exceeds 6 months, affects about 0.5 to 1 percent of the world population (McGrath et al., 2004). The incidence is roughly equally distributed among male and female, albeit that female tend to have a later disease onset (Häfner et al., 1992), better social functioning and better course of illness (Mueser & McGurk, 2004).

10

AETIOLOGY AND PATHOPHYSIOLOGY

More than one-hundred years after the term ‘schizophrenia’ has first been coined, its exact aetiology is still unknown. Apart from the complexity of the disorder, it has been argued that the progress in schizophrenia research may be complicated by the fact that the disorder ‘schizophrenia’ - diagnosed by means of clinical observations – is unlikely to be a separate categorical construct, corresponding to one biological entity (Meyer-Lindenberg, 2010). Instead, psychotic disorders have been hypothe-sized to be the severe end of a continuum (van Os et al., 2009). For decades, attempts have been made to unravel the pathophysiology of schizophrenia through explanatory models from different perspectives. Hypotheses have varied widely, ranging from schizophrenia being a reaction to the rejection of an ambivalent mother, through to schizophrenia as a purely genetic disorder in the first years of the present century (Insel, 2010).Amongst other factors, the consistently reported modest effect sizes of the individual candidate genes in years of schizophrenia research however (Insel, 2010; Meyer-Lindenberg, 2010), has led to the current prevailing view that genetic and environmental factors most likely interact (van Os, 2010). A growing body of literature suggests that environmental factors like urbanicity (Krabbendam & van Os, 2005; Kelly et al., 2010), social adverse experiences (Selten & Cantor-Graae, 2007) and cannabis use (Henquet et al., 2008; Dragt et al., submitted) can unfavourably interact with genetic vulnerability on developing brains during certain sensitive periods, but mainly in people with a par-ticular genetic make-up (Korver & Quee et al., submitted; van Os et al., 2010). The global interest in these complex interactions has fuelled the need for multi-disciplinary research paradigms, which are currently being carried out in worldwide collaborations (EU-GEI, European Network for Schizophrenia Studies; European Community’s Seventh framework Program). At the same time, the neurodevelopmental model has regained wide interest (Lewis, 2002; Cornblatt et al., 2003; Cannon et al., 2003; De Haan and Bakker, 2004; Insel, 2010). The model, positing that pathogenetic biological characteristics are present long before onset of clinical symptoms (Lewis et al., 2002), has led to the current perspective that a first psychosis is not the disease onset but rather a later stage in the course of a developmental disorder (Insel, 2010). This new perspective meant an increasing recognition of the earlier stages of schizophrenia.Indeed, the prodrome and what is termed ‘early psychosis’ has become one of the most enthusiastically researched fields in modern psychiatry (Simon & Velthorst et al., this thesis). Research into the schizophrenia prodrome may not only contribute to a better understanding of the pathogenesis of schizophrenia, enhancing our ability to predict a first psychosis might eventually lead to delaying, improving outcome, or even the prevention of a full-blown psychosis (Yung et al., 1996; Yung et al., 1998).

Chapter 1

11

Chapter 1

PREDICTION OF PSYCHOSIS, THE ULTRA HIGH RISK APPROACH The introduction of the ‘Ultra High-Risk’ (from now: UHR) criteria (Yung et al., 1996; Miller et al., 1999) in the second half of the 90’s has formed a milestone in the creation of operationally defined criteria to identify individuals at risk for psychosis (Simon & Velthorst et al., this thesis). The UHR phase, retrospectively called prodromal phase, recognizable by subtle behavioural changes and a decline in functioning, occurs prior to the development of the characteristic signs and symptoms of the illness that permit definitive diagnosis and lasts on average between 1 and 5 years (Phillips et al., 2002; Yung et al., 2003). UHR criteria are not intended to detect all prodromal pa-tients, but try to capture those at imminent risk of conversion to psychosis, i.e. the phase directly preceding onset of a first episode (Simon & Velthorst et al., this thesis). With the introduction of the Melbourne (Comprehensive Assessment of at Risk Mental States; CAARMS. Yung et al., 2002) and Yale (Structured Interview of Prodromal Syndromes; SIPS, McGlashan, 2002) scales, the psychometric operationalization of the prodromal phase was substantially facilitated (Simon & Velthorst et al., this thesis). In short, the UHR approach, as defined by the PACE clinic (Yung et al., 1996) consists of three alternative criteria: (i) Attenuated Positive Psychotic Symptoms (APPS) group: having experienced sub- threshold, attenuated positive psychotic symptoms during the past year; (ii) Brief Limited Intermittent Psychotic Symptoms (BLIPS) group, defined by hallucina-tions, delusions or formal thought disorders resolving spontaneously within 1 week; and (iii) Trait plus State Risk Factor Group: a first-degree relative with a psychotic disorder or a DSM–IV schizotypal personality disorder of the index person, combined with a significant decrease in functioning during the past year. In addition, several research groups -in particular from Germany- have added a fourth crite-rion to the risk-criteria: the so-called ‘basic symptoms’ (Ruhrmann et al., 2010). Klosterkötter et al. (2001) emphasized the presence of basic symptoms as an important risk-factor for the development of a first psychotic episode. Basic symptoms are self-reported disturbances in cognition and perception that were found to be predictive of a transition to psychosis over a 10 year follow up period. These symptoms include for example thought interference, thought pressure and thought blockages. With these sets of criteria, UHR patients have a chance between 10 and 40% of making the transition to psychosis (Yung et al., 2003, 2006; Cannon et al., 2008; Ruhrmann et al., 2010). Depending on the criteria and instruments used, the at risk cohort is variably termed the ‘Ultra High Risk’/ UHR- group (SIPS), ‘Clinical High Risk’/ CHR-group or AR-group (SIPS plus the basic symptom criterion), or the At Risk Mental State/ARMS- group (CAARMS).

12

Chapter 1

CURRENT STATE OF UHR-RESEARCH

The pioneering work of Yung, McGorry and colleagues over the past 15 years has substantially improved our knowledge about the UHR-phase, the possibilities of clinical care and pharmacotherapy (e.g. Yung et al., 1996; Yung et al., 1998; Yung et al., 2010; McGorry et al. 2002; Amminger et al.,2010). In addition, the enthusiastic use of UHR criteria has raised public awareness of the importance of early intervention and led to a significant increase in early psychosis services. As a result, the spectrum of symptoms with which putative at-risk subjects now present in these services has become broader, and transition rates have declined in recent studies (Yung et al., 2010; Simon & Velthorst et al., this thesis). The drop in transition rates to psychosis has raised concerns about the specificity of UHR criteria to correctly identify patients at risk for a transition to psychosis. With the current UHR-criteria, mainly based on positive symptoms, we may capture a high proportion of false-positives, potentially leading to unnecessarily stigmatization and treatment (Yung et al., 2010). Furthermore, the question has arisen whether the UHR subjects not converting to a psychosis continue to present with UHR symptoms and to be at risk for psychosis, or whether they eventually mature out of them or develop other disorders (Insel, 2010).

AIM AND OUTLINE OF THIS THESIS

In this light, the main objective of the studies described in this thesis was to increase our knowledge about the UHR phase. We have sought to identify factors that could eventually diminish the number of false-positives in UHR research. The current inclusion criteria are mainly based on positive symptoms, and the role of other symptoms has remained largely unexplored. Some have hypothesized that positive symptoms might only have limited discriminative power in schizophrenia research (Carpenter et al., 1999; van Os et al., 2010). In this thesis, the predictive role of certain negative symptoms and social disability has therefore been examined in more detail. In addition, the course of UHR symptoms of young people not converting to psychosis was further investigated.

The overall objective of this thesis has directed to the following three aims:

- To acquire a better understanding of the demographic characteristics and baseline symptomatology in the Ultra High Risk (UHR) phase- To identify factors that predict transition to psychosis and thereby eventually could diminish the number of false-positives (subjects without transition to psychosis) in UHR research. - To investigate the course of UHR subjects who eventually do not convert to a first psychotic episode

13

Chapter 1

AIM I: ACQUIRING A BETTER UNDERSTANDING OF THE DEMOGRAPHIC CHARACTERISTICS AND BASELINE SYMPTOMATOLOGY IN THE ULTRA HIGH RISK PHASE

Chapter 2, the first part of the thesis, focuses on this primary aim and presents a study on the ethnic differences in baseline UHR symptomatology, a matter still widely unexplored. Subsequently, in chapter 3, we investigated whether the negative symptoms reported in the UHR phase do in fact not represent depression symptoms. Is it indeed possible to differentiate between these two types of symptoms in the UHR phase?

AIM II: TO IDENTIFY VARIABLES THAT COULD POSSIBLY CONTRIBUTE TO A BETTER PSYCHOSIS PREDICTION

Part II of this theses starts off with a study about differences in baseline clinical symptoma-tology, general level of functioning (GAF-score) and genetic risk between UHR patients who did (UHR+T) or did not (UHR+NT) make a transition to psychosis. The results are presented in chapter 4. Following the findings of this study, in chapter 5 the role of social disability in the prediction of a first psychosis is examined. In chapter 6, the utility of the Strauss Carpenter Scale as predictive instrument in the UHR phase, a scale normally used in first-episode studies, will be investigated. Could social disability during admission for a first psychosis also contribute to the prediction of disease outcome? This subject will be discussed in chapter 7.

AIM III: TO INVESTIGATE THE COURSE OF UHR SUBJECTS EVENTUALLY NOT CONVERTING TO A FIRST PSYCHOSIS

Chapter 8 and 9 report on a prospective study and review to the course of UHR symptoms in patients who eventually do not convert to a first psychosis.

To conclude, in chapter 10, the findings of this thesis are summarized. Implications and suggestions for future directions will be made.

14

15

Chapter 1

References

American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders.

American Psychiatric Association, Washington DC.

Amminger, G. P., Schafer, M. R., Papageorgiou, K. et al. Long-chain -3 fatty acids for indicated prevention of psychotic

disorders: A randomized, placebo-controlled trial (2010). Archives of General Psychiatry; 67:146-154.

Becker, H. E., Nieman, D. H., Wiltink, S., Dingemans, P. M., van de Fliert, J. R., Velthorst, E., de Haan,

L., van Amelsvoort, T., Linszen, D. H. (2010). Neurocognitive functioning before and after the first psychotic episode:

does psychosis result in cognitive deterioration? Psychological Medicine; 40 (10): 1599-1606.

Bleuler, E. (1950). Dementia Praecox or the Group of Schizophrenias (International Universities Press).

Cannon, T. D., van Erp, T. G., Bearden, C. E., Loewy, R., Thompson, P., Toga, A. W., Huttunen,

M. O., Keshavan, M. S., Seidman, L. J., Tsuang, M. T. ( 2003). Early and late neurodevelopmental influences in the

prodrome to schizophrenia: contributions of genes, environment, and their interactions. Schizophrenia Bulletin;

29 (4): 653-669.

Carpenter, W., T. Jr., Arango, C., Buchanan, R. R. W., Kirkpatrick, B. (1999). Deficit psychopathology and a paradigm

shift in schizophrenia research. British Journal of Psychiatry; 180: 35-38.

Codey, D. (2010). “When one’s soul’s at a white heat”: Dickinson and the “Azarian School”.

The Emily Dickinson Journal; 19 (1): 30-59.

Cornblatt, B. A., Lencz, T., Smith, C. W., Correl, C. U., Auther, A. M., Nakayama, E. (2003).

The schizophrenia prodrome revisited: a neurodevelopmental perspective. Schizophrenia Bulletin; 29 (4): 633-651.

Dominguez de, M. G., Viechtbauer, W., Simons, C. J., van Os., J., Krabbendam., L. (2009).

Are psychotic pathology and neurocognition orthogonal? A systematic review of their associations. Psychological

Bulletin; 135 (1): 157-171.

Dragt, S., Nieman, D. H., van Duin, E. D. A., de Haan, L., Rietdijk, J., Ising, H. K., Klaassen, R. M. C, Wunderink, L.,

van der Gaag, M., Linszen, D. H. (submitted). COMT Val158Met genotype and cannabis use in people with an At Risk

Mental State for psychosis: exploring Gene x Environment interactions.

Haan de, L. & Bakker, J. M. (2004). Overview of theories of schizophrenia: from degeneration

to progressive developmental disorder. Psychopathology; 37: 1-7.

Häfner, H., Nowotny, B., Löffler, W. (1995). When and how does schizophrenia produce

social deficits? European Archives of Psychiatry Clinical Neuroscience; 246: 17–28.

Häfner, H., Riecher-Rössler, A., Hambrecht, M., Maurer, K., Meissner, S., Schmidtke, A.,

Fätkenheuer, B., Löffler, W., van der Heiden, W. (1992). IRAOS: an instrument for the assessment

of onset and early course of schizophrenia. Schizophrenia Research; 6 (3): 209-223.

16

Chapter 1

Harvey, P. D. (2009). When does cognitive decline occur in the period prior to the first episode

of schizophrenia? Psychiatry (Edgmont); 6(7): 12-14.

Henquet, C., Di Forti, M., Morrison, P., Kuepper, R., Murray, R. M. (2008). Gene-environment interplay

between cannabis and psychosis. Schizophrenia Bulletin; 34(6):1111-1121.

Insel, T. R. (2010). Rethinking schizophrenia. Nature; 468: 187-193.

Johnson, T. H. (1961). The complete poems of Emily Dickinson, edited by Thomas H. Johnsonpage 144,

Back Bay Books.

Kelly, B. D., O’Callaghan, E., Waddington, J. L., Feeney, L., Browne, S., Scully, P. J., Clarke, M., Quinn,

J. F., McTigue, O., Morgan, M. G., Kinsella, A., Larkin, C. (2010). Schizophrenia and the city: A review of literature

and prospective study of psychosis and urbanicity in Ireland. Schizophrenia Research; 116 (1):75-89.

Korver, N. & Quee, P. J., Boos, H. B. M., Simons, C., G.R.O.U.P (Submitted). Genetic Risk and Outcome

of Psychosis (GROUP), a multi site longitudinal cohort study focused on gene-environment interaction:

Objectives, Sample Characteristics, Recruitment and Assessment Methods.

Krabbendam, L., van Os, J. (2005). Schizophrenia and urbanicity: a major environmental

influence--conditional on genetic risk. Schizophrenia Bulletin; 31(4): 795-799.

Kraepelin, E. (1919). Dementia Praecox and Paraphrenia (Chicago Medical Book Co.)

Marwaha, S., Johnson, S., Bebbington, P., Stafford, M., Angermeyer, M. C., Brugha, T., Azorin, J. M., Kilian, R.,

Hansen, K., Toumi, M. (2007). Rates and correlates of employment in people with schizophrenia in the UK,

France and Germany. British Journal of Psychiatry;191:30-7.McGorry, P. D., Yung, A. R., Phillips, L. J., Yuen, H. P.,

Francey, S., Cosgrave, E. M., Germano, D., Bravin, J., McDonald, T., Blair, A., Adlard, S., Jackson, H. (2002).

Randomized Controlled Trial of interventions designed to reduce the risk of progression to first-episode psychosis

in a clinical sample with subthreshold symptoms. Archives of General Psychiatry; 59: 921-928.

McGrath, J., Saha, S., Welham, J., El Saadi, O., MacCauley, C., Chant, D. (2004). A systematic review of the incidence

of schizophrenia: the distribution of rates and the influence of sex, urbanicity, migrant status and methodology.

BMC Medicine; 28: 2-13.

McGurk, S. R., Mueser, K. T., DeRosa, T. J., Wolfe, R. (2009). Work, recovery, and comorbidity in schizophrenia:

a randomized controlled trial of cognitive remediation. Schizophrenia Bulletin; 35 (2): 319-335.

Meyer-Lindenberg, A. (2010). From maps to mechanisms through neuroimaging of schizophrenia.

Nature; 468: 194-202.

17

Chapter 1

Mueser, K. T. & McGurk, S. R. (2004). Schizophrenia. The Lancet; 363; 2063-2072.

Os van, J. (2009). A salience dysregulation syndrome. The British Journal of Psychiatry: 194; 101-103.

Os van, J. (2010). Are psychiatric diagnoses of psychosis scientific and useful?

The case of schizophrenia. Journal of Mental Healh; 19 (4) 305-317.

Os van, J. & Kapur, S. (2009). Schizophrenia. Lancet: 374; 634-45.

Os van, J., Kenis, G., Rutten, B. F. (2010) The environment and schizophrenia.

Nature: 468; 203-212.

Selten, J. P., Cantor-Graae, E., Kahn, R. S. (2007). Migration and schizophrenia.

Current Opinion in Psychiatry; 20(2):111-115.

Shapiro, D. I., Marenco, S., Spoor, E. H., Egan, M. F., Weinberger, D. R., Gold, J. M. (2009).

The Premorbid Adjustment Scale as a measure of developmental compromise in patients with schizophrenia

and their healthy siblings. Schizophrenia Research; 112(1-3):136-142.

Simon, A. E. & Velthorst, E., Nieman, D. H., Linszen, D. H., Umbricht, D., de Haan, L. (Submitted).

Remission from an initial ultra high-risk state for psychosis: a systematic review.

Üstün, T. B., Rehm, J., Chatterji, S., Saxena, S., Trotter, R., Room, R., Bickenbach, J. and the WHO/NIH Joint

Project CAR Study Group (1999). Multiple-informant ranking of the disabling effects of different health conditions

in 14 countries. The Lancet; 354: 111-115.

Yung, A. R. & McGorry, P. D. (1996). The prodromal phase of first-episode psychosis:

past and current concepualization. Schizophrenia Bulletin; 22: 353-370.

Yung, A. R., Phillips, L. J., McGorry, P. D., McFarlane, C. A., Francey, S., Harrigan, S., Patton, G. C., Jackson,

H. J. (1998). Prediction of psychosis. A step towards indicated prevention of schizophrenia. British Journal

of Psychiatry; 192: 14-20.

Yung, A. R, Nelson, B., Thompson, A. D., Wood, S. J. (2010).

Should a “Risk Syndrome for Psychosis” be included in the DSMV? Schizophrenia Research; 120(1-3):7-15.

Weinberg, D., Shahar, G., Davidson, L., McGlashan, T. H., Fennig, S. (2009). Longitudinal associations between

negative symptoms and social functioning in schizophrenia: the moderating role of employment status and setting.

Psychiatry; 72(4):370-381.

18

19

01P A R T

20

21

Chapter 2

ETHNICITY AND BASELINE SYMPTOMATOLOGY IN PATIENTS WITH AN “AT RISK MENTAL STATE” FOR PSYCHOSIS.

Eva Velthorst, Dorien Nieman, Wim Veling, Rianne Klaassen, Sara Dragt, Judith Rietdijk, Helga Ising, Lex Wunderink, Don Linszen, Lieuwe de Haan, Mark van der Gaag

PSYCHOLOGICAL MEDICINE, IN PRESS

22

Chapter 2

ABSTRACT

Background Ethnicity has been associated with different incidence rates and different symptom profiles in young patients with psychotic-like disorders. No studies so far have examined the effect of ethnicity on symptoms in people with an At Risk Mental State.

Method In this cross-sectional study, we analyzed the relationship between ethnicity and baseline data on the severity of psychopathology scores in 201 help-seeking patients who met the At Risk Mental State criteria (ARMS; Yung et al., 1996, 2003) and agreed to participate in the Dutch Early Detection and Intervention (EDIE-NL) trial. 87 of these patients had a non-Dutch ethnicity. We explored the possible mediating role of ethnic identity.

Results Higher rates of negative symptoms – anhedonia in particular- were found in the ethnic minority group. This result could be mainly attributed to the Moroccan-Dutch and Turkish-Dutch subgroups, who also presented with more depression symptoms when the groups were examined separately. The ethnic minority group displayed a lower level of ethnic group identity as compared to the immigrants of the International Comparative Study of Ethnocultural Youth. Ethnic identity was inversely related to symptoms in the Moroccan-Dutch patient group.

Conclusions The prevalence of more severe negative symptoms and depression symptoms in ethnic minority groups deserves more attention, as the experience of attenuated positive symptoms when accompanied by negative symptoms or distress has proven to be predictive for transition to a first psychotic episode.

23

1. Introduction

Psychotic illness is more prevalent in migrant and minority ethnic groups (Veling et al. 2007; Kirkbride et al. 2008; Morgan et al. 2010). The association between being a member of an ethnic minority group and incidence of psychotic disorders is typically linked to a greater exposure to adverse social experiences among minorities, such as racial discrimination or social isolation during the course of life (Selten & Cantor-Graae, 2005; Veling et al, 2007; Morgan & Hutchinson, 2010). Whereas higher incidences of psychotic disorders in ethnic minority groups are consistently reported, the literature about ethnic variation in clinical characteristics is less unequivocal. Several studies suggest that ethnic differences are not limited to differences in incidence rates, but that variations could also be observed in the symptom profile. Higher rates of hallucinations, paranoid delusions and affective symptoms have been found in psychotic patients of certain ethnic minority groups as compared to the native patient group (Strakowsky et al. 1996; Bhugra et al., 2000; Barrio et al. 2003; Arnold et al. 2004; Veling et al. 2007). Two studies from the UK however, have reported greater similarities than differences in terms of symptom profile (Harvey et al. 1990, Hutchinson et al. 1999). In fact, no major ethnic differences were found in any of the schizophrenia core features. In the Netherlands, one study of a first-psychosis cohort has examined the differences between ethnic minorities and native Dutch in symptom profile in a group of patients with a first-mental health contact for a psychotic disorder (Veling et al. 2007), and found the Moroccan-Dutch patient group to report significantly higher total psychopathology-, positive- and negative symptom scores compared to the Dutch patient population. In particular, patients originating from Moroccan presented more often with persecutory delusions, bizarre behaviour and visual hallucinations. In addition, Moroccan-Dutch and Turkish-Dutch patients both had higher levels of depression. In other ethnic minority groups, no significant differences were found. To the best of our knowledge, the relationship of ethnicity with symptoms in adolescents and young adults with an At Risk Mental State (ARMS; Yung et al. 2005) for developing psychosis has not been investigated yet. Since variation in psychotic phenomenology might reflect different etiological pathways to psychosis in migrant and minority ethnic groups (Morgan et al. 2010), differences in ARMS symptomatology in young people might implicate different treatment targets for psychosis prevention in future treatment models. In the present cross-sectional study, we aimed to investigate whether ethnic differences found in patients with a psychotic disorder (Veling et al., 2007) are also present in patients with an ARMS for developing a first psychosis. We hypothesized (1) that no significant results would be found when examining symptomatic differences between Dutch patients and the whole ethnic minority group; (2) that the Moroccan-Dutch group would display a. more severe total psychopathology, b. more negative symptoms and c. more paranoid ideas than the native Dutch patient group; (3) that the Moroccan- Dutch and Turkish-Dutch would report more depression symptoms. This study also examined possible factors that could contribute to symptomatic differences in a more explorative way. In a large matched case-control study of first-episode schizophrenia

Chapter 2

24

Chapter 2

among non-Western ethnic minorities in the Netherlands (Veling et al. 2010), a higher group identity has been found to protect against the onset of psychosis (see for an elaboration on the concept of ethnic identity: Berry et al. 2006; Veling et al. 2010). In the present study, we therefore sought to investigate whether ethnic group identity could also affect the manifestation of symptoms in subjects with an ARMS.

2. Methods2.1. Classification of Ethnicity

In accordance with the study by Veling et al. (2010) we used the classification of ethnicity as defined by the Dutch Bureau of Statistics: If a citizen, or (one of) his/her parents was born abroad, he or she is assigned to this foreign ethnicity group. If the parents were born in different foreign countries, the country where the mother was born determined the assignment to a particular group. We first divided our patient cohort in a ‘Dutch’ group and a group of all adolescents with another ethnicity: the ‘Ethnic Minority’ group. For explorative purposes, we then divided our study population into six categories: native-Dutch, Moroccan-Dutch, Turkish-Dutch, Surinamese-Dutch, other Western –, and other Non-Western minority groups.

2.2. Recruitment

Between February 2008 and March 2010, baseline data were collected from 201 help-seeking patients who met the At Risk Mental State criteria (ARMS; Yung et al. 1996, 2003) and agreed to participate in the Early Detection and Intervention (EDIE-NL) trial (Rietdijk et al. 2010). EDIE-NL is a longitudinal multi-centre randomized controlled trial comparing treatment as usual (TAU) with an add-on cognitive behavioural therapy (CBT), aiming at the prevention of a first psychosis. For a comprehensive description of the EDIE-NL treatment model, see Rietdijk et al. (2010). Young people were eligible for the study if aged between 14 and 35 years and if meeting the criteria for an at- risk mental state (ARMS) for the development of a first psychotic episode, as defined by the Personal Assessment and Crisis Evaluation Clinic (PACE) criteria (Yung et al. 1998, 2003). The ARMS approach consists of three alternative criteria: 1) A schizotypal personality disorder or a first degree relative with psychosis, also the ‘Vulnerability group’; 2) ‘Attenuated’ positive symptoms, such as ideas of reference, odd beliefs, magical thinking or unusual perceptual experiences; and 3) A brief psychotic episode of less than one week duration that resolves without antipsychotic medication (Brief Limited Intermittent Psychotic Symptoms, BLIPS).

25

Chapter 2

In addition in all three inclusion groups, there had to be impairment in social functioning as assessed with the Social and Occupational Functional Assessment Scale (SOFAS, Goldman et al. 1992), i.e. a SOFAS- score of < 50 in the past 12 months or longer and/or a drop in SOFAS score of 30% for at least a month in the past year. Exclusion criteria were: 1) current or previous usage of antipsychotic medication defined as a total cumulative dosage above 15 milligram haloperidol equivalents; 2) severe learning impairment (IQ < 70); 3) psychiatric symptoms due to somatic aetiology; 4) insufficient competence in the Dutch language and 5) history of psychosis. The study design was approved by the Dutch Union of Medical- Ethics Trial Committees for mental health organizations and the Medical Ethical Committees of all participating centres. The trial is registered at the International Standard Randomised Controlled Trial registered with number as ISRCTN21353122. Written consent from participants and parents or guardians (if the participant was below the age of 16 years) was obtained after the procedure had been fully explained.

2.3. Instruments

ARMS SYMPTOMATOLOGY The Comprehensive Assessment of At-Risk Mental States (CAARMS; Yung et al., 2005), including SOFAS, was employed to determine the presence, severity, frequency and type of ARMS symptoms. This instrument consists of a semi-structured interview de-signed to assess the ARMS criteria and has proven excellent validity and reliability (Yung et al. 2005). The CAARMS consists of seven subscales that include: four Positive Symptom items, two Cognitive- and three Emotional Disturbances items, three Negative Symptoms items, four Behavioural Change items, four Motor/Physical Changes items and eight General Psychopathology items. Symptomatic criteria for at risk mental state are exclusively based on positive symptom items. EDIE-NL investigators received an extensive two-days training by Prof. Alison Yung, who developed the CAARMS criteria. Reliability checks of the Dutch version of the CAARMS were performed approximately every 3 months during the study. The preliminary pair wise interrater concordance of the intensity- subscales of the CAARMS was .81 and considered acceptable by the training team. The analyses of the present study were based upon the sum score of the positive- and negative items of the CAARMS. In addition, the total sum score was used, representing a global psychopathology rating. The SOFAS (Goldman et al. 1992) was used to determine the level of social and occupational functioning. This scale, ranging from 0 to 100, is a modified version of the Global Assessment of Functioning Scale, separating the measures of social and occupational functioning from the measures of symptoms and psychological functioning.

26

Chapter 2

COMORBIDITY Comorbid diagnoses were examined by means of the Schedule for Clinical Assessment in Neuropsychiatry (SCAN; Wing et al. 1990). In addition, depression was self-reported on the Beck Depression Inventory II (Beck et al. 1996) and assessed by means of the Calgary Depression Scale (CDS); a 9-item structured interview scale, designed to assess the severity of depression in people with psychotic disorders (Addington et al. 1992). ETHNIC AND NATIONAL IDENTITY Ethnic identity and national identity was assessed with the ordinal International Comparative Study of Ethnocultural Youth (ICSEY) scale. This is a 10-item version of the Multigroup Ethnic Identity Measure (Phinney, 1992) with response options ranging from ‘‘strongly disagree’’ (1) to ‘‘strongly agree’’ (5), assessing ethnic and national affirmation, sense of belonging and feelings about being a member of a group. For example: ‘‘being part of ethnic culture is embarrassing to me’’. National identity on the other hand, is the identity as a member of the larger society (in this study: Dutch identity). We compared our mean scores per item with the mean scores of a group of young immigrants (13-18 years old, mean age 15.4) in the Dutch population of the ICSEY – study, a study among more than 10.000 adolescents from 13 countries, which included Surinamese, Turkish and Antillean immigrants in the Netherlands (Berry et al. 2006).

2.4. Statistical analysis

Statistical analysis was performed with SPSS (version 18.0) for Windows. Comparisons of baseline characteristics between the ethnic minority group and the Dutch group were made with Pearson’s chi square tests and independent sample t-tests. We used Cohen’s d-tests to examine differences in the level of ethnic and national identity between our ethnic minority population and the immigrants of the ICSEY study. Partial correlation was used to explore the relationship between ethnic and national identity and the different symptom clusters. To assess baseline differences in symptomatology between the ethnic minority and native Dutch population while controlling for possible confounders, a one-way between-group analysis of covariance was conducted. The independent variable was the ethnic background, and the dependent variables consisted of ARMS symptomatology scores. Age, gender, employment status, educational level, cannabis use over the past 12 months, and the use of benzodiazepines and anti-depressants were used as covariates in the analyses. Preliminary checks were conducted to ensure that there was no violation of the assumptions of normality. Potential confounders were initially tested one by one in univariate analyses for their effect on the outcome variable. Exploratively, we examined the differences in symptomatology compared to the Dutch group by means of univariate analyses in the five different ethnic subgroups. Missing data (≤ 5%) were imputed using the Expectancy Maximization procedure of SPSS for windows. Post hoc analyses were conducted with Tukey HSD tests. A p-value of ≤.05 was considered statistically significant.

27

Chapter 2

3. Results3.1 Subject Characteristics

Table 1 lists the characteristics of our study sample. Over the two years of inclusion, 201 subjects met the inclusion criteria and agreed to participate in the EDIE-trial (99 males, overall mean age=22.7 years ±SD 5.5). 87 participants had a foreign ethnicity (43%). Of these participants, 30 (34.5%) were first generation and 57 (65.5%) were second generation migrants. 192 (90%) patients displayed Attenuated symptoms, of whom 28 also met the ‘Vulnerability’ criterion and 1 reported BLIPS-symptoms. 6 (3.0%) patients only met the ‘Vulnerability’ criterion and another 3 (1.5%) only belonged to the BLIPS group. The ethnic minority group did not differ significantly from the Dutch patient group in terms of the inclusion group they were assigned to, gender, cannabis use during the 12 months prior to baseline, employment status, positive symptoms and SOFAS score. The percentage of ethnic minority patients in each participating centre corresponded to the ethnic variation in the local population. Compared to the ethnic minority group (24.5; s.d.= 5.5), the Dutch group was significantly younger at baseline (21.4; s.d. = 5.2). In addition, we found the ethnic minority group be more often diagnosed with an anxiety disorder (X2=5.6, p=.02) and eating disorder (X2=5.5, p=.02).

28

Chapter 2

Age (mean, ±SD)

Male, sex (n, %)

First degree relative with psychosis (n, %)

Paid full-or part time job/ education (n, %)

First generation (n, %)

Cannabis use (last 12 month) (n, %)2

Medication use (n, %)

Antidepressants

Benzodiazepines

Level of education (n, %)

No education/ primary

Secondary school

Higher education

Site (n, %)

The Hague (Parnassia/VU)

Amsterdam (AMC)

Friesland (GGZ)

Leiden

Amsterdam (PsyQ)

Utrecht

DSM-IV diagnoses (n, %)

Mood disorder

Anxiety disorder

Sleeping disorder

OCD

Eating disorder

Bipolar disorder

Other

Ethnic identity

National Identity

N

201

201

201

201

87

201

201

201

201

201

197

87

87

Ethnic minority group1

N=87

24.5 (5.5)

39 (44.8)

20 (23.0)

60 (69.0)

30 (34.5)

27 (31.0)

25 (28.7)

14 (16.1)

0

31 (35.6)

56 (64.4)

52 (55.9)

19 (47.5)

4 (13.3)

7 (35.0)

5 (45.5)

0

45 (53.6)

53 (63.1)

21 (25.0)

3 (3.6)

4 (4.8)

2 (2.3)

4 (4.8)

3.81 (1.0)

3.90 (1.1)

Dutch group

N=114

21.4 (5.2)

60 (52.6)

15 (13.2)

85 (74.6)

-

43 (37.7)

31 (27.2)

12 (10.5)

0

53 (46.5)

61 (53.5)

41 (44.1)

21 (52.5)

26 (86.7)

18 (65.0)

6 (55.5)

2 (100)

52 (46.0)

52 (46.0)

19 (16.8)

5 (4.4)

0

1 (.8)

8 (7.1)

Statistics

t=-4.0, p<.0001

x2=1.2, p=.27

x2=3.3, p=.07

x2=.78, p=.38

x2=.97, p=.32

x2=.06, p=.81

x2=1.4, p=.24

x2=2.4, p=.12

x2=21.2, p=.001

x2=1.1, p=.29

x2=5.6, p=.02

x2=2.0, p=.16

x2=.09, p=.76

x2=5.5, p=.02

x2=.72, p=.40

x2=.45, p=.50

Table 1 Characteristics of the study sample*

* Significant at the .05 level

1. Ethnic minority group = Moroccan (n=22), Turkish (n=13), Surinamese (n=16), other Western (n=14), Other non- Western (n=22)

2. Cannabis use is defined as having used cannabis > 5 times during lifetime & at least 1 time in the past 12 months (CIDI; WHO, 1993)

3. Other include: Multiple substance abuse, pain disorder, conversion disorder

** Ethnic and national identity was assessed with the ordinal ICSEY Scale of Ethnic and National Identity (Phinney, 1992).

29

Chapter 2

3.2 Baseline symptomatology

Univariate analyses showed highly significant associations between being a member of an ethnic minority group and total psychopathology scores (p=.001) and negative symptoms (p<.0001; see table 2). The difference in negative symptoms was mainly due to higher rates of anhedonia (p=.003) within the ethnic minority group compared to the Dutch group. In addition, the ethnic minority group reported more depression- (p=.002) and more social anxiety symptoms (p=.01). The higher depression rates were not only self-reported, but also confirmed by clinical assessment (CDS; p=.02). In the model, controlled for age, gender, employment status, educational level, cannabis use over the past 12 months, and the use of benzodiazepines and anti-depressants, the significant associations remained for the negative symptoms sum score (p=.01) and subscale anhedonia (p=.007) only. Controlling for BDI-II sum score (negative symptoms sum score F(1,186)=5.83, p=.02; Anhedonia F(1,186)=5.83, p=.02) , these differences were still significant.

Table 2 Differences in ARMS symptomatology between young people with a Dutch and ethnic minority background

CAARMS total psychopathology

Positive symptoms

Negative symptoms

- Alogia

- Avolition

- Anhedonia

SOFAS

BDI-II sum score

SIAS sum score

CDS sum score

Ethnic minority1

(Mean,± SD)

N=87

56.5 (13.9)

10.2 (2.7)

8.1 (3.1)

1.4 (1.2)

3.4 (1.3)

3.4 (1.5)

45.5 (5.2)

45.7 (13.1)

54.3 (16.7)

7.1 (4.9)

Dutch

(Mean, ±SD)

N=114

49.4 (16.6)

10.4 (2.8)

6.4 (3.6)

1.1 (1.2)

2.7 (1.6)

2.5 (1.8)

46.5 (4.8)

40.4 (11.3)

48.1 (16.3)

5.4 (4.5)

Unadjusted

t, P

T=3.25, p=.001

T=-.55, p=.58

T=3.76, p<.0001

T=1.34, p=.18

T=3.06, p=.003

T=3.80, p<.0001

T=-1.45, p=.15

T=3.08, p=.002

T=2.56, p=.01

T=2.43, p=.02

Adjusted 2

F, P

F(1,191)=2.42, p=.12

F(1,192)=1.65, p=.20

F(1,191)=6.36, p=.01

F(1,191)=.75, p=.39

F(1,191)=3.28, p=.07

F(1,191)=7.41, p=.007

F(1,192)=3.56, p=.06

F(1,190)=2.80, p=.10

F(1,190)=1.52, p=.22

F(1,186)=1.35, p=.25

Ranges: CAARMS Total psychopathology ( 0- 168); CAARMS Positive symptoms (0-24); CAARMS negative symptoms (0-18); CAARMS alogia (0-6);

CAARMS Avolition (0-6); CAARMS Anhedonia (0-6); SOFAS (0-100); BDI-II Sumscore (21-84); SIAS Sumscore (0-76); CDS sumscore (0-27).

1 Ethnic minority = Moroccan (n=22), Turkish (n=13), Surinamese (n=16), other Western (n=14), Other non- Western (n=22)

2 Adjusted for potential confounders: age, gender, medicine use, cannabis use, employment status and level of education.

Significant variables are shown in bold

30

3.3 Explorative analyses

3.3.1. Role of possible confounding factors

By examining the effect on the outcome variables of the potential covariates in the model (i.e. age, gender, employment status, educational level, cannabis use over the past 12 months, and the use of benzodiazepines and anti-depressants) we found the older age of the other ethnicity group to have the largest impact on baseline symptomatology; all other factors entered in the model did not significantly change any of the symptom scores. After controlling for age, only the ethnic differ-ences in negative symptoms remained significant (F(1,192)=6.39, p=.01). The differences in total psychopathology (F=2.60, p=.11), BDI-II (F=2.70, p=.10), and CDS (F=1.65, p=.20) lost their significance. No main effect of gender or interaction effects between gender and ethnicity were found.

3.3.2. Symptomatology in different ethnic minority groups

We examined the symptomatic differences after dividing our ethnic minority group into five smaller subgroups (see table 3). We observed overall statistically significant differences in the negative symp-toms total score (F(5,185)=2.24,p=.05), the negative symptom cluster ‘anhedonia’ (F(5,185)=2.48,p=.03) and in self-reported symptoms of depression (F(5,184)=2.47, p=.03). Post-hoc comparisons using the Tukey HSD test indicated that the differences in negative symptom scores can mainly be attributed to the higher scores in the Moroccan-Dutch as compared to the Dutch population (Negative symptoms; total score - mean difference =2.78, SE= .72, p=.002, Anhedonia; mean difference=1.58, SE=.37, p=.001 and Avolition; mean difference=1.04, SE=.30, p=.009). The significance levels did not change after controlling for depression, as assessed with the BDI-II total score. The total score of the negative symptom cluster was also increased in the other Non-Western minority group (mean difference=2.14, SE=.72, p=.04). The overall significant differ-ence in self-reported depression scores was caused by the higher scores of both the Moroccan-Dutch and Turkish-Dutch population (BDI-II; mean difference = 8.50, SE=2.33, p=.005 & mean difference = 12.74, SE= 2.93, p<.0001, respectively). Although no other overall differences were found, post hoc analyses concerning all 5 ethnic groups independently showed higher levels of total psychopa-thology scores within the Turkish-Dutch population (mean difference=12.03, SE=3.98, p=.03). In ad-dition, the Moroccan-Dutch patient group showed higher scores on the CDS (mean difference = 3.17, SE=.99, p=.02).

Chapter 2

31

Chapter 2

Table 3 Severity of psychopathology in patients from different ethnic groups making first contact for ARMS- symptoms

3.3.3. Ethnic and national identity

The mean (SD) score of ethnic and national identity was compared with multi-ethnic cohort of 349 young people from the Netherlands (ICSEY study, Berry et al., 2006) showing an ethnic identity score of 4.54 (.06) and national identity score of 3.19 (.04). In comparison to the youth of the ICSEY study, our ethnic minority group reported scores that resembled a lower ethnic identity and a higher na-tional identity (Cohen’s d -1.0 and .9 respectively; see table 1). The level of ethnic identity was not associated to any of the symptom clusters when looking at the ethnic minority groups taken together. Explorative analyses of the largest subgroups (i.e. the Moroccan-Dutch and the other Non-Western minority group) suggested relations between a higher ethnic identity in the Moroccan-Dutch group and less severe total psychopathology scores (r=-.69, df=13, p=.005), a lower score on the negative symptoms sum score (r=-.63, df=13, p=.01), a lower score on anhedonia (r=-.64, df=13, p=.01) and alogia (r=-.53, df=13, p=.04). No correlations were found in the other Non-Western minority group.

N=201

Total psychopathol.

Positive symptoms

Negative symptoms

Alogia

Avolition

Anhedonia

SOFAS

BDI-II

SIAS

CDS

NativeDutch

n=114

Mean (SD)

49.4 (16.6)

10.4 (2.8)

6.4 (3.6)

1.1 (1.2)

2.7 (1.6)

2.5 (1.8)

46.5 (4.8)

40.4 (11.3)

48.1 (16.3)

5.4 (4.5)

Moroccan- Dutch

n=22

Mean (SD)

58.4 (14.7)

9.4 (2.5)

9.1 (3.1)

1.3 (1.4)

3.8 (1.1)

4.1 (1.3)

45.3 (4.9)

48.9 (11.0)

56.5 (17.1)

8.6 (4.7)

Turkish- Dutch

n=13

Mean (SD)

61.5 (15.2)

11.2 (2.9)

8.7 (2.4)

1.4 (1.3)

3.5 (1.1)

3.9 (1.4)

47.4 (3.5)

53.2 (12.5)

61.1 (17.6)

8.1 (4.8)

Surinamese-Dutch

n=16

Mean (SD)

55.5 (12.5)

10.0 (2.6)

7.7 (3.5)

1.8 (1.3)

3.0 (1.4)

2.9 (1.5)

45.6 (4.8)

38.7 (13.6)

48.9 (19.0)

6.5 (5.3)

Other Western

n=14

Mean (SD)

46.6 (11.8)

9.8 (3.3)

5.9 (2.9)

.6 (.9)

2.6 (1.4)

2.7 (1.7)

44.3 (5.3)

40.1 (13.5)

48.4 (14.1)

5.6 (5.6)

Other Non-western

n=22

Mean (SD)

58.5 (12,5)

10.6 (2.4)

8.5 (2.6)

1.6 (1.1)

3.6 (1.3)

3.3 (1.6)

45.1 (6.4)

46.6 (12.0)

55.5 (14.5)

6.3 (4.4)

Adjusted*

F, df, P

F(5,185)=.91,p=.48

F(5,186)=.89, p=.49

F(5,185)=2.24,p=.05

F(5,185)=1.91, p=.09

F(5,185)=1.36, p=.24

F(5,185)=2.48,p=.03

F(5,185)=.98, p=.43

F(5,184)=2.47, p=.03

F(5,184)=.70, p=.62

F(5.180)=.66, p=.66

* Unadjusted means are shown. Statistical analyses were corrected for age, gender, cannabis use, use of medication, employment status and

education level. Significant outcomes are shown in bold. Outcomes significant with Post Hoc analyses are shown in Italic.

32

4. Discussion4.1 Main findings

To our knowledge, this is the first study examining ethnic differences in baseline symptomatology in a large cohort of patients with an ARMS for developing psychosis. In accordance with a recent Dutch first psychosis-study (Veling et al. 2007), we hypothesized that no significant differences in symptomatology would be found between the native Dutch and the whole ethnic minority group. This hypothesis was partially supported. Although no baseline differences were found in total psychopathology scores, positive symptoms and depression symptoms, we did find higher negative symptom scores in the ethnic minority group after controlling for possible confounders. First- and second generation immigrants from Morocco largely accounted for the difference in negative symptoms between the ethnic minority group and the native Dutch group, particularly because of their higher rates of avolition and anhedonia. The Moroccan- Dutch and Turkish-Dutch ARMS subjects were found to report more depression symptoms as compared to the native Dutch. Although higher levels of negative symptoms and depression were found in non-Western groups within the Netherlands, we were not able to replicate the findings regarding higher levels of symptomatology and persecutory ideas. An explanation might be the heterogeneity of our sample. Not only are the paranoid ideations of our whole sample by definition less severe than those in a first psychosis group, a large part of the ARMS group will most likely eventually recover from their symp-toms (Simon & Umbricht, 2010; Velthorst et al. 2011) or will go on to develop another disorder.

4.2. Ethnic identity

One of the mechanisms that is considered to be a possible contributor to symptomatic differences is the concept of ethnic identity. In our study, we therefore compared the level of ethnic group identity with a large group of immigrants in the Netherlands and found our group displaying a weaker ethnic group identity, which is in congruence with the idea that weak group identity can cause distress and symptoms that possibly appear by (vulnerability for) social adverse experiences. Higher levels of distress might reduce someone’s interest to participate in daily activities and may cause one to withdraw from social contact. Albeit very preliminary, the association between lower negative symp-tom levels and higher levels of group identity within the Moroccan-Dutch group supports this idea, and is in accordance with the theory previously suggested by Veling et al. (2010). At this moment however, the findings are still equivocal with respect to the direction of the possible effect of ethnic group identity and psychotic symptomatology. Interestingly, both a weak (Veling et al. 2010) as well as an increased identification with the own ethnic group (Reininghaus et al. 2010) have been suggested to increase the negative effect of ethnicity on psychosis. A possible explanation for this contradiction may be a varying relationship between strong ethnic identity and access to support networks (Reininghaus et al. 2010). While a strong ethnic identity may be associated with easy access to social support networks in one group/country (e.g. the Netherlands), strong ethnic identity may instead represent compound risk in other groups or countries.

Chapter 2

33

Chapter 2

The mechanism behind the complex association between ethnic identity, distress and symptoms warrants future research, as the found association could have an explanation in two directions: whereas Reininghaus et al. (2010) and Veling et al. (2010) both underline the influence ethnic identity may have on distress, it is not inconceivable that distress or experiences of defeat could alternatively lead to weak group identity.

4.3 Cultural background

Notwithstanding the abovementioned ethnic identity hypothesis, it is plausible that the manifestation of symptoms may also -at least partly- be influenced by someone’s cultural background (Veling et al. 2007; Zandi et al. 2010). While our results show that suffering from depression symptoms seems to be a common phenomenon within the Moroccan population, it has been argued that giving in to such feelings is a taboo in this culture (Zandi et al. 2010), and this might even be more applicable in case of psychotic-like experiences. Feelings of shame may even further reinforce (social) indifference, leading to high scores on the subscale ‘anhedonia’ on the CAARMS for example. On the other hand, shame and stigma in families might also interfere with seeking help early (Rathod et al. 2010), preventing one to seek help before distress and complaints are alarmingly worsening. This hypothesis would account for the older age and higher psychopathology scores among the ethnic minority group.

4.4 Methodological considerations and limitations

A limitation of this study is the fact that it is still unknown whether the migrants in this sample of ARMS patients do represent those who develop a psychotic disorder. Indeed, the sex ratio in our ARMS sample (39% male) differs widely from the sex ratio in the Veling study (72 %). However, our aim was to detect patients susceptible for psychosis in any spectrum. Females are prone for psychosis to a similar extent as male, but rather for psychosis as part of a mood disorder as opposed to a schizophrenia spectrum disorder. Future transition data shall reveal whether a higher percentage of males as opposed to females will eventually convert to a psychosis in the schizophrenia spectrum. The primary aim of the present study was to examine whether symptomatic differences could be found in ethnic minority subjects with a high risk of developing a psychotic disorder compared to Dutch subjects. Growing up as or being a member of an ethnic minority group may cause feelings of social defeat and/ or altered ethnic identity that in turn can play a role in the development of certain symptoms. It is for this reason, as well as the small sample sizes of the separate ethnic groups, that we first investigated the total ethnic minority group. However, some groups may be more vulnerable to certain experiences and therefore we also reported on the results of the different ethnicities separately. The results concerning specific minority groups should be interpreted with caution given the small sample size of certain minority groups.

34

Chapter 2

We recognize that the fact that reviewers were not blind for ethnicity may have accounted for some of the reported differences within our study cohort. At the same time, we do not believe this has affected our results substantially. To avoid cross-cultural bias, ethnic and cultural background was taken into account when measuring symptoms (Zandi et al. 2010). We asked about the self-experienced change instead of merely asking about the current complaints. In addition, within most centers we also consulted parents about the unusual changes they observed in their child over the last year and asked whether they believed the complaints might be culture- specific. Finally, the culture-bias seems to involve hallucinations and dissociative symptoms in particular (Zandi et al. 2010; Blom et al.2010); symptoms in which we did not find any differences. It may be possible that a different manner of presenting complaints could have biased our results. Social anxiety and depression symptoms both were self-reported in our study. Emphasizing as opposed to minimizing complaints may just be a different way of asking for help. This latter argument is less likely to hold for the Moroccan-Dutch subgroup, where depression symptoms were not only self-reported, but also proved to be significantly more severe compared to the native Dutch group when assessed by means of a clinical assessment. The exclusion of some young people with insufficient competence of the Dutch language may have also biased the data of our ethnic minority sample. Finally, the ICSEY study to which we compared our baseline ethnic identity measures did not include a Moroccan group, which in fact was the largest ethnic group of our ARMS cohort. In addi-tion, the ICSEY cohort was somewhat younger. To date, no known study has been conducted to ethnic identity among Moroccan young immigrants, and merits attention in the future.

4.5 Conclusions

Irrespective of the underlying mechanisms, the prevalence of more negative symptoms and depression symptoms in certain ethnic minority groups deserves attention, because attenuated positive symptoms when accompanied by negative symptoms or distress have proven to be predictive for a first psychotic episode (e.g. Cannon et al. 2008; Velthorst et al. 2009; Yung et al. 2006). Feelings of shame and stigma as well as ethnic identity may be important targets in future UHR studies. In addition, athough the correlation between reduced negative symptoms andincreased group identity in the Moroccan-Dutch group are tentative, they are in line with earlier finding in the Netherlands (Veling et al. 2010) and definitely need verification in larger samples. After completing the follow-up of our trial we will be able to evaluate whether the young members of the ethnic minority group with the weakest ethnic group identity and the highest baseline negative symptom scores will eventually be the ones who go on to develop a first psychotic episode.

35

Chapter 2

Acknowledgements

We thank all EDIE-NL researchers and all individuals who participated in the study and were essential for its successful completion.

36

Chapter 2

References

Addington, D., Addington, N., Maticka-Tyndale, E. & Joyce, J. (1992). Reliability and validity of a depression rating

scale for schizophrenics. Schizophrenia Research; 6: 201-208.

Arnold, L. M., Keck, P. E. Jr., Collins, J., Wilson, R., Fleck, D. E., Corey, K. B., Amicone, J., Adebimpe, V. R. &

Strakowski, S. M. (2004). Ethnicity and first-rank symptoms in patients with psychosis. Schizophrenia Research;

67: 207-212.

Barrio, C., Yamada, A. M., Atuel, H., Hough, R. L., Yee, S., Berthot, B. & Russo, P. A. (2003). A thri-ethnic

examination of symptom expression on the positive and negative syndrome scale in schizophrenia spectrum

disorders. Schizophrenia Research; 60: 259-269.

Beck, A. T., Steer, R. A., Ball, R. & Ranieri, W. (1996). Comparison of Beck Depression Inventories–IA and –II

in psychiatric outpatients. Journal of Personality Assessment; 67: 588–597.

Berry, J. W., Phinney, J. S., Sam, D. L., Vedder, P. E. (2006). Immigrant Youth in Cultural Transition.

Acculturation, Identity, and Adaptation Across National Contexts. Mahwah, NJ: Lawrence Erlbaum Associates.

Bhugra, D., Hilwig, M., Corridan, B., Neehall, J., Rudge, S., Mallett, R., Leff, J. (2000). A comparison of symptoms

in cases with first onset of schizophrenia across four groups. European Journal of Psychiatry; 14: 241-249.

Blom, J. D., Eker, H., Basalan, H., Aouaj, Y. & Hoek, H. W. (2010). Hallucinaties toegedicht aan djinns.

Nederlands Tijdschrift voor de Geneeskunde; 154: A973.

Cannon, T. D., Cadenhead, K., Cornblatt, B., Woods, S. W., Addington, J., Walker, E., Seidman, L. J., Perkins,

D., Tsuang, M., McGlashan, T. & Heinssen, R. (2008). Prediction of Psychosis in Youth at High Clinical Risk:

A Multisite Longitudinal Study in North America. Archives of General Psychiatry; 65: 28–37.

Goldman, H. H., Skodol, A. E. & Lave, T. R. (1992). Revising axis V for DSM-IV: a review of measures of social

functioning. American Journal of Psychiatry; 149: 1148-1156.

Harvey, I., Williams, M., McGuffin, P. & Toone, B. K. (1990). The functional psychoses in Afro-Carribeans.

British Journal of Psychiatry; 157: 515-522.

Hutchinson, G. & Sharpley, M. (1999). Ethnicity and first year of contact with psychiatric services.

British Journal of Psychiatry; 175: 492.

Kirkbride, J. B., Barker, D., Cowden, F., Stamps, R., Yang, M., Jones, P. B. & Coid, J. W. (2008). Psychosis,

Ethnicity and socio-economic status. British Journal of Psychiatry; 193: 18-24.

Morgan, C., Charalambides, M., Hutchinson, G., Murray, R. M. (2010). Migration, ethnicity, and psychosis:

toward a sociodevelopmental model. Schizophrenia Bulletin; 36: 655-664.

37

Chapter 2

Morgan, C. & Hutchinson, G. (2010). Letter to the editor: Prevention is better than cure: a reply to McKenzie,

March et al. and Selten & Cantor-Graae. Psychological Medicine; 40: 876-877.

Morgan, C., Kirkbride, J., Hutchinson, G., Craig, T., Morgan, K., Dazzan, P., Boydell, J., Doody, G. A., Jones, P. B.,

Murray, R. M., Leff, J. & Fearon, P. (2008). Cumulative social disadvantage, ethnicity and first-episode psychosis:

a case-control study. Psychological Medicine; 38: 1701-1715.

Phinney, J. S. (1992). The Multigroup Ethnic Identity Measure: A new scale for use with adolescents and young

adults from diverse groups. Journal of Adolescence; 7: 156-175.

Rathod, S., Kingdon, D., Phiri, P. & Gobbi, M. (2010). Developing Culturally Sensitive Cognitive Behaviour Therapy for

Psychosis for Ethnic Minority Patients by Exploration and Incorporation of Service Users’ and Health Professionals’

Views and Opinions. Behavioural and Cognitive Psychotherapy; 38: 511-533.

Reininghaus, U., Craig, T. K., Fisher, H. L., Hutchinson, G., Fearon, P., Morgan, K., Dazzan, P., Doody, G. A., Jones,

P. B., Murray, R. M. & Morgan, C. (2010). Ethnic identity, perceptions of disadvantage, and psychosis: findings from

the ÆSOP study. Schizophrenia Research; 124: 43-48.

Rietdijk, J., Dragt, S., Klaassen, R., Ising, H., Nieman, D., Wunderink, L., Delespaul, P., Cuijpers, P., Linszen,

D. & van der Gaag, M. (2010). A single blind randomized controlled trial of cognitive behavioural therapy in

a help-seeking population with an At Risk Mental State for psychosis: the Dutch Early Detection and Intervention

Evaluation (EDIE-NL) trial. Trials; 22: 11:30.

Selten, J. P., Cantor-Graae, E. (2005). Social defeat: risk factor for schizophrenia. British Journal of Psychiatry;

187: 101-102.

Simon, A. E., Umbricht, D. (2010). High remission rates from an initial ultra-high risk state for psychosis.

Schizophrenia Research; 116: 168-172.

Strakowski, S. M., Flaum, M., Amador, X., Bracha, H. S., Pandurangi, A. K., Robinson D. & Tohen,

M. Racial differences in the diagnosis of psychosis. Schizophrenia Research; 23: 117-124.

Veling, W., Hoek, H. W., Wiersma, D. & Mackenbach, J. P. (2010). Ethnic identity and the risk of schizophrenia

in ethnic minorities: a case-control study. Schizophrenia Bulletin; 36: 1149-1159.

Veling, W., Selten, J. P., Mackenbach, J. P., Hoek, H. W. (2007). Symptoms at First contact for psychotic disorder:

comparison between native Dutch and ethnic minorities. Schizophrenia Research; 95: 30-38.

Velthorst, E., Nieman, D. H., Becker, H. E., Fliert van de, R., Dingemans, P. M., de Haan, L., van Amelsvoort,

T. & Linszen, D. H. (2009). Baseline differences in clinical symptomatology between ultra high risk subjects with

and without a transition to psychosis. Schizophrenia Research; 109: 60-65.

38

Chapter 2

Velthorst, E., Nieman, D. H., Klaassen, R. M., Becker, H. E., Dingemans, P. M., Linszen, D. H., de Haan, L. (2010).

Three-year course of clinical symptomatology in young people at ultra high risk for transition to psychosis.

Acta Psychiatrica Scandinavica; 133: 36-42.

World Health Organization. (1993). Composite international diagnostic interview, Core Version 1.1. Arlington,

VA: American Psychiatric Publishing.

Wing, J. K., Babor, T., Brugha, T., Burke, J., Cooper, J. E., Giel, R., Jablenski, A., Regier, D. & Sartorius, N. (1990).

SCAN. Schedules for Clinical Assessment in Neuropsychiatry. Archives of General Psychiatry; 47: 589-593.

Yung, A. R., McGorry, P. D., McFarlane, C. A., Jackson, H. J., Patton, G. C. & Rakkar, A. (1996).

Monitoring and care of young people at incipient risk of psychosis. Schizophrenia Bulletin; 22: 283–303.

Yung, A. R., Phillips, L. J., Yuen, H. P., Francey, S. M., McFarlane, C. A., Hallgren, M., & McGorry, P. D. (2003).

Psychosis prediction: 12-month follow up of a high-risk (“prodromal”) group. Schizophrenia Research; 60: 21-32.

Yung, A. R., Yuen, H. P., McGorry, P. D., Philips, L. J., Kelly, D., Dell’Olio, M., Francey, S. M., Cosgrave, E. M., Killackey,

E., Stanford, C., Godfrey, K. & Buckby, J. (2005). Mapping the onset of psychosis: the Comprehensive Assessment of

At-Risk Mental States. Australian and New Zealand Journal of Psychiatry; 39: 964-971.

Yung, A. R., Buckby, J. A., Cotton, S. M., Cosgrave, E. M., Killackey, E. J., Stanford, E., Godfrey, K. & McGorry,

P. D. (2006). Psychotic-like experiences in nonpsychotic Help-seekers: Associations with distress, depression,

and disability. Schizophrenia Bulletin; 32: 352–359.

Zandi, T., Havenaar, J., M., Smits, M., Limburg-Okken, A., G., van Es, H., Cahn, W.,

Algra, A., Kahn, R., S. & van den Brink, W. (2010). First contact incidence of psychotic disorders among native Dutch

and Moroccan immigrants in the Netherlands: Influence of diagnostic bias. Schizophrenia Research; 119: 27-33.

39

40

41

Chapter 3

FACTOR ANALYSIS OF THE SCALE OF PRODROMAL SYMPTOMS: DIFFERENTIATING BETWEEN NEGATIVE AND DEPRESSION SYMPTOMS.

Rianne M. C. Klaassen, Eva Velthorst , Dorien H. Nieman, Lieuwe de Haan, Hiske E. Becker, Peter M. Dingemans, Reinaud J. van de Fliert, Mark van der Gaag, Don H. Linszen

PSYCHOPATHOLOGY; IN PRESS

42

Chapter 3

ABSTRACT

Background This study examines the ability of the Scale Of Prodromal Symptoms (SOPS) to differentiate between negative and depressive symptoms in a young help-seeking ultra high risk (UHR) group.

Methods Data of 77 help-seeking patients at UHR for psychosis were analyzed with an exploratory factor analysis. The extracted Depression factor was validated with the Beck Depression Inventory (BDI). The extracted SOPS Negative symptoms factor was validated with the Negative symptoms subscale of the Positive and Negative Syndrome Scale (PANSS).

Results Four factors were extracted from the SOPS: a negative, depression, disorganised and positive factor. The Negative symptom factor consisted of three items: (N1: social anhedonia and withdrawal, N3: decreased expression of emotion; and N4: decreased experience of emotions and self), and could be validated with the PANSS Negative symptoms subscale. The Depression factor was also made up by three items: (G2: dysphoric mood, G4: impaired tolerance to normal stress, and D4: personal hygiene/social attentiveness), and could be validated with the BDI.

Conclusions Our results suggest that 3 items of the Negative symptoms subscale of the SOPS, 2 items of the General and 1 item of the Disorganisation subscale differentiate validly between nega-tive and depression symptoms in an UHR population.

43

Chapter 3

1. Introduction

The success of early detection and intervention in first episode psychosis of schizophrenia-like disorders has led to an increased interest in the period prior to a first episode of psychosis. [1,2,3]. Commonly, the pre-psychotic phase is denoted as either the Ultra High Risk phase (UHR), or the at risk mental state phase (ARMS). Their criteria can be classified into three categories: attenuated positive symptoms (APS), brief limited intermittent psychotic symptoms (BLIPS) and having a ge-netic risk combined with reduced social functioning (GRRF). Yung and others found that the psycho-sis transition risk within a year is 10-40% [4,5,6]. Miller and collaborators [7] constructed the Struc-tured Interview for Prodromal Symptoms (SIPS) in order to assess the criteria used to identify UHR. Embedded within the SIPS is the Scale of Prodromal Symptoms (SOPS), which is designed to rate the severity of relevant symptoms. The SOPS comprises four factors: positive, negative, disor-ganized and general symptoms [8]. The four symptom domains were constructed on a priori grounds [8] to identify UHR patients. To our knowledge, only one validation study has been performed [9], and this study found three instead of four factors: positive, negative, and a general factor. Although the inclusion criteria of UHR studies are typically based on the positive domain of the SOPS, authors in both retrospective [10, 11] and prospective cohort studies [12, 5] have highlighted the substantial prevalence of negative (up to 40%) and depression symptoms (30-40%). The depression and negative symptoms clusters are sometimes difficult to distinguish [9, 12, 6, 13, 14]. However, several reasons underline the importance of differentiating between these two clusters: Firstly, differentiation is required for diagnostic purposes in the UHR phase. Negative symptoms, specifically social anhedonia and withdrawal, have been found to predict first psychosis transition in a UHR-group [15, 16, 17], whereas depression symptoms seem to play a minor role as predictor. Young adults suffering from depression can be at risk for psychosis as well as for depression or a bipolar disorder [18]. Secondly, differentiation between negative symptoms and depression may lead to targeted treatment strategies in the UHR phase, specifically when depression is better diagnosed. Treatment strategies for depression symptoms are well established (for example antidepressant medication or cognitive behavioural therapy (CBT) [19] while specific interventions for negative symptoms are still under development [20,21].Hawkins and collaborators [9]concluded that it is possible to differentiate between depression and negative symptoms in an UHR population by means of the SOPS, since the item ‘dysphoric mood’ did not load on the negative factor. No further arguments were given. We therefore believe that a replication of their factor analysis is needed with specific attention to SOPS negative and depression items. This study focuses on the depression and negative symptom items in an UHR group by performing an exploratory factor analysis of the SOPS. The extracted depression and negative symptoms factors are validated with the golden standard for the Positive And Negative Symptom Scale (PANSS) [22] and with the total score on the Beck Depression Inventory (BDI) [23].

44

Chapter 3

2. Methods2.1 Subjects

Help-seeking adolescents were referred by local mental health services to the Early Psychosis Department of the Academic Medical Center (AMC) of the University of Amsterdam, a unit specialized in diagnosis, treatment and research of early psychosis and subjects at UHR for psychosis transition. The inclusion criteria were: ages between 12 and 35 years, being able and willing to give informed consent and falling into one of the following groups:

1: Attenuated Positive Symptom Group (APS): subjects who have experienced one of the following symptoms in a mild to moderate extent: unusual thought content (delusional ideas), suspiciousness (persecutory ideas), grandiosity, perceptual abnormalities (hallucinations), disorganised communications and odd behaviour/appearance.

2: Brief limited intermittent psychotic symptoms (BLIPS): subjects who have experienced episodes of frank psychotic symptoms during the previous year which spontaneously resolved within a week.

3: Genetic risk and reduced functioning (GRRF): subjects who have a first degree relative with a psychotic disorder, or who themselves have a schizotypical personality disorder. Moreover, the subjects had experienced a decrease in functioning, namely a reduction on the Global assessment of Functioning Scale (GAF) (American Psychiatric Association, 1994) of at least 30%, compared to the highest level of their previous functioning for at least one month within the previous year. The exclusion criteria were a previous psychotic period for more than one week, a premorbid IQ<85, symptoms due to a known general medical disorder or intoxication with drugs or alcohol. ‘Cannabis use’ in the past or the present was defined as the use of cannabis at least 5 times in a lifetime. Cannabis use was not an exclusion factor in itself, but patients were excluded if they used any other drug or if the cannabis seemed to have caused the UHR symptoms. In order to establish the relation between cannabis use and psychotic symptoms, cannabis-using patients were required to stop taking drugs in the month following intake. Subsequently, they were then assessed again with the SIPS to investigate whether their symptoms remained. If so, they were in-cluded in the study. The investigation was carried out in compliance with the latest version of the Declaration of Helsinki. The study design was approved by the Medical Ethical Committee of the AMC. Informed consent of the participants (and parents if subjects were under 16 years old) was obtained after the nature of the procedures had been fully explained.

45

Chapter 3

2.2 Instruments

The SIPS was used to assess prodromal symptoms [8]. The SOPS was used to rate the severity of symptoms. The SOPS rating scale consists of 19 items (5 positive, 6 negative, 4 disorganisation and 4 general symptoms). Each item is rated on a 7 point scale which covers severity variance in the sub psychotic or attenuated range, varying from 0 (Never, absent) to 6 (Severe/Extreme and Psychotic). Scores of 3 to 5 on the positive items are considered as indicative of the UHR phase (APS). A score of 6 on one of the positive symptom scores means the person is psychotic. All inter-viewers received a two-day training by Dr. T. J. Miller, key SIPS/SOPS developer and trainer, followed by a reliability check after six months. The pair-wise inter-rater reliability concordance of the SOPS was 77 percent which was deemed acceptable by the training team. Dysphoric mood is one of the General Symptoms (G2) rated on the SOPS. The score on this item is based on 6 specific questions about mood and mood-related symptoms (i.e. crying, hopelessness, eating problems, sleeping problems, suicidal thoughts, irritability/ aggressiveness and anxiety). The General Assessment Functioning (GAF) of the DSM was used to determine the current and highest level of functioning in the previous year. The severity of depression symptoms was examined with the BDI [23]. The BDI is a 21 item self-rating scale. Each item comprises 4 statements that describe symptom severity. It is validated for general and depressed populations [24, 25] Negative symptoms were assessed with the PANSS [22]. It is validated to measure negative symptoms in schizophrenia and related disorders [26]. The PANSS is composed of 30 items, (7 negative symptom items) each of which is given an anchored score of 1 to 7. All instruments were administered by trained residents and researchers with extensive clinical experience. The intraclass correlation coefficients for the PANSS positive, negative, and general psychopathology subscales were 0.91, 0.84 and 0.76, respectively.

2.3 Procedure

After referral to the AMC, the adolescents and young adults were interviewed by a psychiatrist and psychologist. Parents were interviewed separately about their child’s lifetime history, complaints, family history of psychiatric disorders, substance and medicine (ab)use. The SOPS was rated and each subject was discussed in a case conference. Patients considered to be at UHR were then asked to sign a written informed consent to participate in the Dutch Prediction of Psychosis Study (DUPS). They received an initial assessment after admission into the study in which, amongst other question-naires, the BDI-I and the PANSS were assessed. The follow-up period was three years.

2.4 Statistical analysis Statistical analysis was performed using the Statistical Package for Social Science (SPSS), version 16. Differences in symptom severity between the patients using antidepressants and those who did

46

Chapter 3

not were examined with two-tailed t-tests. Construct validity of the SOPS was tested by entering the data into a principal component analysis, using Kaiser’s criterion (eigenvalues of 1.0 or more) and scree plot analysis. Varimax rotation was used to rotate, and validation was then carried out by a correlation analysis (Spearman). As negative symptoms show overlap with depression symptoms we wanted to correct for artificial inflation of the size of the correlation coefficient obtained. Partial correlations were there-fore used to control for common variance in depression and negative symptoms scales.

3. Results

Of the 285 referrals to the AMC 108 (37.9%) met the UHR criteria. Of these, 77 adolescents gave written informed consent to take part in the study:

Table 1: flow chart of referrals to DUPS project

Total referrals 285

No show 17

SCREENED WITH THE SIPS 268

Exlusions + reason why excluded from study

Psychosis 109

No symptoms 42

Life time symptoms 4

Age <12 or >35 2

Organic illness 2

Mild intellectual disability 1

UHR 108

No informed consent 31

UHR cohort 77

47

Chapter 3

3.1 Baseline characteristics of the sample

The included group consisted of 77 subjects: the mean age was 19.2 years, (S.D. =3.8) and 51 (66.2%) were men. The mean GAF at intake was 50.1 (S.D. =11.4) and patients on medication were included: the medication prescription at the first diagnostic evaluation was measured by dividing it into four categories: (i) antipsychotic medication: patients using antipsychotic medication and antidepressants, and/or other medication such as benzodiazepines were also assigned to this category, (ii) antidepressants or antidepressants with medication other than antipsychotic medication, (iii) other medication, for instance benzodiazepines, methylphenidate, and/or lithium carbonate, and (iv) no medication. Of the patients, 45 (58.4%) did not use any medication (category iv), 9 (11.7%) patients used an antidepressant (ii), 16 (20.8%) used antipsychotic medication (i); 7 patients (9.1%) fell into category iii. Patients who used antidepressants did not differ from those who did not in terms of negative-, positive- disorganised and depression symptomatology on the SOPS. Most subjects met the ‘Attenuated positive symptoms criteria’ (n=69), including APS and overlap with one of the other groups. The mean SOPS ratings were 2.20 for positive-, 2.22 for negative-, 2.15 for general- and 1.18 for disorganised symptoms. Items with the highest mean ratings are listed in Table 2.

Symptom

P1 Unusual thought content

P2 Suspiciousness/persecutory ideas

P3 Grandiose ideas

P4: Perceptual abnormalities/hallucinations

P5: Disorganized communication

N1 Social isolation and withdrawal

N2 Avolition

N3 Decreased expression of emotion

N4 Decreased experience of emotions and self

N5 Decreased ideational richness

N6 Deterioration in role functioning

G1 Sleep disturbance

G2 Dysphoric mood

G3 Motor disturbances

G4 Impaired tolerance to normal stress

D1 Odd behaviour and appearance

D2 Bizarre thinking

D3 Trouble with focus and attention

D4 Personal hygiene/social attentiveness

Mean

3.74

2.56

1.17

2.52

1.03

2.73

2.91

1.26

1.66

1.18

3.57

2.18

3.06

0.81

2.53

0.44

0.92

2.61

0.75

SD

1.58

1.74

1.46

1.90

1.52

2.00

2.00

1.46

1.77

1.54

1.77

1.76

1.51

1.43

1.98

0.88

1.49

1.21

1.23

Range

0-6

0-6

0-5

0-5

0-6

0-6

0-6

0-5

0-5

0-5

0-6

0-6

0-6

0-5

0-6

0-4

0-6

0-6

0-6

Table 2: Mean scores per SIPS item (N=77)

48

Chapter 3

Symptom

N3. Decreased expression of emotion

N4. Decreased experience of emotions and self

N1. Social anhedonia and withdrawal

N2. Avolition

N6. Deterioration in role functioning

G4. Impaired tolerance to normal stress

D4. Personal hygiene/social attentiveness

G2. Dysphoric mood

P5. Disorganized communication

D2. Bizarre thinking

D1. Odd behaviour or appearance

N5. Decreased ideational richness

P1. Unusual thought content/ delusional ideas

P2. Suspiciousness/ persecutory ideas

P4. Perceptual abnormalities/ hallucinations

Factor

1

0.76

0.75

0.70

0.66

0.60

0.75

Factor

2

0.41

0.52

0.76

0.58

0.50

Factor

3

0.72

0.71

0.59

0.55

0.75

Factor

4

0.49

0.64

0.63

0.44

0.75

3.2 Principal component factor analysis of the SOPS

The initial analysis generated 7 components with eigenvalues >1.0. Based upon scree-plot analysis further extractions were limited to four factors, accounting for 46.5 % of the variance. These factors bore a reasonable congruence with the SOPS a priori content areas:

Table 3: Rotated Component Matrix of SOPS items, sample (N=77)

Extraction Method: Principal Component Analysis with Varimax rotation. Loadings <0.40 are not printed for clarity. Primary loadings are in bold interface.

Therefore P3, D3, D4, G1 and G3 are not in the table. a Rotation converged in 5 iterations.

Of the negative symptom items of the SOPS, three had a single loading on Factor 1. These items were: Social anhedonia and withdrawal (N1), Decreased expression of emotion (N3), and Decreased experience of emotions and self (N4). We denote this as the SOPS Negative symptom factor. Factor 2 comprises: Dysphoric mood (G2), Impaired tolerance to normal stress (G4) and Personal hygiene/social attentiveness (D4). We call this the SOPS Depression factor. Factor 3 defines the SOPS Disorganisation factor with the items: Disorganised communication (P5), Bizarre thinking (D2), Odd behaviour or appearance (D1), and Decreased ideational richness (N5).

49

Chapter 3

The items Unusual thought content/ delusional ideas (P1), Suspiciousness/ persecutory ideas (P2) and Perceptual abnormalities/ hallucinations (P4) loaded on factor 4: the SOPS Positive symptom factor.

3.3. Validation Table 4 presents the validation of the SOPS sub scale for depression and negative symptoms by means of correlations and partial correlations.

Table 4: Spearman’s Rho’s correlation and partial correlation between found SOPS depression factor and negative factor, and BDI total score and the PANSS negative items total score

SOPS depressive factor = G2 + G4 + D4

SOPS negative factor NEG = N1 + N3 + N4

BDI = BDI total sum score

PANSS negative items =N1+N2+N3+N4+N5+N6+N7

*= partial correlation SOPS depression factor and BDI negative factor corrected for PANSS Negative Items

**=partial correlation SOPS negative factor and PANSS negative factor corrected for BDI total sum score

The SOPS Depression factor correlates with the BDI total score, while the SOPS Negative factor correlates with the PANSS Negative symptoms total score. The SOPS Negative factor is also associated with the BDI total score. Similarly the SOPS Depression factor correlates with the PANSS Negative symptoms total score. Depression and negative symptoms have a substantial variance in common. When we par-tial out the variance from PANSS Negative factor from the association between SOPS Depression factor and the BDI total score, the association still remains significant. The association between both depression scales is not moderated by negative symptoms. When the common variance with BDI total score is partialled out, the strength of the association between SOPS Negative factor scale and PANSS Negative symptoms total score hardly changes. The association between both negative symptoms scale is not moderated by depression symptoms.

BDI

PANSS Negative items

0.59 p<.001

0.34 p=0.04

0.59 p<.001

-

SOPS Depression factor

0.36 p=0.02

0.63 p<0.001

-

0.50 p<0.001**

SOPS Negative factor

50

Chapter 3

4. Discussion

The results of this study suggest that it is possible to differentiate validly between negative symptoms and depression symptoms in a group at UHR for a psychosis by means of the SOPS. In particular, ‘social anhedonia and withdrawal’, ‘decreased expression of emotion’ and ‘decreased experience of emotions and self’ appear as specific negative symptoms. On the other hand, ‘dysphoric mood’, ‘impaired tolerance to normal stress’ and ‘personal hygiene/social attentiveness’ can be considered unique depression symptoms. However, ‘avolition’ and ‘deterioration in role functioning’ have moderate loadings on both factors, arguing against a strict negative/depressive dichotomy for these items. A possible explanation for this finding is that these two items are part of the described overlap between negative and depression symptoms. Contrary to our expectations, the item ’dysphoric mood’ loaded secondarily on the positive items cluster. Both UHR and psychosis literature provide possible explanations. Yung et al [5] described the following possible chain of reactions: depression can lead to a negative evaluation of experiences, increasing the level of distress, anxiety and depression, which can subsequently cause an exacerbation of positive psychotic symptoms. Another possible explanation can be derived from research with patients with a first psychosis. Drake and colleagues [27] found that paranoid symptoms are strongly correlated with depression. In addition to the item ‘dysphoric mood’, in our study the item ‘Avolition’, an important negative symptom of schizophrenia [28, 29] showed an unexpected second loading. Besides the loading on the negative factor (in accordance with Hawkins et al. and Miller et al.[9,8]), we found avolition to load on the depression factor. A possible explanation could be the nuance difference between the absence of intention (negative symptoms), and mood induced inactivity (depression). This subtle difference may become clearer in longitudinal studies, as depression symptoms vary more over time than negative symptoms with their trait-like nature [30, 31, 32]. The results of this study are in accord with those of Hawkins et al. [9]. They also found ‘social anhedonia and withdrawal’, ‘decreased expression of emotion’ and ‘decreased experience of emotions and self’ to load on a negative symptoms factor. This is in concordance with the negative subscale of the SOPS-instrument. Moreover, in accordance with Hawkins et al., the items ‘dysphoric mood’ and ‘impaired tolerance to normal stress’ load together, suggesting that these two items to be specific for depression (especially since they also correlate with the BDI). It should be noted that some items do not consistently load on the same factor. For instance, ‘impairment in personal hygiene and/or social attentiveness’ is a disorganized item in the SOPS and part of the depression factor in our study, whereas the Hawkins’ study suggested it to be part of the negative symptom cluster. The items ‘deterioration in role functioning’ and ‘trouble with focus and attention’ show similar differences in a comparison of the results from the different studies.

51

Chapter 3

The problem with construct validity of some items is not unique for the UHR group. The same confusion in psychosis and schizophrenia, led researchers in this field to conclude that double loadings do not automatically point to a diffuse item, but rather point to a complex model of causation of symptoms [33, 34]. This complex model of causality might also be applicable to our UHR group. The choice for the BDI as the instrument to rate depression is that it is validated in general populations and in depressed patient groups, and, being a self- report scale, it has good clinical usage. The Calgary Depression Rating Scale (CDS) [35] might be a good candidate for rating depression in the UHR patient group, given the high prevalence of negative symptoms. It is currently investigated in an ongoing early detection and intervention study (EDIE) trial, in the Netherlands. Finally, we need to point out limitations in our study. We based our results upon a relatively small patient group. Nonetheless, the overlap of our main findings with those reported by Hawkins et al [9] taken in combination with the correlation with standard measurements factor analysis, reinforces the validity of our construct. In summary, we conclude that the SOPS negative symptoms can be differentiated from de-pression symptoms in patients at UHR for psychosis. A confirmatory factor analysis in a larger pa-tient sample to validate the results is needed.

Acknowledgments

We thank Ron Laport, Jean Paul Selten, Henrikje Klasen and Sandra Schrama for commenting on earlier versions of this article.

52

Chapter 3

References

1 Reading, B., Birchwood, M. (2005). Early intervention in psychosis, rationale, and evidence for effectiveness.

Disease Management & Health Outcomes; 13(1):53-63.

2 Ricciardi, A., McAllister, V., Dazzan, P. (2008). Is early intervention in psychosis effective? Review.

Epidemiologia e Psichiatria Sociale; 17(3):227-235.

3 Marshall, M., Lewis, S., Lockwood, A., Drake, R., Jones, P., Croudace, T. (2005). Association between duration

of untreated psychosis and outcome in cohorts of first-episode patients: a systematic review. Archives of General

Psychiatry; 62(9):975-983.

4 Yung, A. R., McGorry, P. D., McFarlane, C. A., Jackson, H. J., Patton, G. C., Rakkar, A. (1996).

Monitoring and care of young people at incipient risk of psychosis. Schizophrenia Bulletin; 22(2):283-303.

5 Yung, A. R., Buckby, J. A., Cosgrave, E. M., Killackey, E. J., Baker, K., Cotton, S. M., McGorry, P. D. (2007).

Association between psychotic experiences and depression in a clinical sample over 6 months.

Schizophrenia Research; 91:246-253.

6 Olsen, K. A., Rosenbaum, B. (2005). Prospective investigations of the prodromal state of schizophrenia:

review of studies. Acta Psychiatrica Scandinavica; 113:247-272.

7 Miller, T. J., McGlashan, T. H., Woods, S. W., Stein, K., Driesen, N., Corcoran, C. M., Hoffman, R., Davidson,

L. (1999). Symptom assessment in schizophrenic prodromal states. Psychiatric Quarterly; 70(4):273-287.

8 Miller, T. J., McGlashan, T. H., Rosen, J. L., Cadenhead, K., Cannon, T., Ventura, J., McFarlane, W., Perkins,

D. O., Pearlson, G. D., Woods, S. W. (2003). Prodromal assessment with the Structured Interview for Prodromal

Syndromes and the Scale of Prodromal Symptoms: predictive validity, interrater reliability, and training to

reliability. Schizophrenia Bulletin; 29(4):703-715.

9 Hawkins, K. A., McGlashan, T. H., Quinlan, D., Miller, T. J., Perkins, D. O., Zipursky, R. B., Addington, J.,

Woods, S. W. (2004). Factorial structure of the scale of prodromal symptoms. Schizophrenia Research; 68:339-347.

10 Birchwood, M. (2003). Pathways to emotional dysfunction in first-episode psychosis. Br J Psychiatry; 182:373-375.

11 Hafner, H., Loffler, W., Maurer, K., Hambrecht, M., an der Heiden, W. (1999). Depression, negative symptoms,

social stagnation and social decline in the early course of schizophrenia. Acta Psychiatrica Scandinavica;

100:105-118.

12 Lencz, T., Smith, C. W., Auther, A., Correll, C. U., Cornblatt, B. (2004). Nonspecific and attenuated negative

symptoms in patients at clinical high-risk for schizophrenia. Schizophrenia Research; 68:37-48.

13 Simon, A. E., Dvorsky, D. N., Boesch, J., Roth, B., Isler, E., Schueler, P., Petralli, C., Umbricht D. (2006).

Defining subjects at risk for psychosis: a comparison of two approaches. Schizophrenia Research; 81:83-90.

53

Chapter 3

14 Yung, A. R., Philips, L. J., Pan Yuen, H., McGorry, P. D. (2004). Risk factors for psychosis in an ultra high risk

group: psychopathology and clinical features. Schizophrenia Research; 67:131-142.

15 Cannon, T. D., Cadenhead, K., Cornblatt, B., Woods, S. W., Addington, J., Walker, E., Seidman, L. J., Perkins,

D., Tsuang, M., McGlashan, T., Heinssen, R. (2008). Prediction of psychosis in youth at clinical high risk:

a multisite longitudinal study in North America. Archives of General Psychiatry; 65(1):28-37.

16 Correl, C. U., Penzner, J. B., Frederickson, A. M., Richter, J. J., Auther, A. M., Smith, C. W., Kane, J. M., Cornblatt,

B. A. (2007). Differentiating in the preonset phases of schizophrenia and mood disorders: evidence in support

of a bipolar mania prodrome. Schizophrenia Bulletin; (33)3:703-714.

17 Velthorst, E., Nieman, D., Becker, H. E., van de Fliert, R., Dingemans, P. M., Klaassen, R., de Haan,

L., van Amelsvoort, T., Linszen, D. H. (2009). Baseline differences in clinical symptomatology between ultra

high risk subjects with and without a transition to psychosis. Schizophrenia Research;109 (1-3):60-5.

18 Hafner, H., Maurer, K., Trendler, G., an der heiden, A., Schmidt, M., Konnecke, R. (2005). Schizophrenia and

depression: challenging the paradigm of two separate diseases-a controlled study of schizophrenia, depression

and healthy controls. Schizophrenia Research; 77:11-24.

19 Cornblatt, B. A., Lencz, T., Smith, C. W., Olsen, R., Auther, A. M., Nakayama, E., Lesser, M. L., Tai, J. Y., Shah,

M. R., Carmel, A. F., Kane, J. M., Correll, C. U. (2007). Can antidepressants be used to treat the schizophrenia

prodrome? Results of a prospective, naturalistic treatment study of adolescents. Journal of Clinical Psychiatry;

68:546-557.

20 Phillips, L. J., McGorry, P. D., Yung, A. R., McGlashan, T. H., Cornblatt, B., Klosterkotter, J. (2005).

Prepsychotic phase of schizophrenia and related disorders: recent progress and futures opportunities.

British Journal of Psychiatry; 187(48):233-244.

21 Grant, P. M., Beck, A. T. (2009). Defeatist beliefs as a mediator of cognitive impairment, negative symptoms and

functioning in schizophrenia. Schizophrenia Bulletin; 35(4):798-806.

22 Kay, S. R., Fiszbein, A., Opler, L. A. (1987). The positive and negative symptoms scale (PANSS).

Schizophrenia Bulletin; 13(2):261-276.

23 Beck, A. T., Ward, C. H., Mendelson, M., Mock, .J, Ernbaugh, J. (1961).

An inventory for measuring depression. Archives of General Psychiatry; 4:561-571.

24 Richter, P., Werner, J., Heerlein, A., Alfred Kraus, A., Sauer, H. (1998).

On the validity of the Beck Depression Inventory. A review. Psychopathology; 31:160-168.

25 Shafer, A.B. (2006). Meta-analysis of the Factor Structures of Four Depression Questionnaires:

Beck, CES-D, Hamilton, and Zung. Journal of Clinicial Psychology; 62(1):123-146.

54

26 Emsley, R., Rabinowitz, J., Torreman, M. (2003). The factor structure for the Positive and Negative

Syndrome Scale (PANSS) in recent-onset psychosis. Schizophrenia Research; 61(1):47-57.

27 Drake, R. J., Pickles, A., Bentall, R. P., Kinderman, P., Haddock, G., Tarrier, N., Lewis, S. W. (2004).

The evolution of insight, paranoia and depression during early schizophrenia. Psychological Medicine;

34 (2):285-292.

28 American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders (1994);

American Psychiatric Association, Washington D.C.

29 Lindenmayer, J. P., Khan, A. (2006) Psychopathology. In: Lieberman JA, Stroop TS, Perkins DA, editors.

Textbook of schizophrenia. Arlington: American Psychiatric Publishing Inc; 201-204.

30 an der Heiden, W., Könnecke, R., Maurer, K., Ropeter, D., Häfner, H. (2005). Depression in the

long-term course of schizophrenia. European Archives of Psychiatry and Clinical Neuroscience; 255(3):174-84.

31 Faerden, A., Finset, A., Friis, S., Agartz, I., Barrett, E. A., Nesvåg, R., Andreassen, O. A.,Marder, S. R.,

Melle, I. (2010). Apathy in first episode psychosis patients: One year follow up.

Schizophrenia Research;116 (1);20-26.

32 Häfner, H., Maurer, K., Trendler, G., an der Heiden, W., Schmidt, M. (2005). The early course of schizophrenia

and depression. European Archives of Psychiatry and Clinical Neuroscience; 255:167–173.

33 Blanchard, J. J., Cohen, A. S. (2006). The structure of negative symptoms within schizophrenia:

implications for assessment. Schizophrenia Bulletin; 32(2):238-245.

34 Van der Gaag, M., Hoffman, H., Remijssen, M., Hijman, R., de Haan, L., van Meijel, B., van Harten,

P. N., Valmaggia, L., de Hert, M., Cuijper, A., Wiersma, D. (2006). The five-factor model of the

Positive and Negative Syndrome Scale II: a ten-fold cross-validation of a revised model.

Schizophrenia Research; 68: 37-48.

35 Addington, D., Addington, J., Maticka-Tyndale, E. (1994).

Specificity of the Calgary Depression Scale for schizophrenics. Schizophrenia Research; 11:239-244.

Chapter 3

55

56

57

02P A R T

58

59

Chapter 4Chapter 4

BASELINE DIFFERENCES IN CLINICAL SYMPTOMATOLOGY BETWEEN ULTRA HIGH RISK SUBJECTS WITH AND WITHOUT A TRANSITION TO PSYCHOSIS.

Eva Velthorst, Dorien H. Nieman, Hiske E. Becker, Reinaud van de Fliert, Peter M. Dingemans, Rianne Klaassen, Lieuwe de Haan, Thérèse van Amelsvoort, Don H. Linszen

SCHIZOPHRENIA RESEARCH (2009); 109 (1-3): 60-5.

60

Chapter 4

ABSTRACT

Background The chance of transition to psychosis in patients at Ultra High Risk for developing psychosis (UHR) is 10–15%. The aim of present study was to investigate differences in baseline clinical symptomatology, general level of functioning (GAF-score) and genetic risk between UHR patients who did (UHR+T) or did not (UHR+NT) make a transition to psychosis. Sharpening UHR inclusion criteria may aid in improving prediction of transition to psychosis.

Method The study sample was taken from 285 patients who were examined within the Dutch Prediction of Psychosis Study (DUPS) at the Academic Medical Center of the University of Amsterdam, the Netherlands. Out of 73 included UHR subjects, 18 made a transition to psychosis. Psychopathology was investigated with the Structured Interview for Prodromal Syndromes, Bonn Scale for the Assessment of Basic Symptoms and GAF-score. The follow-up period of the study was three years.

Results The UHR+T group showed more social anhedonia and withdrawal, more bizarre thinking and a lower GAF score at baseline than the UHR+NT group.

Conclusions In agreement with the results of Cannon et al. [Cannon, T.D., Cadenhead, K., Cornblatt, B., Woods, S.W., Addington, J., Walker, E., Seidman, L.J., Perkins, D., Tsuang, M., McGlashan, T., Heinssen, R., 2008. Prediction of Psychosis in Youth at High Clinical Risk: A Multisite Longitudinal Study in North America. Arch. Gen. Psychiat. 65 (1) 28–37.], our study indicates that severity of specific symptoms at baseline is related to transition to psychosis in UHR subjects. These findings may contribute to a more accurate prediction of a first psychotic episode. Furthermore, symptoms that are increased at baseline in the UHR+T group could be a focus of cognitive behavioural therapy in the UHR period.

61

Chapter 4

1. Introduction

The early phase of a psychosis has been considered as a critical period, which can dispropor-tionately influence the long term evolution of the disorder (Reading and Birchwood; 2005). In more recent research great importance is attached to an even earlier period. Yung et al. (1996, 2003) and Klosterkötter et al.(2005) described the assessment and treatment of pre-psychotic or ‘ultra high risk’ symptoms. The ultra high risk phase, retrospectively called prodromal phase, recognizable by subtle behavioural changes and a decline in functioning, occurs prior to the development of the characteristic signs and symptoms of the illness that permit definitive diagnosis and lasts on average between 1 and 5 years (Phillips et al., 2002; Yung et al., 2003). Whereas Reading and Birchwood (2005) emphasize intervention during the first episode of psychosis, Yung et al. (2003) argue that intervention within the above mentioned ultra high risk phase of schizophrenia and related psychoses is indicated and may result in attenuation, delay or even prevention of the onset of psychosis in some vulnerable individuals. The identification of people with a psychotic vulnerability is needed to enable preventive approaches. How can this, often non-specific, phase be recognised? ‘The Personal Assessment and Crisis Evaluation (PACE) Clinic’ in Melbourne, Australia, initiated the formation of criteria for UHR for developing psychosis. Their criteria can be subsumed under three headings: Genetic risk and reduced functioning, Attenuated psychosis and Brief limited intermittent psychotic symptoms (BLIPS) (Yung et al., 2003). With these three criteria, UHR patients used to have a chance between 10 and 40% of making the transition to psychosis (Yung et al., 2003, 2006a,b). More recent studies however show transition rates of 10 to 15%, which might partly reflect a reduction in duration of UHR symptoms prior to receiving help (Yung et al., 2007; Cannon et al., 2008). Klosterkötter et al. (2001) emphasized the presence of ‘basic symptoms’ as an important risk-factor for the development of a first psychotic episode. Basic symptoms are self reported disturbances in cognition and perception that were found to be predictive of a transition to psychosis over a 10 year follow up period. These symptoms include for example thought interference, thought pressure and thought blockages. Klosterkötter et al. reported that the absence of basic symptoms excludes a subsequent schizophrenia diagnosis with a probability of 96%. In our naturalistic longitudinal cohort study we assessed UHR symptomatology and basic symptoms in a subsample of the 285 patients who were examined within the Dutch Prediction of Psychosis Study (DUPS) at the Adolescent Clinic of the Academic Medical Center (AMC) of the University of Amsterdam, the Netherlands. Part of this Dutch sample also participated in the European Prediction of Psychosis Study (EPOS, see for an earlier publication: Nieman et al., 2007). The follow up period in the study was three years.

62

Chapter 4

A previous comparable study found that certain symptoms contribute uniquely to the prediction of psychosis (Cannon et al., 2008). In their 2 1/2 year follow-up of 291 treatment-seeking UHR patients, they reported that genetic risk for schizophrenia with recent deterioration, higher levels of unusual thought content, higher levels of suspicion/paranoia, greater social impairment, and a history of substance abuse are key predictors of a subsequent psychosis. If replicated, these findings may contribute to amore accurate prediction of a first psychotic episode. Therefore, following the study of Cannon et al. (2008), the present study focuses on the predictive value for transition to psychosis of baseline UHR symptoms. Furthermore, the predictive value of genetic risk with reduced functioning and recent deterioration (assessed with the GAF score) will be evaluated. Within this framework the following question has been formulated: are differences in baseline clinical symptoms, genetic risk and GAF score present between UHR patients with a transition to a psychotic disorder (UHR+T) and UHR patients without a transition (UHR+NT) at three years follow up?

2. Methods2.1. Participants

Prior to their referral to the AMC, all patients sought help for various complaints at local mental health services and other health services. The patients were referred to the DUPS project for a second opinion because the referring clinician suspected a psychotic development. Inclusion criteriawere age between 12 and 35 years, being able and willing to give informed consent and falling in one of the following groups:1: Genetic risk in combination with reduced functioning: subjects who have a first degree relative with a psychotic disorder, or who themselves have a schizotypical personality disorder and who have experienced a significant decrease in functioning during the past year (i.e. 30% reduction of GAF-score for at least one month).2: Attenuated Positive Symptoms (APS): subjects who have experienced sub threshold, attenuated positive psychotic symptoms, defined by at least 1 of the following symptoms, appearing several times per week for at least 1 week within the last three months: unusual thought content/delusional ideas, suspiciousness /persecutory ideas, grandiosity, perceptual abnormalities/hallucinations, disorganized communication and odd behavior/ appearance.3: Brief limited intermittent psychotic symptoms (BLIPS): subjects who have experienced episodes of frank psychotic symptoms. BLIPS were defined by hallucinations, delusions or formal thought disorders occurring within the last 3 months and resolving spontaneously within 1 week.

63

Chapter 4

4: Basic symptoms: presence of at least 2 of the following 9 self-reported disturbed cognitive and perceptive basic symptoms of at least ‘moderate’ severity during the last 3 months: inability to divide attention, thought interference, -pressure and -blockage, disturbances of receptive and of expressive speech, disturbance of abstract thinking, unstable ideas of reference, captivation of attention by details of the visual field.

Exclusion criteria of this study were the presence or history of a psychotic disorder for more than one week, an IQ 85, symptoms due to a known general medical disorder or intoxication with drugs or alcohol. Patients were allowed to use cannabis, but they were excluded if they used any other drug or if the cannabis caused the UHR symptoms. To sort out the relation between cannabis use and psychotic symptoms, we asked the cannabis-using patients to stop taking drugs during the following month. Subsequently, they were assessed again with the SIPS to investigate if their symptoms remained.

If the symptoms abated, the patient was not included the study. In addition, we asked if the symptoms ever occurred without cannabis use. ‘Cannabis use’ in past or present was defined as having used cannabis at least 5 times in a lifetime. The investigation was carried out in accordance with the latest version of the Declaration of Helsinki. The study design was approved by the Medical Ethical Committee of the AMC. Informed consent of the participants was obtained after the nature of the procedures had been fully explained.

2.2. Instruments

The semi-structured interview, Structured Interview for Prodromal Syndromes (SIPS, Miller et al., 2002), was used to determine the presence, severity and type of prodromal symptoms. The Scale Of Prodromal Symptoms (SOPS), the rating scale of the SIPS, has four 4 SIPS subscales that include five Positive Symptom items, six Negative Symptom items, four Disorganization Symptoms items and four General Symptom items. All symptoms are rated on a 7 point rating scale rating from 0 (Never, absent) to 6 (Severe/Extreme –and Psychotic for the positive items). The diagnosis of a prodromal state is based on the score at the positive items. Scores in the 3 to 5 range are considered as indicative of the UHR phase (APS). A score of 6 signifies psychosis or BLIPS (Miller et al., 2002). The two interviewers (DHN and RvdF) received a two day training workshop by Dr. T. J. Miller, one of the SIPS authors, including a reliability check after approximately six months. The pairwise inter-rater concordance of the SIPS was 77% and determined acceptable by

the training team.

64

Chapter 4

The Bonn Scale for the Assessment of Basic Symptoms Prediction scale (BSABS-P, Klosterkötter et al., 1996) is developed to assess ‘basic symptoms’. The BSABS-P is a semistructured interview that consists of 33 principal items and can be divided into 5 BSABS-subscales: Cognitive thought disorders, Additional symptoms with positive predictive value (i.e. thought perseveration and decreased ability to discriminate between ideas and perception, fantasy and true memories), Visual perception disorders, Acoustic perception disorders and Cognitive motor disorders. The symptoms are rated on a scale rating from 0 (not present/absent) to 6 (severe). The investigators received repeated training by one of the scales’ authors (Dr. F. Schultze-Lutter). Concordance rate with expert rating (F.S.-L.) was 87.9%. To determine the general level of functioning and the reduction in functioning, the GAF-score (American Psychiatric Association, 1994) was used. This score is derived of the GAF scale, a retrospective scale which is rated by a psychiatrist from 1 to 100, for the current situation and for highest level in past year. 1 to 10, for example, signifies ‘a persistent danger of severely hurting self or others (e.g., recurrent violence) OR persistent inability to maintain minimal personal hygiene OR serious suicidal act with clear expectation of death’, whereas 91 to 100 on the other hand, stands for ‘Superior functioning in a wide range of activities, life’s problems never seem to get out of hand, is sought out by others because of his or her many positive qualities. No symptoms.’ The decline in functioning is computed by the subtraction of the current GAF-score from the highest GAF-score in past year. A transition to psychosis was operationalized as a continuation of BLIPS, i.e., as one or more psychotic symptoms persisted for more than one week with a score of four or more on hallucinations, delusions or formal thought disorder on the Positive and Negative Syndromes Scale (PANSS, Kay et al., 1987) for longer than one week. To establish a formal DSM-IV diagnosis, the SCID (Structured Clinical Interview for DSM-IV, Spitzer et al., 1992) was administered to all patients after transition to psychosis.

2.3. Procedure

After their referral to DUPS, putative UHR patients were invited for a first interview with a psychiatrist and a psychologist. In this approximately two hour lasting face to face interview, subjects were asked about their lifelong history of complaints, family history of psychiatric disorders, drug- and medicine use. Subsequently, in a standardized order, the SIPS and BSABS-P were administered. Simultaneously, in another interview the parents or guardians were asked about the lifelong development of their child. The SOPS and BSABS-P were scored and each subject was discussed in a staff meeting. When considered at ‘ultra high risk’, the patients were asked to sign a written informed consent before participating in the DUPS project. In this naturalistic longitudinal cohort study, subjects were followed up for three years. They were referred back to their referring mental health institution. Some received treatment, others were only monitored. Patients, their parents or caretakers and the referring instances were asked to contact the DUPS project in case of increasing symptoms. In addition, they were seen for a SIPS interview at 9, 18 and 24 months and interviewed by telephone at 36 months.

65

Chapter 4

2.4. Statistical analysis

In current study, the differences in scores on the two questionnaires and one rating scale (SIPS, BSABS-P and GAF score) between the two groups were examined with two-tailed t-tests. For the items social anhedonia and withdrawal, decreased expression of emotion, additional symptoms with a positive value and current GAF-score a t-test with pooled variance was used because SD was significantly unequal between the two groups. A p-valueb0.05 was considered statistically significant. The possible difference between groups in sex was evaluated using a chi-square test. With a Spearman’s rank order correlation, we looked at the strength of the relationship between the most robust variables. A multivariate algorithm that optimizes prediction of conversion to psychosis was derived from a Cox proportional hazards model.

3. Results

Table 1 lists the demographics of the included group. The included group consisted of 73 subjects (47 men, mean age 19.2, SD 3.9). Of these 73 UHR subjects, 18 (25%) made a transition to a psychosis (13 men, mean age, 20.4, SD 3.9) and 55 did not (34 men, mean age 18.7, SD 3.8). The mean interval between inclusion and transition of this group was 11.9 months (Range= 2.00–36.00, SD=8.6). Eight subjects (11%) were lost to follow up. The patients with follow-up information did not differ significantly from those lost to follow up in terms of sex, age, severity of positive and negative symptoms, basic symptoms, global functioning and family predisposition.

Table 2 lists the mean scores on the SOPS of the two groups. Two-tailed t-tests showed that there were no overall differences in positive symptoms between the UHR+T group and the UHR+NT group at baseline. There was a significant difference between negative symptoms in the UHR+T group and the UHR+NT group at baseline. This difference is mainly caused by a significant higher score on item N1 of the SIPS ‘social anhedonia and withdrawal’ by the UHR+T group (t=3.1, p=.004) compared to the UHR+NT group.

66

Table 1. Sociodemographic variables of the UHR+T and UHR+NT group

CHARACTERISTICS

Age, mean,(SD), y

Male, No. (%)

Cannabis use, No. (%)

Genetic risk, No. (%)

Global functioning, mean ± SD, score

INCLUSION SYMPTOMS

Attenuated symptoms, No. (%)

Attenuated + basic symptoms, No. (%)

BLIPS, No. (%)

BLIPS + attenuated, No. (%)

BLIPS + attenuated + basic s., No. (%)

BLIPS + basic symptoms, No. (%)

Genetic risk & reduced functioning + basic

symptoms, No. (%)

Genetic risk & reduced functioning + attenuated

symptoms, No. (%)

Genetic risk & reduced funtioning + attenuated +

basic symptoms. No. (%)

COMORBID DIAGNOSES

Affective disorder, No. (%)

Anxiety disorder, No. (%)

Substance abuse (in rem.), No. (%)

Development disorder, No. (%)

Dissociative disorder, No. (%)

None/ postponed, No. (%)

Others, No. (%)

UHR+T

(n = 18)

20.4 (3.9)

13 (72.2)

5 (31.3)

7 (38.9)

45.8 (8.4)

4 (22.2)

8 (44.4)

1 (5.6)

0

1 (5.6)

2 (11.1)

0

1 (5.6)

1 (5.6)

8 (44.4)

4 (22.2)

3 (16.7)

3 (16.7)

1 (5.6)

6 (33.3)

0

UHR+NT

(n = 55)

18.7 (3.9)

34 (62.8)

32 (58.2)

11 (20.0)

51.4 (11.7)

17 (30.9)

29 (52.7)

0

2 (3.6)

4 (7.3)

0

1 (1.8)

0

2 (3.6)

21 (38.2)

8 (14.5)

8 (14.5)

3 (5.5)

1 (1.8)

13 (23.6)

7 (12.7)

Statistics

t=1.7

X2=.6

X2=3.4

X2=2.6

t=-2.2

X2=.22

df

72

1

1

1

40

1

P Value

.10

.42

.07

.11

.03

.64

UHR+NT = Ultra high risk without transition, UHR+T = Ultra high risk with transition.

No. = number of subjects.

Comorbid diagnoses were assessed using the Structured Clinical Interview for DSM disorders (Spitzer et al., 1992).

Chapter 4

67

Table 2: SOPS subscales (Mean and SD) of UHR+NT and UHR+T at baseline (T0)

Although no significant differences were noticeable between the UHR+NT group and the UHR+T group in all disorganisation symptoms taken together, a difference was found in one of its symptoms, namely D2 ‘Bizarre thinking’ (t=2.1, pb.05). Thismeans that more subjects,who eventually made a transition to psychosis, already scored higher on this SIPS item than the subjects who did not make the transition to psychosis. We found trends on the SIPS P2 item ‘suspiciousness’ (t=1.7, p=.09) and on the Negative symptoms item N3 ‘decreased expression of emotion’ (t=1.9,p=.07) with the UHR+T group reporting slightly more suspiciousness and decreased expression of emotion at baseline than the UHR+NT group. Table 3 lists the mean scores on the BSABS in the two groups. None of the BSABS symptoms showed increased severity in the UHR+T group compared to the UHR+NT group. Several basic symptoms were even less severe in the UHR+T group compared to the UHR+NT group.

Chapter 4

Positive symptoms

Negative symptoms

Disorganisation symptoms

General symptoms

UHR+T

N=18

11.7 (4.0)

16.4 (6.8)

5.6 (2.9)

9.4 (4.3)

UHR+NT

N=55

11.1 (3.6)

12.4 (7.2)

4.6 (2.8)

8.6 (4.4)

Statistics

t=.6

t=2.1

t=1.4

t=.6

df

71

70

71

70

P Value

.54

.04

.19

.52

Range: Positive symptoms = 0-30, Negative symptoms = 0-36, Disorganisation symptoms = 0-24, General symptoms = 0-24. (0 = no symptoms).

Table 3: BSABS subscales (Mean and SD) of UHR+NT and UHR+T at baseline (T0)

Cognitive thought disorders

Additional symptoms with positive predictive value

Visual perception disorders

Acoustic perception disorders

Cognitive motor disorders

UHR+T

N=18

10.72(6.13)

1.28 (1.87)

1.94 (3.49)

6.17 (4.73)

2.33 (2.64)

UHR+NT

N=55

12.8 (7.8)

3.0 (2.8)

3.3 (5.1)

7.1 (4.8)

1.6 (2.6)

Statistics

t=-1.0

t=-3.0

t=-1.1

t=-.07

t=1.1

df

70

73

70

70

71

P Value

.31

.005

.29

.47

.79

Range: Cognitive thought disorders = 0-54, Additional symptoms = 0-12, Visual perception disorders = 0-78, Acoustic perception disorders = 0-36,

Cognitive motor disorders = 0-18. (0 = no symptoms).

68

Concerning the analyses of differences in general functioning at baseline, we only found a significant difference in current GAF-score, with a lower score in the UHR+T group compared to the UHR+NT group (t=-2.2, p=.03). Both the difference in amount of decline in functioning (3.73) and the difference in GAF-score highest in the past year (2.46) between the two groups were not significant. Table 4 indicates the strength of the relationship between the variables that showed a difference at baseline between the UHR+T and the UHR+NT group. The most significant correlation at the .01 level was found between ‘Social anhedonia and withdrawal’ and ‘Decreased expression of emotion’. Using the Backward Stepwise method of the Cox proportional hazard model, it can be concluded that reporting social anhedonia and withdrawal symptoms is the best predictor for a subsequent psychosis in our UHR group (Wald=4.5, p=.03, RR=1.3).

Table 4: Spearman’s Rank Order Correlation, N=73

4. Discussion

Our study finds striking similar results as the study of Cannon et al. (2008). As well as the Cannon et al. study, our results show that the UHR subjects who later make the transition to psychosis have increased social anhedonia and withdrawal as assessed with the SIPS compared to the UHR subjects that do not make the transition to psychosis. In addition, our results concerning the lower general level of functioning and the higher amount of bizarre thinking in the subjects that later make a transition to psychosis are comparable with the results of Cannon et al. We found trends for increased suspiciousness and decreased expression of emotion in the UHR+T group. Although no conclusions can be drawn upon these latter relative powerless findings, we believe that they do give some clinical valuable information as they point in the same direction as the larger 2008 Cannon et al. study.

Chapter 4

Measures

(1) P2

(2) N1

(3) N3

(4) D2

(5) Genetic risk

(6) GAF-score

1

.37**

.30*

.02

-.02

-.18

2

.45**

.13

-.10

-.36**

3

.11

.07

-.18

4

.03

-.36**

5

.10

**= correlation is significant at the 0.01 level. *= correlation is significant at the 0.05 level.

P2= suspiciousness, N1= social anhedonia and withdrawal, N3= decreased expression in emotion, D2= bizarre thinking

69

Especially withdrawal from social interaction may be an important factor in the development of a psychosis. Because thoughts are not tested in social interaction, social withdrawal might contribute to the development and maintenance of delusional ideas and suspiciousness. Moreover, social with-drawal occurs at the expense of psychosocial development and it might reinforce other negative symptoms such as alogia and decreased expression of emotion. Otherwise, alogia and decreased expression of emotion may reinforce social withdrawal. Not only was the difference at baseline between the UHR+NT and the UHR+T groups statistically significant, it was also clinically discerna-ble in retrospect. Johnstone et al. (2005) also found social withdrawal to be predictive of a first psychosis. Addington et al. (2007) reported that social functioning in UHR patients was at the level of first episode schizophrenia patients. Social anhedonia and withdrawal may partly reflect a depressive state. However, the absence of a difference in comorbid affective disorders as opposed to the significant difference in social anhedonia and withdrawal between the UHR+T and UHR+NT groups, suggests that social anhedonia and withdrawal does not merely indicate a depressive state. In addition, there was no significant difference in mean score on the SIPS item ‘Dysphoric mood’ between both groups. Disentanglement of negative symptoms and depression remains an issue. Cognitive behavioural therapy is a promising method for intervening in the UHR period (Morrison et al., 2004). An important focus in cognitive behavioural therapy for UHR patients should be reducing social anhedonia and withdrawal and reducing suspiciousness, as these symptoms are often related. Furthermore, antidepressant medications may improve negative symptoms and may even prevent psychosis (Cornblatt et al., 2007; Fusar-Poli et al., 2007). In contrast with the 2001 study of Klösterkotter, we did not find significant more basic symptoms at baseline in the UHR+T group compared to the UHR+NT group. This contradiction may be contri-butable to their older patient group (mean=29.7 years old at baseline) and longer follow-up period. In our study, basic symptoms do not predict psychosis within a follow up period of three years. In the present study, cannabis users did not have a higher transition rate than non-cannabis users. Although Cannon et al. (2008) found a history of substance abuse to be predictive of conversion to psychosis, they did not find any specific subclass of their substances tested (i.e., alcohol, hypnotics, cannabis, amphetamines, opiates, cocaine and hallucinogen) to be significantly associated with risk. Our finding might also be attributable to the fact that we did not make a distinction between present and past use. Yung et al. (2006b) reported on differences in current GAF score within a similar group. In line with Yung et al. and Cannon et al. (2008) our study found that low functioning at baseline was significantly associated with later psychosis within the UHR group. Conversely, we did not find any differences in amount of decline and highest GAF score in the past year. Thus, the UHR+T subjects functioned comparable to the UHR+NT subjects at some point in the year before inclusion in the DUPS study.

Chapter 4

70

In the present study, eight subjects were lost to follow up. Therefore, we cannot be certain about their current state of functioning. A transition to psychosis is highly unlikely, since we asked the practitioner who referred the patients to contact us when they suspected a psychosis. In addition, many patients in the Amsterdam region are referred to our Adolescent Clinic in case of psychosis.In a commentary on the Cannon et al. (2008) study, Yung (2008) underlined that although the findings can be helpful in narrowing down the UHR inclusion criteria, it should be noticed that the predictive value of the newly found predictors of transition to psychosis in the general population is most likely much lower because of the reduction in sensitivity. Yung states that these criteria should therefore not be applied in the general population. More research is needed to determine the value of the newly found predictors as valid UHR inclusion criteria. Our study shows that a few differences are present at baseline in reported symptoms between those patients who eventually do and do not make the transition to psychosis. Our study is the first to replicate the Cannon et al. (2008) reported results. Especially the amount of negative symptoms merits specific attention in the future. Adding the negative item social anhedonia and withdrawal to the UHR inclusion criteria may improve prediction of transition to psychosis in UHR patients.

Acknowledgments

We thank Erik Wagemans for his help with the data.

Chapter 4

71

Chapter 4

References

Addington, J., Penn, D., Woods, S.W., Addington, D., Perkins, D.O. (2007). Social functioning in individuals at clinical

high risk for psychosis. Schizophrenia Research; 99(1–3): 119–124.

American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric

Association, Washington DC.

Addington, J., Walker, E., Seidman, L. J., Perkins, D., Tsuang, M., McGlashan, T., Heinssen, R. (2008).

Prediction of psychosis in youth at high clinical risk: a multisite longitudinal study in North America.

Archives of General Psychiatry; 65(1): 28–37.

Cannon, T. D., Cadenhead, K., Cornblatt, B., Woods, S.W., Addington, J., Walker, E., Seidman, L. J., Perkins, D.,

Tsuang, M., McGlashan, T., Heinssen, R. (2008). Prediction of Psychosis in Youth at High Clinical Risk: A Multisite

Longitudinal Study in North America. Archives of General Psychiatry; 65 (1): 28–37.

Cornblatt, B. A., Lencz, T., Smith, C. W., Olsen, R., Auther, A. M., Nakayama, E., Lesser,M. L., Tai, J. Y., Shah, M. R.,

Foley, C. A., Kane, J. M., Correll, C. U. (2007). Can antidepressants be used to threat the schizophrenia prodrome?

Results of a prospective, naturalistic treatment study of adolescents. Journal of Clinical Psychiatry; 68(4): 546–557.

Fusar-Poli, P., Valmaggia, L., McGuire, P. (2007). Can antidepressants prevent psychosis?

Lancet ; 370 (9601) : 1746–1748.

Kay, S. R., Fiszbein, A., Opler, L. A. (1987). The positive and negative symptom scale (PANSS).

Schizophrenia Bulletin; 13(2): 261–276.

Klosterkötter, J., Ebel, H., Schultze-Lutter, F., Steinmeyer, E. M. (1996). Diagnostic validity of basic symptoms.

European Archives of Psychiatry and Clinical Neuroscience; 246(3): 147–154.

Klosterkötter, J., Hellmich, M., Steinmeyer, E. M., Schultze-Lutter, F. (2001).

Diagnosing schizophrenia in the initial prodromal phase. Archives of General Psychiatry; 58(2): 158–164.

Klosterkötter, J., Ruhrmann, S., Schultze-Lutter, F., Salokangas, R. K. R., Linszen, D., Birchwood, M., Juckel,

G., Morrison, A., Vázquèz-Barqueri, J. L., Hambrecht, M., von Reventlow, H. (2005). The European Prediction of

Psychosis Study (EPOS): integrating early recognition and intervention in Europe. World Psychiatry; 4(3): 161–167.

Johnstone, E. C., Ebmeier, K. P., Miller, P., Owens, D. G. C., Lawrie, S. M. (2005). Predicting schizophrenia:

findings from the Edinburgh high risk study. British Journal of Psychiatry; 186(1): 18–25.

Miller, T. J., McGlashan, T. H., Rosen, J. L., Somjee, L., Markovich, P. J., Stein, K., Woods, S. W. (2002). Prospective

diagnosis of the initial prodrome for schizophrenia based on the structured interview for prodromal syndromes:

preliminary evidence of interrater reliability and predictive validity. American Journal of Psychiatry; 159(5): 863–865.

72

Chapter 4

Morrison, A. P., French, P., Walford, L., Lewis, S. W., Kilcommons, A., Green, J., Parker, S., Bentall, R. P. (2004).

Cognitive therapy for the prevention of psychosis in people at ultra-high risk: randomised controlled trial. British

Journal of Psychiatry; 185(4): 291–297.

Nieman, D., Becker, H., Fliert van der, R., Plat, N., Bour, L., Koelman, H., Klaassen, M., Dingemans, P., Niessen,

M., Linszen, D. (2007). Antisaccade task performance inpatients at ultra high risk for developing psychosis.

Schizophrenia Research; 95(1–3): 54–60.

Phillips, L. J., Leicester, S. B., O’Dwyer, L. E., Francey, S. M., Koutsogiannis, J., Abdel-Baki, A., Kelly, D., Jones, S.,

Vay, C., Yung, A. R.,McGorry, P. D. (2002). The PACE Clinic: identification and management of young people at “ultra”

high risk of psychosis. Journal of Psychiatric Practice; 8(5): 255–269.

Reading, B., Birchwood, M. (2005). Early intervention in psychosis, rationale and evidence for effectiveness.

Disease Management and Health Outcomes; 13(1): 53–63.

Spitzer, R. L., Williams, J.B., Gibbon, M., First, M. B. (1992). The Structured Clinical Interview for DSM-III-R (SCID).

I: history, rationale, and description. Archives of General Psychiatry; 49(8): 624–629.

Yung, A. (2008). Prognostic factors for progression to psychosis in high risk youth.

Evididence Based Mental Health; 11: 72.

Yung, A. R., McGorry, P. D., McFarlane, C. A., Jackson, H. J., Patton, G. C., Rakkar, A. (1996).

Monitoring and care of young people at incipient risk of psychosis. Schizophrenia Bulletin; 22(2): 283–303.

Yung, A. R., Phillips, L. J., Yuen, H. P., Francey, S. M., McFarlane, C. A., Hallgren, M., McGorry, P. D. (2003).

Psychosis prediction: 12-month follow up of a high risk (“prodromal”) group. Schizophrenia Research; 60(1): 21–32.

Yung, A. R., Buckby, J. A., Cotton, S. M., Cosgrave, E. M., Killackey, E. J., Stanford,C., Godfrey, K., McGorry, P. D.

(2006a). Psychotic-like experiences in nonpsychotic help-seekers: associations with distress, depression, and

disability. Schizophrenia Bulletin; 32(2): 352–359.

Yung, A. R., Stanford, C., Cosgrave, E., Killackey, E., Phillips, L., Nelson, B., McGorry, P. D. (2006b). Testing the Ultra

High Risk (prodromal) criteria for the prediction of psychosis in a clinical sample of young people. Schizophrenia

Research; 84(1): 57–66.

Yung, A. R., Yuen, H. P., Berger, G., Francey, S., Hung, T. C., Nelson, B., Phillips, L., McGorry, P. D. (2007). Declining

transition rate in ultra high risk (prodromal) services: dilution or reduction of risk? Schizophrenia Bulletin; 33(3):

673–681.

73

74

75

Chapter 5

DISABILITY IN PEOPLE CLINICALLY AT HIGH RISK OF PSYCHOSIS.

Eva Velthorst, Dorien H. Nieman, Don Linszen, Hiske Becker, Lieuwe de Haan, Peter M. Dingemans,Max Birchwood, Paul Patterson, Raimo K. R. Salokangas, Markus Heinimaa, Andreas Heinz, Georg Juckel, Heinrich Graf von Reventlow, Paul French, Helen Stevens, Frauke Schultze-Lutter, Joachim Klosterkötter, Stephan Ruhrmann, on behalf of the EPOS group

BRITISH JOURNAL OF PSYCHIATRY (2010); 197: 278-284

76

ABSTRACT Background Decline in social functioning occurs in individuals who later develop psychosis.

Aims To investigate whether baseline differences in disability are present in those who do and those who do not make a transition to psychosis in a group clinically at high risk and whether disability is a risk factor for transition.

Method Prospective multicentre, naturalistic field study with an 18-month follow-up period on 245 help-seeking individuals clinically at high risk. Disability was assessed with the Disability Assessment Schedule of the World Health Organization (WHODAS–II).

Results At baseline, the transition group displayed significantly greater difficulties in making new friends (z =73.40, P = 0.001), maintaining a friendship (z =73.00, P = 0.003), dealing with people they do not know (z =72.28, P = 0.023) and joining community activities (z =72.0, P = 0.05) compared with the non-transition group. In Cox regression, difficulties in getting along with people significantly contributed to the prediction of transition to psychosis in our sample (b = 0.569, s.e. = 0.184, Wald = 9.548, P = 0.002, hazard ratio (HR) = 1.767, 95% CI 1.238–2.550).

Conclusions Certain domains of social disability might contribute to the prediction of psychosis in a sample clinically at high risk.

Chapter 5

77

Chapter 5

Introduction

Psychotic disorders are associated with multiple social disabilities in work, study, independent living, interpersonal relations and self-care,1–4 and serious disability in functioning is one of the core features of the DSM–IV diagnosis of schizophrenia.2 Disability refers to the limitations in an indi-vidual’s ability to perform the expected socially and culturally defined roles and tasks.5 In schizophrenia, disability is thought to be mainly associated with negative symptoms and neurocognitive and social cognitive deficits.6,7 Functional decline has been found to occur before the first psychotic episode. In retrospective studies, Häfner et al8 and Larsen et al 9 found that deficits were already noticeable in the prodromal and premorbid phase of schizophrenia, respectively. Assessment scales frequently employed in prospective studies and developed for detecting individuals at an ultra-high risk of psychosis include the criterion of impaired functioning connected with genetic risk, as defined by either family history of psychosis or schizotypical personality disorder in the index person.10,11 Preliminary findings from recent studies indicate that decline in functioning may be a predictor of psychosis transition. However, these studies mainly focused on social contacts.7,8,12–14 In the present 18-month follow-up study of a large European population of individuals clinically at high risk of psychosis, disability was investigated in more detail by examining the following six domains: understanding of and interaction with the world; moving and getting around; self-care; getting along with people; life activities; and participation in society. In line with the major focus of this multicentred, prospective naturalistic field study to identify risk factors for the transition from pro-drome to a first psychotic episode,15 the aim of the present analyses is to investigate: first, differ-ences in disability at baseline between those who do and those who do not make the transition to psychosis in a group at high clinical risk. And second, whether disability is a risk factor for transition to a first psychosis. For exploratory purposes, we further sought to investigate the association between social disability and clinical symptomatology. We hypothesised that the transition group would display significantly greater disability at baseline than the non-transition group.

78

Chapter 5

Figure 1: Flow diagram of participants referred to the European Prediction of Psychosis Study.

Assessed for eligibility (n=661)

Fulfilling ‘clinically at high risk criteria (n=513)

Reasons for non-participation

- Refusal (140)

- Physical illness (7)

- Symptoms due to substance abuse (29)

- Verbal IQ<85 (12)

- Opposition from relative/ profess. (11)

Reasons for dropout:

- Not available for further assessment (38)

- Consent withdrawn (10)

- Suicide (1)

- Other (e.g. unable to locate participant (9))

Informed consent (n=245)

WHODAS-II data (n=239)

Lost to follow-up (n=58)

- Within 9 months (n=35)

- Between 9 and 18 months (n=23)

79

Chapter 5

Instruments

‘Clinically at high risk’ symptomatologyThe SIPS 3.0,17 including GAF–M, was employed to determine the presence, severity and type of ultra-high-risk criteria. The Scale of Prodromal Symptoms (SOPS), the rating scale of the SIPS, has four subscales that include five positive symptom items, six negative symptom items, four disorganisation symptoms items and four general symptom items. All symptoms are rated on a seven-point rating scale anchored from 0 (never, absent) to 6 (severe/extreme – and psychotic for the positive items). Symptomatic criteria for prodromal state are exclusively based on positive symptom items and the disorganisation symptom odd behaviour/appearance. The EPOS investigators received extensive training from Dr Tandy Miller, one of the SIPS authors, including a reliability check after approximately 6 months. The pair-wise interrater concordance of the SIPS was 77% and determined acceptable by the training team. The Bonn Scale for the Assessment of Basic Symptoms – Prediction list (BSABS–P)18 was compiled to assess cognitive disturbances. To the item collection of the BSABS–P, three theoretical subscales were defined totalling 33 cognitive, perceptual and motor disturbances assessed on a sev-en-point severity scale (0–6) with maximum frequency of occurrence during the preceding 3 months as the guiding criterion. The investigators received repeated training by the scale’s first author (F. Schultze-Lutter). Concordance rate with expert rating (F.S.L.) was 87.9%.

Disability

Disability was assessed with the World Health Organization’s Disability Assessment Schedule (WHODAS–II).3,24 The WHODAS– II assesses disability in individuals irrespective of diagnosis. The nature of disability is rated directly from individuals’ responses.24 The questionnaire has been conceptually linked to the International Classification of Functioning, Disability and Health, a model of functioning and disability that systematically organises the consequences of disease into three dimensions: body functions and structure, activities, and participation.5 The 36-item interview-er-administered WHODAS–II assesses functioning and disability during the past 30 days and covers the following activity domains:(a) domain 1: understanding and interacting with the world (six items); (b) domain 2: moving and getting around (five items); (c) domain 3: self-care (four items);(d) domain 4: getting along with people (five items);(e) domain 5: life activities (eight items):(i) household activities (four items)(ii) work activities (four items);(f) domain 6: participation in society (eight items).

80

Chapter 5

The domain life activities is divided in household activities (four items) and work activities (four items) and because of the small percentage of participants to whom this latter subdomain applied we looked at the subdomain household activities separately. In both subdomains, the amount of time spent on the activities was additionally recorded. The scale’s general items were rated on a five-point severity scale, i.e. from 1 (no dysfunction) to 5 (serious to maximum dysfunction).In addition to the scores on each single item and in congruence with Chopra et al,3 we determined an average score for each domain, thereby allowing a comparison between groups in the WHODAS–II profiles. Hence, the average score for each participant could range from 1 (indicating no reported difficulty) to 5 (indicating severe difficulty) in each domain.3 Until now, no study has been conducted on the specific psychometric properties of the WHODAS–II when used with people at clinically high risk for psychosis. However, the validity of the instrument has been tested in numerous other patient populations, suggesting acceptable internal consistency, test–retest reliability and convergent validity.5,24–26 In a validation study of 904 participants across ten different conditions (for example depression), the WHODAS–II has been shown to have good reliability (Cronbach’s a ranged from 0.70 to 0.97 for the different subscales).25 Further, Chopra et al24 and McKibbin et al27 have found the WHODAS–II to have fair test–retest reliability in people with long-term psychotic disorders despite deficiencies in reality testing. In spite of this demonstrated satisfying reliability, the two items ‘getting dressed’ and ‘problems because of barriers or hindrances in the world around you’ were thought to induce equivocal inter-pretation within our young sample. ‘Hindrances in the world around you’, for example, may have been misinterpreted as hindrances caused by other people. Only those who interpreted hindrances as physical (for example, hindrances caused by a wheelchair) understood the question correctly. Because of the questionable reliability of these two items in our group, they were not entered in the analyses and not considered further.

Follow-up

Follow-up assessments with the SIPS and BSABS–P took place at 9 and 18 months. Transition to psychosis was operationalised as a continuation of BLIPS, i.e. any single item on the positive subscale of SIPS (SIPS–Positive) with a score of 6 for more than 7 days.13,28,29 Following identification of full-blown psychotic symptoms in the SIPS interview, the diagnostic category on transition was determined by applying DSM–IV criteria22 of psychotic disorders and affective disorders with psychotic features. Thereby, the different time threshold of Criterion B of ‘brief psychotic disorder’ was adapted to the BLIPS definition. Past and present psychosis as parts of the exclusion criteria as well as psychotic diagnosis on transition were assessed with the SCID–I.23

81

Chapter 5

Statistical analysis

Statistical analyses were performed using SPSS (version 16.0.2) for Windows. Comparisons between the transition group and non-transition group were made with Pearson’s chi-squared tests and an independent sample t-test for the GAF score. To examine differences in the categorical domains as well as in the separate categorical items of the WHODAS–II, Mann– Whitney U-tests were used. Kaplan–Meier survival analysis was used to calculate survival time, i.e. time to transition. The effect of different covariates on time to transition was estimated with the Cox proportional hazard model. In this model, transition to psychosis was entered as the status variable, time to onset of psychosis or to follow-up as the time variable and the weighted sum scores of the WHODAS–II domains (with the exception of work activities) as covariates. Weighted sum scores of each domain were computed by dividing the sum score of all items (except for the items ‘getting dressed’ and ‘problems because of barriers or hindrances in the world around you’) with the number of items. Subsequently, these scores were entered into the analysis as ordinal data using the backward Wald method. Treatment was entered as an additional covariate in the model using the blockwise method. The recruiting centre was entered as a stratum variable in the regression model. P<0.05 was considered statistically significant. A survival curve was obtained using Kaplan–Meier survival analysis. Bivariate correlations between social disability and SIPS symptoms were evaluated using Spearman’s rank correlation. We adjusted the type I error, i.e. alpha, for multiple testing by dividing the overall alpha level of 0.05 by the number of items. After this correction, P≤.0013 was considered statistically significant.

82

Chapter 5

ResultsParticipant characteristics

At 18-month follow-up, 37 participants had made a transition to psychosis. The transition group did not differ significantly from the non-transition group in terms of age, gender and drug use. In congru-ence with results of previous studies,13,15 we found a difference in the level of global functioning (GAF score) at baseline (P<0.005), with a significantly lower GAF score in the transition group (42.4, s.d.= 1.6) compared with the non-transition group (52.7, s.d.= 0.8). Only four participants met all three ultra-high-risk criteria, three of whom also met cognitive disturbances criteria. In total 143 participants displayed both ultra-high-risk and cognitive distur-bances symptoms, 72 met the ultra-high-risk criteria, and 24 reported only cognitive disturbances symptoms. These three cohorts did not significantly differ in transition rates. The mean time to tran-sition from baseline examination was 496.8 days (s.e.= 8.5, 95% CI 480.2–513.6). A total of 239 participants completed the WHODAS–II at baseline (131 males, mean age 22.5, s.d.= 5.3). Of these, 35 were lost to follow-up at 9 months and altogether 58 at 18 months. Participants did not differ significantly from those lost to follow-up in terms of age, gender, severity of negative symptoms, severity of disability, global functioning and familial risk. However, differences were found on two positive symptom items of the SIPS; those lost to follow-up had a lower score on suspiciousness (t =72.07, P= 0.040) and a higher score on hallucinations at baseline (t = 2.15, P= 0.030). As regards medication, antidepressants were prescribed to 45 (18.8%) of the 239 partici-pants, antipsychotics to 29 (12.1%) and both antidepressants and antipsychotics to 22 (9.2%); for 30 (12.6%) participants no reliable information on their medication was available.

Disability

Online Table 1 displays the overall scores of the domains as assessed by the WHODAS–II. Within the entire group, the highest disability scores were found in domain 6: participation in society (mean 2.38, s.d.= 0.96, median 2.29). Mann–Whitney U-tests indicate that the baseline ratings of the transition group were significantly higher than those of the non-transition group only in domain 4, getting along with people (z =73.03, P = 0.002). This difference was mainly caused by the items making new friends (z =73.40, P= 0.001), maintaining a friendship (z =73.00, P= 0.003) and dealing with people you do not know (z =72.28, P= 0.023).

83

Chapter 5

Table 1: Mean ratings of the transition group and the non-transition group at baselinea

Disabilities domains

1. Understanding and communication

Concentrating on doing something

Remembering to do important things

Analysing and finding solutions for problems

Learning a new task

Generally understanding

Starting/ maintaining a conversation

2. Getting around

Standing for long periods

Standing up from sitting down

Moving around inside your home

Getting out of your home

Walking a long distance

3. Self-care

Washing your whole body

Eating

Staying by yourself for a few days

4. Getting along with people

Dealing with people you do not know

Maintaining a friendship

Getting along with people who are close to you

Making new friends

Sexual activities

5. Life activities*

Taking care of your household responsibilities

Doing your most important householdtasks well

Getting all the housework done that you needed to do

Getting your household work done as quickly as needed

6. Participation in society

Joining in community activities

Living in dignity because of attitudes/actions of others

Amount of time spend on health conditions, or its conseq.**

Emotionally affected by your health condition

Drain on the financial resources of you/ your family

Problems of your family because of your health problems

Doing things by yourself for relaxation or pleasure

Transition group

(N=37)

2.11

2.54

2.08

2.22

1.70

1.78

2.35

1.58

1.86

1.41

1.24

1.86

1.54

1.51

1.19

1.57

1.78

2.49

2.62

2.73

2.14

3.14

1.81

1.91

2.08

1.68

2.00

1.86

2.71

2.54

2.89

2.76

3.03

2.43

2.62

2.70

Non-transition group

(N=202)

2.02

2.02

2.12

2.23

1.78

1.69

2.16

1.39

1.64

1.23

1.10

1.58

1.37

1.46

1.21

1.49

1.68

1.97

2.06

1.96

1.94

2.21

1.70

1.90

1.97

1.78

1.90

1.94

2.31

2.14

1.57

2.59

3.18

1.85

2.39

2.49

StatisticsZ

-0.82

-1.76

-0.11

-0.24

-0.09

-1.16

-0.67

-1.56

-1.12

-0.88

-1.50

-1.39

-0.85

-0.29

-0.78

-0.86

-0.72

-3.03

-2.28

-3.00

-0.91

-3.40

-0.48

-0.25

-0.68

-0.53

-0.60

-0.37

-1.53

-1.97

-1.12

-0.74

-0.89

-1.84

-0.87

-0.78

P

0.41

0.08

0.92

0.81

0.93

0.25

0.51

0.12

0.27

0.38

0.13

0.16

0.40

0.77

0.43

0.39

0.47

0.002

0.02

0.003

0.37

0.001

0.63

0.80

0.50

0.59

0.55

0.72

0.13

< 0.05

0.27

0.46

0.37

0.07

0.39

0.44

**= correlation is significant at the 0.01 level. *= correlation is significant at the 0.05 level.

P2= suspiciousness, N1= social anhedonia and withdrawal, N3= decreased expression in emotion, D2= bizarre thinking

84

Chapter 5

Although no significant difference was found for average scores in domain 6, participation in society, a difference was found within its component item, joining community activities (z =72.0, P= 0.05), with the transition group reporting more difficulties in this area. Questions related to work activities could not be answered by 33% of the participants (37% of the transition group and 31% of the non-transition group), mostly because of participants not having been employed or otherwise working (i.e. self-employed, studying, in school or training or doing housework etc.) in the past 30 days before the interview. For the remaining participants, the difference between the two groups in number of hours spent on work or study is approaching significance (t =71.93, P =0.056), with the transition group working less total hours than the non-transition group. The amount of missing data because of the absence of work/study in the subdomain work activities renderedfurther interpretation of this section impracticable.

Cox regression analysis

The backward stepwise Cox proportional hazard model was used to identify the predictive value of the weighted sum scores of the six domains, except for the subdomain work activities. In this analysis, domain 4, getting along with people, was retained in the model as a predictor of a first psychotic episode (b = 0.672, s.e.= 0.187, Wald = 12.964, P<0.001, hazard ratio (HR) = 1.958, 95% CI 1.358–2.822). Thus the relative risk of developing a psychosis nearly doubled with increasing difficulties in contact with other people. None of the other covariates further contributed independently to the prediction of a first psychotic episode. Adding treatment to the model, the use of antipsychotics (b =71.682, s.e.= 0.534, P= 0.002) and antidepressants (b =72.125, s.e.= 0.743, P= 0.004) were kept in the equation. Domain 4, getting along with people, continued to contribute significantly to the equation, and the change in hazard ratio was only minor (b = 0.569, s.e.= 0.184, Wald = 9.548, P= 0.002, HR= 1.767, 95% CI 1.238–2.550). The survival curves of the two disability groups for the domain getting along with people are presented in Fig. 2; the rate of transition differed significantly between the group with moderate to severe disability (n = 72) and the group with no to mild disability (n = 167) in this domain (X2 = 11.91, P= 0.001).

85

Chapter 5

Figure 2: Survival analysis for 18-month follow-up (n=239).

Spearman’s rank correlation

Table 2 displays the correlations between symptoms as assessed with the SIPS and social disability before (P<0.05) and after Bonferroni adjustment (Padjusted<0.0013). We found disability in the fourth domain, getting along with people, and the sixth domain, participation in society, to be related to both negative and general symptoms of the SIPS. Although not surviving Bonferroni adjustment, there was an additional association between social disability and the SIPS positive item suspiciousness. The highest correlations, however, were found between getting along with people and the negative symptom social anhedonia and withdrawal as well as between participation in society and the general symptom dysphoric mood.

DIFFICULTIES IN GETTING ALONG WITH PEOPLE

NONE OR MILD (n = 167)

MODERATE TO SERVERE (n = 72)

TIME (DAYS IN STUDY)

CU

MU

LATI

VE

SUR

VIV

AL

86

Chapter 5

Measures a

P1. Unusual thought content/ delusional ideas

P2. Suspiciousness/ persecutory ideas

P3. Grandiose ideas

P4. Perceptual abnormalities/ hallucinations

P5. Disorganized communication

N1. Social anhedonia and withdrawal

N2. Avolition

N3. Decreased expression of emotion

N4. Decreased experience of emotions and self

N5. Decreased ideational richness

N6. Deterioration in role functioning

D1. Odd behaviour or appearance

D2. Bizarre thinking

D3. Trouble with focus and attention

D4. Personal hygiene/ social attentiveness

G1. Sleep disturbance

G2. Dysphoric mood

G3. Motor disturbances

G4. Impaired tolerance to normal stress

Getting along with people b Participation in society b

Rho

0.11

0.20

-0.01

-0.003

0.09

0.36

0.24

0.30

0.23

0.09

0.24

0.22

0.17

0.17

0.16

0.16

0.27

0.18

0.27

P

0.11

0.002*

0.90

0.97

0.15

<0.001**

<0.001**

<0.001**

<0.001**

0.16

<0.001**

0.001**

0.01*

0.009*

0.02

0.01*

<0.001**

0.006*

<0.001**

Rho

0.24

0.17

-0.05

0.05

0.14

0.29

0.33

0.23

0.17

0.08

0.29

0.17

0.09

0.22

0.12

0.21

0.36

0.27

0.34

P

0.71

0.008*

0.44

0.41

0.04

<0.001**

<0.001**

<0.001**

0.01*

0.20

<0.001**

0.008*

0.16

0.001**

0.06

0.001**

<0.001**

<0.001**

<0.001**

Table 2. Spearman’s Rank Order Correlations between Social Disability and Clinical symptomatology, (n=239).

a P1 to G4 the items from the Structured interview of Prodromal Syndromes.

b “Getting along with people” and “Participating in the society” are two domains of the Disability Assessment Schedule

of the World Health Organization (WHODAS–II) focusing on social contact.

Results in bold are significant: * significant at the 0.01 level; ** significant after Bonferroni correction (P50.00013).

87

Chapter 5

Discussion

In the present study on the role of disability in the transition to psychosis, we employed a validated and reliable questionnaire to measure a wide range of disability in a large sample of young people clinically at high risk for a psychosis. To our knowledge, our study is not only the first prospective study to use the WHODAS–II in a large clinical high-risk sample but also the first prospective study that in detail examined disability as a risk factor of a transition to psychosis. As expected, we found that within this help-seeking population, the transition group was experiencing significantly greater social disability at baseline compared with the non-transition group. Difficulties in the domain getting along with people appeared to contribute most to the prediction of the onset of a first psychotic episode within a study of people clinically at high risk. Within this category, developing or maintaining friendships appeared to be particularly problematic for those later developing psychosis, since those who made a transition reported significantly greater difficulties at baseline in both developing and maintaining friendships compared with those who did not make the transition. Although we only compared disability in cross-section, such difficulties may well be progressive. In a representative group of young people with early-onset psychosis, Boeing et al30 found that up to 82% of the sample experienced a moderate to severe ‘lack’ of friendships during their first psychotic episode. This is a noteworthy finding, since strong social networks are related to both better health status and better quality of life.31 Furthermore, our correlation analyses suggest that social disability in individuals clinically at high risk may be mainly related to negative and general symptoms although a modest association was found between social disability and the positive SIPS item suspiciousness. Difficulties in getting along with people showed high correlations with negative symptoms, and disturbances in participating in society were mainly associated with general symptoms such as dysphoric mood. Since recent studies have found strong associations between negative symptoms and transition to psychosis in a sample clinically at high risk,13,14 it is not surprising that difficulties in getting along with people was the best predictor of a first psychotic episode in our sample. Our findings on social disability in our sample are also in line with previous studies of individuals clinically at high risk, in which transition was associated with more severe social anhedonia and withdrawal.13,14,32,33 In a retrospective study, participants diagnosed with schizophrenia frequently increasingly lagged behind their healthy peers in social development from an early, pre-psychotic state on.8,34 Using the Premorbid Adjustment Scale, Shapiro et al35 found social abnormalities in childhood even before the age of 11.

88

Chapter 5

These studies suggest that social deficits may precede positive ultra-high-risk symptoms. Such deficits may be caused by the early onset of attenuated negative symptoms as suggested by the clinical high-risk model of Cornblatt et al.36 Negative and general affective symptoms may cause the adolescent to withdraw from increasingly demanding and complex social contacts when adult roles have to be taken up.8 In addition, the reported negative attitude of others when the individual is in the clinically at high risk phase37 may also contribute to this social withdrawal. Irrespective of the possible nature of any interaction between negative and positive symptoms and social disability, poor premorbid social functioning has been related to elevated risk and poor outcomes in psychotic disorders,6 thus making it a significant target for an early intervention in the at-risk state. Improving social functioning, for example through cognitive– behavioural therapy in the clinically at high risk period, may protect the adolescent or young adult from social isolation, and maintaining reality testing with peers may reduce the risk of later delusional ideation. In addition, to improve psychosis prediction, an assessment of disability in people clinically at high risk may be relevant in identifying important needs, for both the individual and their families.38 Help in meeting these needs through appropriate early intervention is likely to also improve the quality of life of individuals clinically at high risk and, thereby, to diminish the likelihood of transition to psychosis.

Limitations

In spite of the outlined strengths of our study, some critical issues regarding the use of the WHODAS–II in a clinically at high risk sample need to be addressed. First, the WHODAS–II questionnaire does not take into account some of the unique social issues that occur in adolescence. In their study of individuals at ultra-high risk, Cornblatt et al7 employed two new, broad measures (social and role scale) based on the Social and Occupational Functioning Assessment Scale (SOFAS) and GAF respectively to identify disability, incorporating specific issues such as peer acceptance and dating. An additional consideration of these scales in future research may help to decide whether these are superior to the WHODAS–II in younger samples and/or whether they further contribute to improvements in psychosis prediction. Second, a methodological issue must be considered. As fitted models always perform in an ‘optimistic manner’39 in the model-development data, cross-validation in an independent sample is needed to control for tailor-made modelling. Although sample splitting in theory is an option for model validation in large samples, the limited number of transitions did not allow it for statistical reasons. Existing or future samples of comparable size and risk definition are required to validate our findings.

89

Chapter 5

Another potential limitation concerns the 9-month gap in between assessments that may have introduced some inaccuracies in recall, particularly of transition. Indeed, some participants may have had some difficulties in distinguishing their attenuated symptoms from full-blown psychotic symp-toms after several months, whereas others may have failed to recall a past psychotic episode at the time of the interview. This latter possibility, however, seems unlikely as researcher staff were trained to probe for transition at follow-up assessments. Additionally, if the participant had not remained in the care of the specialised centre but was back-referred, the referring practitioner as well as the participant and his/her family had been asked to re-establish contact as soon as they suspected psy-chosis or if the mental state of the individual deteriorated notably. Finally, once participants made a transition to a psychosis, they were not included in follow-up assessments of the WHODAS–II. Thus it will not be possible to further examine the course and sever-ity of disability after the onset of psychoses. Addressing this issue in future research is recommended.In conclusion, our study gives further evidence that a decline in social functioning may be a risk factor for a first psychosis within samples clinically at high risk.13,15 In combination with other predictors (neurobiological or psychopathological parameters), social disability measures may make a valuable contribution to a more accurate prediction of first psychotic episodes.

Acknowledgements

The International Advisory Board ‘European Prediction of Psychosis’ (EPOS) group comprises: Patrick D. McGorry, Australia; Thomas H. McGlashan, USA; Martin Knapp, UK.

The authors thank the following scientists for their work for EPOS: Reinaud van der Fliert, MSc, Rianne Klaassen, MD (Department of Psychiatry, AMC, University of Amsterdam, the Netherlands); Heinz Picker, PhD, Meike Neumann, Dipl-Psych, Anke Brockhaus-Dumke, MD, Ralf Pukrop, PhD (Department of Psychiatry & Psychotherapy/ FETZ, University of Cologne, Cologne, FRG); Tanja Svirskis, MD, PhD, Jukka Huttunen, BM, Tiina Laine, MSc, Tuula Ilonen, PhD, Terja Ristkari, MA, Jarmo Hietala, MD, PhD, (Department of Psychiatry, University of Turku, Turku, Finland); Amanda Skeate, DClin (ED:IT, Early Intervention Service, Birmingham, UK); Yehonala Gudlowski, Dipl-Psych (Charite´, Berlin, FRG); Seza Ozgu¨ rdal, Dipl-Psych, Henning Witthaus, MD (Charite´ , Berlin, and Ruhr-University, Bochum, FRG); Shoˆ n Lewis, MD, PhD, Antony Morrisson, PhD (School of Psychological Sciences, University of Manchester, Manchester, UK).

90

Chapter 5

References

1 Wiersma, D., Wanderling, J., Dragomirecka, E., Ganev, K., Harrison, G., van der Heiden ,W., et al. (2000).

Social disability in schizophrenia: its development and prediction over 15 years in incidence cohorts in six

European centres. Psychological Medicine; 30: 1155–67.

2 Burns, T., Patrick, D. (2007). Social functioning as an outcome measure in schizophrenia studies.

Acta Psychiatrica Scandinavica; 116: 403–18.

3 Chopra, P., Herrman, H., Kennedy, G. (2008). Comparison of disability and quality of life measures in patients

with long-term psychotic disorders and patients with multiple sclerosis: an application of the WHO Disability

Assessment Schedule II and WHO quality of Life-BREF. International Journal of Rehabilitation Research; 31: 141–9.

4 Juckel, G., Schaub, D., Fuchs, N., Naumann, U., Uhl, I., Witthaus, H., et al. (2008). Validation of the Personal

and Social Performance (PSP) Scale in a German sample of acutely ill patients with schizophrenia.

Schizophrenia Research; 104: 287–93.

5 Chwastiak, L., Von Korff, M. (2003). Disability in depression and back pain.

Evaluation of the World Health Organization Disability Assessment Schedule (WHODAS II) in a primary

care setting. Journal of Clinical Epidemiology; 56: 507–14.

6 Gureje, O., Herrman, H., Harvey, C., Morgan, V., Jablensky, A. (2002).

The Australian National Survey of Psychotic Disorders: profile of psychosocial disability and its risk factors.

Psychological Medicine; 32: 639–47.

7 Cornblatt, B. A., Auther, A. M., Niendam, T., Smith, C. W., Zinberg, J., Bearden, C. E., et al. (2007).

Preliminary findings for two new measures of social and role functioning in the prodromal phase of

schizophrenia. Schizophrenia Bulletin; 33: 668–702.

8 Häfner, H., Nowotny, B., Löffler, W. (1995). When and how does schizophrenia produce social deficits?

European Archives of Psychiatry and Clinical Neuroscience; 246: 17–28.

9 Larsen, T. K., McGlashan, T. H.. Johannessen, J. O., Vibe-Hansen, L. (1996). First-episode schizophrenia: II.

Premorbid patterns by gender. Schizophrenia Bulletin; 22: 257–69.

10 Yung, A. R., McGorry, P. D., McFarlane, C. A., Jackson, H. J., Patton, G. C., Rakkar, A. (1996).

Monitoring and care of young people at incipient risk of psychosis. Schizophrenia Bulletin; 22: 283–303.

11 Miller, T. J., McGlashan, T. H., Rosen, J. L., Somjee, L., Markovich, P. J., Stein, K., et al. (2002)

Prospective diagnosis of the initial prodrome for schizophrenia based on the Structured Interview for Prodromal

Syndromes: preliminary evidence of interrater reliability and predictive validity. American Journal of Psychiatry;

159: 863–5.

91

Chapter 5

12 Yung, A. R., Buckby, J. A., Cotton, S. M., Cosgrave, E. M., Killackey, E. J., Stanford, C., et al. (2006).

Psychotic-like experiences in nonpsychotic help-seekers: associations with distress, depression, and disability.

Schizophrenia Bulletin 2006; 32: 352–9.

13 Cannon, T. D., Cadenhead, K., Cornblatt, B., Woods, S. W., Addington, J., Walker, E., et al. (2008) Prediction

of psychosis in youth at high clinical risk. Archives of General Psychiatry; 65: 28–37.

14 Velthorst, E., Nieman, D. H., Becker, H. E., van der Fliert, R., Dingemans, P. M., Klaassen, R., et al. (2009).

Baseline differences in clinical symptomatology between ultra high risk subjects with and without a transition

to psychosis. Schizophrenia Research; 109: 60–5.

15 Ruhrmann, S., Schultze-Lutter, F., Salokangas, R. K. R., Heinimaa, M., Linszen, D., Dingemans, P., et al. (2010).

Prediction of psychosis in adolescents and young adults at high risk: results from the Prospective European

Prediction of Psychosis Study (EPOS). Archives of General Psychiatry; 67: 241–51.

16 Klosterkötter, J., Ruhrmann, S., Schultze-Lutter, F., Salokangas, R. K., Linszen, D., Birchwood, M., et al. (2005).

The European Prediction of Psychosis Study (EPOS): integrating early recognition and intervention in Europe.

World Psychiatry; 4: 161–7.

17 McGlashan, T. H., Miller, T. J., Woods, S. W., Rosen, J. L., Hoffman, R. E., Davidson, L. (2001).

Structured Interview for Prodromal Syndromes (Version 3.0). PRIME Research Clinic, Yale School of Medicine.

18 Schultze-Lutter, F., Addington, J., Ruhrmann, S., Klosterkötter, J. (2007).

Schizophrenia Proneness Instrument, Adult Version (SPI-A). Giovanni Fioriti Editore.

19 Kay, S. R., Fiszbein, A., Opler, L. A. (1987) The positive and negative symptom scale (PANSS).

Schizophrenia Bulletin; 13: 261–76.

20 Hall, R. C. (1995).Global assessment of functioning.

A modified scale. Psychosomatics; 36: 267–75.

21 Hall, R. C., Parks, J. (1995).

The modified global assessment of functioning scale: addendum. Psychosomatics; 36: 416–7.

22 American Psychiatric Association (1994).

Diagnostic and Statistical Manual of Mental Disorder (4th edn) (DSM–IV). APA.

23 First, M. B., Spitzer, R. L., Gibbon, M., Williams, J. B. W. (1997).

Structured Clinical Interview for DSM–IV Axis I Disorders (SCID–I). American Psychiatric Publishing.

24 Chopra, P. K., Couper, J. W., Herrman, H. (2004). The assessment of patients with longterm psychotic disorders:

application of the WHO Disability Assessment Schedule II. The Australian and New Zealand Journal of

Psychiatry; 38: 753–9.

92

25 Pösl, M., Cieza, A., Stucki, G. (2007).

Psychometric properties of the WHODAS-II in rehabilitation patients. Quality of Life Research; 16: 1521–31.

26 Federici, S., Meloni, F., Mancini, A., Lauriola, M., Belardinelli, M. O. (2009). World Health Organisation Disability

Assessment Schedule II: contribution to the Italian validation. Disability Rehabilitation; 31: 553–64.

27 McKibbin, C., Patterson, T. L., Jeste, D. V. (2004). Assessing disability in older patients with schizophrenia.

Results from the WHODAS-II. Journal of Nervous and Mental Disease; 192: 405–13.

28 Yung, A. R., Phillips, L. J., Yuen, H. P., McGorry, P. D. (2004). Risk factors for psychosis in an ultra high-risk

group: psychopathology and clinical features. Schizophrenia Research; 67: 131–42.

29 Yung, A. R., Nelson, B., Stanford, C., Simmons, M. B., Cosgrave, E. M., Killackey, E., et al. (2008).

Validation of ‘‘Prodromal’’ criteria to detect individuals at ultra high risk of psychosis: 2 year follow-up.

Schizophrenia Research; 105: 10–7.

30 Boeing, L., Murray, V., Pelosi, A., McCabe, R., Blackwood, D., Wrate, R. (2007).

Adolescent onset psychosis: prevalence, needs and service provision. British Journal of Psychiatry; 190: 18–26.

31 Becker, T., Leese, M., McCrone, P., Clarkson, P., Szmukler, G., Thornicroft, G. (1998).

Impact of community mental health services on users’ social networks. PRiSM Psychosis Study 7.

British Journal of Psychiatry; 173: 404–8.

32 Mason, O., Startup, M., Halpin, S., Schall, U., Conrad, A., Carr, V. (2004).

Risk factors for transition to first episode psychosis among individuals with ’at-risk mental states’.

Schizophrenia Research; 71: 227–37.

33 Riecher-Rössler, A., Gschwandtner, U., Aston, J., Borgwardt, S., Drewe, M., Fuhr, P., et al. (2007).

The Basel early-detection-of-psychosis (FEPSY)-study – design and preliminary results.

Acta Psychiatrica Scandinavica; 115: 114–25.

34 Rössler, W., Salize, H. J., van Os, J., Riecher-Rössler, A. (2005).

Size of burden of schizophrenia and psychotic disorders. European Neuropsychopharmacology; 15: 399–409.

35 Shapiro, D. I., Marenco, S., Spoor, E. H., Egan, M. F., Weinberger, D. R., Gold, J. M. (2009).

The Premorbid Adjustment Scale as a measure of developmental compromise in patients with schizophrenia

and their healthy siblings. Schizophrenia Research; 112: 136–42.

36 Cornblatt, B. A., Lencz, T., Smith, C. W., Correll, C. U., Auther, A. M., Nakayama, E. (2003).

The schizophrenia prodrome revisited: a neurodevelopmental perspective. Schizophrenia Bulletin; 29: 633–51.

Chapter 5

93

Chapter 5

37 Salokangas, R. K. R., Heinimaa, M., Svirskis, T., Laine, T., Huttunen, J., Ristkari, T., et al

and the EPOS Group (2009). Perceived negative attitude of others as an early sign of psychosis.

European Psychiatry; 24: 233–8.

38 Herrman, H., Hawthorne, G., Thomas, R. (2002). Quality of life assessment in people living with psychosis.

Social Psychiatry and Psychiatry Epidemiology; 37: 510–8.

39 Hosmer, D. W., Lemeshow, S. (2000). Applied Logistic Regression (2nd edn). John Wiley & Sons.

94

95

Chapter 6

THE STRAUSS AND CARPENTER PROGNOSTIC SCALE PREDICTS TRANSITION TO PSYCHOSIS IN SUBJECTS CLINICALLY AT HIGH RISK FOR PSYCHOSIS.

Dorien H. Nieman, Eva Velthorst, Hiske Becker, Lieuwe de Haan, Peter Dingemans, Björn Krefft, Don Linszen, Max Birchwood, Paul Patterson, Raimo KR Salokangas, Markus Heinimaa, Andreas Heinz, Georg Juckel, Heinrich Graf von Reventlow, Anthony Morrison, Shôn Lewis, Frauke Schultze-Lutter, Joachim Klosterkötter, Stephan Ruhrmann.

SUBMITTED

96

ABSTRACT

Objective: To investigate the predictive value of the Strauss and Carpenter Prognostic Scale (SCPS) for transition to a first psychotic episode in subjects clinically at-risk (AR) for psychosis.

Methods: 244 AR subjects participating in the European Prediction of Psychosis Study (EPOS) were assessed with the SCPS, an instrument that has been shown to predict outcome in schizophrenia patients reliably. The follow up period was 18 months.

Results: At 18 months, 37 participants had made the transition to psychosis. SCPS items that remained as independent predictors in the Cox proportional hazard model were: Most usual quality of useful work in the past year (Wald=7.61, p=.006, hazard ratio (HR)= 1.49, 95% CI=1.12 / 2.0), Quality of social relations (Wald=7.39, p=.007, HR=1.39, 95% CI=1.10 / 1.77), Presence of thought disorder, delusions, hallucinations, or hallucinations in the past year (Wald=10.33, p=.001, HR=1.88, 95% CI=1.28 / 2.75) and Reported severity of subjective distress in past month (Wald=8.51, p=.004, HR=.62, 95% CI=.45 / .86). The SCPS total score was also highly predictive of a first psychotic episode by itself (Wald=18.40, p<.0001, HR=1.10, 95% CI=1.05 / 1.15).

Conclusions: To the best of our knowledge, the present study is the first to reveal that the SCPS has predictive value for transition to psychosis within 18 months in an AR sample, suggesting temporal continuity of predictors of outcome. In combination with other variables, the SCPS could make a valuable contribution to a more accurate prediction of a first psychotic episode in vulnerable populations.

Chapter 6

97

Introduction

Predicting the course of illness in schizophrenia patients has been the topic of numerous papers.1-10 Several variables with good predictive validity have been identified repeatedly such as gender, premor-bid functioning, age of illness onset, previous psychiatric illness and life stress.3,4 One of the most commonly used questionnaires in outcome prediction research is the Strauss and Carpenter Prognostic Scale (SCPS).10 Items of this scale have been found to predict outcome at 2, 5 and 11 years follow up.7-9 In these studies, the majority of the patients was diagnosed with schizophrenia and the patients were interviewed at their first admission. At the 11 year follow up, Carpenter and Strauss9 investigated intercorrelations between the four dimensions of outcome (i.e. duration of hospitalization in the previous year, frequency of social contact, time spent employed during the past year and symptom severity during the past month). They reported that the concept of loosely linked domains of outcome functioning based on 2 and 5-year follow up observations of this cohort, persist at 11 year follow up, suggesting a temporal continuity in outcome domains that assess functioning and symptoms. Concerning predic-tors of outcome, patients with more frequent social contacts and more stable heterosexual relation-ships at initial assessment showed a more favourable outcome with respect to frequency of social contacts. Furthermore, the presence and severity of either thought disorders, delusions or hallucina-tions at first hospital admission was significantly related to symptom severity at follow up. For some decades, the focus on prediction of outcome in schizophrenia patients has been extended to prediction of a first psychotic episode.11-14 The mainly applied research strategy for revealing possible predictors of a first psychotic episode is following up at-risk (AR) subjects who have been clinically identified by certain criteria that can differ in definition, operationalization and terminology.13-16 The transition rates for these AR subjects varies widely (10-56%), which underlines the need to identify signs within this somewhat heterogeneous group that further improve prediction of psychosis.17 Recent attempts have been made to improve the prediction of a first psychotic episode in young people meeting the AR-criteria. In addition to information-processing (e.g. reduction of P300 event- related potential amplitude18 and mismatch negativity19) and neurocognitive deficits (e.g. reduced verbal fluency20), it has been shown that among others21 a low level of baseline functioning, and in particular social impairments could contribute to differentiating those who eventually do and who do not develop psychosis.12, 22, 23 Furthermore, severity of (attenuated) positive symptoms has also been found to be predictive of a first psychotic episode.12, 23 To the best of our knowledge, the value of the SCPS scale in predicting transition to psychosis in subjects at-risk for developing psychosis has not yet been investigated. It is still unclear if a temporal continuity of outcome can already be predicted in the period prior to transition. Unlike other scales used in AR research, the SCPS encompasses all domains of functioning and symptoms relevant for the prediction of outcome in schizophrenia patients. Examining individual SCPS items that demonstrate predictive value for a first psychotic episode may provide insight into factors that play a role in enhanced risk for transition to psychosis and that separate converters form non-converters at baseline.

Chapter 6

98

In addition, better understanding of the predictors of a first psychotic episode may help to identify targets for intervention in individuals AR for psychosis, which in turn may lead to better outcomes. We hypothesize that SCPS items assessing both positive symptoms and also SCPS items assessing (social) functioning are predictive of a first psychotic episode in AR subjects. Additionally, we hypothesize that the predictors of outcome as assessed with the SCPS in recent onset schizophrenia patients in previous studies are already present in the period prodromal to a first psychotic episode.

Method2.1. Recruitment

Between August 2002 and April 2006, data were collected from 245 help-seeking individuals (age range 16-35) who met ultra-high-risk and/or ‘cognitive disturbances’ criteria and agreed to participate in the European Prediction of Psychosis Study (EPOS).12, 24 EPOS is a European collaboration of six centres in four countries: Germany, Finland, The Netherlands and the UK. Referral to the early detection services came from psychiatrists, psychologists, general practitioners, outreach clinics, counselling services or teachers, or was self-initiated. Inclusion criteria comprised slightly extended ultra-high-risk criteria as assessed with the Structured Interview for Prodromal Syndromes (SIPS 3.0)25 and cognitive disturbances as assessed by the Bonn Scale for the Assessment of Basic Symptoms – Prediction List (BSABS–P), an abbreviated version of the Schizophrenia Proneness Instrument, Adult version (SPI–A).26The ultra-high-risk approach consists of three alternative criteria:(a) attenuated psychotic symptoms defined by at least one of the following symptoms with SIPS score ‘moderate’ to severe but not psychotic’ (3–5), appearing several times per week for at least 1 week within the past 3 months: unusual thought content/delusional ideas, suspiciousness/persecutory ideas, grandiosity, perceptual abnormalities/hallucinations and/or disorganised communication and, as an extension of the usual SIPS UHR criteria, odd behaviour/appearance;(b) brief limited intermittent psychotic symptoms (BLIPS) defined by hallucinations, delusions or formal thought disorders occurring within the past 3 months and resolving spontaneously within 1 week scoring ‘severe and psychotic’ (6) on the SIPS and achieving at least a ‘moderate’ score on the respective item of the Positive and Negative Syndrome Scale for Schizophrenia;27 (c) genetic risk and functional deterioration defined by a 30% or greater reduction on the Global Assessment of Functioning Scale, modified version (GAF–M)28-30 compared with the highest level of previous functioning for at least 1 month within the previous year in combination with a first- or second-degree relative with a history of any DSM–IV psychotic disorder30 or a DSM–IV schizotypal personality disorder of the index person.

Chapter 6

99

Chapter 6

The ‘cognitive disturbances’ criterion requires the presence of at least two of nine cognitive basic symptoms of at least ‘moderate’ severity (≥3) during the last 3 months and, independent of severity, first occurrence at least 1 year before intake. The nine basic symptoms were: inability to divide attention, thought interference, pressure, and blockage, disturbances of receptive and of expressive speech, disturbance of abstract thinking, unstable ideas of reference, captivation of attention by details of the visual field. Exclusion criteria were: a low verbal IQ (IQ<85); past or present psychotic episode lasting longer than 1 week (i.e. fulfilling DSM–IV30 criteria of a brief psychotic episode for at least 7 days, assessed by the Structured Clinical Interview for DSM–IV31); and symptoms relevant for inclusion arising from a known general medical disorder or drugs or alcohol dependency. On account of the naturalistic design of the present study, (prior) use of antipsychotics was not considered an exclusion criterion. The investigation was carried out in accordance with the latest version of the Declaration of Helsinki. The study design was approved by the medical ethics committees of all participating centres. Informed written consent from participants was obtained after the procedure had been fully explained.

2.2. Subject characteristics

Of the total of 245 EPOS participants, 244 participants were interviewed with the SCPS at baseline (137 males, mean age=22.5, SD=5.23). Of these subjects, 35 were lost to follow-up at 9 months and alto-gether 58 at 18 months. The patient characteristics of the EPOS sample were described previously by Ruhrmann et al12 (see this article for an elaborate description of the sample). At the end of the 18-month follow-up period, 37 participants had made the transition to psychosis. The mean time to transition from baseline examination was 496.8 days (SE= 8.5, 95% CI 480.2–513.6). Diagnoses according to DSM-IV after transition were brief psychotic disorder (n=2 [5.4%], time criterion adapted to BLIPS definition); schizophreniform disorder (n = 3 [8.1%]), schizophrenia (n=23 [62.2%]), schizoaffective disorder (n=3 [8.1%]), and mood disorder with psychotic features (n=6 [16.2%]). Converters group did not differ significantly from non-converters in terms of age, gender and cannabis use. 2.2. Instruments

‘At-risk’ symptomatology The SIPS 3.0,25 including GAF–M, was employed to determine the presence, severity and type of the extended ultra-high-risk criteria. The Scale of Prodromal Symptoms (SOPS), the rating scale of the SIPS, has four subscales that include five positive symptom items, six negative symptom items, four disorganisation symptoms items and four general symptom items. All items are rated on a seven-point rating scale anchored from 0 (never, absent) to 6 (severe/extreme – and psychotic for the positive items). Symptomatic criteria for an at-risk state are exclusively based on positive symptom items and, in our extension, the disorganisation symptom odd behaviour/appearance. The EPOS investigators received extensive training from Dr. Tandy Miller, one of the SIPS authors, including a reliability check after approximately 6 months. The pair-wise interrater concordance of the

100

Chapter 6

SIPS was 77% and determined acceptable by the training team. The Bonn Scale for the Assessment of Basic Symptoms – Prediction list (BSABS–P)26, 32 was compiled to assess cognitive disturbances. To the item collection of the BSABS–P, three theoretical subscales were defined totalling 33 cognitive, perceptual and motor disturbances assessed on a seven-point severity scale (0–6) with maximum frequency of occurrence during the preceding 3 months as the guiding criterion. The investigators re-ceived repeated training by the scale’s first author, Dr. Frauke Schultze-Lutter. Concordance rate with expert rating (F.S.L.) was 87.9%.

2.3. Strauss and Carpenter Prognostic Scale

This scale is composed of 21 items referring to the following variables: quantity and quality of useful work in the past year, social class, social relationships, heterosexual relationships, family history of psychiatric symptoms, action problems (violence and suicidal or homicidal gestures), flat affect, dura-tion of previous hospitalisations, time since first occurrence of psychotic symptoms, ‘presence of thought disorder, delusions or hallucinations in the past year’, ‘presence of depression, hypomania or mania in the past year’, precipitating events, reported severity of subjective distress, ability to meet own basic needs, and usual fullness of life.6-10 Wherever appropriate, the temporal reference for premorbid functioning is the year prior to evaluation. Each item was rated by the interviewer on a 5-point severity scale from 0 (rating indicates a poor prog-nosis) to 4 (rating indicates a favourable prognosis). For example, item 1B, ‘Most useful quality of useful work in the past year’, considering in regard to person’s age and education would be rated as follows:

4: Very competent 3: Competent 2: Moderately competent 1: Marginally competent0: Incompetent

Similarly, the 12th item, ‘Presence of thought disorder, delusions or hallucinations in the past year’ is scored:

4: None of any of the above. 3: Minimal presence of any or all of above. 2: Moderate amount or any of above. 1: Relatively severe and/or continuous presence of any or all of above. 0: Severe and/or continuous presence of any or all of above.

Until now, no study has been conducted on the specific psychometric properties of the SCPS in people at-risk for a first psychotic episode. However, the validity of the instrument has been tested in several studies of schizophrenia patients and has shown acceptable internal consistency, test-retest reliability and convergent validity.1

101

Chapter 6

2.4. Follow up

Follow-up assessments with the SIPS and BSABS–P took place at 9 and 18 months. Transition to psychosis was operationalised as a continuation of BLIPS, i.e. any single item on the positive subscale of SIPS (SIPS–Positive) with a score of 6 for more than 7 days.33, 34 Following identification of full-blown psychotic symptoms in the SIPS interview, the diagnostic category on transition was determined by applying DSM–IV criteria30 of psychotic disorders and affective disorders with psychotic features. The time threshold of the Criterion B of ‘brief psychotic disorder’ was adapted to the BLIPS definition. Past and present psychosis as parts of the exclusion criteria as well as psychotic diagnosis on making transition were assessed with the SCID–I.31

2.5. Statistical analysis

Statistical analysis was performed using SPSS (version 18.0) for Windows. Comparisons between the AR group with transition to psychosis (converters) and the AR group without transition to psychosis (non-converters) were made with Pearson’s chi square tests and Mann-Whitney U tests. Predictors were selected in several steps.35 Firstly, of the 19 potential predictor variables examined, 12 were associated with conversion to psychosis in univariate analyses when changes of the -2 log- likelihood of the model and the Wald statistic became significant (p<0.15). As shown in Table 1, when multivariate analysis was applied to predictors from each assessment domain, which effectively removes redundancy among related measures, the number of predictors meeting the cut-off for inclusion fell to 10. Secondly, we entered retained covariates into a multivariate backward stepwise Cox regression model, p<.05 across domains. To assess the effect of SCPS total score on survival time, a separate Cox proportional hazards analysis was employed. The center that participants attended was entered as a strata variable in all Cox regressions.36 A survival curve was obtained for SCPS total score using Kaplan Meier survival analysis. For this test, a p value of <.05 was considered statistically significant. SCPS items 2 (Social class) and 11b (What is the longest period that SEVERE psychiatric problems have ever persisted more or less continuously, at least once a week) had > 30% missing values. Therefore, these items could not be imputed with multiple imputations and these two items were not included in any of the statistical analyses. Social classes as defined by SCPS item 2 (based on the Hollingshead-Redlich publication in 1958)37 were difficult to transpose onto modern European AR subjects. Item 11b showed many missing values due to its non-applicability in a large percentage of the AR subjects who did not have persisting severe psychiatric problems. The other SCPS items had 0.8-3% missing values; these values were imputed with the multiple imputations function of SPSS 18.

102

Chapter 6

Results

Ten variables were included in the final Cox model (see Table 1). Of these variables, the items that remained in the model as independent predictors were: Most usual quality of useful work in the past year (Beta = -.41, SE = .15, Wald = 7.68, p = .006, hazard ratio (HR) = 1.50, 95% confidence interval (CI) for HR = 1.13 / 2.0), Quality of social relations (Beta = -.33, SE = .12, Wald = 7.39, p = .007, HR = 1.39, 95% CI = 1.10 / 1.77), Presence of thought disorder, delusions, hallucinations, or hallucinations in the past year (Beta = -.63, SE = .20, Wald = 10.33, p = .001, HR = 1.88, 95% CI = 1.28 / 2.75) and Reported severity of subjective distress in past month (Beta = .48, SE = .16, Wald = 8.51, p = .004, HR = .62, 95% CI = .45 / .86). A higher score on this latter SCPS item means more severe subjective distress, indicating a better prognosis as assessed with the SCPS.

Domain

Work

Social relations

Hetero-sexual relations

Treatment

Family history

Onset psychiatric symptoms

Action problems

Flattened affect

Hospitalization

Duration of psychiatric symptoms

Positive symptoms

Mood symptoms

Precipitating events

Subjective distress

Ability to meet basic needs

Fullness of life

No. of items

2

2

1

1

1

1

1

1

1

2

1

1

1

1

1

1

SCPS item

1A: Quantity of useful work in past year,

1B: Most usual quality of useful work in past year

3A: Number of social relations most usual in past year

3B: Quality of social relationships

4: Heterosexual relations most usual in past year

5: Treatment facilities used currently

6: Family history of psychiatric hospitalization

7: Earliest age of onset of psychiatric symptoms

8: Action problems since age 12

9: Flattened, diminished expression of feeling or emotion in past month

10: Previous hospitalization (or intense family surveillance)

11A: Length of time since first occurrence of hallucinations or delusions,

11C: Longest period that any significant psychiatric symptoms have ever persisted

12: Presence of thought disorder, delusions, or hallucinations in past year

13: Presence of depression, hypomania, or mania in past year

14: Precipitating events for most recent psychiatric upset

15: Reported severity of subjective distress in past month

16: Most usual ability to meet own basic needs in past year

17: Most usual fullness of life in past year

Table 1. Potential predictor variables by domain Bold indicates that the predictor variable met statistical criteria for association with conversion to psychosis

The SCPS total score was also highly predictive of a first psychotic episode by itself (Beta = -.098, SE = .023,

103

Chapter 6

The SCPS total score was also highly predictive of a first psychotic episode by itself (Beta = -.098, SE = .023, Wald = 18.40, p< .0001, HR = 1.10, 95% CI = 1.05 / 1.15). This means that the relative risk of developing a psychosis increased by 10% for every decrease in the SCPS total score of 1. In the total at-risk group SCPS total scores ranged between 26 and 68 and the mean SCPS score for the UHR group was 48.1 (SD = 7.72, median = 49). When the total group was split according to the mean SCPS total score, the survival distribution differed significantly between the groups with above (n = 117) and below (n= 127) the mean (log rank [Mantel Cox] = 12.86, p < .0001) with those below the mean showing more transitions (Figure 1).

Figure 1: Survival analysis for 18 month follow up.

SCPS TOTAL SCORE

BELOW MEAN (n = 117)

ABOVE MEAN (n = 127)

TIME (DAYS IN STUDY)

CU

MU

LATI

VE

SUR

VIV

AL

104

Chapter 6

Discussion

To our knowledge, the present study is the first prospective follow-up study examining the utility of the SCPS as a tool for improving prediction of a first psychotic episode. Within this help-seeking popula-tion, the SCPS total score contributed to a better prediction of a first psychosis. Individual items that remained in the Cox regression model as independent predictors for conversion to psychosis in our AR group were ‘Most usual quality of useful work in the past year’, ‘Quality of social relations’, ‘Presence of thought disorder, delusions, or hallucinations in the past year’ and ‘Reported severity of subjective distress in the last month’. The majority of the converters received a diagnosis in the schizophrenia spectrum. In previous re-search, the SCPS has been used to predict the course of the illness for patients diagnosed with schizophrenia. In contrast, the present study investigates the predictors at an earlier stage, i.e. preceding a first psychotic episode. As expected, the results of the present study indicate a temporal continuity of the predictive value of variables often used in schizophrenia studies, extended to the period before the onset of a first psychotic episode. Our results with regard to the SCPS item ‘Quality of social relations’, which assesses the number and closeness of social relationships, is in congruence with earlier findings in the same study sample.22 These findings show that the item ‘difficulties in getting along in with people’ of the World Health Organisation - Disability Assessment Schedule-II was a significant predictor for a first psychotic episode. Also in the large independent sample of the North American Prodrome Longitudinal Study (NAPLS), social dysfunction was predictive for a first psychotic episode.23, 38 These findings combined support the hypothesis that the likelihood of transition within an AR group is increased when attenuated positive symptoms are accompanied by social withdrawal. It may be that young people are more prone to make the transition to psychosis when they stop discussing thoughts and experiences with other people, which may reinforce their delusional ideas and paranoia for example.22 Increased severity of subjective distress in the converters compared to the non-converters is also described by Yung et al.39 As opposed to first episode schizophrenia samples,6-10 more severe subjective distress in AR subjects seems to be a negative prognostic factor, i.e. increases the likelihood of transition to psychosis. In AR subjects, increased severity of subjective distress may mirror their relative intact reality testing compared to first episode patients. Further research could focus on examining whether targeted early interventions directed at the predictive domains may improve outcomes. The reduced work quality may represent the pivotal deficits in schizophrenia and its prodrome in information-processing17,18 (leading to cognitive deficits) and social functioning.22, 23, 38 On the one hand, such deficits may contribute to making it difficult to work. On the other hand, the life structuring effect that is usually provided by work or study may reduce the likelihood of making a transition. At-risk subjects often indicate that they want to work or study, and so encouraging these patients with finding work or schooling tailored to their abilities may help to reduce transition. Individual Placement and Support (IPS) has shown promise for vocational development in patients with a first psychotic episode40 and this approach may also be warranted for at-risk subjects.

105

Chapter 6

Further research is also vital to explore appropriate treatments for positive symptoms in at-risk subjects. Best treatment options for AR positive symptoms are currently unclear.41 The positive symptoms may be the expression of a underlying biological vulnerability present from early on in life, similarly to fever being a non-specific expression of an underlying illness.42 Investigating this biological vulnerability, that may incorporate deficits in information-processing,17,18 may eventually lead to more accurately targeted treatments of the positive symptoms. Our results suggest that the SCPS may be useful in AR research. It is valid, is not time-consuming (it can be administered and scored within 10 minutes) and no extensive training is needed to be able to administer this practical and useful scale reliably.

Limitations

In spite of the previously described strengths of our study, some critical issues regarding the use of the SCPS in an at-risk sample need to be addressed. Firstly, previous research has shown that outcome in schizophrenia research is not a unitary concept. In the present study, we did approach outcome in at-risk subjects as a unitary concept (transition to psychosis). Future studies should investigate the value of the SCPS in predicting outcomes across diverse domains (i.e. duration of hospitalization in the previous year, frequency of social contact, time spent employed during the past year and symptom severity during the past month). A proportion of non-converters psychosis shows persistently poor social functioning38 and/or negative symptoms.43 Secondly, a methodological issue must be considered. As fitted models always perform in an ‘optimistic manner’35 in the model-development data, cross-validation within an independent sample would be useful to control for tailor-made modelling. Although in theory, sample-splitting is an option for model validation in large samples, the limited number of converters in the present study did not allow this for statistical reasons. Existing or future samples of comparable size and risk definition are required to validate our findings. Thirdly, two items from the SCPS were omitted from the total score due to a large percentage of missing values. The SCPS total score in our study may therefore not be comparable to the SCPS total score reported in other studies investigating schizophrenia patients. In our at-risk group, we only experienced difficulty with two of the 21 items. With this in mind, perhaps the SCPS could be modified slightly to better fit the at-risk patient group. In conclusion, our study reveals that the SCPS can improve prediction of transition in an already clinically identified at-risk sample in the period before a first psychotic episode. In addition to positive symptoms, quality of work and social relationships as well as subjective distress were independent predictors of psychosis in at-risk subjects. In combination with other variables, the SCPS could make a valuable contribution to a more accurate prediction of a first psychotic episode in vulnerable populations.

106

Chapter 6

Acknowledgements

The International Advisory Board ‘European Prediction of Psychosis’ (EPOS) group comprises: Patrick D. McGorry, Australia; Thomas H. McGlashan, USA; Martin Knapp, UK.

The authors thank the following scientists for their work for EPOS: Reinaud van der Fliert, MSc, Rianne Klaassen, MD (Department of Psychiatry, AMC, University of Amsterdam, the Netherlands); Heinz Picker, PhD, Meike Neumann, Dipl-Psych, Anke Brockhaus-Dumke, MD, Ralf Pukrop, PhD (Department of Psychiatry & Psychotherapy/ FETZ, University of Cologne, Cologne, FRG); Tanja Svirskis, MD, PhD, Jukka Huttunen, BM, Tiina Laine, MSc, Tuula Ilonen, PhD, Terja Ristkari, MA, Jarmo Hietala, MD, PhD, (Department of Psychiatry, University of Turku, Turku, Finland); Amanda Skeate, DClin (ED:IT, Early Intervention Service, Birmingham, UK); Yehonala Gudlowski, Dipl-Psych (Charite´ , Berlin, FRG); Seza Ozgürdal, Dipl-Psych, Henning Witthaus, MD (Charite´ , Berlin, and Ruhr-University, Bochum, FRG); Paul French, PhD, helen Stevens, PhD (School of Psychological Sciences, University of Manchester, Manchester, UK).

107

Chapter 6

References

1 Händel, M., Bailer, J., Brauer, W., Laubenstein, D., Rey, E. R. (1996). The prognostic scale by Strauss and Carpenter

and its validity. European Archives of Psychiatry and Clinical Neuroscience; 246:203-208.

2 Möller, H. J., Jäger, M., Riedel, M., Obermeier, M., Strauss, A., Bottlender, R. (2010). The Munich 15-year follow-up

study (MUFUSSAD) on first-hospitalized patients with schizophrenic or affective disorders: comparison of

psychopathological and psychosocial course and outcome and prediction of chronicity. European Archives

of Psychiatry and Clinical Neuroscience; 260:367-384.

3 Harder, D. W., Strauss, J. S., Greenwald, D. F., Kokes, R. F., Ritzler, B. A., Gift, T. E. (1990). Predictors of outcome

among adult psychiatric first-admissions. Journal of Clinical Psychology; 46:119-128.

4 Whitty P, Clarke M, McTigue O, Browne S, Kamali M, Kinsella A, Larkin C, OÇallaghan E. Predictors of outcome

in first-episode schizophrenia over the first 4 years of illness (2008). Psychological Medicine; 38:1141-1146.

5 Menezes, M. N., Arenovich, T., Zipursky, R. B. (2006). A systematic review of longitudinal outcome studies

of first-episode psychosis. Psychological Medicine; 36:1349-1362.

6 Strauss, J. S., Carpenter, W. T. Jr, Bartko, J. J. (1974). An approach to the diagnosis and understanding of

schizophrenia: Part III. Speculations on the processes that underlie schizophrenic symptoms.

Schizophrenia Bulletin; 11:61-70.

7 Strauss, J. R., Carpenter, W. T. (1974). The prediction of outcome in schizophrenia. II.

Relationships between predictor and outcome variables. Archives of General Psychiatry; 37:42.

8 Strauss, J. S., Carpenter, W. T. Jr. (1977). Prediction of outcome in schizophrenia. III.

Five-year outcome and its predictors. Archives of General Psychiatry; 34:159-163.

9 Carpenter, W. T. Jr., Strauss, J. S. (1991). The prediction of outcome in schizophrenia. IV:

Eleven-year follow-up of the Washington IPSS cohort. Journal of Nervous and Mental Disease; 179:517-525.

10 Kokes, R. F., Strauss, J. S., Klorman, R. (1977). Premorbid adjustment in schizophrenia. Part II.

Measuring premorbid adjustment: the instruments and their development. Schizophrenia Bulletin; 3(2):186-213.

11Q Addington, J., Cadenhead, K. S., Cannon, T. D., et al. (2007). North American Prodrome Longitudinal Study:

A Collaborative Multisite Approach to Prodromal Schizophrenia Research. Schizophrenia Bulletin; 33:665-672.

12 Ruhrmann, S., Schultze-Lutter, F., Salokangas, R. K. R., et al. (2010). Prediction of Psychosis in Adolescents and

Young Adults at high risk: Results from the Prospective European Prediction of Psychosis Study (EPOS).

Archives of General Psychiatry; 67:241-251.

13 McGorry, P. D., Yung, A. R., Phillips, L. J. (2003). The “close-in” or ultra high-risk model:

A safe and effective strategy for research and clinical intervention in prepsychotic mental disorder.

Schizophrenia Bulletin; 29:771-790.

108

Chapter 6

14 Yung, A. R., Stanford, C., Cosgrave, E., et al. (2006). Testing the Ultra High Risk (prodromal) criteria for the

prediction of psychosis in a clinical sample of young people. Schizophrenia Research; 84:57-66.

15 Schultze-Lutter, F., Michel, C., Ruhrmann, S., Klosterkötter, J., Schimmelmann, B. G. (2011).

Prediction and Early Detection of First-Episode Psychosis. In: Ritsner M (ed.). Textbook of Schizophrenia

Spectrum Disorders, Volume II. Dordrecht; Springer Science+Business Media B.V.: pp. 207-268.

16 Schultze-Lutter, F., Schimmelmann, B. G., Ruhrmann, S. (2011).

The Near Babylonian Speech Confusion in Early Detection of Psychosis. Schizophrenia Bulletin; May 9.

17 Ruhrmann, S., Schultze-Lutter, F., Klosterkötter, J. (2010). Probably at-risk, but certainly ill--advocating

the introduction of a psychosis spectrum disorder in DSM-V. Schizophrenia Research; 120(1-3): 23-37.

18 Van Tricht, M. J., Nieman, D. H., Koelman, J. H. T. M., et al. (2010). Reduced parietal P300 amplitude is

associated with an increased risk for a first psychotic episode. Biological Psychiatry; 68(7):642-648.

19 Bodatsch, M., Ruhrmann, S., Wagner, M., Müller, R., Schultze-Lutter, F., Frommann, I., Brinkmeyer, J., Gaebel,

W., Maier, W., Klosterkötter, J., Brockhaus-Dumke, A. (2011). Prediction of psychosis by mismatch negativity.

Biological Psychiatry; 69 (10): 959-66.

20 Becker, H. E., Nieman, D. H., Dingemans, P. M., van de Fliert, J. R., De Haan, L., Linszen, D. H. (2010).

Verbal fluency as a possible predictor for psychosis. European Psychiatry; 25(2):105-110.

21 Correll, C. U., Hauser, M., Auther, A. M., Cornblatt, B. A. (2010). Research in people with psychosis risk syndrome:

a review of the current evidence and future directions. Journal of Child Psychological Psychiatry; 51(4):390-431.

22 Velthorst, E., Nieman, D., Linszen, D., et al. (2010). Disability in patients clinically at high risk for a psychosis.

British Journal of Psychiatry; 197:278-284.

23 Cannon, T. D., Cadenhead, K., Cornblatt, B., Woods, S. W., Addington, J., Walker, E., Seidman, L. J., Perkins,

D.,Tsuang, M., McGlashan, T., Heinssen, R. (2008). Prediction of Psychosis in Youth at High Clinical Risk.

Archives of General Psychiatry; 65 (1): 28-37.

24 Klosterkötter, J., Ruhrmann, S., Schultze-Lutter, F., et al. (2005) The European Prediction of Psychosis Study

(EPOS): integrating early recognition and intervention in Europe. World Psychiatry; 4:161-167.

25 McGlashan, T. H., Miller, T. J., Woods, S. W., Rosen, J. L., Hoffman, R. E., Davidson, L. (2001).

Structured interview for prodromal syndromes. New Haven, PRIME Research Clinic: Yale School of Medicine.

26 Schultze-Lutter, F., Klosterkötter, J. (2002). Bonn Scale for the Assessment of Basic Symptoms - Prediction List,

BSABS-P. Cologne: University of Cologne.27 Kay, S. R., Fiszbein, A., Opler, L. A. (1987).

The positive and negative symptom scale (PANSS). Schizophrenia Bulletin; 13:261–276.

28 Hall, R. C. (1995).Global assessment of functioning. A modified scale. Psychosomatics; 36:267-275.

29 Hall, R. C., Parks, J. (1995). The modified global assessment of functioning scale: Psychosomatics; 36:416-417.

109

Chapter 6

30 APA. Diagnostic and Statistical Manual of Mental Disorders (1994). 4th edition ed. Washington,

DC: American Psychiatric Association.

31 First, M. B., Spitzer, R. L., Gibbon, M., Williams, J. B. W. (1997). Structured Clinical Interview for DSM-IV Axis I

Disorders (SCID-I). Arlington, VA: American Psychiatric Publishing.

32 Schultze-Lutter, F., Addington, J., Ruhrmann, S., Klosterkötter, J. (2007). Schizophrenia Proneness Instrument,

Adult version (SPI-A). Giovanni Fioriti Editore s.r.l., Rome.

33 Yung, A. R., Phillips, L. J., Yuen, H. P., McGorry, P. D. (2004). Risk factors for psychosis in an ultra high-risk group:

psychopathology and clinical features. Schizophrenia Research; 67:131-142.

34 Yung, A. R., Nelson, B., Stanford, C., et al. (2008). Validation of “Prodromal” criteria to detect individuals at

ultra high risk of psychosis: 2 year follow-up. Schizophrenia Reserch; 105:10-7.

35 Hosmer, D. W., Lemeshow, S. (2000). Applied logistic regression. 2nd ed. New York: John Wiley & Sons.

36. Twisk, J. W. R. (2006). Applied Multilevel analysis. Cambridge, England: Cambridge University Press.

37. Hollingshwead, A. B., Redlich, F. C. (1958). Social class and mental illness:

a community study. New York: John Wiley & Sons.

38 Cornblatt, B. A., Auther, A. M., Niendam, T. et al. (2008). Preliminary findings for two new measures of social

and role functioning in the prodromal phase of schizophrenia. Schizophrenia Bulletin; 33; 688-702.

39 Yung, A. R., Buckby, J. A., Cotton, S. M. et al. (2006). Psychotic-like experiences in nonpsychotic help-seekers:

Associations with distress, depression, and disability. Schizophrenia Bulletin; 32:352–359.

40 Killackey, E., Jackson, H. J., McGorry, P. D. (2008). Vocational intervention in first-episode psychosis:

individual placement and support v. treatment as usual. British Journal of Psychiatry; 193:114-120.

41 De Koning, M. B., Bloemen, O. J. N., van Amelsvoort, T. A. M. J., et al. (2009). Early intervention in patients at

ultra high risk of psychosis: benefits and risks. Acta Psychiatrica Scandinavica; 119(6):426-442.

42 Häfner, H., Löffler, W., Maurer, K., Hambrecht, M., an der Heiden, W. (1999).

Depression, negative symptoms, social stagnation and social decline in the early course of schizophrenia.

Acta Psychiatrica Scandinavica; 100;105–118,

43 Ruhrmann, S., Schulze-Lutter, F., Salokangas, R. K. R., et al. (2010). Course of psychopathology in at-risk

mental states of psychosis – what happens to the non-converters? Early Intervention in Psychiatry; 4 (Suppl. 1):16

110

111

Chapter 7

SOCIAL DISABILITY AT ADMISSION FOR A FIRST PSYCHOSIS DOES NOT PREDICT CLINICAL OUTCOME AT 5-YEAR FOLLOW-UP (BRIEF REPORT).

Eva Velthorst, Dorien H Nieman, Carin Meijer, Don Linszen, Lieuwe de Haan

JOURNAL OF NERVOUS AND MENTAL DISEASE (2011); 199: 510-512

112

Chapter 7

ABSTRACT

Although it has often been reported that premorbid social deficits are associated with clinical outcome in schizophrenia, the association between clinical outcome and social disabilities during admission for a first psychosis is still unclear. We examined whether a detailed assessment of social disability (assessed with the Groninger Social Disabilities Schedule-II) in the month prior to admission for a first psychotic episode contributed to the prediction of disease outcome in terms of psychopathology in 82 patients with schizophrenia. After controlling for PANSS-sum score at baseline, none of the social disability domains significantly predicted the number of relapses or the severity of clinical symptomatology at five-year follow-up. Our results suggest that poor social functioning at admission does not necessarily predict poor disease outcome. Following Di Michele and Bolino (2004), we hypothesize that, in order to reliably predict the course of schizophrenia it may be necessary to assess social functioning during clinical stabilization.

113

Chapter 7

Introduction

Since the early 1900”s, evidence has accumulated that premorbid social deficits are associated with both clinical and functional outcome in schizophrenia (e.g. Addington et al., 2003; Brill et al., 2009; Flyckt, 2006; Haro, 2006; Häfner et al., 1992; Haim 2006; Gaebel, 1987; Kraepelin, 1919; Lauronen, 2007; Malla et al., 2006; Möller, 1986; Möller et al., 2010; Simonsen, 2010; Üçok et al., 2006; White at al., 2009). The focus of these studies, examining premorbid functioning in retrospect, varies from early childhood to the year prior to admission for a first psychosis. Recent prospective studies have shown that social disability prior to disease onset is in fact linked to higher transition rates in a group Clinically at High Risk (Fusar- Poli et al., 2010; Velthorst et al., 2010). Although social deficits before admission for a first psychosis have been consistently associated with poor functional outcome (Siegel et al., 2006; Wiersma et al, 2000), it is still unclear if the level of social disabilities in the month prior to admission could also predict the course of symptomatology. Whereas several studies did report associations of various socio-demograpic variables during a first psychosis with clinical outcome (e.g. employment and marital status; Bobes et al, 2009, Di Michele et al., 2004; Haro et al., 2008; Lambert et al., 2008), only one of them embedded a structured assessment of baseline social disability (Wunderink et al., 2009). In this two-year follow up study, Wunderink et al. found the level of social disability shortly after the start of treatment for a first psychotic episode (as assessed with the Groningen Social Disabilities Schedule, GSDS-II; Wiersma et al., 1990) to be one of two independent predictors for recovery, next to duration of untreated psychosis. In this study, recovery was defined as symptomatic and functional remission, sustained for a prolonged period. In the present longitudinal study of a cohort of young people recently admitted with a first psychosis, we investigated the association between social disability in the month prior to admission and clinical outcome. Social disability was assessed in more detail (examining seven different social roles) and a longer follow up period than the study of Wunderink et al. (2009). The aim of the study was to investigate whether difficulties in social functioning as assessed during admission for a first psychosis could contribute to the five-year prediction of relapse and symptom recovery.

114

Chapter 7

MethodSubjects

We recruited 82 in- and outpatients from 15 to 28 years of age, who met DSM-IV-R criteria (APA, 1994) for schizophrenia or a related disorder, and who agreed to participate in a prospective five-year follow-up study at the Department of Early Psychosis of the Academic Medical Center in Amsterdam, the Netherlands. Socio demographic, clinical, and predictor variables assessed at baseline are shown in table 1. Subjects were eligible for the study if they: 1) met DSM-IV-R criteria for schizophrenia or a related disorder, including schizoaffective and schizophreniform disorders, 2) were in- or outpatients aged between 15 to 28 years and 3) were living in close contact with parent(s) or other relatives. Exclusion criteria were a neurological or endocrine disease, and mental retardation. The study was approved by the local ethical review board. The most common types of antipsychotic medication used at baseline were olanzapine (45), risperidone (23) and clozapine (5). Type of antipsychotic use (F=.42, p=.83) or level of compliance (F=.43, p=.65) were not associated with disability at baseline.

Measurements

Baseline assessments on psychopathology and disability were conducted during a first diagnostic interview and covered the month prior to admission for a first psychosis. Psychopathology at baseline and follow-up was assessed with the Positive and Negative Syndrome Scale (PANSS; Kay, 1987). Psychotic relapse was defined as: a) recurrence or exacerbation of florid psychotic symptoms (hallucinations, delusions, formal thought disorder and/or inappropriate/bizarre behaviour) with duration of >-1 week, with or without rehospitalisation and (b) a significant increase of prescribed antipsychotic medication (WHO, 1992). Ratings of disability at baseline were made with the Groningen Social Disability Schedule (see table 2; GSDS; Wiersma, 1990). GSDS-II assesses functioning and disability during the past four weeks and covers the following eight domains: Self-care-, Household-, Family-, Social-, Partner-, Citizen-, Employment-, and Parent Role. Because of the small percentage of patients to whom the latter domain applied, this domain was left out of analyses in the present study. The GSDS-II has been found to have good psychometric properties in the Dutch population (Wiersma, 1988, 1990, 2005). Ratings of the GSDS-II domains are made on a 4-point scale ranging from 0 (no disability) to 3 (severe disability). Each role is composed of two or three dimensions (see table 1). The Social role for example, is made up of the dimensions „Quality of the contacts”, and „Frequency and number of contact”. Depending on the socio-demographic characteristics of an individual, role- scores can be based on different anchor points. The Partner Role for example, consists of two reference groups: Individuals with and individuals without a steady relationship. The latter category implies „the frequency and quality of attempts to find a partner”. A similar subdivision can be found in the Occupational role, where different questions are used for workers, housewives/men, the unemployed and retired population.

115

Chapter 7

CharaCteristiCs

Diagnosis accorDing to DsM iV

- schizophrenia

- schizoaffectiVe DisorDer

- other psychotic DisorDers1

age ±sD

Male (n, %)

panss, Mean ±sD score

- positiVe subscale score

- negatiVe subscale score

- total panss score

gsDs suM score ( Mean ±sD score)1

- faMily role

- partner role

- social role

- self care role

- occupational role

- householD role

- citizen role

Marital status, not MarrieD (n, %)

liVing alone2 (n, %)

Working situation, eMployeD (paiD) /stuDent2 (n, %)

Duration of untreateD psychosis (Mean no. Months ±sD score )

67 (82)

9 (11)

6 (7)

21.2 ±2.9

69 (84)

20.1 ±7.4

21.0 ±7.2

80.9 ±20.2

9.1±4.2

1.4 ±1.0

1.2 ±1.5

1.1 ±.9

.7 ±.8

1.7 ±.9

1.4 ±.8

1.5 ±.9

82 (100)

52 (73)

29 (40)

13.6 ±22.2

table 1. Patient CharaCteristiCs at admission

1 the Partner role was not inCluded in the sum sCore.

2 data were available for 69 Patients.

116

Chapter 7

GSDS

1. Self-care

2. Household

3. Family role

4. Partner role

5. Citizen role

6. Social role

7. Occupational role (daily activities)

Sub dimensions

(i) personal hygiene

(ii) self-presentation

(i) contribution and maintenance of atmosphere

(ii) share in the economic independence

(iii) single household

(i) emotional connection with parents

(ii) actual connection

(iii) emotional and actual connection with siblings

(i) emotional connection

(ii) sexual relationship

(iii) obtaining a relationship with a partner (for those without a partner)

(i) general interest in the society

(ii) participation in social groups, organizations and/or associations

(iii) interests (of others) in the society

(i) quality of the contacts

(ii) frequency and quantity of the contacts

(i) daily routine

(ii) performance at work

(iii) contact with others

(iv) (other) daily activities

Table 2. Description of the different (sub) domains of the GSDS

Statistical analysis

Hierarchical multiple regressions were used with the seven social disability scales as predictors of psychotic phenomenology and number of relapses after controlling for baseline symptomatology (PANSS sum score). A p-value of <.05 was considered statistically significant.

117

Chapter 7

Results

Of 71 patients with data available, 45 (65%) had used anti-psychotic medication for more than 2 years with at least 75 % compliance. Forty-nine percent of the participants (n=39) experienced at least one relapse during follow-up. After having controlled for PANSS-sum score at baseline, we found none of the social disability domains to remain significantly predictive for the number of relapses at five-year follow-up (see table 3; R squared change = .09; F change (7, 65) = .92; p = .50); nor for the level of clinical symptomatology. No significant associations were found with regard to PANSS-sum score (R-squared change = .05, F change (7, 68) = .55, p=.80), the amount of positive- (R-squared change = .06, F change (7, 69) = .94, p=.48), negative-( R-squared change = .07, F change (7, 69) = 1.59, p=.15), or general symptoms (R-squared change = .04, F change (7, 69) = .59, p=.76) at five year follow-up. We did find a significant association between the sub domain self care, and worse self care being related to better outcome. This unexpected finding is most likely due to the limited number of subjects who reported moderate to severe impairment in this domain (13 and 1 respectively).

Table 3. Hierarchical multiple regression analysis of baseline disability in seven social roles as predictor of number of relapses and PANSS sum score over a five-year follow-up

Predictor variablea

Self care Role

Household Role

Family Role

Partner Role

Citizen Role

Social Role

Occupational Role

GSDS Sum score

Number of relapses

B-coeffient part correlation p

PANSS sum score

B-coeffient part correlation p

-.272

.124

-.092

-.074

-.110

.087

.135

.040

-.238

.092

-.073

-.066

-.086

.073

.106

.038

.05

.44

.54

.58

.47

.54

.37

.75

-.015

-.148

.129

.038

-.187

.057

.019

-.031

-.013

-.112

.104

.034

-.147

.050

.016

-.030

.908

.337

.374

.770

.209

.671

.892

.795

a The analysis was corrected for baseline PANSS- sum score

b Of 80 patients, relapse data were available. 6 of those patients were recurrent psychotic during follow-up, and there data were therefore left out of this analysis

b The last available PANSS-assessment was used (n=77). Time since first interview: Mean=50 months, SD±17.7, median= 59 months, range= 14- 77 months.

118

Chapter 7

Discussion We examined whether a detailed assessment of social disability in the period around first psychosis onset predicted disease outcome in terms of psychopathology, after correcting for symptomatology at baseline. In contrast to the study of Wunderink et al. (2009), the results of our study do not support a direct relationship between baseline social disability and clinical outcome at five year follow-up. Whereas social functioning in the year prior to a first psychosis has proven a strong and reliable predictor for disease outcome (see for example Möller et al., 2010; Möller et al., 1986; Gaebel and Pietzcker, 1987), social disability closer to admission seems to lose its predictive power. It might be that social disabilities during this period of rapidly worsening symptoms may in fact reflect state-dependent characteristics instead of stable prognostic traits.

Conclusions

Our results suggest that poor social functioning at admission does not necessarily mean poor disease outcome. Following Di Michele and Bolino (2004), we hypothesize that a measurement of social func-tioning during clinical stabilization may be necessary to reliably predict the course of schizophrenia.

119

Chapter 7

References

Addington, J., van Mastrigt, S., Addington, D. (2003). Patterns of premorbid functioning in first-episode psychosis:

initial presentation. Schizophrenia Research; 62:23-30.

American Psychiatric Association (1994). DSM-IV-R: Diagnostic and Statistical Manual of Mental Disorders. IV,

American Psychiatric Association, Washington, DC.

Bobes J, Ciudad A, Álvarez E, San L, Polavieja P, Gilaberte I (2009) Recovery from schizophrenia:

Results from a 1-year follow-up observational study of patients in symptomatic remission.

Schizophrenia Research 115:58-66.

Brill N, Levine SZ, Reichenberg A, Lubin G, Weiser M, Rabinowitz J (2009)

Pathways to functional outcomes in schizophrenia: The role of premorbid functioning, negative symptoms

and intelligence. Schizophrenia Research 110:40-46.

Di Michele V, Bolino F (2004) The natural course of schizophrenia and psychopathological predictors of outcome.

A community-based cohort study. Psychopathology 37:98-104.

Flyckt L, Mattsson M, Edman G, Carlsson R, Cullberg J (2006) Predicting 5-year outcome in first-episode psychosis

construction of a prognostic rating scale. J Clin Psychiatry 67:916-924.

Fusar-Poli P, Byrne M, Valmaggia L, Day F, Tabraham P, Johns L, McGuire P, OASIS Team (2009) Social dysfunction

predicts two year clinical outcome in people at ultra high risk for psychosis. J Psychiatr Res. 44:294-301.

Gaebel W, Pietzcker A (1987) Prospective study of course of illness in schizophrenia: part II.

Prediction of outcome. Schizophr Bull. 13:299-306.

Häfner H, Riecher-Rössler A, Maurer K, Fätkenheuer B, Löffler W (1992) First onset and early symptomatology

of schizophrenia. A chapter of epidemiological and neurobiological research into age and sex differences.

Eur Arch Psychiatry Clin Neurosci 242:109-118.

Haim R, Rabinowitz J, Bromet E (2006) The relationship of premorbid functioning to illness course in schizophrenia

and psychotic mood disorders during two years following first hospitalization. J Nerv Ment Dis 791-795.

Haro JM, Novick D, Suarez D, Alonso J, Lépine JP, Ratcliffe M, and the SOHO Study Group (2006)

Remission and relapse in the outpatient care of schizophrenia. Three-year results from the schizophrenia outpatient

health outcomes study. J Clin Psychopharmacol 26:571-578.Haro JM, Novick D, Suarez D, Ochoa S, Roca M (2008)

Predictors of the course of illness in outpatients with schizophrenia: A prospective three year study. Prog Neuropsy-

chopharmacol Biol Psychiatry 32:1287-1292.

Kay SR, Fiszbein A, Opler LA (1987). The positive and negative symptom scale

(PANSS). Schizophr. Bull. 13: 261-276.

120

Kraepelin E (1919) Dementia Praecox and Paraphrenia (Vol) GM Robertson (Ed.) Edinburgh: E. & S. Livingstone.

Lambert M, Naber D, Schacht A, Wagner T, Hundemer H-P, Karow A, Huber CG, Suarez D, Haro JM, Novick D,

Dittman RW, Shcimmelmann BG (2008) Rates and predictors of remission and recovery during 3 years in 392

never-treated patients with schizophrenia. Acta Psychiatr Scand 118:220-229.

Lauronen E, Miettunen J, Veijola J, Karhu M, Jones PB, Isohanni M (2007)

Outcome and its predictors in schizophrenia within Northern Finland 1966 Birth Cohort. Eur Psychiatry 22:129-136.

Malla A, Norman R, Schmitz N, Manchanda R, Béchard-Evans L, Takhar J, Haricharan R (2006)

Predictors of rate and time to remission in first-episode psychosis: a two-year outcome study.

Psychol Med 36:649-658.

Möller HJ, Schmid-Bode W, von Zerssen D (1986) Prediction of long-term outcome in schizophrenia by prognostic

scales. Schizophr. Bull 12:225-235.

Möller HJ, Jäger M, Riedel M, Obermeier M, Strauss A, Bottlender R (2010).

The Munich 15-year follow-up study (MUFUSSAD) on first-hospitalized patients with schizophrenic or affective

disorders: comparison of psychopathological and psychosocial course and outcome and prediction of chronicity.

Eur Arch Psychiatry Clin Neurosci 260:367-384.

Siegel SJ, Irani F, Brensinger CM, Kohler CG, Bilker WB, Ragland JD, Kanes SJ, Gur RC, Gur RE (2006)

Prognostic cariables at intake and long-term level of function in schizophrenia. Am J Psychiatry 163:433-441.

Simonsen E, Friis S, Opjordsmoen S, Mortensen EL, Haahr U, Melle I, Joa I, Johannessen JO, Larsen TK,

Røssberg JI, Rund BR, Vaglum P, McGlashan TH (2010). Early identificatnion of non-remission in first-episode

psychosis in a two-year outcome study.

Üçok A, Polat A, Çakir S, Genç A (2006) One year outcome in first episode schizophrenia/ Predictors of relapse.

Eur Arch Psychiatry Clin Neurosc 256:37-43.Velthorst E, Nieman DH, Linszen D, Becker H, de Haan L, Dingemans PM,

Birchwood M, Patterson P, Salokangas RK, Heinimaa M, Heinz A, Juckel G, von Reventlow HG, French P, Stevens H,

Schultze-Lutter F, Klosterkötter J, Ruhrmann S (2010) Disability in people clinically at high risk of psychosis. Br J

Psychiatry 197:278-84.

White C, Stirling J, Morris J, Montague L, Tantam D, Lewis S (2009) Predictors of 10-year outcome of first-episode

psychosis. Psychol Med 39:1447-1456.

World Health Organization (WHO) 1992. WHO Coordinated Multi-Center Study on the Course and Outcome

of Schizophrenia (LCS). Geneva, WHO

Wiersma D, Wanderling J, Dragomirecka E, Ganev K, Harrison G, van der Heiden W, Nienhuis FJ, Walsh D (2000)

Social disability in schizophrenia: its development and prediction over 15 years in incidence cohorts in six European

centres. Psychol med 30:1155-1167.

Chapter 7

121

Wiersma D, de Jong A, Kraaijkamp HJM, Ormel J (1990) GSDS II,The Groningen Social Disabilities Schedule,

Second Version. Groningen, The Netherlands: Department of Psychiatry, Rijksuniversiteit Groningen.

Wunderink L, Sytema S, Nienhuis FJ, Wiersma D (2009) Clinical recovery in first-episode psychosis. Schizophr.

Bull 35:362-369.

Chapter 7

122

123

03P A R T

124

125

Chapter 8

THREE-YEAR COURSE OF CLINICAL SYMPTOMATOLOGY IN YOUNG PEOPLE AT ULTRA HIGH RISK FOR TRANSITION TO PSYCHOSIS.

Velthorst E, Nieman DH, Klaassen RMC, Becker HE, Dingemans PM, Linszen DH, de Haan L.

ACTA PSYCHIATRICA SCANDINAVICA (2011); 123(1):36-42.

126

Chapter 8

ABSTRACT

Objective The investigation into the course of ultra high risk (UHR) symptomatology of those patients who eventually do not meet the psychosis-threshold criteria within the 3-year timeframe of the study.

Method The course of UHR symptoms, GAF score and employment status was investigated in 57 patients who did not make a transition to psychosis and who were examined within the Dutch Prediction of Psychosis Study in Amsterdam, the Netherlands.

Results At the 3- year follow-up, 75% of the patients who did not make a transition to psychosis had remitted from UHR status. With a Generalized Estimation Equation Model it was shown that this group recovered from positive (F = 52.7, P < 0.0001), negative (F = 24.3, P < 0.0001), disorganization (F = 14.4, P < 0.0001) and general symptoms (F = 25.0, P < 0.0001) within the time-frame of the study. In addition, the level of global functioning and likelihood of having a job and ⁄ or education significantly improved. The largest improvements occurred within the first year. UHR symptoms did not re-occur after improvement.

Conclusion With the current UHR criteria, a large percentage of the included subjects appear to have transitory complaints and dysfunctioning. A refinement of the UHR criteria may diminish the chance of including ‘false positives’ in future UHR studies.

Significant outcomes• A substantial percentage of patients initially at ultra high risk (UHR) for psychosis report significantly fewer UHR symptoms within 1 year.• Recovery is not limited to a reduction in symptomatology, but also involves improvement in global functioning as well as employment status.• The large number of ‘false positives’ indicates the necessity of refining the current UHR criteria for psychosis.

Limitations• Results are based on a limited sample size• Transition to other comorbid disorders was not investigated

127

Chapter 8

Introduction

Although our knowledge about the prodromal phase of schizophrenia is improving, with the current ‘UHR’ criteria (1), the reported transition rate from UHR state to psychosis dropped to 10 to 25 per cent in recent studies (2–5). The UHR criteria, which are mainly based on positive symptoms, have led to the inclusion of a considerable number of ‘false positives’ in UHR studies. Questions thus arose about the stability of psychotic-like symptoms in young people who eventually do not meet the psychosis threshold (6). The clinical positive symptoms reported for this group may in fact represent transitory age-specific symptoms rather than stable phenomena (6, 7). On the other hand, the symptoms of these patients may remain stable over the time span of the UHR studies. So far, one study investigated the naturalistic course of clinical symptomatology in young UHR patients who did not convert into a full-blown psychosis, showing substantial clinical remission over a 1- year period (6). In their prospective study, Simon and Umbricht found the positive- negative-, disorganized- and general-item clusters of the Structured Interview for Prodromal Syndromes (SIPS) to decrease significantly in the non-converters. In the absence of a protocolized intervention, the chance of remission to a non-risk state was over four-fold higher than the chance of transition to psychosis within a year of establishing UHR status (6). The results of their prospective study suggest that in the vast majority of young people initially at UHR, the reported problems are only temporary, or at least temporarily of clinical severity.Following Simon and Umbricht (6), the present naturalistic longitudinal cohort study focuses on the course of clinical symptomatology in UHR patients who do not reach the psychosis-threshold within the timeframe of the study (UHR Non- Transition; UHR-NT). In addition, we investigate the course of global functioning and employment status. We extended the follow-up period to 3 years, to capture any fluctuations in seemingly remitted symptoms.

Aims of the study

To replicate previous findings regarding the course of UHR symptomatology in a sample without a transition to psychosis with a longer follow-up period, to rule out transitory phenomena that may have occurred after 1 year. We hypothesized that a considerable number of patients would show reduced positive, negative, disorganization and general symptoms within 3-year follow-up, as well as a significant increase in global functioning (as assessed with the GAF-score). In addition, we expected the likelihood of having a job and ⁄ or education to be significantly improved compared to baseline.

128

Chapter 8

Materials and MethodsSample

Between August 2002 and July 2009, data were collected from 77 UHR patients; a subsample of the 285 help-seeking patients who were examined within the Dutch Prediction of Psychosis Study (DUPS) at the Adolescent Clinic of the Academic Medical Center (AMC) of the University of Amsterdam, the Netherlands. The main reasons for non-participation of the remaining 208 patients were the presence of a psychosis, symptoms not severe enough to meet UHR criteria, and a general refusal to participate in research activities. Part of our Dutch sample also participated in the European Prediction of Psychosis Study (EPOS, see earlier publications: 5, 8). Prior to their referral to the AMC, all patients sought help for various complaints at local mental health services and other health services. The patients were referred to the DUPS project for a second opinion because the referring clinician suspected a development toward a psychosis. Inclu-sion criteria were ages in the range between 12 and 35 years, ability and willingness to give informed consent and falling in one or more of the following groups:

i) Genetic risk in combination with reduced functioning: subjects who have a first-degree relative with a psychotic disorder, or who themselves have a schizotypal personality disorder and who have experienced a significant decrease in functioning during the past year (i.e. 30% reduction in GAF score for at least 1 month).

ii) Attenuated Positive Symptoms (APS): subjects who have experienced subthreshold, attenuated positive psychotic symptoms, defined by at least one of the following symptoms appearing several times per week for at least one week within the previous 3 months: unusual thought content ⁄ delusional ideas, suspiciousness ⁄ persecutory ideas, grandiosity, perceptual abnormalities ⁄ hallucinations and disorganized communication.

iii) Brief limited intermittent psychotic symptoms (BLIPS): subjects who have experienced episodes of frank psychotic symptoms. BLIPS were defined by hallucinations, delusions or formal thought disorders occurring within the previous 3 months which resolved spontane- ously within a week.

129

Chapter 8

The exclusion criteria of this study were the presence or history of a psychotic disorder for more than one week, an Intelligence Quotient < 85, symptoms because of a known medical disorder or intoxication with drugs or alcohol. Patients were allowed to use cannabis, as is widely accepted in UHR research. However, if the patients used hard drugs, or if the cannabis caused the UHR symptoms, they were not included in the study. To establish the relation between cannabis use and UHR symptoms, we asked cannabis-using patients whether the symptoms ever occurred during periods of cannabis abstinence. If symptoms did not occur without cannabis use (as was the case in a few subjects), we asked those patients to stop taking cannabis during the following month. Because they were help seeking, the patients were usually mtivated to do so. Subsequently, they were assessed again with the SIPS to investigate if their symptoms were still present. If so, they were included in the study. If the symptoms abated, the patients were not included. ‘Cannabis use’ in the past or present was defined as having used cannabis at least five times in a lifetime (9). The investigation was carried out in accordance with the latest version of the Declaration of Helsinki. The study design was approved by the Medical Ethical Committee of the AMC. The informed consent of the participants was obtained after the nature of the procedures had been fully explained.

Instruments

The semi-structured interview SIPS (10, 11) was used to determine the presence, severity and type ofprodromal symptoms. The Scale Of Prodromal Symptoms (SOPS), the rating scale of the SIPS, hasfour subscales that include five positive, six negative, four disorganization and four general symptom items. All symptoms are rated on a seven-point rating scale ranging from 0 (Never, absent) to 6 (Se-vere ⁄ Extreme –and Psychotic for the positive items). The diagnosis of a prodromal state is based on the score at the positive items. Scores in the 3–5 range are considered to be indicative of the UHR phase (APS). A score of 6 signifies psychosis or BLIPS (9). Each interviewer (DHN, PD, and two other senior investigators) received a 2-day training workshop by Dr. T.J. Miller, one of the SIPS au-thors, including reliability check after approximately 6 months. The pairwise inter-rater reliability con-cordance of the SIPS was 77% and deemed acceptable by the training team. The GAF score (12) was used to determine the global level of functioning. This score is de-rived from the GAF scale, a scale ranging from 1 to 100, both for the current situation and for the highest level in previous year. A transition to psychosis was defined as the persistence of one or more positive psychotic symptoms for more than one week with a score of four or more on hallucinations, delusions or formal thought disorder on the Positive and Negative Syndromes Scale (13). The Structured Clinical Inter-view for DSM-IV (SCID-I; 14) was administered to all patients after transition to psychosis to establish a formal DSM-IV diagnosis.

130

Chapter 8

The treatment was documented using the Treatment Documentation Sheet (Patterson & Birchwood), an instrument designed for the EPOS project to assess treatments received during the previous 9 months. In our study, treatment was defined as any form of psychotherapy for at least five sessions and ⁄ or psychotropic medication prescription for at least 1 month in the previous 9 months. Data of employment status were collected with the standardized Client Socio-demographic and Service Receipt Inventory (15). A meaningful employment status was defined as having a full or part-time job and ⁄ or study.

Procedure

After their referral to the AMC, putative UHR patients were invited for a first interview with a psychiatrist and a psychologist. In this face to face interview which lasted approximately 2 h, subjects were asked about their lifetime history of complaints, family history of psychiatric disorders, as well as drug and medicine use. Subsequently, the SIPS was administered. Simultaneously, in another in-terview, parents or guardians were separately asked about the lifetime development of their child. All the diagnostic information for each subject was discussed in a staff meeting. Those patients considered to be at ‘UHR’ were asked to sign a written informed consent before participating in the DUPS project. Written informed consent was also obtained from parents or guardians if the participant was below the age of 18 years. They were referred back to their referring mental health professionals. Of these, some received treatment while others were only monitored. Patients, their parents or caretakers and the referring instances were asked to contact the DUPS project in case of increasing symptoms. In addition, a SIPS interview was carried out at 1-, 2- and 3- year follow-up.

Statistical analysis

All data were analysed using the spss statistical package for Windows (version 17.0; IBM Company, Chicago, IL, USA). Differences in age, clinical symptomatology and GAF score at baseline between those with and without follow-up data were examined by means of independent two-tailed t tests. Gender, treatment and cannabis use differences, as well as differences in family predisposition were analysed using Chi-square tests. It was not possible to screen all UHR-NT patients at each assessment point. To correct for study drop-outs and missing assessments, we conducted a linear mixed model analysis. With this analysis, we measured the impact of time on the participants’ scores on the SOPS and GAF, across four different time points (at baseline, after 1-, 2- and 3-year follow-up). A Generalized Estimating Equation model (GEE) was conducted to examine the development in employment status over time (i.e. the increase or decline in the likelihood of a significant job and ⁄ or study). Cannabis use was included as a covariate in the GEE and linear mixed model.

131

Chapter 8

ResultsBaseline characteristics

Table 1 displays the demographics of the included group. The included group consisted of 77 subjects(51 men, mean age 19.2, SD 3.8). Twenty patients (26%) had made a transition to a psychosis (14 men, mean age, 20.2, SD = 4.0) and 57 had not (37 men, mean age 18.9, SD 3.7). The mean inter-val between inclusion and transition for this group was 13 months (Range = 2.00– 37.00, SD = 9.0). Seven UHR-NT patients (12%) were lost to the follow-up after initial assessment. The patients with follow-up information did not differ significantly at baseline from those lost to follow-up in terms of gender, age, severity of positive, negative, general and disorganization symptoms, global functioning and family predisposition. They did differ in cannabis use, with a higher percentage of cannabis users in the group lost to follow-up (-2 (1) = 5.0, P = 0.03). In total, 100% of the patients lost to follow-up (as opposed to 56% in the follow-up cohort) reported to have used cannabis at least five times during their lives.

Table 1: Baseline characteristics of the 77 followed-up patients with UHR status

Age (mean; ± SD)

Gender (male/female)

First-degree relative with psychosis (%)

Global functioning (mean; ± SD)

Cannabis use (n; %)

Employment status (n; %)

Paid/ self- employed

Unemployed

Student

Other

Missing

SOPS (mean; ± SD)

Positive Symptom Subscale total

Negative Symptom Subscale total

Disorganized Symptom Subscale total

General Symptom Subscale total

UHR-group (n=77)

9.19 (3.8)

51/26

16 (20.8)

50.1 (11.4)

42 (54.5)

14 (18.2)

15 (19.5)

36 (46.8)

4 (5.2)

8 (10.4)

11.0 (3.8)

13.3 (7.3)

4.7 (2.8)

8.6 (4.3)

132

Chapter 8

Follow-up analysis

Table 2 lists the characteristics of the UHR-NT group (n = 57). We were able to conduct at least one follow-up assessment of 50 UHR-NT patient (87.8%). On average, 2.2 follow-up assessments were carried out at varying time points (SD=0.8, Range=1-3). Complete follow-up data were available for 23 UHR-NT patients (40.4%). Of the 50 UHR-NT patients with a minimum of a single follow-up assessment, at least 17 patients (34%) received medication with or without psychotherapy at some point during the follow-up period. Eleven patients (22%) reported a period in which they received a protocolized form of psychotherapy without medication. At least 9 patients (18%) reported a period of therapy with medication, as well as a period in which they received therapy without medication. Thirteen patients did not have any kind of treatment in the period(s) between assessments. The most com-mon types of prescribed medication were antidepressants (n = 12 patients), antipsychotics (n = 11), anxiolitics (n = 5) and methylphenidate (n = 5). The dose of prescribed antipsychotics varied from 50 to 200 mg chlorpromazine equivalents.

Table 2: Characteristics of the UHR-NT group (n=57)

UHR-NT group (n=57)

Global functioning ( M; ± SD)

Employment status (No.; %)

Paid/self employed or student

Unemployed/ other*

SOPS (mean; ± SD)**

Positive Symptom

Subscale total

Negative Symptom

Subscale total

Disorganized Symptom

Subscale total

Baseline

n=57

51.3 (12.0)

n=51

37 (72.5)

14 (27.5)

n=57

10.8 (3.7)

12.5 (7.3)

4.5 (2.8)

1-year follow-up

n=37

66.5 (12.8)

n=31

26 (83.9)

5 (16.1)

n=37

4.8 (3.3)

6.8 (5.8)

2.4 (1.7)

2-year follow-up

n=34

68.4 (12.0)

n=31

23 (74.2)

8 (25.8)

n=38

3.5 (2.9)

5.6 (6.1)

2.0 (1.5)

>30 months

follow-up

n=33

70.7 (11.1)

n=36

30 (83.3)

6 (16.7)

n=36

3.1 (3.1)

3.9 (6.1)

1.7 (1.8)

Statistics

F=26.3

X2=11.0

F=52.7

F=24.3

F=14.4

p

<.0001

.012

<.0001

<.0001

<.0001

1 year follow-up [12.2 months; SD=2.7; range 8-18], 2 year follow-up [23.79, SD=2.4; range18-30), > 30 month follow-up [39.42, SD=9.6; range 31-58]

* Other: sheltered employment, sickness leave, reintegration project

** Ranges: Positive = 0-30; Negative = 0-36, Disorganized = 0-24, General = 0-24. Cannabis use was included as a covariate in the model.

133

Chapter 8

As expected, the results of the SOPS in the UHR-NT group showed a significant decline in positive (F (75) = 52.7, P < 0.0001), negative (F (75) = 24.3, P < 0.0001), disorganization (F (75) = 14.4, P < 0.0001), and general symptoms (F (77) = 25.0, P < 0.0001; see Fig. 1). Our results reveal that the largest decline in symptomatology occurs within the first year after baseline assessment. In addition, the GAF score significantly increased over time (F (78) = 26.3, P < 0.0001). No main or interaction effects of cannabis were found. At the 3-year follow-up, 75% of the UHR-NT group had remitted from UHR status (i.e. no score of 3 or more on items P1 t ⁄m P5 of the SIPS), while 25% still reported UHR symptoms. Of the patients who remitted from UHR status, 85.1% did also find a job and ⁄ or study. When examining the course of employment status, we found that the likelihood ratio for having a job and ⁄ or study increased at the 3- year follow-up (x2 (4) = 11.0, P = 0.01). Despite the unexpected decline at the 2- year follow-up, our data indicate a significant increase in the number of young people with a full- or part-time job and ⁄ or study over 3 years. Here, we did find a main effect for cannabis use (x2 (1) = 4.4, P = 0.04). Although the amount of studying and ⁄ or working of cannabis-using UHR-NT patients (61.4%) also increased with time, the frequency of a significant employment status within this group was consistently lower compared to the group who never used cannabis.

Figure 1: Course of SOPS symptomatology in the UHR-NT group

POSITIVE

NEGATIVE

DISORGANIZED

GENERAL

ASSESSMENT

MEA

N S

OPS

SC

OR

E

134

Chapter 8

Discussion

In this 3-year follow-up study of a cohort initially at UHR for developing a psychosis, we sought to investigate the course of clinical symptomatology and level of functioning of those who eventually did not make the transition to a psychosis. Our study indicates that a large part of the young helpseeking people show remission from UHR status 1 year after baseline assessment, which is in accordance with the results of Simon and Umbricht (6). We found the UHR-NT group to recover from positive, negative, disorganization and general symptoms within the first year of follow-up. In addition, the level of global functioning, as well as the likelihood of having a paid job and ⁄ or being a student in this group significantly improved over time. The UHR symptoms did not reoccur after improvement during the 3- year follow-up. The high remission rate in the UHR-NT group raises questions about the origin of the reported symptoms at baseline. As Simons et al. (6) already suggested, the baseline symptoms may in fact have represented transitory age-related psychotic- like experiences (7, 16). Transitory psychotic-like experiences such as extrasensory perceptions appear rather frequently in early adulthood (7, 17) possibly partly related to an inability to distinguish between relevant and irrelevant stimuli (17), as well as to stressful age-related events. In spite of the central role that the famous psychiatrist Klaus Conrad ascribed to “delusional mood” in the prodromal phase (18), it seems that even delusional mood does not necessarily result in a first psychotic episode. Another explanation of the UHR symptom remission may be found in the early treatment received in the UHR-NT cohort. In spite of the naturalistic design of our study, most young people received treatment (psychotherapy and ⁄ or treatment with medication) sometime during follow-up, and this might have contributed to the prevention of a worse outcome. However, the fact that most of the 20 UHR people who did make a transition to a psychosis also received some form of treatment contradicts this hypothesis. Moreover, as we rated the period with the highest level of symptoms somewhere since last assessment point, we do not know whether subjects improved because of the therapy. Randomized control trials to the effects of therapy are needed to unravel the cause and effect of any form of treatment. Although various randomized trials of the effects of medication on the development of a psychosis have already been carried out (19–21), at the moment definite conclusions about the efficacy of UHR interventions cannot be drawn (22). Preliminary results concerning the prescription of omega-3 fatty acids, however, seem promising and this subject is the focus of current research (23). Alongside the ‘true positives’, it seems that with the current UHR criteria we include a large number of young people with transitory complaints. This issue fuels the need for further research into the core features of the prodromal phase of psychotic disorders. Yung et al. (16) already emphasized that certain psychotic-like experiences make a transition more likely when accompanied by distress, depression and disability. In addition, recent studies underlined the predictive value of negative symptoms as ‘social anhedonia and withdrawal’, and a lower level of global functioning (3, 4).

135

Chapter 8

The addition of the criterion, a moderate score (≥3) on the item ‘social anhedonia and withdrawal’ to the current criteria (of a score of 3–5 or a score of 6 less than a week on at least one positive item of the SIPS), may contribute to a larger transition percentage in UHR groups.

Limitations

Our results should be considered in the context of some limitations. In our study, we did not assesscomorbid disorders at each time point; therefore, it is unclear as to whether the complaints of the UHR-NT group in fact represented prodromal symptoms for the transition to other psychiatric disorders as a bipolar or a depressive disorder. However, the increased GAF score at follow-up seems to reflect only a modest occurrence of other severe psychiatric disorders. A second limitation is our small sample size. Seven UHR-NT patients were lost to follow-up.Although we used a linear mixed model to correct for our unbalanced sample, we cannot completely rule out that we based our results upon a biased sample. A transition to psychosis in the patients lost to follow-up is unlikely however, as we asked the practitioner who referred the patients to contact us when they suspected a psychosis. In addition, most patients with a first psychotic episode in the Amsterdam region were referred to our Early Psychosis Department in case of psychosis. To our knowledge, this study is the first to examine the course of clinical symptoms in a UHR group with a follow-up period of 3 years. With this relatively long follow-up period, we were able to reduce the possibility of missing fluctuating, and seemingly remitted symptoms. The most substan-tial improvement in clinical symptoms as well as in global functioning was seen in the first year after baseline assessment. As well as Simon and Umbricht (6) we believe that the complaints in the UHR-NT group most likely represented age related transitory dysfunction.

Acknowledgements

The authors thank Dr. M. W. Koeter for his valuable help with the statistical analysis. We also thank Erik Wagemans, Suzanne Wiltink, Anne Hilde Krips and Inger van der Kooij for their help with the collection and management of our data.

136

References

1 Yung, A. R., Phillips, L. J., Yuen, H. P. et al. (2003). Psychosis prediction: 12-month follow up of a high-risk

(‘‘prodromal’’) group. Schizophrenia Research; 60:21–32.

2 Yung, A. R., Yuen, H. P., Berger, G. et al. (2007). Declining transition rate in ultra high risk (prodromal) services:

dilution or reduction of risk? Schizophrenia Bulletin; 33:673–681.

3 Cannon, T. D., Cadenhead, K., Cornblatt, B. et al. (2008). Prediction of psychosis in youth at high clinical risk:

a multisite longitudinal study in North America. Archives of General Psychiatry; 65:28–37.

4 Velthorst, E., Nieman, D. H., Becker, H. E. et al. (2009). Baseline differences in clinical symptomatology

between ultra high risk subjects with and without a transition to psychosis. Schizophrenia Research; 109:60–65.

5 Ruhrmann, S., Schultze-Lutter, F., Salokangas, R. K. R. et al. (2010). Prediction of psychosis in adolescents and

young adults at high risk: results from the prospective European prediction of psychosis study (EPOS).

Archives of General Psychiatry; 67:241–251.

6 Simon AE, Umbricht D. (2010).High remission rates from an initial ultra-high risk state for psychosis.

Schizophrenia Research; 116:168–172.

7 Verdoux, H., van Os, J. (2002). Psychotic symptoms in non-clinical populations and the continuum of psychosis.

Schizophrenia Research; 54:59–65.

8 Klosterkötter, J., Ruhrmann, S., Schultze-Lutter, F. et al. (2005). The European Prediction of Psychosis Study

(EPOS): integrating early recognition and intervention in Europe. World Psychiatry; 4:161–167.

9 World Health Organization (1993). Composite international diagnostic interview, Core Version 1.1.Arlington,

VA: American Psychiatric Publishing.

10 Mcglashan, T. H., Miller, T. J., Woods, S. W., Rosen, J. L., Hoffman, R. E., Davidson, L. (2001).

Structured Interview for prodromal syndromes (Version 3.0). New Haven: PRIME Research Clinic, Yale School

of Medicine.

11 Miller TJ, Mcglashan TH, Rosen JL et al. (2002). Prospective diagnosis of the initial prodrome for schizophrenia

based on the Structured Interview for Prodromal Syndromes: preliminary evidence of interrater reliability and

predictive validity. American Journal of Psychiatry; 159:863–865.

12 American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders, 4th edn.

Washington, DC: APA.

13 Kay, S. R., Fiszbein, A., Opler, L. A. (1987). The positive and negative symptom scale (PANSS).

Schizophrenia Bulletin; 13:261–276.

14 First, M. B., Spitzer, R. L, Gibbon, M., Williams, J. B. W. (1997). Structured clinical interview for DSMIV Axis I

Disorders (SCID-I). Arlington, VA: American Psychiatric Publishing.

15 Becker, T., Knapp, M., Knudsen, H. C. et al. (1999). The EPSILON study of schizophrenia in five European

countries. Design and methodology for standardising outcome measures and comparing patterns of care

and service costs. British Journal of Psychiatry; 175:514–521.

16 Yung, A. R., Buckby, J. A., Cotton, S. M. et al. (2006). Psychotic-like experiences in nonpsychotic Help-seekers:

Associations with distress, depression, and disability. Schizophrenia Bulletin; 32:352–359.

17 Wigman, J. T. W., Vollebergh, W. A. M., Raaijmakers, Q. A. W. (2009). The structure of the extended psychosis

phenotype in early adolescence – a cross-sample replication. Schizophrenia Bulletin.

DOI:10.1093/schbul/sbp154 [Epub ahead of print].

18 Mishara, A. L. (2010). KLAUS CONRAD (1905-1961):

Delusional mood, psychosis, and beginning schizophrenia. Schizophrenia Bulletin; 36:9–13.

Chapter 8

137

Chapter 8

19 Mcgorry, P. D., Yung, A. R., Phillips, L. J. et al. (2002).

Randomized controlled trial of interventions designed to reduce the risk of progression to first-episode psychosis

in a clinical sample with subthreshold symptoms. Archives of General Psychiatry; 59:921–928.

20 Mcglashan, T. H., Zipursky, R. B., Perkins, D. et al. (2006). Randomized, double-blind trial of olanzapine versus

placebo in patients prodromally symptomatic for psychosis. American Journal of Psychiatry; 163:790–799.

21 Ruhrmann, S., Bechdolf, A., Kühn, K. U. et al. (2007). Acute effects of treatment for prodromal symptoms

for people putatively in a late initial prodromal state of psychosis. British Journal of Psychiatry Suppl; 191

(Suppl. 51): s88–s95.

22 De Koning, M. B., Bloemen, O. J. N., van Amelsvoort, T. A. et al. (2009). Early intervention in patients at ultra

high risk of psychosis: benefits and risks. Acta Psychiatrica Scandinavica;119:426–442.

23 Amminger, G. P., Schäfer, M. R., Papageorgiou, K. et al. (2010). Longchain omega-3 fatty acids for indicated

prevention of psychotic disorders: a randomized, placebo-controlled trial. Archives of General Psychiatry; 67:146–154.

138

139

Chapter 9

ULTRA HIGH-RISK STATE FOR PSYCHOSIS AND NON-TRANSITION: A SYSTEMATIC REVIEW.

Andor E. Simon*, Eva Velthorst*, Dorien H. Nieman , Don Linszen, Daniel Umbricht, Lieuwe de Haan

* Joint first authorship

SCHIZOPHRENIA RESEARCH, IN PRESS

140

ABSTRACT

Background Most effort in ultra high-risk (UHR) research has been directed at defining the clinical and neurobiological characteristics of those UHR subjects who go on to develop psychosis. The characteristics and outcome of the remaining UHR subjects have remained relatively unexplored.

Method We performed a systematic review of clinical UHR studies to investigate whether information was available on the characteristics and outcome of UHR subjects who did not convert to psychosis.

Results Of 2462 potentially relevant papers, 31 met inclusion criteria, i.e. 20 naturalistic and 11 intervention studies. On average 76% (range 46 - 92.6%) of the UHR patients made no transition to psychosis during follow-up (range 6 to 40 months). Nearly half of the studies provided no charac-teristics of those UHR subjects who did not develop psychosis. Six studies reported remission rates from initial UHR status (range 15.4% to 54.3%). Linear regression showed that more recent studies reported significantly lower transition rates as compared to earlier publications. An older mean age at baseline was associated with significant lower transition rates in publications with follow-ups exceeding 1 year.

Conclusions Our review illustrates that the long-term outcome of UHR subjects that do not develop psychosis is to date under-investigated. The studies reporting remission rates suggest that UHR criteria capture a non-negligible proportion of subjects that do not convert to psychosis.

Chapter 9

141

Introduction

It has long been recognized that the onset of overt psychosis is preceded by a prodromal phase in the vast majority of cases. The introduction of the ‘ultra high-risk’ (UHR) criteria (Yung et al., 1996; Miller et al., 1999) in the second half of the 90’s formed a milestone in the creation of operationally defined criteria to identify individuals at risk for psychosis. Following a ‘close-in’ strategy, the UHR criteria comprised the combination of trait and state risk factors for psychosis. They were not intended to detect all prodromal patients, but tried to capture those at imminent risk of conversion to psychosis, i.e. the phase directly preceding onset of a first episode by 1 year. With the introduction of the Melbourne (Comprehensive Assessment of at Risk Mental States; CAARMS) (Yung et al., 2002) and Yale (Structured Interview of Prodromal Syndromes; SIPS) (McGlashan et al., 2001) scales the psychometric operationalization of the prodromal phase was substantially facilitated and ignited an impressive surge around the globe to enhance early recognition, secondary prevention and intervention of psychotic disorders. Research over the last 15 years, however, has unequivocally shown that transition rates from a UHR state to full-blown psychosis have been decreasing from above 50% in the initial studies using UHR criteria (Miller et al., 2002) to rates as low as 10- 15% in some recent studies (Haroun et al., 2006). This drop in reported transition rates raised concerns that these criteria may capture a high proportion of falsepositives, potentially causing fear, stigma and leading to unnecessary treatment (Yung et al., 2010). This may be of particular concern in adolescents and young adults as first onset of psychotic-like symptoms typically occurs in these age groups but more often represent transitory phenomena that do not necessarily foreshadow psychosis (Simon et al., 2009;van Os et al., 2009; Dominguez et al., 2011). The decreasing transition rates to psychosis arguably raise the question whether the remaining UHR subjects continue to present with UHR symptoms and to be at risk for psychosis, or even whether they actually fully remit from an initial UHR state. To date, this question has remained relatively unexplored; as the UHR literature has predominantly focussed on reporting transition rates to psychosis and on studying characteristics of UHR subjects that went on to develop psychosis. The present paper is a review of clinical UHR studies with a specific focus on whether information on rates of non-converting or remitting UHR subjects is provided.

Chapter 9

142

Chapter 9

Methods2.1 Search method

Database searches of publications targeted at subjects with an UHR state for a first episode of psychosis were conducted in Medline, PsychINFO and Embase, by combining the following two sets of keywords: (1) the keyword search terms ‘ultra high risk’, ‘uhr’, ‘(prodrome OR prodromal)’, ‘at high risk’, ‘clinical* at risk’, ‘(at clinical* risk), ‘at risk mental state’, ‘arms’, ‘attenuated symptoms’, ‘psychosis risk syndrome’, ‘(blips OR brief limited intermittent psychotic symptoms)’, ‘high risk patient’, ‘high risk population’, ‘high risk group’, ‘conversion’, ‘transition’, ‘remission’; and (2) the keyword search terms ‘psychosis’, ‘psychotic’, ‘psychotic disorder’, ‘prepsychosis’, ‘pre-psychosis’, ‘pre-psychotic’, ‘schizophreni*’. As the SIPS (McGlashan et al., 2001) and the CAARMS (Yung et al., 2002) nearly always refer to the assessment of subjects at UHR for developing psychosis, these terms were combined with the first set of keywords in order to find similar relevant articles not covered by the initial search criteria. To refine the search, we applied a ‘prospective research filter’ for Medline and Embase. Using such a filter for PsychINFO would have entailed the possibility of exempting some relevant records and was therefore not applied to this particular database. Our combined search eventually yielded 2462 potentially relevant papers. Initially the title and abstract were read by the authors to validate inclusion, and if necessary the entire article to identify studies that mentioned transition rates of subjects at UHR for developing psychosis.

2.2 Inclusion criteria

We included those papers in the review which: (1) were original research papers published since 1948 to 2011;(2) were published in English;(3) contained follow-up data;(4) reported transition/ conversion rates to a first psychotic episode;(5) specifically aimed at putative prodromal states of first episode psychosis, applying the UHR criteria as defined by the PACE clinic (Yung et al., 1996). This UHR approach consists of three alternative criteria: (i) Attenuated (Positive) Psychotic Symptoms (APS) group: having experienced sub- threshold, attenuated positive psychotic symptoms during the past year; (ii) Brief Limited Intermittent Psychotic Symptoms (BLIPS) group, defined by hallucinations, delusions or formal thought disorders resolving spontaneously within 1 week; and (iii) Trait plus State Risk Factor Group: a first-degree relative with a psychotic disorder or a DSM–IV schizotypal personality disorder of the index person, combined with a significant decrease in functioning during the past year;

143

Chapter 9

(6) and focussed primarily on measures of clinical or social outcomes, while studies that primarily assessed neurobiological measures, i.e. cognitive functioning or brain status applying structural and/or functional imaging data, were not included. Although the latter studies generally report transition rates as well (Pantelis et al., 2003;Seidman et al., 2010), not all of them systematically state whether they included all study subjects or just subsamples of larger clinical study cohorts. It cannot be ruled out that studies with a primary neurobiological focus or combining clinical with neurobiological measures request an additional level of consent from subjects and thus sample a narrower range of subjects. In contrast, studies assessing clinical and social outcomes sample a broader range of subjects and may be more representative in mirroring rates of transition and non-transition. For the sake of comparability and optimal homogeneity between studies, we chose to review studies that used UHR criteria only. Therefore, we did not include studies that used other sets of at-risk criteria (Klosterkötter et al., 2001), or UHR in combination with other at-risk criteria, e.g. basic symptoms (Ruhrmann et al., 2010), although these sets of at-risk criteria have importantly influenced and contributed to prodromal research. For the same reason, we did not include non-clinical general population studies. When more than one publication was found for the same UHR sample, we retained those with clinical and social outcome measures. When more than one publication reported clinical and social outcome measures, we included the most recent publication in our review. Where only transition rates were reported, we derived non-transition rates by assuming that non-transition was the outcome in the remaining subjects by taking account of the attrition rates.

A total of 31 studies, i.e. 20 naturalistic studies and 11 intervention studies from 13 different research groups met inclusion criteria.

2.3 Statistical analyses

SPSS for Windows (Version 18) was used to examine factors that may have contributed to the varying transition rates reported in different studies. Logistic regression was conducted to examine possible main and interaction effects of mean age, duration of follow-up and design of the study on non-transition rate.

Results

All studies with a naturalistic design are listed in Table 1. The term ‘naturalistic’ is used to delineate studies in which subjects are followed-up without any specific intervention. Studies with the primary aim to examine the effects of intervention are shown in Table 2.

144

Author

(Addington et al., 2011)

(See also:(Cannon et al., 2008;Woods

et al., 2009) /(Addington et al., 2007)

(Nelson et al., 2011)

(Velthorst et al., 2011)

(Simon and Umbricht, 2010)

(Fusar-Poli et al., 2010)

(Demjaha et al., 2010)

(Bechdolf et al., 2010)

(Nelson and Yung, 2010)

(Thompson et al., 2010)

(Broome et al., 2010)

Study

NAPLS

USA

PACE

Melbourne,

Australia

DUPS

Amsterdam,

Netherlands

Bruderholz

Switzerland

OASIS

London,

UK

OASIS

London,

UK

PACE

Melbourne,

Australia

PACE

Melbourne,

Australia

PACE

Melbourne,

Australia

OASIS

London,

UK

N

370

817

77

72

152

122

92

168

104

35

Attrition rate

18.1%

n=67

38%

n=307

12%

n=7

= attrition rate of

the UHR NT Group

41.7%

n=30

17.8%

n=27

28.8%

n=30*

* Attrition rate of

the UHR-NT sample

NR

None3

NR

20%

N=28

Follow- up

2.5 years

6 months

3 years

1 year

2 years

2 years

1.7 years

1 year

21.3

months

21.3

months

Mean age yrs

(range)

18

(SD=4.9)

Median

=18

(14 - 29)

19.2

(12 – 35)

20.3

(14 – 40)

23.5

(14-35)

23

(16 - 35)

18.0

(15 – 24)

18.3

(14 – 26)

20%

N=28

UHR-T:

24.0 (SD=3.8)

-

UHR-NT:

24.5

(SD=4.3)

18.1*

Table 1: Prospective UHR studies with a naturalistic design

145

Instruments1

SIPS

CAARMS

SIPS

SIPS

CAARMS

CAARMS

CAARMS

CAARMS

CAARMS

CAARMS

Inclusion group

APS: 100%

APS: 83%

BLIPS: 4%

Trait: 27%

NR

APS: 93.1%

BLIPS: 4.2%

Trait: 2.8%

NR

APS: 100%

BLIPS: 14.8%

Trait: 13.9%

APS: 80.5%

BLIPS: 5.5%

Trait: 30.5%

APS: 85.1%

BLIPS: 4.8%

Trait: 20.2%

APS: 66.3%

BLIPS: 27.9%

Trait: 37.5%

NR

Remission rate2

24%

* of the total sample

with complete

follow-up data

(n=303)

NR

54.3%*

n=38

* of the total sample

with complete

follow-up data

(n=70)

52.4%*

n=22

* of the total sample

with complete 1-yr

follow-up data

(n=42)

NR

NR

NR

NR

NR

NR

NR

Transition rate

(in %) by

inclusion group

APS: 35%

APS alone: 9.4%

BLIPS: 13.9%

APS + Trait: 8.7%

Trait alone: 4.3%

NR

NR

NR

NR

APS: 24.3%

BLIPS: 80.0%

Trait: 21.4%

NR

APS: 40.6%

BLIPS: 44.8%

Trait: 46.1%

NR

Non-Transit.

rate

65%

N=209

91.2%

n=745

74%

n=57

86.5%

n=65

84.4%

n=129

85.2%

n=104

78.3%

N=72

91.1%

n=153

60.6%

n=63

82.3%

n=23

Transition rate

35 %

n=82

8.8%

n=72

26%

n=20

13.5%

n=7

15.6%

n=23

14.8%

n=18

21.7%

N=20

8.9%

n=15

39.4%

n=41

17.9%

n=5

146

(Yung et al., 2008)

(further follow-up of (Yung et al., 2006)

(Yung et al., 2007)

(Cornblatt et al., 2007a)

(Haroun et al., 2006)

(see also (Kristensen and Cadenhead, 2007)

(Lam et al., 2006)

(Broome et al., 2005)

(Yung et al., 2004)

(continuation of research

(Yung et al., 2003)

(Mason et al., 2004)

(see also(Carr et al., 2000)

(Miller et al., 2002)

(see also (Miller et al., 2003)

(Yung et al., 1998)

(interim report of first 20 recruited of 38

included patients)

OHY

Melbourne,

Australia

PACE

Melbourne,

Australia

RAP

CAPPS

New York,

USA

CARE

Santiago,

USA

EASY

Hong Kong,

China

OASIS

London,

UK

PACE

Melbourne,

Australia

PAS

Melbourne,

Australia

PRIME

New Haven,

USA

PACE

Melbourne,

Australia

119

142

121

50

62

58

104

74

13

20

= sub

Sample

None3

None3

NR

(63.6%

n=77 missing

GAF-data)

20%

n=10

None3

27.6%

n=16

5.7%

n=6

None3

None3

Of the total sample

(n=38):

31.6%

n=12

2 years

3 years

1 year

1 year

6 months

30 months

6 months

2 years

1 year

6 months

(15 – 24)

(no separate

data available

on UHR

subsample)

None3

19.3

(SD=3.4)

RAP: 16.0

-

CAPP: 17.3

(12-29)

18.7

(12 – 30)

16.2

(7 – 24)

24.1

(SD=4.2)

19.4

(14 – 28)

17.3

(13 – 28)

17.8

(SD=6.1)

16-30

147

APS: 93.3%

BLIPS: 0%

Trait: 10.9%

NR

APS: 100%

NR

APS: 82.3%

BLIPS: 19.4%

Trait: 19.4%

APS: 84.5%

BLIPS: 20.7%

Trait: 13.8%

APS: 66.3%

BLIPS: 27.9%

Trait: 37.5%

APS: 58.1%

BLIPS: 31.1%

Trait: 25.7%

APS: 100%

BLIPS: 0%

Trait: 14.3%

NR

NR

NR

NR

23.1%*

n=6

* of the total sample

with complete 1-yr

follow-up data

(n=26)

NR

NR

NR

NR

15.4%

n=2

* of the total sample

with complete 1-yr

follow-up data

(n=13)

NR

NR

NR

APS: 19.1%

NR

APS: 33.3%

BLIPS: 25.0 %

Trait: 16.7%

NR

NR

APS: 51.2%

BLIPS: 60.9%

Trait: 10.5%

NR

NR

84%

n=100

64%

n=91

80.9%

n=98

85%

n=34

71%

n=44

85.7%

n=52

65.4%

n=68

50%

n=37

46%

n=6

60%

n=12

16%

n=19

36%

n=51

19.1%

n=23

15%

n=6

29%

n=18

14.3%

n=6

34.6%

n=36

50%

n=37

54%

n=7

40%

n=8

CAARMS

CAARMS

SIPS

SIPS

CAARMS

CAARMS

BPRS

APSS

BPRS

SIPS

CAARMS

148

Author

(Amminger et al.,2010)

(Lemos-Giraldez

et al., 2009)

(Walker et al., 2009)

Study about the effects of

medication in a subsample of

(Cannon et al., 2008)

(Shim et al., 2008)

(see also (Kwon et

al., 2010)

(Morrison et al.,

2007)

See also (Morrison

et al., 2004

(Phillips et al., 2007)

Further follow-up of(McGorry

t al., 2002)

Instruments1

PANSS

SIPS

SIPS

CAARMS

GHQ

PANSS

CAARMS

Study

Vienna,

Austria

P3

Cantabria,

Spain

NAPLS

USA

SEOUL

YOUTH

CLINIC

Seoul,

Korea

EDIE

Manchester,

UK

PACE

Melbourne,

Australia

N

Ω-3

PUF

A

41

-

Place

bo

40

61

191

27

58

59

Attrition rate

6.2%

n=5

27.4%

n=16

NR

44.4%

n=12

47%

n=31

30.5%

n=18

Follow- up

1 year

3 years

6 months

2 years

3 years

3-4 yr

Mean age yrs

(range)

Ω-3 PUFAs

16.8

-

Placebo

16.0

21.7

(15 – 31)

18.7

(SD=4.7)

21.5

(16 – 35)

22

(16 – 36)

NR

(16 – 36)

Table 2: Prospective UHR treatment studies

149

Inclusion group

APS: 90.1%

BLIPS:43.2%

(=transient

psychosis)

Trait: 7.4%

APS: 85.2%

BLIPS: 4.9%

Trait: 9.8%

NR

APS: 88.9%

BLIPS: 0%

Trait: 22.2%

APS: 82.8%

BLIPS:10.3%

Trait: 6.9%

NR

Medication

51% Ω-3

PUFAs

-

9.9%

antidepressant

s

-

3.7% Benzo’s

Total: 78.7 %

-

78.7 %

antipsychotics

Total: 90%

25.1%

antipsychotics

-

58.1%

antidepressant

s

Total: 100%

100%

antipsychotics

+

antidepressant

s or

anxiolytics on

need

None

Non-

Transit.

Rate

Ω-3

95.1%

n=36

-

Placebo

72.5%

n=27

87.5%

n=47

77.5%

n=148

92.6%

n=25

CT: 80%

n=28

-

Monitor

78%

n=18

SPI:

69%

n=21

-

NBI:

57.1%

n=16

Transition

rate

Ω-3

4.9%

n=2

-

Placebo

27.5%

n=11

23%

n=14

22.5%

n=43

7.4%

n=2

CT: 20%

n=7

-

Monitor

22%

n=5

SPI:

31%

n=10

-

NBI:

42.9%

n=12

Transition rate

(in %) by

inclusion group

NR

APS: 19.7%

BLIPS: 3.3%

Trait: 0%

NR

NR

NR

NR

Therapy

Needbased

treatment

(n=NR)

82%

CBT

none

Needbased

Treatment

(e.g. CBT)

CT +

monitoring

n=35

-

Monitoring

alone

n=23

Remission

rate1

NR

29.0%

n=9

* of the

total

sample

with

complete

1-yr

follow-up

data

(n=31)

NR

NR

NR

NRAfter cessation of RCT

(see McGorry et al., 2002):

74% sought treatment

(e.g. MHS, GP).

> 50% used either antidepressants/

anxiolytics &

> 50% antipsychotics

150

(Cornblatt et al.,

2007b)

(Amminger et al.,

2006)

(McGlashan et al.,

2006)

See also et al.

(Woods et al., 2003)

(McGorry et al.,

2002)

See also (Phillips et al., 2002)

(Morrison et al.,

2002)

SIPS

CAARMS

SIPS

CAARMS

PANSS

GHQ

GAF

RAP

New York,

USA

PACE

Melbourne,

Australia

PRIME

New Haven,

USA

PACE

Melbourne,

Australia

EDIE

Manchester,

UK

48

92

60

59

31

None3

6.5%

n=6

35.1%

n=27

None3

25.8%

n=8

30.5

months

1 year

1 year

1 year

6 months

15.8

(12-18)

19.74

(14–28)

Placebo:

17.2

SD=4.0

-

Olanzap

18.2

SD=5.5

20

(14 – 28)

23.2

SD=4.78

Subscript Table 1: Prospective UHR studies with a naturalistic design

1 I.e. instruments used to determine UHR status. 2 Remission = no score ≥ 3 on the SOPS & CAARMS positive items.

Remission rate is calculated with the available data of the UHR-NT sample. 3 ‘None’ might mean that a subsample with available follow-up data is drawn

from an unknown larger sample Abbreviations: APS= Attenuated Psychotic Symptoms, APSS= The Assessment of Prodromal and Schizotypal Symptoms,

ARMS= At Risk Mental State, BLIPS= Brief Limited Intermittent Psychotic Symptoms, BPRS= Brief Psychiatric Rating Scale, CAARMS = Comprehensive

Assessment for At Risk Mental States, CAPPS= Center for the Assessment and Prevention of Prodromal States, CBT= Cognitive Behavioural Therapy,

DUPS= Dutch Prediction of Psychosis Study, FIGS= Family Interview for Genetic Studies, NAPLS= North American Prodrome Longitudinal Study,

NR = not-reported, RCT= Randomized Control Trial, OASIS=Outreach And Support in South London, OHY= ORYGEN Youth Health, PACE=Personal

Assessment and Crisis Evaluation Clinic, PAS=Psychological Assistance Service, PRIME= Prevention through Risk Identification, Management, and

Education, RAP= The Recognition and Prevention, SIPS= Structured Interview for Prodromal Syndromes, SOFAS= Social and Occupational Functioning

Assessment, SSRI = Selective Serotonin Uptaker, UHR-NT = Ultra High Risk no transition.

151

After cessation of RCT

(see McGorry et al., 2002):

74% sought treatment

(e.g. MHS, GP).

> 50% used either antidepressants/

anxiolytics &

> 50% antipsychotics

APS: 100%

APS: 67.4%

BLIPS:30.2%

Trait: 32.6%

(1 or 2)&3:

NR

NR

APS: 83.9%

BLIPS: 6.5%

Trait: 9.7%

ADP:

41.7%

SGAP:

58.3%

30.4%

antipsyc

hotics

-

45%

SSRIs

31 pt.

received

olanzapine-

Placebo

n=29

none

75%

n=36

74.4%

N=64

65%

n=39

SPI:

81%

n=25

-

NBI:

64%

n=18

88%

n=18

25%

n=12

35%

n=21

35%

n=21

SPI: 19%

n=6

-

NBI: 36%

n=10

22%

n=5

APS: 25%

NR

NR

NR

APS: 22%

NR

69.6 %

Needbased

treatment

-

30.4%

CBT+risperidon

e

Supportive

psychosocial

interventions

Monitoring

-

CBT +

monitoring

NR

NR

NR

NR

NR

Subscript Table 2: Prospective UHR treatment studies

1 I.e. instruments used to determine UHR status.

2 Remission = no score ≥ 3 on the SOPS & CAARMS positive items. Remission rate is calculated on the base of the available data of the UHR-NT sample.

3 ‘None’ might mean that a subsample with available follow-up data is drawn from an unknown larger sample Abbreviations: APS= Attenuated Psychotic

Symptoms, ARMS= At Risk Mental State, BLIPS= Brief Limited Intermittent Psychotic Symptoms, BPRS= Brief Psychiatric Rating Scale, CAARMS =

Comprehensive Assessment for At Risk Mental States, CBT= Cognitive Behavioural Therapy, CT=Cognitive Therapy, EDIE= Early Detection and Intervention

Evaluation, GAF= General Assessment of Functioning, GHQ= General Health Questionnaire, NAPLS= North American Prodrome Longitudinal Study,

NBI= Needs Based Intervention, NR = not-reported, PACE=Personal Assessment and Crisis Evaluation Clinic, PANSS= Positive and Negative Syndrome

Scale, P3=Preventive Program for Psychosis, PRIME= Prevention through Risk Identification, Management, and Education, PUFA’s = polyunsaturated

fatty acids, RAP= The Recognition and Prevention Centre, RCT= Randomized Control Trial, SIPS= Structured Interview for Prodromal Syndromes,

SPI= Specific Preventive Intervention, SSRI = Selective Serotonin Uptaker, UHR-NT = Ultra High Risk no transition.

152

3.1 Non-transition rates

Overall, the 31 studies that were reviewed reported transition rates from 7.4 to 54% (mean rate= 24%, SD=12.1); thus on average 76% of the UHR patients had not made a transition to psychosis during follow-up (range 46 - 92.6%). Follow-up periods ranged from 6 to approximately 40 months. Performing linear regression, neither main nor interaction effects of mean age and duration of follow-up on transition rate was found (see Figure 1). However, an older mean age at baseline was associated with significant lower transition rates when examining publications with longer follow-up (>1 year) separately (adjusted R2=.28, F (1, 14) = 6.71, p=.02). The percentage of subjects not converting to a first psychotic episode did not significantly differ between subjects in treatment studies and subjects participating in studies with a naturalistic design (Beta=.07, t=.45, p=.66). Figure 2 shows the association between non-transition rate and year of publication. Signifi-cantly lower transition rates were reported in more recent naturalistic studies (adjusted R2=.27, F (1.18) = 6.61, p=.02), but not in intervention studies (adjusted R2=.18, F (1.9) = 1.96, p=.19). Overall, significantly lower transition rates were associated with more recent publication date in earlier publications (R2=.18, F(2.28)=4.32, p=.02).

3.2 Remission rates

Six studies (19.4%) (Miller et al., 2002;Haroun et al., 2006;Velthorst et al., 2011;Lemos-Giraldez et al., 2009;Simon and Umbricht, 2010; Addington et al., 2011), of which 5 with a naturalistic design, reported remission rates from initial UHR status (see Tables 1 and 2). Remission was defined as absence of a score of 3 (moderate) or more on any of the SOPS positive symptom items.

3.3 UHR groups

Twenty-one studies indicated the distribution of UHR groups (see ‘Inclusion group’ in Tables 1 and 2). In all of these studies, the APS group was by far the most common UHR group. Less than 10% BLIPS subjects were sampled in more recently published studies compared to earlier studies. Only 10 studies reported group-related transition rates. In 4 of 7 studies that sampled BLIPS subjects and provided group-related transition rates, transitions to psychosis from BLIPS groups outweighed those reported in the other UHR groups.

Chapter 9

153

Chapter 9

Figure 1. UHR- Non-Transition rates (UHR-NT), considering age and duration of follow-up.

AGE

ALL STUDIES

UH

R N

T (%

)

100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -

0 -

15 16 17 18 19 20 21 22 23 24 25

AGE

≥1 YEAR FOLLOW-UP

UH

R N

T (%

)

100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -

0 -

15 16 17 18 19 20 21 22 23 24 25

AGE

≤1 YEAR FOLLOW-UP

UH

R N

T (%

)

100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -

0 -

15 16 17 18 19 20 21 22 23 24 25

154

Figure 2. UHR- Non Transition rates (UHR-NT), considering year of publication and type of study

YEAR OF PUBLICATION

ALL STUDIES

UH

R N

T (%

)

100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -

0 -

1997 1999 2001 2003 2005 2007 2009 2011

YEAR OF PUBLICATION

NATURALISTIC DESIGN STUDIES

UH

R N

T (%

)

100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -

0 -

1997 1999 2001 2003 2005 2007 2009 2011

YEAR OF PUBLICATION

TREATMENT STUDIES

UH

R N

T (%

)

100 -

90 -

80 -

70 -

60 -

50 -

40 -

30 -

20 -

10 -

0 -

1997 1999 2001 2003 2005 2007 2009 2011

Chapter 9

155

Chapter 9

4 Discussion

We reviewed 20 naturalistic and 11 interventions studies of UHR subjects. Based on the reported transition rates in those studies that do not explicitly indicate nontransition rate, and on those studies that report a non-transition rate, we found that on average 76% of UHR subjects do not convert to psychosis (range 46% to 92.6%) during follow-up. Higher non-transition rates were found in some more recently published studies in comparison to early UHR studies. Only 6 studies report the rate of subjects that actually remitted from initial UHR status, with rates of 15.4% (Miller et al., 2002), 20% (Lemos-Giraldez et al., 2009), 23.1% (Haroun et al., 2006), 24% (Addington et al., 2011), 52.4% (Simon and Umbricht, 2010), and 54.3% (Velthorst et al., 2011).

4.1 Understanding the increasing non-transition rates

The finding that more recent studies were associated with higher non-transition rates is in line with a previous report (Yung et al., 2007). The global increase of the number of early detection and inter-vention services over the last 15 years since introduction of the UHR criteria and the efforts to increase awareness in general population have undoubtedly led to a broader spectrum of symptom severity being seen in early psychosis services. Subsequently, whereas the initial UHR studies sampled a narrower spectrum of UHR subjects, more recent studies most likely sample a clinical phenotype that embraces the entire continuum between non-clinical and clinical manifestation (van Os et al., 2000) and who thus may inherently be at lower risk. This seems to be corroborated by our additional finding that transition rates dropped in natural, but not in intervention studies, mirroring that naturalistic studies may sample a lower-risk population. Also, our review shows that fewer subjects with brief limited intermittent psychotic symptoms are being sampled in more recent studies. Therefore, clinical UHR samples may now present with symptoms that are closer to the subclinical symptom spectrum occurring in general population (Hanssen et al., 2005;van Os et al., 2009), in particular in adolescents (Poulton et al., 2000; Dominguez et al., 2011). As adolescence, however, is often precisely the phase of life when first signs of impending psychosis emerge, the question whether these phenomena actually express inherent risk for psychosis or whether they are of benign, transitory nature is of paramount relevance. In the seminal Dunedin birth cohort study by Poulton et al., self-reported psychotic-like symptoms at age 11 were associated with a 16-fold higher risk to develop a schizophreniform disorder at age 26 (Poulton et al., 2000). Along the same line, Dominguez et al. (2011) reported that persist-ence of subclinical psychotic symptoms in German adolescents aged 14 to 17 years was associated with higher transition rates to clinical psychosis over a follow up period of 8.4 years.

156

However, it is noteworthy that the majority of subjects enrolled in these studies never developed psychosis; in fact in most cases these subclinical psychotic symptoms were transitory. This is in agreement with high rates of transitory hallucinatory symptoms in children (Escher et al., 2002) and in adolescents (Simon et al., 2009), as well as with meta-analytic findings by van Os et al. (van Os et al., 2009) who reported 75-90% of developmental psychotic experiences to be transitory and to disappear over time. In summary, the clinical phenotypes of subjects that are assessed for putative at-risk states range from transitory and mostly benign through to severe and disabling phenomena and mirror a vast array of underlying mental vulnerability. The use of broader recruitment strategies in early psychosis services results in more subjects with a low psychosis vulnerability being sampled (Simon et al., 2007). A promising perspective to reduce non-transition rates is offered by combining UHR criteria with additional clinical and neurobiological variables offers, i.e. basic symptoms (Ruhrmann et al., 2010), negative symptoms (Riecher-Rössler et al., 2009), a stronger emphasis on functional and social impairment (Cannon et al., 2008; (Ruhrmann et al., 2010), and cognitive functioning (Riecher-Rössler et al., 2009).

4.2 How reliable are the high rates of non-transition?

The high non-transition and remission rates must be interpreted with caution. First, it may be that subjects that were lost to attrition went on to develop psychosis, thus decreasing the actual non-transition and remission rates, respectively. However, if we hypothesized that all subjects who were lost to attrition indeed went on to develop psychosis, the corrected non-transition rates still outweigh transition rates. Moreover, it has recently been shown that the less symptomatic subjects are more likely to be lost to attrition (Simon and Umbricht, 2010). Of note here that not all studies reported attrition rates and that only some studies corrected the rate of transition for attrition rate. Second, it may be that the follow-up intervals were too short to cover all transitions to psychosis. Transitions to psychosis have been observed even up to 55 months after initial assessment in one study (Riecher-Rossler et al., 2009), and in a very recent study with a maximum follow-up of 15 years, transitions to psychosis were reported even up to 10 years. It is noteworthy that 3 of the 6 reported remission rates were relatively low, i.e. 15.4% (Miller et al., 2002), 20% (Lemos-Giraldez et al., 2009), and 23.1% (Haroun et al., 2006). At least in a number of studies most subjects with an initial UHR status go on to present with UHR symptoms even after longer follow-up periods. This may particularly be the case for those subjects who are now seen in earlier phases of the prodrome as a result of earlier referral strategies (Yung et al., 2007). These subjects may convert to psychosis after a longer period of time, or alternatively, such early intervention may be more effective and may prevent further disease progression (Yung et al., 2007). The question about the further course of subjects with persisting UHR symptoms is pivotal and implies the question how long they need to be followed-up.

Chapter 9

157

Chapter 9

In terms of nosology, a state of persisting UHR symptoms strongly overlaps with schizotypal personality disorder; e.g. if a subject develops UHR symptoms but is seen in an early psychosis service only years later, the diagnosis would probably be schizotypal personality disorder. On the one hand, schizotypal personality disorder in adolescents has been associated with a higher transition rate to psychosis (Woods et al., 2009). On the other hand, remission from a persisting UHR state has been reported up to 3 years (Velthorst et al., 2011). Also, transition to psychosis is inversely related to length of follow-up periods (Cannon et al., 2008; Riecher-Rossler et al., 2009; Ruhrmann et al., 2010). Moreover, we found that an older mean age at baseline was associated with significant lower transition rates in studies with follow-up periods exceeding 1 year. Third, it could be argued that UHR symptoms may remit but later re-occur before a subsequent first episode of psychosis. A similar observation has been reported for basic symptoms which are thought to describe subtle, sub-clinical self-experienced and self–reported disturbances of thought processes, perception, drive and affect. Huber & Gross (Huber and Gross, 1989) observed so-called ‘out-post symptoms’ in 15% of their subjects, i.e. basic symptoms that disappeared and re-occurred at later stages. In contrast to these basic symptoms that capture the earliest phases of the initial prodrome, the UHR criteria, however, were modelled to capture the late prodrome immediately preceding overt psychosis and thus to characterize an increase in symptom severity and not the opposite. Further, in the studies by Simon & Umbricht (Simon and Umbricht, 2010) and by Velthorst et al. (Velthorst et al., 2011), it could be shown that level of functioning as measured with the GAF also improved with symptomatic remission. Such phenomenon would be highly improbable in subjects that are at true ultra high-risk for psychosis, as decreasing levels of functioning have robustly been documented to be a persistent core phenomenon along the entire prodrome of at-risk subjects who later converted to psychosis (Hafner et al., 1999). In a most recent analysis of the North American Prodrome Longitudinal Study (NAPLS), however, although the group of subjects who remitted in terms of symptoms did improve in social and role functioning, they still showed more severe functional impairment as compared to nonpsychiatric controls (Addington et al., 2011).

4.3 Implications of our findings

Our review illustrates that the long-term outcome of UHR subjects that do not develop psychosis is to date under-investigated. As the population seen in early psychosis services has shifted from a higher to a lower risk profile since introduction of the UHR criteria, their application should occur within consideration that a substantial proportion of these subjects will not convert to psychosis or may not even be at true risk for psychosis. At the same time we are aware that transition rate to psychosis, in a relative short period of time, in help-seeking UHR samples remains 400 fold higher than the incidence rate in the general population (Cannon et al., 2008).

158

Moreover the UHR approach offers the opportunity of minimizing the duration of untreated psychosis by early detection and intervention when the transition to a first episode psychosis occurs (Melle et al., 2004) Therefore, UHR research remains important in early recognition, secondary prevention and intervention of psychosis.

4.4 Limitations

We only included prospective UHR studies that set out to investigate clinical characteristics and did not include those that primarily assessed neurobiological or combined clinical and neurobiological measures. Latter studies request an additional level of consent from subjects and not always system-atically state whether they included the entire sample or only subsamples of a clinical study cohort. Thus, it cannot be ruled out that they sample a narrower range of UHR subjects and may not be as representative in mirroring rates of transition and non-transition as studies primarily focusing on measures of clinical and social outcome. Further, initial versions of the CAARMS did not include a one-year criterion for symptom onset or progression and thus may have captured subjects with sustained UHR symptoms, theoretically resulting in lower non-transition rates. Thus, we considered comparing the non-transition rates between studies using the CAARMS and those using the SIPS. Comparisons, however, would have been biased by the application of different versions of the CAARMS in the reviewed studies. Also, sample sizes of SIPS and CAARMS studies would have been too small to perform comparisons with required power.

Acknowledgements

The authors thank Joost Daams for assisting with the database searches and Maarten Koeter for his statistical advice.

Chapter 9

159

Chapter 9

References

Addington, J., Cadenhead,K.S., Cannon,T.D., Cornblatt,B., McGlashan,T.H., Perkins,D.O., Seidman, L.J., Tsuang,M.,

Walker,E.F., Woods,S.W., Heinssen,R. and North American Prodrome Longitudinal Study. (2007). North American

Prodrome Longitudinal Study: a collaborative multisite approach to prodromal schizophrenia research.

Schizophrenia Bulletin; 33: 665-672.

Addington, J., Cornblatt, B. A., Cadenhead, K. S., Cannon,T. D., McGlashan,T.H., Perkins,D.O., Seidman,L.J.,

Tsuang, M.T., Walker,E.F., Woods,S.W. and Heinssen,R. (2011). At Clinical High Risk for psychosis:

Outcome for Nonconverters. American Journal of Psychiatry

Amminger, G.P., Leicester,S., Yung, A.R., Phillips, L.J., Berger, G.E., Francey, S.M., Yuen, H.P. and McGorry, P.D.

(2006). Early-onset of symptoms predicts conversion to non-affective psychosis in ultra-high risk individuals.

Schizophrenia Research; 84: 67-76.

Amminger, G.P., Schafer, M.R., Papageorgiou, K., Klier, C.M., Cotton, S.M., Harrigan, M.S.M., Mackinnon, A.,

McGorry, P. D. and Berger, G.E. (2010). Long-chain -3 fatty acids for indicated prevention of psychotic disorders:

A randomized, placebo-controlled trial. Archives of General Psychiatry; 67: 146-154.

Bechdolf, A., Thompson, A., Nelson, B., Cotton, S., Simmons, M. B., Amminger, G.P., Leicester, S., Francey,S .M.,

McNab, C., Krstev, H., Sidis, A., McGorry, P.D. and Yung, A.R. (2010). Experience of trauma and conversion to

psychosis in an ultra-high-risk (prodromal) group. Acta Psychiatrica Scandinavia; 121: 377-384.

Broome, M.R., Day, F., Valli, I., Valmaggia, L., Johns, L.C., Howes, O., Garety, P. and McGuire, P. K., (2010).

Delusional ideation, manic symptomatology and working memory in a cohort at clinical high-risk for psychosis:

A longitudinal study. European Psychiatry.

Broome, M.R., Woolley, J.B., Johns, L.C., Valmaggia, L.R., Tabraham, P., Gafoor, R., Bramon, E. and Mcguire,

P.K. (2005). Outreach and support in south London (OASIS): Implementation of a clinical service for prodromal

psychosis and the at risk mental state. European Psychiatry; 20 (5-6): 372-378.

Cannon, T.D., Cadenhead, K., Cornblatt, B., Woods, S.W., Addington, J., Walker, E., Seidman, L.J., Perkins, D.,

Tsuang, M., McGlashan, T. and Heinssen, R. (2008). Prediction of psychosis in youth at high clinical risk:

a multisite longitudinal study in North America. Archives of General Psychiatry; 65: 28-37.

Carr, V., Halpin, S., Lau, N., O’Brien, S., Beckmann, J. and Lewin, T. (2000). A risk factor screening and assessment

protocol for schizophrenia and related psychosis. Australian and New Zealand Journal of Psychiatry; 34: S170-S180.

Cornblatt, B.A., Auther, A.M., Niendam, T., Smith, C.W., Zinberg, J., Bearden,C.E. and Cannon,T.D., (2007a).

Preliminary findings for two new measures of social and role functioning in the prodromal phase of schizophrenia.

Schizophrenia Bulletin; 33: 688-702.

Cornblatt, B. A., Lencz, T., Smith, C. W., Olsen, R., Auther, A. M., Nakayama, E., Lesser, M. L., Tai, J. Y., Shah, M. R.,

Foley, C. A., Kane, J. M. and Correll, C. U. (2007b). Can antidepressants be used to treat the schizophrenia prodrome?

Results of a prospective, naturalistic treatment study of adolescents. Journal of Clinical Psychiatry; 68: 546-557.

160

Chapter 9

Demjaha, A., Valmaggia, L., Stahl, D., Byrne, M. and McGuire, P. (2010). Disorganization/Cognitive and Negative

Symptom Dimensions in the At-Risk Mental State Predict Subsequent Transition to Psychosis. Schizophrenia Bulletin

Dominguez, M., Wichers, M., Lieb, R., Wittchen, H. U. and van Os.J. (2011). Evidence that onset of clinical psychosis

is an outcome of progressively more persistent subclinical psychotic experiences: an 8-year cohort study.

Schizophrenia Bulletin; 37: 84-93.

Escher, S., Romme, M., Buiks, A., Delespaul, P. and van Os J. (2002). Independent course of childhood

auditory hallucinations: a sequential 3-year follow-up study. British Journal of Psychiatry; Suppl. 43: s10-s18.

Fusar-Poli, P., Byrne, M., Valmaggia, L., Day, F., Tabraham, P., Johns, L., McGuire, P. and OASIS Team. (2010).

Social dysfunction predicts two years clinical outcome in people at ultra high risk for psychosis. Journal of Psychiatry

Research; 44: 294-301.

Hafner, H., Loffler, W., Maurer, K., Hambrecht, M. and an der Heiden, W. (1999). Depression, negative symptoms,

social stagnation and social decline in the early course of schizophrenia. Acta Psychiatrica Scandinavica; 100: 105-118.

Hanssen, M., Bak, M., Bijl, R., Vollebergh, W. and van Os, J. (2005). The incidence and outcome of subclinical

psychotic experiences in the general population. British Journal of Clinical Psychology; 44: 181-191.

Haroun, N., Dunn, L., Haroun, A. and Cadenhead, K. S. (2006). Risk and protection in prodromal schizophrenia:

ethical implications for clinical practice and future research. Schizophrenia Bulletin; 32: 166-178.

Huber, G. and Gross, G. (1989). The concept of basic symptoms in schizophrenic and schizoaffective psychoses.

Recenti Progressi in Medicina; 80: 646-652.

Klosterkotter, J., Hellmich, M., Steinmeyer, E. M. and Schultze-Lutter, F. (2001). Diagnosing schizophrenia in the initial

prodromal phase. Archives of General Psychiatry; 58: 158-164.

Kristensen, K. and Cadenhead, K.S. (2007). Cannabis abuse and risk for psychosis in a prodromal sample.

Psychiatry Research; 151; 151-154.

Kwon, J. S., Shim, G., Park, H. Y. and Jang, J. H. (2010). Current concept of prodrome from the experience of the seoul

youth clinic high risk cohort in Korea. Clinical Neuropsychiatry; 7: 56-62.

Lam, M. M., Hung, S. F. and Chen, E. Y. (2006). Transition to psychosis: 6-month follow-up of a Chinese high-risk

group in Hong Kong. Australian and New Zealand Journal of Psychiatry; 40: 414-420.

Lemos-Giraldez, S., Vallina-Fernandez, O., Fernandez-Iglesias,P., Vallejo-Seco, G., Fonseca-Pedrero, E.,

Paino-Pineiro, M., Sierra-Baigrie, S., Garcia-Pelayo, P., Pedrejon-Molino, C., Alonso-Bada, S., Gutierrez-

Perez, A. and Ortega-Ferrandez, J.A. (2009). Symptomatic and functional outcome in youth at ultra-high risk

for psychosis: a longitudinal study. Schizophrenia Research; 115: 121-129.

Mason, O., Startup, M., Halpin, S., Schall, U., Conrad,A. and Carr,V. (2004). Risk factors for transition to first episode

psychosis among individuals with ‘at-risk mental states’. Schizophrenia Research; 71 (2-3): 227-237.

161

Chapter 9

McGlashan, T.H., Miller, T. J., Woods, S. W. et al. (2001). Structured Interview for Prodromal Syndromes

(Version 3.0, unpublished manuscript) PRIME Research Clinic, Yale School of Medicine, New Haven, Connecticut.

McGlashan, T. H., Zipursky, R. B., Perkins, D., Addington, J., Miller, T., Woods, S. W., Hawkins, K. A., Hoffman,

R. E., Preda, A., Epstein, I., Addington, D., Lindborg, S., Trzaskoma, Q., Tohen, M. and Breier, A. (2006).

Randomized, double-blind trial of olanzapine versus placebo in patients prodromally symptomatic for psychosis.

American Journal of Psychiatry; 163: 790-799.

McGorry, P. D., Yung, A. R., Phillips, L. J., Yuen, H. P., Francey, S., Cosgrave, E. M., Germano, D., Bravin, J., McDonald,

T., Blair, A., Adlard, S. and Jackson, H. (2002). Randomized controlled trial of interventions designed to reduce the risk

of progression to first-episode psychosis in a clinical sample with subthreshold symptoms. Archives of General

Psychiatry; 59: 921-928.

Melle, I., Larsen, T.K., Haahr, U., Friis, S., Johannessen, J. O., Opjordsmoen, S., Simonsen, E., Rund, B. R., Vaglum, P.

and McGlashan, T. (2004). Reducing the duration of untreated first-episode psychosis: effects on clinical presentation.

Archives of General Psychiatry; 61: 143-150.

Miller, T. J., McGlashan, T. H., Rosen, J. L., Cadenhead, K., Cannon, T., Ventura, J., McFarlane, W., Perkins, D. O.,

Pearlson, G. D. and Woods, S. W. (2003). Prodromal assessment with the structured interview for prodromal

syndromes and the scale of prodromal symptoms: predictive validity, interrater reliability, and training to reliability.

[Erratum appears in Schizophrenia Bulletin. 2004; 30(2):following 217]. Schizophrenia Bulletin; 29: 703-715.

Miller, T. J., McGlashan, T. H., Rosen, J. L., Somjee, L., Markovich, P. J., Stein, K. and Woods, S. W. (2002).

Prospective diagnosis of the initial prodrome for schizophrenia based on the Structured Interview for Prodromal

Syndromes: preliminary evidence of interrater reliability and predictive validity. American Journal of Psychiatry;

159: 863-865.

Miller, T.J., McGlashan, T.H., Woods,S.W., Stein, K., Driesen, N., Corcoran, C. M., Hoffman, R. and Davidson, L. (1999).

Symptom assessment in schizophrenic prodromal states. Psychiatric Quarterly; 70: 273-287.

Morrison, A. P., Bentall, R. P., French, P., Walford, L., Kilcommons, A., Knight, A., Kreutz, M. and Lewis, S. W. (2002).

Randomised controlled trial of early detection and cognitive therapy for preventing transition to psychosis in high-risk

individuals: Study design and interim analysis of transition rate and psychological risk factors. [References].

British Journal of Psychiatry; 181: s78-s84.

Morrison, A. P., French, P., Parker, S., Roberts, M., Stevens, H., Bentall, R. P. and Lewis, S. W. (2007).

Three-year follow-up of a randomized controlled trial of cognitive therapy for the prevention of psychosis in people

at ultrahigh risk. Schizophrenia Bulletin; 33: 682-687.

Morrison, A. P., French, P., Walford, L., Lewis, S. W., Kilcommons, A., Green, J., Parker, S. and Bentall, R. P. (2004).

Cognitive therapy for the prevention of psychosis in people at ultra-high risk: Randomised controlled trial. British

Journal of Psychiatry; 185: 291-297.

Nelson, B., Yuen, K. and Yung, A. R. (2011). Ultra high risk (UHR) for psychosis criteria: are there different levels of

risk for transition to psychosis? Schizophrenia Research; 125: 62-68.

162

Chapter 9

Nelson, B. and Yung, A. R. (2010). Can clinicians predict psychosis in an ultra high risk group?

Australian and New Zealand Journal of Psychiatry; 44: 625-630.

Pantelis, C., Velakoulis, D., McGorry, P. D., Wood, S. J., Suckling, J., Phillips, L. J., Yung, A. R., Bullmore, E. T., Brewer,

W., Soulsby, B., Desmond, P. and McGuire, P. K. (2003). Neuroanatomical abnormalities before and after onset of

psychosis: a cross-sectional and longitudinal MRI comparison. Lance; 361: 281-288.

Phillips, L. J., Curry, C., Yung, A. R., Yuen, H. P., Adlard, S. and McGorry, P. D. (2002). Cannabis use is not associated

with the development of psychosis in an ‘ultra’ high-risk group. Australian and New Zealand Journal of Psychiatry;

36: 800-806.

Phillips, L. J., McGorry, P. D., Yuen, H. P., Ward, J., Donovan, K., Kelly, D., Francey, S. M. and Yung, A. R. (2007).

Medium term follow-up of a randomized controlled trial of interventions for young people at ultra high risk of

psychosis. Schizophrenia Research; 96: 25-33.

Poulton, R., Caspi, A., Moffitt, T. E., Cannon, M., Murray, R. and Harrington, H. (2000). Children’s self-reported

psychotic symptoms and adult schizophreniform disorder: a 15-year longitudinal study. Archives of General Psychiatry;

57: 1053-1058.

Riecher-Rossler, A., Pflueger, M. O., Aston, J., Borgwardt, S. J., Brewer, W. J., Gschwandtner,U. and Stieglitz,

R. D. (2009). Efficacy of using cognitive status in predicting psychosis: a 7-year follow-up. Biological Psychiatry; 66:

1023-1030.

Ruhrmann, S., Schultze-Lutter,F., Salokangas, R.K., Heinimaa,M., Linszen, D., Dingemans,P., Birchwood, M.,

Patterson, P., Juckel, G., Heinz, A., Morrison, A., Lewis, S., von Reventlow,H.G. and Klosterkotter, J. (2010).

Prediction of psychosis in adolescents and young adults at high risk: results from the prospective European

prediction of psychosis study. Archives of General Psychiatry; 67: 241-251.

Seidman, L. J., Giuliano, A. J., Meyer, E. C., Addington, J., Cadenhead, K. S., Cannon, T. D., McGlashan, T. H., Perkins,

D. O., Tsuang, M. T., Walker, E. F., Woods, S. W., Bearden, C. E., Christensen, B. K., Hawkins, K., Heaton, R., Keefe, R.

S., Heinssen, R. and Cornblatt, B. A. (2010). Neuropsychology of the prodrome to psychosis in the NAPLS consortium:

relationship to family history and conversion to psychosis. Archives of General Psychiatry; 67: 578-588.

Shim, G. S., Kang, D.-H., Choi, J.-S., Jung, M. H., Kwon, S. J., Jang, G. E. and Kwon, J. S. (2008). Prospective outcome

of early intervention for individuals at ultra-high-risk for psychosis. Early Intervention in Psychiatry; 2: 277-284.

Simon, A. E., Cattapan-Ludewig, K., Gruber, K., Ouertani, J., Zimmer, A., Roth, B., Isler, E. and Umbricht, D. (2009).

Subclinical hallucinations in adolescent outpatients: an outcome study. Schizophrenia Research; 108: 265-271.

Simon, A. E. and Umbricht, D. (2010). High remission rates from an initial ultra-high risk state for psychosis.

Schizophrenia Research; 116: 168-172.

Simon, A. E., Cattapan-Ludewig, K., Zmilacher, S., Arbach, D., Gruber, K., Dvorsky, D. N., Roth, B., Isler, E., Zimmer, A.

and Umbricht, D. (2007). Cognitive functioning in the schizophrenia prodrome. Schizophrenia Bulletin; 33: 761-771.

163

Chapter 9

Thompson, A., Nelson, B. and Yung, A. R. (2010). Predictive validity of clinical variables in the “at risk” for psychosis

population: International comparison with results from the North AmericanProdrome Longitudinal Study.

Schizophrenia Research; 126; 51-57.

Van Os. J., Hanssen, M., Bijl, R. V. and Ravelli, A. (2000). Strauss (1969) revisited: a psychosis continuum

in the general population? Schizophrenia Research; 45: 11-20.

Van Os. J., Linscott, R. J., Myin-Germeys, I., Delespaul, P. and Krabbendam, L. (2009). A systematic review and

meta-analysis of the psychosis continuum: evidence for a psychosis proneness-persistence-impairment model

of psychotic disorder. Psychological Medicinel; 39: 179-195.

Velthorst, E., Nieman, D. H., Klaassen, R. M., Becker, H. E., Dingemans, P. M., Linszen, D. H. and de, Haan.L. (2011).

Three-year course of clinical symptomatology in young people at ultra high risk for transition to psychosis. Acta

Psychiatrica Scandinavica; 123: 36-42.

Walker, E. F., Cornblatt, B. A., Addington, J., Cadenhead, K. S., Cannon, T. D., McGlashan, T. H., Perkins,

D. O., Seidman, L. J., Tsuang, M. T., Woods, S. W. and Heinssen, R. (2009). The relation of antipsychotic and

antidepressant medication with baseline symptoms and symptom progression: a naturalistic study of the

North American Prodrome Longitudinal Sample. Schizophrenia Research; 115: 50-57.

Woods, S. W., Addington, J., Cadenhead, K. S., Cannon, T. D., Cornblatt, B. A., Heinssen, R., Perkins, D. O.,

Seidman, L. J., Tsuang, M. T., Walker, E. F. and McGlashan, T. H. (2009). Validity of the prodromal risk syndrome for

first psychosis: Findings from the north american prodrome longitudinal study. Schizophrenia Bulletin; 35: 894-908.

Woods, S. W., Brier, A., Zipursky, R. B., Perkins, D. O., Addington, J., Miller, T. J., Hawkins, K. A., Marquez, E., Lindborg,

S. R., Tohen, M. and McGlashan, T. H. (2003). Randomized Trial of Olanzapine versus Placebo in the Symptomatic

Acute Treatment of the Schizophrenic Prodrome. Biological Psychiatry; 54: 453-464.

Yung, A. R., McGorry, P. D., McFarlane, C. A., Jackson, H. J., Patton, G. C. and Rakkar, A. (1996).

Monitoring and care of young people at incipient risk of psychosis. Schizophrenia Bulletin; 22: 283-303.

Yung, A. R., Nelson, B., Stanford, C., Simmons, M. B., Cosgrave, E. M., Killackey, E., Phillips, L. J., Bechdolf, A., Buckby,

J. and McGorry, P. D. (2008). Validation of “prodromal” criteria to detect individuals at ultra high risk of psychosis: 2

year follow-up. Schizophrenia Research; 105: 10-17.

Yung, A. R., Nelson, B., Thompson, A. D. and Wood, S. J. (2010). Should a “Risk Syndrome for Psychosis” be included

in the DSMV? Schizophrenia Research; 120: 7-15.

Yung, A. R., Phillips, L., McGorry, P., Ward, J., Donovan, K. and Thompson, K. (2002). Comprehensive assessment

of at risk mental states (CAARMS) PACE clinic, Department of Psychiatry, University of Melbourne, Melbourne.

Yung, A. R., Phillips, L. J., McGorry, P. D., McFarlane, C. A., Francey, S., Harrigan, S., Patton, G. C. and Jackson, H. J.

(1998). Prediction of psychosis. A step towards indicated prevention of schizophrenia. British Journal of Psychiatry

Suppl. 172; 14-20.

164

Chapter 9

Yung, A. R., Phillips, L. J., Yuen, H. P., Francey, S. M., McFarlane, C. A., Hallgren, M. and McGorry, P. D. (2003).

Psychosis prediction: 12-month follow up of a high-risk (“prodromal”) group. Schizophrenia Research; 60: 21-32.

Yung, A. R., Phillips, L. J., Yuen, H. P. and McGorry, P. D. (2004). Risk factors for psychosis in an ultra high-risk group:

psychopathology and clinical features. Schizophrenia Research; 67: 131-142.

Yung, A. R., Stanford, C., Cosgrave, E., Killackey, E., Phillips, L., Nelson, B. and McGorry, P. D. (2006).

Testing the Ultra High Risk (prodromal) criteria for the prediction of psychosis in a clinical sample of young people.

Schizophrenia Research; 84: 57-66.

Yung, A. R., Yuen, H. P., Berger, G., Francey, S., Hung, T. C., Nelson, B., Phillips, L. and McGorry, P. (2007). Declining

transition rate in ultra high risk (prodromal) services: dilution or reduction of risk? Schizophrenia Bulletin; 33: 673-681.

165

166

167

Chapter 10

SUMMARY, CONCLUSIONS AND DISCUSSION

168

Summary

The objective of this thesis was to increase our knowledge about the Ultra High Risk (UHR) phase for developing a first psychotic episode. In this light, our aims were (1) to acquire a better understanding of the demographic characteristics and baseline symptomatology of this helpseeking (distressed) group and (2) to identify factors that predict transition to psychosis and thereby eventually could diminish the number of false-positives (subjects without transition to psychosis) in UHR research. In particular, the presence and predictive value of negative symptoms and social disability was examined in more detail. The third aim (3) was to investigate the course of UHR symptoms of young people who eventually do not convert to psychosis.

1. Demographic characteristics and baseline symptomatology of subjects at UHR for developing a first psychosis (chapters 2 and 3)

After introducing the subject of this thesis in chapter 1, chapter 2 presents findings about ethnic differences in baseline UHR symptomatology in 201 help-seeking subjects participating in the Dutch Early Detection and Intervention (EDIE-NL) trial. The results demonstrate that among the subjects from Moroccan and Turkish descent, more negative symptoms – anhedonia in particular- and more depression symptoms were reported. By examining possible contributing factors that could have accounted for (part of) these ethnic differences, it was found that ethnic identity was found to be inversely related to symptoms in the subjects from Moroccan descent. Elaborating on the findings of chapter 2 the question arose whether it is possible to differentiate between negative and depression symptoms in the UHR group. Chapter 3 focuses on this differentiation by performing an exploratory factor analysis of the Scale Of Prodromal Symptoms (SOPS) in the Dutch Prediction of Psychosis Study (DUPS). It was demonstrated that it is indeed possible to differentiate between the two symptom clusters by some items of the SOPS. In particular, ‘social anhedonia and withdrawal’, ‘decreased expression of emotion’ and ‘decreased experience of emotions and self’ appeared as specific negative symptoms. On the other hand, within our group, ‘dysphoric mood’, ‘impaired tolerance to normal stress’ and ‘personal hygiene/social attentiveness’ came forward as depression symptoms. Instead, ‘avolition’ and ‘deterioration in role functioning’ only had moderate loadings on the depression and negative symptoms factors,

Chapter 10

169

2. Identifying variables that could contribute to improved prediction of a first psychotic episode (chapters 4, 5, 6 and 7)

In chapter 4, global differences in baseline clinical symptomatology and general level of functioning (GAF-score) are investigated between UHR patients who did (UHR+T) or did not (UHR+NT) make a transition to psychosis. The 72 subjects were examined at the Academic Medical Center of the University of Amsterdam, the Netherlands (DUPS). This study revealed that severity of specific symptoms at baseline is related to transition to psychosis in UHR subjects. In particular, the UHR+T group showed more social anhedonia and withdrawal, more bizarre thinking and a lower GAF score at baseline compared to the UHR+NT group.The highly significant difference in the negative symptom item ‘social anhedonia and withdrawal’ at baseline between those who did and did not convert to a first psychosis, raised interest in the possible contributing role of impaired social functioning in the prodromal phase. Therefore, amongst other disabilities in a variety of domains, the role of impaired social functioning in the prediction of a first psychosis has been further examined in chapter 5. In the European Prediction of Psychosis Study (EPOS), a prospective multicentre, naturalistic field study with an 18-month follow-up period in 245 help-seeking individuals clinically at high risk, disability was assessed with the Disability Assessment Schedule of the World Health Organization (WHODAS–II). As we expected, it was found that baseline difficulties in social disability – especially in the areas of keeping and making friendships - significantly contributed to the prediction of transition to psychosis within our sample. In chapter 6, the predictive value of the Strauss and Carpenter Prognostic Scale (SCPS), an instrument that is widely used in first- psychosis research, is investigated in a UHR population (EPOS). This study revealed that the SCPS has predictive value for transition to psychosis in the UHR period, suggesting temporal continuity of predictors of outcome. More severe positive symptoms, a reduced ability to work or study, a low quality of social relationships and high levels of subjective distress were independent predictors of psychosis in UHR subjects. In combination with other variables, the SCPS could make a valuable contribution to a more accurate prediction of a first psychotic episode. The important role of social functioning in the prediction of psychosis leads to the question whether social disability during admission for a first psychosis also contributes to the prediction of future disease outcome once the psychosis threshold has already been crossed. In chapter 7, we therefore examine in a Dutch cohort of 82 patients with schizophrenia, whether a detailed assessment of social disability (assessed with the Groninger Social Disabilities Schedule-II) in the month prior to admission for a first psychotic episode, contributes to the prediction of disease outcome in terms of psychopa-thology. After controlling for PANSS-sum score at baseline, none of the social disability domains significantly predicted the number of relapses or the severity of clinical symptomatology at five-year follow-up. This result suggests that poor social functioning at admission for a first psychosis does not necessarily predict poor disease outcome, and perhaps indicates that in order to predict the course of schizophrenia in a reliable way, it may be necessary to assess social functioning during clinical stabilization.

Chapter 10

170

3. Investigating the course of UHR subjects who eventually do not convert to a first psychotic episode (chapters 8 and 9)

In chapter 8 the course of UHR symptoms, GAF score and employment status is investigated in 57 patients who did not make a transition to psychosis and who were examined within the Dutch Prediction of Psychosis Study in Amsterdam, the Netherlands. At the 3- year follow-up, 75% of the patients who did not make a transition to psychosis had remitted from UHR status. A substantial percentage of patients initially at UHR for psychosis reported significantly fewer UHR symptoms within 1 year. Recovery was not limited to a reduction in symptomatology, but also involved improvement in global functioning as well as in employment status. UHR symptoms did not re-occur after improvement, signifying that with the current UHR criteria, a substantial percentage of the included subjects appears to have only transitory complaints and dysfunction. Combining UHR criteria with additional clinical and biological variables may offer a promising perspective to improve the prediction of transition and thereby reduce non-transition rates.In chapter 9, we review clinical UHR studies systematically and summarize available information on the characteristics and outcome of UHR subjects who did not convert to psychosis. We found that nearly half of the relevant 31 studies did not provide any characteristics of those UHR subjects who did not develop psychosis. Five studies reported remission rates from initial UHR status (range 15.4% to 54.3%). Linear regression showed that more recent studies reported significantly lower transition rates as compared to earlier publications. In addition, an older mean age at baseline was associated with significant lower transition rates in publications with follow-ups exceeding 1 year. The clinical phenotypes of subjects who were assessed for putative at-risk states varied widely, and ranged from transitory and mostly benign through to severe and disabling phenomena.

Conclusions

The central objective of this thesis was to increase our knowledge about the Ultra High Risk (UHR) phase for developing a first psychotic episode. We believe that the results of the various sub-studies have provided new perspectives on the baseline characteristics of this group and on criteria that could contribute to a better psychosis prediction. To conclude, I will summarize our main results point by point, finishing with a discussion and some recommendations for future research:

1. Help seeking UHR subjects of Moroccan-Dutch and Turkish-Dutch origin present with more negative symptoms (anhedonia in particular) and with more depression symptoms compared to the native Dutch group; in the Moroccan-Dutch group probably partly related to a lower level of ethnic identity.

Chapter 10

171

2. It may possible to differentiate between negative and depression symptoms in the UHR phase with the Scale Of Prodromal Symptoms (SOPS).

3. Severity of ‘social anhedonia and withdrawal’, ‘bizarre thinking’ and a lower GAF score may be related to transition to psychosis in UHR subjects.4. UHR subjects who convert to a psychosis report significantly more social impairments compared to those who do not.

5. With the current UHR criteria that are mainly based on positive symptoms, a large percentage of the included subjects appear to have only transitory complaints and dysfunction.

6. The Strauss and Carpenter Prognostic Scale has predictive value for transition to psychosis in the UHR period, indicate a temporal continuity of the predictive value of variables often used in schizophrenia studies.

7. Social functioning at admission for a first psychosis does not necessarily predict poor disease outcome. This result may indicate that in order to predict the course of schizophrenia in a reliable way, it may be necessary to assess social functioning during clinical stabilization.

In conclusion, our results indicate that in young people meeting the UHR criteria, in particular the severity of baseline negative symptoms and social impairments differentiate between the young people who do and who do not convert to a first psychosis. These findings give way to hypothesize about the complex mechanisms associated with the development of a first psychosis. Especially withdrawal from social interaction may be an important factor. Although it is likely that withdrawal from social interaction is not merely a causal factor in the development of psychotic symptoms, but could also be the result of e.g. paranoia, we believe that if thoughts are not tested in social interaction, social withdrawal (either through anhedonia or social (cognitive) impairments) might also further reinforce the development and maintenance of delusional ideas and suspiciousness: The lonely soul wanders…

Chapter 10

172

Discussion

Since the mid nineties, UHR research predominantly focused on attenuated positive symptomatology and the importance of negative symptoms remained largely unexplored. In the present thesis we have found that, in order to fully describe the UHR phase and to reliable predict its course, more emphasis should be put on social impairments and the negative item social anhedonia. These impairments could cause withdrawal from contact with peers and in turn may further reinforce delusional ideas. The importance of negative symptomatology in the prediction of the course of the UHR phase is in concordance with studies on the course of schizophrenia in which it has been argued that, given the low discriminative power of positive symptoms, more focus on negative symptoms is necessary to differentiate between various syndromes (e.g. bipolar disorder, brief psychotic disorder, schizophreniform disorder, and psychotic depression; van Os, 2009). During the period in which the studies that contributed to this thesis were performed, several manuscripts in the UHR field have been published, stipulating the importance of social anhedonia in particular, which has led to a growing awareness that this symptom may be considered an important core feature of the phase preceding a first psychosis (Riecher-Rossler et al., 2009; van Tricht et al., 2010; Ruhrmann, IEPA, 2010). Altogether, these findings have strengthened us in our view that social impairments and negative symptoms might indeed play a more important role than suspected initially. With the ongoing debate as to whether the UHR criteria should be incorporated as a ‘Psychosis Risk Syndrome’ in the DSM-V, research on the UHR criteria is as pivotal as ever. Pro’s and con’s of the Psychosis Risk Syndrome, emphasizing the attenuated positive criteria, have been de-bated in recent conferences and various manuscripts. On the one hand, amongst others, proponents have argued that a formal diagnosis would help to promote the needed treatment for -and research to- this currently ill patient group generally with a low level of functioning (GAF-score <50 mostly) (Ruhrmann et al., 2010; Woods et al., 2010; Woods et al., 2009). They stated that without a formal diagnosis, most people will probably not receive help during this time when help could prevent fur-ther deterioration (Carpenter & van Os, 2011). In this line of thought, McGorry proposed a clinical staging model in which a Psychosis Risk Syndrome would form the second stage following an initial ‘entry’ stage characterized by general distress and risk for various mental disorders (McGorry, 2010). On the other hand, opponents argue that there is still no proven treatment for the Psychosis Risk Syndrome (Carpenter & van Os, 2011). Also, they find it unwise to diagnose a potentially large number of ‘false positives’ and fear that the base rate of transition to psychosis will even be lower in populations outside tertiary research settings (Carpenter & van Os, 2011; Yung et al., 2011; Yung et al., 2010). This apprehension has led the DSM-V collaborators to include the help-seeking criterion (Carpenter & van Os, 2011). However, also within help-seeking samples in research groups, the transition rates have been declining and a proportion of young people presents with transitory complaints (Velthorst et al., this thesis; Simon & Velthorst et al., this thesis).

Chapter 10

173

For this reason, recent studies have tried to follow a ‘close-in’ strategy, where prediction models are being developed in which various factors are combined in order to improve the prediction of a the development of a first psychosis (Cannon et al., 2008; Ruhrmann et al., 2010). Also, efforts have been made to examine endophenotypical (Wood et al., 2010; Fusar-Poli et al., 2011; van Tricht et al., 2010), and sociodemographical factors that could enhance psychosis prediction (Dragt et al., 2011; Fusar-Poli et al., 2010). Combining UHR criteria with additional clinical and biological variables may offer a promising perspective to improve prediction of psychosis and reduce non-transition rates. At the moment, the complex DSM-V debate is still indecisive. As was stated during an international conference in Florence (2010), it seems that “all have won and all deserve prices”. In fact, in a recent commentary Prof. Carpenter and Prof. van Os (two of the boardmembers of the DSM-V Psychosis work group) mentioned that it may be that the diagnosis will only be added in the appendix for research purposes until required field studies have been carried out (Carpenter & van Os). Irrespective of the final outcome of the current debate, it is indeed safe to say that the use of the current UHR criteria for scientific purposes is of vital importance in enhancing our understanding of schizophrenia and its development. We propose that future research on the role of social anhedo-nia and withdrawal as well as on the associated social disabilities, could significantly contribute to the enhancement of psychosis prediction.

Implications and future research

The assumption that social withdrawal may play a significant role in the pathway to a first psychosis raises the question how to investigate the validity of this assumption. One suggestion would be to refine the current UHR criteria by adding the criterion “ a moderate score (>3) on the SIPS item ‘social anhedonia and withdrawal’ ”. If social withdrawal indeed would contribute to the prediction of a first psychosis, this would result in a substantially higher transition rate at follow-up in comparison to the current UHR-studies. Also, to gain insight in the underpinnings of this association, a more elaborate assessment of the self-reported reasons of social withdrawal should be conducted in a prospective follow-up study. Retrospective assessment of the self-reported causes of withdrawal may not be sufficient to fully disentangle the mechanism underlying the relationship between social withdrawal and psychosis. Although the findings of this thesis still need replication (for example with a prospective study as suggested above), we have taken the liberty to hypothesize about this subject with a neurodevelop-mental perspective. To this end, based on the findings of this thesis and on findings from studies concerning the development of schizophrenia, our hypothetical model may represent a possible pathway to a first psychosis.

Chapter 10

174

Figure 1: Social impairment and the prediction of psychosis

Poor premorbid

adjustment

Social withdrawal

Social disability

Biological Vulnerability

e.g.

(social) cognitive deficits

- affect recognition

- working memory

Social demanding tasks

1st Psychotic Episode

Attenuated

Psychotic Symptoms

Transient complaints

AdulthoodAdolescenceChildhood

Social

stress

As can be seen in the model above, and in congruence with the neurodevelopmental model of Cornblatt et al., (2002), we propose that in most of the people

developing a psychosis biological vulnerabilities are already present. Amongst others, deficits in (social) cognition (e.g. affect recognition, van Rijn et al., 2011;

Kim et al., 2010 and working memory, Becker et al., 2010; Wood et al., 2003; Keshavan et al., 2011) have been reported years before disease onset. These

already present social deficits could lead to a poor premorbid adjustment, and early withdrawal (deficits before the age of 11 were reported by Shapiro et al.,

2009). The limited amount of social contact in childhood could in turn reinforce the cognitive impairments earlier described. Early impairments and poor

social adjustment might enhance the social stress often experienced in adolescence when social tasks become more demanding (Myin-Germeys et al., 2002;

Lataster et al., 2010) and could lead to a further withdrawal and social deficits (this thesis). Social stress has found to be associated with Attenuated

Psychotic Symptoms (i.e. sub-clinical psychotic symptoms) (Collip et al., 2009), experiences that are quite common in the general population (Dominguez et

al., 2011) and could be of transitory character. However, we propose that social withdrawal may increase the chance that these Attenuated Psychotic

Symptoms transit into a full-blown first psychotic episode.

Chapter 10

175

One should bear in mind that the proposed model only displays one possible pathway to psychosis. The pathway to psychosis most likely constitutes of various roads, holding that whereas social withdrawal might imply a worsening of symptoms in one subgroup, it may be that for others social withdrawal is an adequate coping strategy that may prevent overstimulation from the environment and may even diminish the chance of subsequent psychosis development. In addition, it could be that social withdrawal could cause one person to go over the psychosis-threshold, while someone else may withdraw from contact as a consequence of e.g. paranoia. To conclude, the exact relationship between social withdrawal and psychosis may vary across different subgroups and warrants further investigation in future samples.

Strengths

To our knowledge, manuscripts in this thesis were the first to report on social disabilities as assessed with the WHO-DAS, on ethnic differences in a UHR sample and on the long-term course of UHR symptoms in young people not transiting to a first psychotic episode. The manuscripts in this thesis yielded converging results leading to our hypothetical model about the role social withdrawal in the pathway to a first psychosis. Furthermore, our results add to the body of knowledge about the impor-tance of the inclusion of negative symptoms in UHR criteria which is hopefully a first step in the reduction of false positives in UHR research.

Chapter 10

176

Limitations

Our results should be considered in the context of some limitations. Most of them have been discussed in the relevant chapters, but a few that may have general implications are mentioned here. Firstly, one of the current UHR inclusion criteria is “having a first-degree relative with a psychotic disorder or a DSM–IV schizotypal personality disorder of the index person, combined with a significant decrease in functioning during the past year”. The 30% reduction in GAF score may partly overlap with the social impairments found in our studies. However, besides the fact that very few young people only met this UHR-criterion, with the GAF it is hard to disentangle to which extent the lower level of functioning could be attributed to social impairments as opposed to UHR symptomatology. To this end, the WHODAS and SCPS may be more specific in determining the exact role of social deterioration in the months prior to a first psychosis. Secondly, although in some studies findings were presented on large-scale collaborations in multicenter studies (EPOS, EDIE), in other studies the sample sizes were relatively small and these results therefore warrant replication in future studies. In addition, our longest follow-up period was 3 years. Recently, it has been found that al-though most transitions take place in the first year of follow-up, transitions to psychosis are reported even up to 10 years (Yung et al., 2011). Therefore it is possible that some of our subjects made a tran-sition after our follow-up period. Finally, in most studies, co morbid disorders were not examined at the end of the follow-up period. It would have been of interest to investigate whether subjects developed other psychiatric disorders than schizophrenia.

The findings described in this thesis may have extended our knowledge about the UHR phase for developing a first psychosis. These results could not only contribute to a better psychosis prediction, they may be helpful in choosing adequate treatment objectives as well. The ongoing Dutch Early Detection and Intervention (EDIE-NL) trial already incorporates social impairments as one of their UHR treatment targets. In order to improve the quality of life in this disabled patient group, they try to stimulate social contact as well as the maintenance of work and study. Future research will reveal whether the focus on these impairments will indeed contribute to reducing the likelihood of making the transition to a first psychotic episode.

Chapter 10

177

Chapter 10

References

Becker, H. E., Nieman, D. H., Dingemans, P. M., van de Fliert, J. R., De Haan, L., Linszen, D. H. (2010).

Verbal fluency as a possible predictor for psychosis. European Psychiatry; 25(2):105-10.

Cannon, T. D., Cadenhead, K., Cornblatt, B., Woods, S. W., Addington, J., Walker, E., Seidman, L. J., Perkins, D.,

Tsuang, M., McGlashan, T. & Heinssen, R. (2008). Prediction of Psychosis in Youth at High Clinical Risk:

A Multisite Longitudinal Study in North America. Archives of General Psychiatry; 65: 28–37.

Collip, D., Myin-Germeys, I., Van Winkel, R., van Os, J. (2009).

Stress and psychosis: is sensitisation the underlying mechanism?. Tijdschrift voor de Psychiatrie; 51 (8): 559-567.

Cornblatt, B. A., Lencz, T., Smith, C. W., Correll, C. U., Auther, A. M., Nakayama, E (2003).

The schizophrenia prodrome revisited: a neurodevelopmental perspective. Schizophrenia Bulletin; 29: 633–651.

Dominguez, M., D., Saka, M., C., Lieb, R., Wittchen, H., U., van Os, J. (2010). Early expression of negative/disorganized

symptoms predicting psychotic experiences and subsequent clinical psychosis: a 10-year study. American Journal of

Psychiatry; 167(9):1075-1082.

Dragt, S., Nieman, D. H., Veltman, D., Becker, H. E., van de Fliert, R., de Haan, L., Linszen, D. H. (2011).

Environmental factors and social adjustment as predictors of a first psychosis in subjects at ultra high risk.

Schizophrenia Research;125 (1): 69-76.

Fusar-Poli, P., Byrne, M., Valmaggia, L., Day, F., Tabraham, P., Johns, L., McGuire, P.; OASIS Team (2010).

Social dysfunction predicts two years clinical outcome in people at ultra high risk for psychosis. Journal of Psychiatry

Research; 44 (5):294-301.

Fusar-Poli, P., Crossley, N., Woolley, J., Carletti, F., Perez-Iglesias, R., Broome, M., Johns, L., Tabraham, P., Bramon, E.,

McGuire, P. (2011). Gray matter alterations related to P300 abnormalities in subjects at high risk for psychosis:

longitudinal MRI-EEG study. Neuroimage; 55(1):320-8.

Keshavan, M., S., Kulkarni, S., Bhojraj, T., Francis, A., Diwadkar, V., Montrose, D. M., Seidman, L. J., Sweeney, J (2010).

Premorbid cognitive deficits in young relatives of schizophrenia patients. Frontiers in Human Neuroscience; 9 (3):62.

Kim, H. S., Shin, N. Y., Choi, J., S., Jung, M. H., Jang, J. H., Kang, D. H., Kwon, J. S. (2010).

Processing of facial configuration in individuals at ultra-high risk for schizophrenia. Schizophrenia Research;

118(1-3):81-7.

Lataster, T., Collip, D., Lardinois, M., van Os, J., Myin-Germeys, I. (2010). Evidence for a familial correlation between

increased reactivity to stress and positive psychotic symptoms. Acta Psychiatrica Scandinavica; 122(5):395-404.

McGorry, P. D (2010). Risk syndromes, clinical staging and DSM V: new diagnostic infrastructure for early intervention

in psychiatry. Schizophrenia Research; 120 (1-3): 49-53.

178

Myin-Germeys, I., Krabbendam, L., Delespaul, P., van Os, J. (2003). Can cognitive deficits explain differential sensitivity

to life events in psychosis? Social Psychiatry and Psychiatric Epidemioligy; 38(5): 262-268.

Os van, J. (2009). ‘Salience syndrome’ replaces ‘schizophrenia’ in DSM-V and ICD-11: psychiatry’s evidence-based

entry into the 21st century? (2009). Acta Psychiatrica Scandinavica; 120: 363-372.

Riecher-Rössler, A., Pflueger M. O., Aston J., Borgwardt S. J., Brewer W.J., Gschwandtner U., Stieglitz R. D. (2009).

Efficacy of using cognitive status in predicting psychosis: a 7-year follow-up. Biological Psychiatry; 66 (11):1023-30.

Rijn van, S., Aleman, A., de Sonneville, L., Sprong, M., Ziermans, T., Schothorst, P., van Engeland, H., Swaab,

H. (2011). Misattribution of facial expressions of emotion in adolescents at increased risk of psychosis:

the role of inhibitory control. Psychological Medicine; 41(3):499-508.

Ruhrmann, S., Bodatsch, M., Schultze-Lutter, F., Salokangas, R. K. R., Linszen, D., Birchwood, M., Juckel, G., Lewis,

S., Klosterkötter, J. (2010). Persistent negative symptoms and the risk for psychosis. Abstract ( Presented at the 7th

international conference on early psychosis). Early Intervention in Psychiatry; 4 (suppl. 1): 95.

Ruhrmann, S., Schultze-Lutter, F., Klosterkötter, J. (2010). Probably at-risk, but certainly ill--advocating the introduction

of a psychosis spectrum disorder in DSM-V. Schizophrenia Research; 120(1-3): 23-37.

Shapiro, D. I., Marenco, S., Spoor, E. H., Egan, M., F., Weinberger, D. R., Gold, J. M. (2009).

The Premorbid Adjustment Scale as a measure of developmental compromise in patients with schizophrenia

and their healthy siblings. Schizophrenia Research; 112 (1-3):136-42.

Simon, A. E. & Velthorst, E., Nieman, D. H., Linszen D., Umbricht, D., de Haan, L. Ultra high-risk state for psychosis

and non-transition: a systematic review, submitted

Tricht van, M. J., Nieman D. H., Koelman J. H., van der Meer J. N., Bour L. J., de Haan L., Linszen D., H. (2010).

Reduced parietal P300 amplitude is associated with an increased risk for a first psychotic episode.

Biological Psychiatry; 68 (7):642-648.

Velthorst, E., Nieman, D. H., Klaassen, R. M. C., Becker, H. E., Dingemans, P. M., Linszen, D. H., de Haan,

L. (2011). Three-year course of clinical symptomatology in young people at ultra high risk for transition to psychosis.

Acta Psychiatrica Scandinavica; 123 (1): 36-42.

Wood, S. J., Pantelis, C., Proffitt, T., Phillips, L. J., Stuart, G. W., Buchanan, J. A., Mahony, K., Brewer, W., Smith,

D. J., McGorry, P. D. (2003). Spatial working memory ability is a marker of risk-for-psychosis. Psychological Medicine;

33 (7):1239-1247.

Woods, S. W., Walsh, B.C., Saksa, J. R., McGlashan, T. H. (2010). The case for including Attenuated Psychotic

Symptoms Syndrome in DSM-5 as a psychosis risk syndrome. Schizophrenia Research; 123 (2-3):199-207.

Chapter 10

179

Woods, S. W., Addington, J., Cadenhead, K. S., Cannon, T. D., Cornblatt, B. A., Heinssen, R., Perkins, D. O., Seidman,

L. J., Tsuang, M. T., Walker, E. F., McGlashan, T. H. (2009). Validity of the prodromal risk syndrome for first psychosis:

findings from the North American Prodrome Longitudinal Study. Schizophrenia Bulletin; 35(5): 894-908.

Yung, A. R., Nelson, B., Thompson, A. D. and Wood, S. J. (2010). Should a “Risk Syndrome for Psychosis” be included

in the DSMV? Schizophrenia Research; 120: 7-15.

Yung, A., Nelson. B., Yuen, H. P., Spiliotacopoulos, D., Lin, A., Simmons, M., Bruxner, A., Broussard, C., Thompson,

A., McGorry, P. (2011). Long term outcome in an ultra high risk (“prodromal”) group. Abstract (Presented at the

International Congress on Schizophrenia Research (ICOSR)). Schizophrenia Bulletin; 37 (supll. 1): 6.

Chapter 10

180

181

NEDERLANDSE SAMENVATTING EN CONCLUSIE

182

Samenvatting

Het doel van dit proefschrift was het vergroten van de kennis over de Ultra Hoog Risico (UHR) fase voor het ontwikkelen van een eerste psychose. Binnen dit kader kwamen wij tot de volgende subdoelen: (1) het verkrijgen van meer kennis over de demografische kenmerken en baseline symptomen van deze hulpzoekende groep en (2) het identificeren van factoren die een psychose voorspellen, waardoor uiteindelijk het aantal fout-positieven (personen zonder transitie naar een psychose) in UHR onderzoek zou kunnen verminderen. Hierbij werd in het bijzonder de invloed van negatieve symptomen en van problemen met sociale interactie onderzocht. Tenslotte (3) onderzochten wij het beloop van de UHR symptomen bij jonge mensen die uiteindelijk géén transitie naar een psychose doormaken.

1. Demografische kenmerken en baseline symptomatologie van jongeren met een UHR voor het ontwikkelen van een eerste psychose (hoofdstukken 2 en 3).

Na de introductie van het onderwerp van dit proefschrift in hoofdstuk 1, presenteren wij in hoofdstuk 2 onze bevindingen over etnische verschillen in UHR baseline symptomatologie bij 201 hulpzoekende jongeren die deelnamen aan de Dutch Early Detection and Intervention Trial (EDIE). De resultaten tonen aan dat jongeren van Marokkaanse en Turkse afkomst, meer negatieve symptomen - anhedonie in het bijzonder- en depressie rapporteerden. Tevens onderzochten wij factoren die mogelijk zouden kunnen bijdragen aan deze etnische verschillen, en vonden dat een lage etnische identiteit gerelateerd is aan een groter aantal symptomen in de Marokkaans-Nederlandse patiëntengroep. Naar aanleiding van de bevindingen van hoofdstuk 2, rees de vraag of het in de UHR-groep mogelijk is om onderscheid te maken tussen negatieve symptomen en depressie. Hoofdstuk 3 richt zich daarom op deze differentiatie door het uitvoeren van een exploratieve factoranalyse van de Scale of Prodromal Symptoms (SOPS) in de Dutch Prediction of Psychosis Study (DUPS). Er werd aanget-oond dat het inderdaad mogelijk is om met sommige SOPS-items tussen de twee symptoomclusters te differentiëren. In het bijzonder kwamen de items ‘sociale anhedonie en terugtrekgedrag’, ‘verminderde expressie van emotie’ en ‘verminderde ervaring van emoties en zelf’ naar voren als specifiek negatieve symptomen. Anderzijds kwamen binnen onze groep de items ‘dysfore stemming’, ‘ver-minderde tolerantie voor normale stress’ en ‘verminderde persoonlijke hygiëne / sociale aandacht’ juist naar voren als symptomen van depressie. ‘Avolitie’ en ‘vermindering van het rolfunctioneren’ laadden slechts in lichte mate op de depressie- en negatieve symptoom clusters.

Nederlandse samenvatting en conclusie

183

Nederlandse samenvatting en conclusie

2. Het identificeren van variabelen die zouden kunnen bijdragen aan een betere voorspelling van een eerste psychose (hoofdstukken 4, 5, 6 en 7).

In hoofdstuk 4 worden de baseline verschillen in klinische symptomen en het algemeen niveau van functioneren (GAF-score) onderzocht tussen UHR patiënten die wél (UHR + T) of géén (UHR + NT) transitie naar een psychose doormaken. De 72 geïncludeerde patiënten werden onderzocht binnen het Academisch Medisch Centrum van de Universiteit van Amsterdam, Nederland (DUPS). De studie toonde aan dat de ernst van specifieke symptomen at baseline gerelateerd is aan een transitie naar psychose in UHR patiënten. De UHR + T groep rapporteerde voornamelijk meer sociale anhedonie en terugtrekgedrag, meer bizarre gedachten en een lagere GAF score at baseline in vergelijking met de UHR+NT groep. Het zeer significante verschil in het negatieve symptoom ‘sociale anhedonie’ tussen degenen die uiteindelijk wél en zij die géén transitie doormaken, wees ons op de mogelijk voorspellende waarde van verminderd sociaal functioneren in de prodromale fase. Daarom is, naast een verscheidenheid aan overige tekorten in verschillende levensdomeinen, de rol van een verminderd sociaal functioneren in de voorspelling van een eerste psychose nader onderzocht in hoofdstuk 5. In de European Prediction of Psychosis Study (EPOS), een prospectieve multi-center, naturalistische veldstudie met een 18-maanden follow-up periode bij 245-hulp zoekende personen met een ultra hoog risico op een eerste psychose, werden tekorten beoordeeld met de ‘Disability Schedule’ van de World Health Organization (WHODAS-II). In lijn met onze verwachtingen, bleek dat binnen onze steekproef baseline tekorten in het sociaal functioneren – in het bijzonder op het gebied van het maken en onderhouden van vriendschappen - in belangrijke mate bijdroegen aan de voorspelling van de overgang naar een eerste psychose. In hoofdstuk 6 wordt de voorspellende waarde van de Strauss and Carpenter Prognostic Scale (SCPS), een instrument dat vaak wordt gebruikt in schizofrenie onderzoek, onderzocht in een UHR groep (EPOS). De studie toonde aan dat de SCPS voorspellende waarde heeft voor transitie naar psychose tijdens de UHR periode, hetgeen duidt op temporele continuïteit van deze uitkomstvoorspellers. Ernstigere positieve symptomen, een verminderde mogelijkheid om te kunnen werken of studeren, een lage kwaliteit van sociale relaties en een hoge mate van subjectieve lijdensdruk bleken onafhankelijke voorspellers voor een psychose in UHR patienten. In combinatie met andere factoren, zou de SCPS een waardevolle bijdrage kunnen leveren aan een betere voorspelling van een eerste psychose.

De belangrijke rol van het sociaal disfunctioneren in de voorspelling van psychose wierp de vraag op of sociale tekorten tijdens opname voor een eerste psychose ook zouden kunnen bijdragen aan de voorspelling van het verdere beloop van de stoornis.

184

Nederlandse samenvatting en conclusie

In hoofdstuk 7 hebben wij daarom in een Nederlands cohort van 82 patiënten met schizofrenie

onderzocht of een gedetailleerde beoordeling van de sociale tekorten (gemeten met de Groninger Social

Disability Schedule-II) in de maand voorafgaand aan de opname voor een eerste psychose, zou kunnen helpen de voorspelling van het verdere beloop van de stoornis (met betrekking tot de psychopatholo-gie) te voorspellen. Na correctie voor PANSS-somscore bij aanvang, bleek geen van de sociale tekorten- domeinen significant voorspellend voor het aantal recidieven of de ernst van de klinische symptomatologie na vijf jaar follow-up. Dit resultaat zou kunnen betekenen dat een slecht sociaal functioneren bij opname voor een eerste psychose niet noodzakelijkerwijs een slecht beloop van de stoornis hoeft te betekenen. Mogel-ijk heeft het sociaal functioneren tijdens een klinisch stabielere fase een grotere voorspellende waarde voor het beloop van schizofrenie.

3. Onderzoeken van het beloop van de UHR symptomen bij jonge mensen die uiteindelijk géén transitie naar een psychose doormaken (hoofdstukken 8 en 9).

In hoofdstuk 8 wordt het beloop van de UHR symptomen, de GAF score en werkstatus onderzocht bij 57 jongeren die uiteindelijk geen transitie naar een psychose doormaakten en die werden onderzocht binnen DUPS in Amsterdam. Na 3 jaar follow-up, was 75% van de patiënten die geen transitie naar een psychose doormaakten hersteld van hun UHR status. Een aanzienlijk percentage van de patiënt-en die aanvankelijk in de UHR groep vielen, rapporteerden al binnen 1 jaar significant minder UHR symptomen. Herstel was niet beperkt tot een vermindering van symptomen, maar betrof ook een verbetering van het algemene functioneren en de werkstatus. UHR symptomen deden zich na herstel niet opnieuw voor binnen 3 jaar, hetgeen suggereert dat met de huidige UHR criteria, een aanzienlijk percentage van de geïncludeerde jongeren klachten rapporteert van voorbijgaande aard en slechts tijdelijk disfunctioneert. Het combineren van UHR criteria met klinische en biologische factoren biedt mogelijk een veelbelovend perspectief om de voorspelling van een eerste psychose te verbeteren en zodoende het aantal ‘fout-positieven’ te verminderen. In hoofdstuk 9 bespreken we alle klinische UHR studies tot dusver systematisch en vatten alle beschikbare informatie samen over de kenmerken en het beloop van UHR patiënten die géén transitie doormaken naar een psychose. Wij vonden dat bijna de helft van de 31 relevante studies geen enkele informatie bood over jongeren die uiteindelijk geen psychose ontwikkelden. Vijf studies rapporteerden percentages over de remissie van UHR status (range 15.4% tot 54.3%). Lineaire regressie toonde aan dat de meer recente studies significant lagere transitiepercentages rapporteren in ver-gelijking met eerdere studies. Daarnaast was een oudere leeftijd at baseline geassocieerd met significant lagere transitiepercent-ages in publicaties met een follow-up duur langer dan een jaar. De klinische fenotypen van proef-personen die werden gezien voor een mogelijke at-risk status liepen sterk uiteen, en varieerden van voornamelijk van voorbijgaande aard tot ernstig invaliderende verschijnselen.

185

Nederlandse samenvatting en conclusie

Conclusies

De algemene doelstelling van dit proefschrift was onze kennis over de UHR fase voor het ontwikkelen van een eerste psychotische episode te vergroten. We zijn van mening dat de resultaten van de verschillende deelonderzoeken hebben bijgedragen aan nieuwe inzichten over de baseline kenmerken van deze groep en over criteria die een rol zouden kunnen spelen in een betere voorspelling van een psychose. Een betere voorspelling van transitie naar psychose is een eerste stap in de ontwikkeling van secundaire preventie van schizofrenie. Hierna een puntsgewijze opsomming van onze belangrijkste resultaten, gevolgd door een korte conclusie. 1. Hulpzoekende UHR patiënten van Marokkaans-Nederlandse en Turks-Nederlandse afkomst rapporteren meer negatieve symptomen(anhedonie in het bijzonder) en meer depressie in vergelijking met de autochtone Nederlandse groep; in de Marokkaans-Nederlandse groep is dit mogelijk deels gerelateerd aan een verminderde identificatie met de etnische groep.2. Het is wellicht mogelijk om met de Scale Of Prodromal Symptoms (SOPS) te differentiëren tussen negatieve symptomen en depressie in de UHR fase. 3. De ernst van de symptomen ‘sociale anhedonie en terugtrekgedrag’, ‘bizar denken’ en een lagere GAF score hangen mogelijk samen met een grotere kans de transitie naar eerste psychose door te maken in UHR patiënten. 4. UHR patiënten die een transitie doormaken naar een psychose rapporteren significant meer problemen met sociale interactie dan degenendie geen transitie doormaken.5. De klachten van een groot percentage van geïncludeerde UHR patiënten die worden gerapporteerd op basis van de huidige UHR criteria,hoofdzakelijk gebaseerd op positieve symptomen, blijken van voorbijgaande aard evenals het sociaal disfunctioneren. 6. De Strauss and Carpenter Prognostic Scale heeft voorspellende waarde voor transitie naar psychose in de UHR fase, hetgeen duidt optemporele continuïteit van de uitkomstvoorspellers vaak gebruikt in schizofrenie onderzoek.7. Sociaal disfunctioneren in de maand voor opname voor een eerste psychose voorspelt niet noodzakelijkerwijs een slecht ziektebeloop. Dit resultaat kan erop duiden dat, om het beloop van schizofrenie op een betrouwbare manier te voorspellen, het nodig is om sociaal functioneren te beoordelen tijdens een klinisch stabiele situatie.Uit onze resultaten blijkt dat bij jongeren die aan de UHR criteria voldoen, voornamelijk ernstige negatieve symptomen en sociale beperkingen at baseline bepalen of de jongeren wel of geen transitie doormaken naar een eerste psychose. Deze bevindingen geven een indicatie van de complexe mechanismen die mogelijk ten grondslag liggen aan de ontwikkeling van een eerste psychose. Hoewel het waarschijnlijk is dat terugtrekken uit sociaal contact niet alleen een causale factor is in de ontwikkeling van psychotische symptomen, maar tevens het gevolg is van bijvoorbeeld paranoïde gedachten, denken wij dat wanneer gedachten niet getest worden in sociale situaties, sociaal terugtrekgedrag (hetzij door anhedonie hetzij als gevolg van sociaal-cognitieve tekorten) wellicht ook de verdere ontwikkeling en de instandhouding van waanideeën en achterdocht kan versterken. The lonely soul wanders…

186

187

Dankwoord

Het dankwoord! Een onmisbare en allerlaatste inspanning van een fijn onderzoekshoofdstuk. Allereerst wil ik uiteraard graag alle proefpersonen bedanken, zonder wier veelvuldige bezoek aan het AMC dit proefschrift nooit tot stand was gekomen. Maar natuurlijk ook: Mijn promotoren, prof. dr. Don Linszen en prof. dr. Lieuwe de Haan, bedankt voor jullie kennis en tomeloze enthousiasme in het doen van onderzoek. Beste Don, jouw inzet en enthousiasme voor de UHR-populatie heeft uiteindelijk de basis gevormd voor mijn promotie. Ik heb veel van je mogen leren tijdens de mini-colleges, wanneer ik weer eens plotseling binnen kwam zetten in je kamer op de eerste. Beste Lieuwe, het is fantastisch om samen te mogen werken met iemand die zo duidelijk, snel en gemakkelijk benaderbaar is. Hartelijk dank ook voor je vertrouwen, en de kans die je me hebt gegeven om met mijn nieuwe aanstelling een weg te kunnen vinden in de onderzoekswereld. Dr. Dorien Nieman, mijn co-promotor. Dorien, dank (!) dat je er al die jaren voor me bent geweest en dat je me wegwijs hebt gemaakt in de wereld van het onderzoek. Ik kon altijd met alles bij je binnenvallen (en hoop dat te mogen blijven doen). Jij hebt me uiteindelijk eigenlijk alles geleerd wat nodig is om een proefschrift tot stand te laten komen. De andere leden van mijn promotiecommissie prof. dr. J. van Os, prof. dr. M. van der Gaag, prof. dr. A.H. Schene, prof. dr. G.M. Schippers, dr. C. Morgan en dr. M.W. Koeter wil ik bedanken voor het beoordelen van dit proefschrift, en alle co-auteurs voor de prettige samenwerking. Mijn lieve AMC-collega’s. Bedankt voor jullie gezelligheid op feestjes, maar vooral ook gewoon tijdens de lunch e.d. Ik kan me niet voorstellen dat ik hier jarenlang had gezeten zonder jullie. Ik wil deze gelegenheid tevens graag aangrijpen om me te verontschuldigen voor het feit dat ik jullie met enige regelmaat van het werk heb afgehouden. Marise, gelukkig blijf je nog even voor onze dagelijkse portie chocola (en hopelijk is dit even je laatste verlof, zo ongezellig…); Sara, wat fijn dat we de laatste loodjes samen hebben kunnen doorlopen, veel leuker zo!; Fabiana en Maarten, lieve kamergenoten, bedankt voor alle leuke gesprekjes over de dagelijkse perikelen; Floor en Daniella, wat zijn we een goed team!; Carin, heel erg bedankt dat ik je altijd voor alles mag lastigvallen. Ik zie heel erg uit naar onze samenwerking de komende tijd!; Julia en Nikie, Florence 2012, zelfde kamer?; en last -but certainly not least- gaat mijn dank natuurlijk ook uit naar Rianne, Lindy, Albertine, Jet, Erik, Oswald, Nienke, Renate, Bouke, Mariken, Therese, Laura, Annick en alle collega’s van de (dag)kliniek en het VIP-team.

Dankwoord. List of Publications, Curriculum Vitae

188

Collega’s van buiten het AMC, EDIE-medewerkers, ik heb erg genoten van onze gezellige vergaderingen. Ik hoop velen van jullie te blijven zien op congressen, of gewoon daarbuiten. Dear EU-GEI colleagues, although not directly involved in the development of this thesis, lots of articles that came out of your hands have inspired me in my own writing. Thanks for your warm welcome in the World of Science. Dan natuurlijk nog alle vrienden buiten de onderzoekswereld, met wie ik het maar getroffen heb. Zonder de avondjes en weekenden aan de kade, in het park, in de kroeg, op het terras, aan tafel en op vakantie waren de afgelopen jaren heel wat eenzamer verlopen. Remco, Skybox.org: een speciaal bedankje voor het prachtige kaft & binnenwerk. Geweldig dat je dit allemaal vrijwillig voor mij hebt willen doen! Dank ook aan mijn familie. Paps en Mams, voor al jullie lieve meedenken, steun en trots! Paps, hoewel je Engelse uitspraak niet bepaald best is, heb ik ontzettend veel aan je grammaticale hulp gehad. Oma Ienemienemutte, papa vertelt u altijd veel over mijn werk, en u bent altijd even geïnteresseerd. Wat fijn dat u dit nog allemaal mee kunt maken! Ten slotte, uiteraard, mijn Paranimfen, Niels en Famke. Niels, broer, als strafpleiter altijd op je gemak in de spotlights, en als mijn ‘verdediger’/ paranimf dan ook volledig op je plek. Bedankt voor je hulp en relativeringsvermogen. Famke, wat heerlijk dat wij altijd onder het genot van een wijntje alle onderzoeksperikelen kunnen doornemen. Toevallig dat we na al die jaren hele andere wegen bewandeld te hebben, beiden in het onderzoek zijn beland. Dat hadden we op 12-jarige leeftijd nooit kunnen bedenken. Ik ben ontzettend trots dat je het zo geweldig doet!

Dankwoord. List of Publications, Curriculum Vitae

189

Dankwoord. List of Publications, Curriculum Vitae

List of publicationsJournal Articles

2011 Eva Velthorst, Dorien Nieman, Wim Veling, Rianne Klaassen, Sara Dragt, Judith Rietdijk, Helga Ising, Lex Wunderink, Don Linszen, Lieuwe de Haan and Mark van der Gaag. Ethnicity and baseline symptomatology in patients with an “At Risk Mental State” for psychosis. Psychological Medicine (in press)

2011 Andor Simon* & Eva Velthorst*, Dorien H Nieman, Don H Linszen, Simon Umbricht, Lieuwe de Haan. Remission from an initial ultra high-risk state for psychosis: a systematic review. Schizophrenia Research (in press). * Combined first authorship

2011 Eva Velthorst, Dorien H Nieman, Carin Meijer, Don Linszen and Lieuwe de Haan. Social disability at admission for a first psychosis does not predict clinical outcome at 5-year follow-up, Journal of Nervous and Mental Disease, 199(7): 510-512.

2011 Eva Velthorst, Dorien H. Nieman, Rianne Klaassen, Hiske E. Becker, Peter M. Dingemans, Don H. Linszen and Lieuwe de Haan. Three year course of clinical symptomatology in young people at high risk for transition to psychosis. Acta Scandinavica Psychiatrica 123 (1): 36-42.

2010 Eva Velthorst, Dorien H. Nieman, Don H. Linszen, Hiske E. Becker, Lieuwe de Haan, Peter M. Dingemans et al. Disability in patients clinically at high risk for a psychosis. British Journal of Psychiatry; 197 (4): 278-84. 2010 Hiske E. Becker, Dorien H. Nieman, Suzanne Wiltink, Peter M. Dingemans, Reinaud van de Fliert, Eva Velthorst, et al. Neurocognitive functioning before and after the first psychotic episode: does psychosis result in cognitive deterioration? Psychological medicine; 40 (10); 1599-606.

2010 Rianne Klaassen, Eva Velthorst, Dorien H. Nieman, Lieuwe de Haan, Hiske E. Becker, Peter M. Dingemans, J. Reinaud van de Fliert, Mark van der Gaag, Don H. Linszen. Negative symptoms and depression can be validly differentiated in a young help-seeking population with an ultra-high risk for developing a psychosis. Psychopathology (in press).

2009 Eva Velthorst, Dorien H. Nieman, Hiske E. Becker, Reinaud van de Fliert, Peter M. Dingemans, Rianne Klaassen, et al. Baseline differences in clinical symptomatology between ultra high risk subjects with and without a transition to psychosis. Schizophrenia Research; 109 (1-3): 60-5.

190

Book Chapters

2011 Don Linzen, Hiske Becker, Sara Dragt, Rianne Klaassen, Mariken de Koning, Eva Velthorst. Diagnostiek bij hoogrisicogroepen. In, Handboek schizofrenie, Red. W. Cahn, L. Krabbendam, I. Myin-Germeys, R. Bruggeman, L. de Haan. De Tijdstroom (Utrecht), 73-99.

Submitted

2011 Dorien H. Nieman, Eva Velthorst, Hiske Becker, Lieuwe de Haan, Peter Dingemans, Björn Krefft et al. The Strauss and Carpenter Prognostic Scale predicts transition to psychosis in subjects clinically at high risk for psychosis.

2011 Dorien H. Nieman, Sara Dragt, Francesca Soen, Mirjam J. van Tricht, Eva Velthorst, Hiske E. Becker, Don H. Linszen, Lieuwe de Haan. Prediction of transition to psychosis: modelling pre-psychotic risk factors in Ultra High Risk subjects.

Refereed Conference Proceedings

2011 Eva Velthorst, Dorien H. Nieman, Don H. Linszen, Hiske E. Becker, Lieuwe de Haan, Peter M. Dingemans et al. Disability in patients clinically at high risk for a psychosis. Presented at the International Congress of Schizophrenia Research (ICOSR), April.

2010 Eva Velthorst, Dorien H. Nieman, Rianne Klaassen, Hiske E. Becker, Peter M. Dingemans, Don H. Linszen and Lieuwe de Haan. Three year course of clinical symptomatology in young people at high risk for transition to psychosis. Presented at the International Early Psychosis Association (IEPA) congress, December.

Dankwoord. List of Publications, Curriculum Vitae

191

Curriculum Vitae

The author of this thesis was born on February 4th 1983 in Hoorn, the Netherlands. After graduating high school in 2001 (VWO) she allowed herself a study-sabbatical in which she worked as logistic employee in the clothing industry. In 2002 she moved from Hauwert to Amsterdam and started her study Clinical Psychology at the University of Amsterdam. She obtained her Masters degree in 2007 with thesis subject: “Baseline differences in clinical symptomatology between ultra high risk subjects with and without a transition to psychosis”. This study, conducted at the Academic Medical Center, Amsterdam, was later adapted to one of the current thesis-chapters. She has been employed at the department of Early Psychosis of the Academic Medical Center ever since; first as research assistant, thereupon as psychologist/data manager of the EPOS-project and from October 2008 as PhD-candidate. She obtained her Master degree in Forensic Educational Sciences in January 2009, with a study on the Dutch Forensic reports advising youth custody. From the end of 2009 she is involved in the European Network of national schizophrenia networks studying Gene-Environment interactions (EU-GEI) as training/research coordinator.

De auteur van dit proefschrift is geboren op 4 februari 1983 te Hoorn, Nederland. Na het behalen van haar middelbare school diploma (VWO) gunde zij zichzelf een studieonderbreking, waarin ze werkte als logistiek medewerkster bij een confectiebedrijf. In 2002 verhuisde zij naar Amsterdam en begon haar studie Klinische Psychologie aan de Universiteit van Amsterdam. Ze behaalde haar masterdiploma in 2007 met de master these: “Baseline differences in clinical symptomatology between ultra high risk subjects with and without a transition to psychosis”. Deze studie, uitgevoerd binnen het Academisch Medisch Centrum te Amsterdam, is later omgevormd tot één van de hoofd-stukken in huidig proefschrift. Sindsdien heeft ze onafgebroken op het Academich Medisch Centrum gewerkt; eerst als onderzoeksassistent, later als psychologe/ datamanager van het EPOS project en sinds oktober 2008 als PhD- student. Zij behaalde haar masterdiploma forensische Orthopedagogische Wetenschappen in januari 2009, met een studie naar de Nederlandse Forensische rapporten waarbij een jeugd PIJ- maatregel werd geadviseerd. Vanaf eind 2009 is zij betrokken bij de studie: “European Network of national schizophrenia networks studying Gene-Environment interactions (EU-GEI)” als training/ onderzoekscoördinator.

Dankwoord. List of Publications, Curriculum Vitae

192