lable at ScienceDirect

Journal of Psychiatric Research 45 (2011) 1140e1145

Contents lists avai

Journal of Psychiatric Research

journal homepage: www.elsevier .com/locate/psychires

A longitudinal study of obsessive-compulsive disorder in individuals atultra-high risk for psychosis

Leonardo F. Fontenelle a,b,c,*,1, Ashleigh Lin a,d, Christos Pantelis a, Stephen J. Wood a,e,Barnaby Nelson d, Alison R. Yung d

aMelbourne Neuropsychiatry Centre, University of Melbourne and Melbourne Health, AustraliabAnxiety and Depression Research Program, Institute of Psychiatry, Universidade Federal do Rio de Janeiro, BrazilcDepartment of Psychiatry and Mental Health, Institute of Community Health, Universidade Federal Fluminense, BrazildOrygen Youth Health Research Centre and Centre for Youth Mental Health, University of Melbourne, Australiae School of Psychology, University of Birmingham, UK

a r t i c l e i n f o

Article history:Received 8 November 2010Received in revised form20 February 2011Accepted 3 March 2011

Keywords:Obsessive-compulsive disorderPsychosisComorbidityFollow-up studies

* Corresponding author. Melbourne NeuropsychiPsychiatry, University of Melbourne and Melbourne Hton South, Victoria 3053, Australia.

E-mail address: lfontenelle@gmail.com (L.F. Fonten1 Present address: Rua Visconde de Pirajá, 547, Sala

CEP 22410-003, Brazil. Tel./fax: þ55 21 22394919.

0022-3956/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.jpsychires.2011.03.005

a b s t r a c t

Background: We evaluated whether (1) a diagnosis of obsessive-compulsive disorder (OCD) at baseline, or(2) the persistence, remission or emergence of de novo OCD at follow-up, were associated with the devel-opment of different psychotic disorders in a cohort of individuals at ultra-high risk (UHR) for psychosis.Methods: Patients were assessed for OCD at baseline and after a mean of 7.4 years follow-up and clas-sified into: (i) Non-OCD group - patients without OCD both at baseline and follow-up (n ¼ 269; 86.2%),(ii) Incident OCD group - patients without OCD at baseline but with OCD at follow-up (n ¼ 17; 5.4%), (iii)Remitting OCD group - patients with OCD at baseline but without OCD at follow-up (n ¼ 20; 6.4%), (iv)Persistent OCD group - patients with OCD both at baseline and at follow-up (n ¼ 6; 1.9%). Rates ofdifferent DSM-IV psychotic disorders at follow-up were compared across these groups.Results: Patients who displayed remitting OCD were not related to the development of any DSM-IVpsychotic disorder. A diagnosis of incident OCD was associated with greater rates of psychotic disordersat follow-up, particularly mood disorders with psychotic features and psychotic disorders not otherwisespecified (PDNOS), and greater baseline severity of general psychopathology, alogia, and avolition-apathy. Two of the six patients (40%) with persistent OCD developed schizophrenia, while only 12.5%,5.0%, and 9.7% of incident, remitting, and non-OCD groups, respectively, exhibited the same condition atfollow-up. Rates of antipsychotic use in the previous two years were not significantly different betweenthe groups.Conclusions: Our findings suggest that, in a cohort of individuals at UHR for psychosis, remission of OCDdoes not increase the risk of psychosis, while de novo OCD was associated with development of mooddisorders with psychotic features and PDNOS.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Observations of patients with obsessive-compulsive disorder(OCD) indicate that the incidence of schizophrenia is low (de Haanet al., 2009), even after long-term follow-up (Pollitt, 1957; Kringlen,1965; Lo, 1967). In contrast, obsessive-compulsive symptoms (OCS)are noted more commonly at various stages of psychosis and

atry Centre, Department ofealth, 161 Barry Street, Carl-

elle).719, Ipanema, Rio de Janeiro,

All rights reserved.

schizophrenia (Lysaker et al., 2009), although their impact on theoutcome is still unclear. Early theorists have argued OCD mightprevent the development of “malignant schizophrenia”, retard the“personality disintegration” associated with schizophrenia, or evenherald remission of schizophrenic illness (Bottas et al., 2005).

In fact, while some studies have suggested that schizophreniapatients with OCD (or OCS) may exhibit less severity of formalthought disorder and affective blunting (Poyurovsky et al., 1999),fewer negative symptoms (Tibbo et al., 2000; de Haan et al., 2005),and greater global functioning (Tibbo et al., 2000), a recent meta-analysis of cross-sectional studies suggested that OCS were asso-ciated with greater severity of global, positive and negativepsychotic symptoms (Cunill et al., 2009). Thus, it remains unclearwhether OCS/OCD bears any relationship to the emergence of

L.F. Fontenelle et al. / Journal of Psychiatric Research 45 (2011) 1140e1145 1141

psychosis and schizophrenia, or whether the two disorders areindependent. In order to assess this relationship, longitudinalstudies from early psychosis that examine the relationship to long-term outcome are needed.

To date, there is only one study in a pre-psychotic sample ofyoung people at ultra-high risk (UHR) for developing psychosis.Niendam et al. (2009) reported that, although UHR individuals withOCS exhibited greater levels of depression, suicidality, and positiveand negative symptoms, baseline OCD (14% of their sample) wasassociatedwith a statistical trend towards lower rates of conversionto psychosis over the follow-up period. Further, while self-reportedOCS remained temporally stable among UHR individuals who didnot convert to psychosis, OCS declined among those UHR individ-uals who did develop psychosis. However, this analysis relied onself-reported symptoms and it remains unclear whether persis-tence, remission, or de novo OCD over time are associated withchanges that contribute to conversion to psychosis.

In this long-term longitudinal study of the UHR for psychosiscohort at the Personal Assessment and Crisis Evaluation (PACE)Clinic, we evaluated whether a diagnosis of OCD at baseline and thepersistence, remission or emergence of de novo OCD are associatedwith the development of different psychotic disorders at follow-up.We predicted that increased transition to psychosis would be seenamong UHR individuals who do not show OCD at baseline or at thefollow-up, or who remit from OCD at follow-up.

2. Methods

2.1. Participants and procedure

The sample consisted of 312 individuals (55.1% female) recruitedconsecutively on admission to the PACE Clinic at Orygen YouthHealth in Melbourne, Australia between 1994 and 2005. At base-line, all were assessed with the Comprehensive Assessment ofAt-Risk Mental States (CAARMS) (Yung et al., 2005) andmet criteriafor UHR for psychosis. These criteria are the attenuated positivesymptoms, brief limited intermittent psychotic symptoms, and/orhave vulnerability for psychotic illness, all of which must beaccompanied by a deterioration in functioning. UHR individualswere aged 15e25 years at baseline and had no prior psychoticepisode (treated or untreated). Detailed criteria for the identifica-tion of the UHR group are described elsewhere (Yung et al., 2005)and are summarized in Table 1.

Table 1PACE Ultra-High Risk criteria: (1) must be aged between 15 and 25 years, (2) have been reffollowing three groups.

Group 1: Attenuated positive psychotic symptoms � Presor mspe

� Freq� Rece� Dur

Group 2: Brief limited intermittent psychotic symptoms � Tranrefethin

� Dur� Freq� Sym� Rece

Group 3: Trait and state risk factors � Schiwit

� Signand

� This

Adapted from Yung et al. (2005).

The CAARMS also includes a screening section for OCD symp-toms, comprising questions onwhether patients have distressing orintrusive thoughts, forced repetitive behaviors, rituals/supersti-tions, need to do things a certain way, or checking compulsions.Along with other symptoms dimensions, OCD symptoms are ratedon [1] a 6-point severity (or interference) scale (0¼ never/absent to6 ¼ extreme), [2] a 6-point frequency and duration scale(0 ¼ absent to 6 ¼ continuous), and [3] its relationship withsubstance abuse (0¼ not in relation, 1¼ sometimes in relation, and2 ¼ only in relation to substance abuse).

Participants were followed-up between 2007 and 2009. Atracking system based on a previously documented system used atOrygen (Henry et al., 2007) was implemented to locate andrecontact participants in this cohort. The mean age of the sample atbaseline was 18.8 (SD ¼ 3.2) and at follow-up assessment was 26.1(SD ¼ 5.04) years old. The mean time to follow-up was 7.4 years(SD ¼ 3.2, range ¼ 2.4e14.8). All participants provided writteninformed consent in accordance with guidelines provided by thelocal mental health service research and ethics committee. Thisresearch was conducted in accordance with the Declaration ofHelsinki.

2.2. Measures

Additional measures of psychiatric symptoms at both baselineand follow-up assessments were the Structured Clinical Inter-view for DSM-IV (SCID; First et al., 2002), the Brief PsychiatricRating Scale (BPRS) (McGorry et al., 1988), and the Scale forAssessment of Negative Symptoms (SANS) (Andreasen, 1982). Aproportion of the participants were also assessed on the Ham-ilton Rating Scale for Anxiety (HARS) (Hamilton, 1959) andHamilton Rating Scale for Depression (HRSD) (Hamilton, 1960).Diagnostic decisions followed the guidelines provided by theSCID/DSM-IV-TR, such that whenever another Axis I disorder waspresent (including a psychotic disorder), the patient was diag-nosed with OCD only if the content of the obsessions orcompulsions was not restricted to it. At both assessments, thefunctioning and disability of participants was determined usingthe Quality of Life Scale (QLS) (Heinrichs et al., 1984). TheWechsler Abbreviated Scale of Intelligence (Wechsler, 1999) oran abbreviated form of the Wechsler Adult Intelligence Scale-Revised (WAIS-R) (Weschler, 1981) were used to estimate currentIQ at baseline and follow-up.

erred to a specialized service for help, and (3) meet the criteria for one or more of the

ence of at least one of the following symptoms: ideas of reference, odd beliefsagical thinking, perceptual disturbance, paranoid ideation, odd thinking and

ech, odd behavior and appearanceuency of symptoms: at least several times a weekncy of symptoms: present within the last yearation of symptoms: present for at least 1 week and no longer than 5 yearssient psychotic symptoms. Presence of at least one of the following: ideas ofrence, magical thinking, perceptual disturbance, paranoid ideation, oddking or speechation of episode: less than 1 weekuency of symptoms: at least several times per weekptoms resolve spontaneouslyncy of symptoms: must have occurred within the last yearzotypal personality disorder in the identified individual, or a first-degree relativeh a psychotic disorderificant decline in mental state or functioning, maintained for at least 1 monthnot longer than 5 yearsdecline in functioning must have occurred within the past year

L.F. Fontenelle et al. / Journal of Psychiatric Research 45 (2011) 1140e11451142

2.3. Study groups

1. Baseline OCD: All individuals who had a lifetime history of OCDat the first assessment (baseline).a) Persistent OCD group: Subjects who had both a lifetime

history of OCD at the first assessment (baseline) anddisplayed OCD between the first and the second assessment(follow-up).

b) Remitting OCD group: Subjects who had a lifetime historyof OCD at the first assessment (baseline) but did not displayOCD between the first and the second evaluation (follow-up).

2. Incident OCD (or de novo OCD): Subjects who did not havea lifetime history of OCD at the first assessment (baseline) butdisplayed OCD somewhere between the first and the secondassessment (follow-up).

3. Non-OCD group: Subjects who had no lifetime history of OCD,either at the first assessment (baseline) or between the firstand the second assessment (follow-up).

2.4. Statistical analysis

Continuous variables were described as means (standard devi-ation, SD) and categorical variables as absolute (n) and relative (%)values. Comparisons between two groups were made with Chi-square (c2) tests and independent samples t-tests or Man-neWhitney tests according to the normality of distribution.Comparisons between four groups were made using c2 and Krus-kalleWallis tests. At this point, the standard level of significancewas .05. Post hoc c2 or ManneWhitney tests with Bonferroni’scorrection followed significant differences.

3. Results

We found that 8.1% (32 of 396) and 8.0% (25 of 314) of our UHRsubjects had a diagnosis of OCD at baseline and at follow-up,respectively. Six subjects who had OCD at baseline but were lost tofollow-up, and two individuals who had a diagnosis of OCD atfollow-up andwere not assessed for OCD at baseline were excludedfrom subsequent analysis, leaving 312 to be classified as incidentOCD (n ¼ 17, 5.4%), persistent OCD (n ¼ 6, 1.9%), remitting OCD(n ¼ 20, 6.4%), or non-OCD (n ¼ 269, 86.2%). A total of 57 (18.2% of312) patients fulfilled criteria for at least one psychotic disorder atfollow-up: 31 (9.9%) developed schizophrenia, 1 (.3%) schizo-affective disorder, 1 (.3%) delusional disorder, 5 (1.6%) substanceinduced psychotic disorder, 13 (4.1%) psychotic disorder nototherwise specified (PDNOS), 4 (1.3%) bipolar disorder withpsychotic features, and 6 (1.9%) major depression with psychoticfeatures. The specific prevalence rates of OCD (including incident,persistent, and remitting cases) among UHR individuals whodeveloped psychosis was 24.5% (n ¼ 14) for any psychotic disorder,16.1% (n ¼ 5) in schizophrenia, 0% in schizoaffective disorder, 100%(n ¼ 1) in delusional disorder, 0% in substance induced psychoticdisorder, 46.1% (n ¼ 6) in PDNOS, and 28.5% (n ¼ 4) in mooddisorders with psychotic features, i.e. 50.0% (n ¼ 2) in bipolardisorders with psychotic features, and 33.3% (n ¼ 2) in majordepressive disorder with psychotic features.

At baseline, available CAARMS OCD symptoms subscale scoreslead to predictable results, i.e. remitting OCD (n ¼ 13) had greaterseverity at baseline than incident OCD (n ¼ 8; p ¼ .005) and thenon-OCD group (n¼ 197; p< .001), and greater frequency/durationof OCD symptoms than the non-OCD group (p< .001). Incident OCDand non-OCD groups were not significantly different both inseverity (p ¼ .89) and frequency/duration (p ¼ .75) of baseline OCD

symptoms. Only one patient with persisting OCD had baselineCAARMS OCD values, thus not allowing us to compare groups interms of baseline severity, frequency, and duration of OCD.

A comparison between the four groups according to the CAARMSOCD symptoms subscale scores at follow-up also generated resultsin the expected direction. Persistent (n ¼ 4) and incident OCD(n ¼ 15) had both greater severity (p < .001 each) and frequency/duration (p¼ .005 and p< .001, respectively) of OCD symptoms thannon-OCD group (n ¼ 241). Persistent OCD had greater severity(p¼ .006) of OCD symptoms than remitting OCD group (n¼ 16) andincident OCD had both greater severity (p ¼ .001) and frequency/duration (p ¼ .001) of OCD symptoms than remitting OCD group.Persisting OCD did not differ from incident OCD in terms of severity(p ¼ .02, non-significant after Bonferroni’s correction) andfrequency/duration (p ¼ .83) of OCD symptoms. Remitting OCDgroup did not differ from non-OCD group in terms of severity(p ¼ .04; non-significant after Bonferroni’s correction) andfrequency/duration (p ¼ .13) of OCD symptoms.

There were no significant differences between UHR subjectswith and without OCD at baseline, either in terms of clinicalaspects or transition to different psychotic disorders (Table 2).However, as seen in Table 3, incident OCD was associated withgreater rates of psychotic disorders at follow-up, particularlymood disorders with psychotic features and psychotic disordersNOS, and greater baseline severity of general psychopathology,alogia, and avolition-apathy. Two out of 6 (40%) patients withpersisting OCD received a diagnosis of schizophrenia at follow-up,although this rate was not significantly different from thosereported in other groups.

4. Discussion

In this long-term follow-up study of subjects at ultra-high riskfor the development of psychosis, 5.4% developed incident OCD,1.9% exhibited persistent OCD, and 6.4% displayed remitting OCD.Further, we found that around 8.0% of our UHR individualsexhibited OCD as a comorbidity either at baseline or at follow-up.Although this rate is more than 3 times greater than the preva-lence of OCD described in the general population (Faravelli et al.,2004), it is much lower than our OCD rates among the totalnumber of UHR individuals (22.4%) who developed psychosis. Thissuggests that the transition to psychosis, rather than the high riskstatus, is associated with greater rates of OCD. Lower rates ofOCD among UHR individuals can also be considered consistentwith the findings from a meta-analysis showing a positivecorrelation between the severity of global, positive, and negativesymptoms and severity of OCS in patients with schizophrenia(Cunill et al., 2009).

The fact that the prevalence of OCD at baseline (8.1%) remainedrelatively stable at the follow-up (8.0%) did not reflect a greatnumber of persistent cases. In fact, remitted (n ¼ 20) and incident(n ¼ 17) OCD patients outnumbered persistent cases (n ¼ 6) in oursample. Accordingly, this pattern of OCD symptoms’ fluctuationwas already reported in epidemiological studies (Nelson and Rice,1997) and was critical in allowing us to examine a dynamic rela-tionship between OCD and psychosis and testing our hypothesis ofa relative antagonism between both states.

Consistent with Niendam et al. (2009), we were unable to findsignificant differences between UHR individuals with and withoutbaseline OCD in terms of psychopathology (BPRS, SANS, HDRS, andHARS scores), quality of life, and conversion rates to different DSM-IV psychotic disorders. While some studies of patients withschizophrenia reported similar results (Berman and pappas, 1997;Poyurovsky et al., 2001; Craig et al., 2002; Byerly et al., 2005;Sevincok et al., 2007), the fact that UHR individuals, by definition,

Table 2Comparison between baseline features and follow-up conversion rates between UHR patients with and without OCD.

Presence of OCD at baseline Absence of OCD at baseline Test results

Age 19.3 (3.1) 19.9 (3.3) t ¼ .4; df ¼ 372; p ¼ .64Gender (female) 16 (61.%) 158 (54.9%) c2 ¼ .43; df ¼ 1; p ¼ .51IQ 101.7 (13.3) 98.0 (14.0) t ¼ 1.4; df ¼ 330; p ¼ .15BPRS

Psychotic subscale 9.6 (2.5) 9.5 (2.9) t ¼ .81; df ¼ 392; p ¼ .93Total 45.6 (7.4) 47.3 (9.6) t ¼ �.98; df ¼ 392; p ¼ .32

SANSAffective flattening 5.1 (4.7) 5.5 (5.4) t ¼ 3.6; df ¼ 393; p ¼ .67Alogia 1.4 (1.5) 2.4 (2.5) ManneWhitneys’s p ¼ .86Avolition-apathy 3.5 (2.7) 4.1 (3.0) t ¼ �.95; df ¼ 393; p ¼ .34Anhedonia-Asociality 7.2 (5.2) 6.0 (4.5) t ¼ 1.4; df ¼ 393; p ¼ .14Attention 1.2 (1.5) 1.8 (2.0) ManneWhitneys’s p ¼ .63Total 18.5 (11.0) 19.8 (12.9) t ¼ �.57; df ¼ 393; p ¼ .57

HARS 15.3 (7.0) 15.3 (8.3) t ¼ �.002; df ¼ 175; p ¼ .99HDRS 18.6 (6.7) 18.9 (10.1) t ¼ �.18; df ¼ 26.8; p ¼ .85QLS 75.4 (26.2) 76.0 (20.5) t ¼ �.14; df ¼ 389; p ¼ .88DSM-IV-TR Psychotic disorders at follow-up

Schizophrenia 4 (15.4%) 34 (12.1%) Fisher’s exact p ¼ .54Schizoaffective disorder 0 (.0%) 1 (.4%) Fisher’s exact p ¼ 1.0Delusional disorder 0 (.0%) 1 (.4%) Fisher’s exact p ¼ 1.0Substance induced psychotic disorder 0 (.0%) 6 (2.2%) Fisher’s exact p ¼ 1.0Psychotic disorder NOS 2 (7.7%) 15 (5.4%) Fisher’s exact p ¼ .64Bipolar disorder with psychotic features 0 (.0%) 5 (1.8%) Fisher’s exact p ¼ 1.0Major depression with psychotic features 0 (.0%) 5 (1.8%) Fisher’s exact p ¼ 1.0Mood disorders with psychotic features 0 (.0%) 10 (3.6%) Fisher’s exact p ¼ 1.0Any psychotic disorder 5 (19.2%) 64 (22.2%) c2 ¼ .12; df ¼ 1; p ¼ .72

IQeIntelligence quotient; BPRSeBrief Psychiatric Rating Scale; SANSeScale for the Assessment of Negative Symptoms; HARSeHamilton Rating Scale for Anxiety;HDRSeHamilton Rating Scale for Depression; QLSeQuality of Life Scale; DSM-IV-TReDiagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision.

L.F. Fontenelle et al. / Journal of Psychiatric Research 45 (2011) 1140e1145 1143

exhibit lower levels of psychopathology than patients withschizophreniamay have lessened the potential differences betweenpatients with and without OCD. Alternatively, the lack of signifi-cance between groups could suggest that OCD may be an intrinsic

Table 3Comparison between baseline features and follow-up conversion rates between UHR pa

IncidentOCD (n ¼ 17)

PersistenOCD (n ¼

Age 18.2 (2.8) 18.6 (3.2Gender (female) 10 (62.5%) 3 (50%)Intelligence quotient 91.0 (14.9) 107.0 (17Brief Psychiatric Rating Scale

Psychotic subscale 10.2 (2.8) 7.8 (2.4)Total 51.0 (7.1) * 39.3 (5.6

Scale for the Assessment of Negative SymptomsAffective flattening 5.3 (5.8) 3.6 (4.8)Alogia 3.4 (3.2) * .1 (.40) *

Avolition-apathy 5.6 (3.1) * 1.3 (1.2)Anhedonia-Asociality 6.6 (5.0) 8.1 (5.8)Attention 2.0 (1.9) .6 (1.6)Total 23 (12.8) 14.0 (9.8

Hamilton Rating Scale for Anxiety 21.8 (9.7) 14.2 (4.5Hamilton Rating Scale for Depression 27.1 (11.2) 16 (.70)Quality of Life Scale 68.6 (19.6) 87.6 (13.Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision. P

Schizophrenia 2 (12.5%) 2 (40.0%)Schizoaffective disorder 0 (.0%) 0 (.0%)Delusional disorder 1 (6.7%) 0 (.0%)Substance induced psychotic disorder 0 (.0%) 0 (.0%)Psychotic disorder NOS 5 (31.3%)z 0 (.0%)Bipolar disorder with psychotic features 2 (13.3%) 0 (.0%)Major depression with psychotic features 2 (13.3%) 0 (.0%)Mood disorder w/psychotic features 4 (26.7%)z 0 (.0%)Any Psychotic disorder 10 (62.5%)yz 2 (40.0%)Any psychotropic at the entire period 10 (66.7%) 4 (80.0%)Antipsychotics at the entire period 7 (46.7%) 2 (50%)Any psychotropic two years before follow-up 9 (64.3%) 3 (75%)Antipsychotics two years before follow-up 5 (35.7%) 1 (25%)

Note: 0-OCD vs. OCDeOCD: *p� .008; 0-OCD vs. OCD-0: yp� .008; 0-OCD vs. 0-0: zp� .00Assessment of Negative Symptoms; HARSeHamilton Rating Scale for Anxiety; HDRSeHamand Statistical Manual of Mental Disorders, 4th edition, text revision.

dimension of the at-risk state, free of additional impairments inpsychopathological or quality of life spheres.

The diagnosis of a psychotic disorder at follow-up was greateramong UHR individuals who exhibited incident OCD, including

tients with different OCD outcomes.

t6)

RemittingOCD (n ¼ 20)

Non-OCDgroup (n ¼ 269)

Test results

) 19.5 (3.4) 18.8 (3.2) KruskalleWallis’ p ¼ .718 (40.0%) 123 (45/7%) c2 ¼ .98; df ¼ 3; p ¼ .80

.0) 99.7 (12.1) 98.3 (13.7) KruskalleWallis’ p ¼ .14

10.3 (2.6) 9.3 (3.0) KruskalleWallis’ p ¼ .20) * 47.3 (7.3) 47.2 (9.8) KruskalleWallis’ p ¼ .03

5.1 (4.9) 5.6 (5.5) KruskalleWallis’ p ¼ .781.4 (1.5) 2.3 (2.5) KruskalleWallis’ p ¼ .01

* 4.5 (2.7) 3.9 (3.0) KruskalleWallis’ p ¼ .016.8 (5.1) 5.9 (4.5) KruskalleWallis’ p ¼ .681.3 (1.6) 1.7 (2.0) KruskalleWallis’ p ¼ .23

) 19.3 (11.0) 19.6 (12.9) KruskalleWallis’ p ¼ .51) 17.1 (6.8) 15.2 (8.5) KruskalleWallis’ p ¼ .28

20.3 (7.4) 18.3 (10.0) KruskalleWallis’ p ¼ .056) 71.8 (30.8) 78.5 (23.5) KruskalleWallis’ p ¼ .11sychotic disorders at follow-up

1 (5.0%) 26 (9.7%) c2 ¼ 5.7; df ¼ 3; p ¼ .120 (.0%) 1 (.4%) c2 ¼ .15; df ¼ 3; p ¼ .980 (.0%) 0 (.0%) c2 ¼ 19.6; df ¼ 3; p < .0010 (.0%) 5 (1.9%) c2 ¼ .75; df ¼ 3; p ¼ .861 (5.0%) 7 (2.6%)z c2 ¼ 31.1; df ¼ 3; p < .0010 (.0%) 2 (.7%) c2 ¼ 18.0; df ¼ 3; p < .0010 (.0%) 4 (1.5%) c2 ¼ 11.0; df ¼ 3; p ¼ .010 (.0%) 6 (2.2%)z c2 ¼ 28.0; df ¼ 3; p < .0012 (10.0%)y 43 (16.0%)z c2 ¼ 24.2; df ¼ 3; p < .00110 (66.7%) 140 (55.3%) c2 ¼ 2.0; df ¼ 3; p ¼ .581 (5.9%) 47 (18.6%) c2 ¼ 10.2; df ¼ 3; p ¼ .028 (53.3%) 95 (42.8%) c2 ¼ 4.4; df ¼ 3; p ¼ .221 (6.7%) 34 (15.4%) c2 ¼ 5.2; df ¼ 3; p ¼ .20

8. IQeIntelligence quotient; BPRSeBrief Psychiatric Rating Scale; SANSeScale for theilton Rating Scale for Depression; QLS: Quality of Life Scale; DSM-IV-TReDiagnostic

L.F. Fontenelle et al. / Journal of Psychiatric Research 45 (2011) 1140e11451144

higher rates of psychotic disorders NOS and mood disorders withpsychotic features. The finding that de novo OCD was particularlylikely to result in non-schizophrenic psychoses is consistent withearly observations suggesting that delusions can arise in the courseof OCD but do not necessarily signify a diagnosis of schizophrenia(Insel and Akiskal, 1986). Rather, these delusions may representreactive, affective, or paranoid psychoses that are generallytransient (Insel and Akiskal, 1986), and may be responsive totreatment with antidepressants. For example, it is interesting tonote that 14 out of 25 patients with poor insight OCD recoveredinsight into their OCS after six months of combined treatment withserotonin reuptake inhibitors and cognitive behavioral therapy,with concomitant improvement of obsessive-compulsive anddepressive symptoms (Matsunaga et al., 2002). Although it istempting to suggest that patients with incident OCD are particu-larly prone to develop major depression with psychosis, we cannotexclude, at the present moment, the opposite phenomenon (assuggested by Gittleson, 1966).

Alternatively, it could be argued that the association betweenincident OCD and non-schizophrenia psychoses could be ascribedto the fact that patients from this group reported greater use ofantipsychotics during the two years before the follow-up assess-ments. In fact, the ability of some antipsychotics, particularly theatypicals, to generate de novo OCS, has been well documented inthe literature (for a review, see (Lykouras et al., 2003). Neverthe-less, it should be emphasized that the differences between thegroups in terms of rates of antipsychotic use during the previoustwo years failed to reach statistical significance and that in no casewas OCD judged by the examiners as being related to anysubstance use on the SCID or on the OCD subscale of the CAARMS.Regardless of the specific reason for these findings, the fact thatthe incident OCD group reported greater severity of generalpsychopathology, alogia and avolition-apathy at baseline suggeststhat UHR individuals with prominent negative symptoms maybe particularly prone to develop late-onset primary or drug-induced OCD.

We have shown that a diagnosis of OCD at baseline did notpredict conversion to DSM-IV psychotic disorders. Nevertheless,two (40%) out of six cases with early-onset persistent OCD haddeveloped schizophrenia at follow-up. Studies with larger numberswill be required to examine this further. Accordingly, we could onlyspeculate about the clinical significance of these results. Forexample, persistent OCD could lead to schizophrenia throughexposure to chronic stress (Takahashi et al., 2009), or continualdysfunction in anatomical regions that are shared by both condi-tions, including the basal ganglia, thalamus, anterior cingulum,orbitofrontal cortex and regions of the temporal cortex (Bottaset al., 2005). Based on a correlation analysis, Guillem et al. (2009)suggested that delusions and obsessions, and auditory hallucina-tions and compulsions might share similar pathophysiologicalmechanisms in patients with schizophrenia.

The results need to be interpreted in light of several limitationsof this study. Firstly, despite assessing a large sample of UHRindividuals, we could identify only a small percentage of persistingOCD cases. Indeed, a greater number of individuals showing OCDboth at baseline and follow-up would help us to reach definitiveconclusions regarding the role of persistent OCD and the futuredevelopment of schizophrenia. Secondly, many potentially relevantOCD features were not assessed. In that sense, it would have beeninteresting to evaluate whether severity or type of symptomdimensions differ between patients showing persistent, remitting,or de novo OCD and whether patients showing a particular OCDphenotype were at greater risk for psychosis. Finally, we wereunable to compare the rates of typical vs. atypical antipsychotic useacross different groups of OCD patients. Therefore, we cannot

exclude the possibility that the use of atypical antipsychotics wasmore common among UHR patients that showed de novo OCD,something that could provide a better explanation for the finding ofrelating mood disorders with psychotic features to that group oflate-onset OCD.

Role of funding source

This manuscript “A longitudinal study of obsessive-compulsivedisorder in individuals at ultra-high risk for psychosis” wassupported by the Colonial Foundation and a National Health andMedical Research Council (NHMRC) Program Grant (# 566529).None of the funding agencies had any role in study design; in thecollection, analysis and interpretation of data; in the writing of thereport; and in the decision to submit the paper for publication.

Contributors

A/Prof. Leonardo F. Fontenelle and Prof. Christos Pantelisdeveloped the study hypothesis. A/Prof. Leonardo F. Fontenellemanaged the literature searches, performed the statistical analysesand wrote the first draft of the manuscript. Dr. Ashleigh Lin assistedwith data analysis and editing of the manuscript. Prof. Alison R.Yung, A/Prof. Stephen J. Wood, and Dr. Barnaby Nelson designed thestudy, wrote the protocol, and edited the manuscript. All authorshave contributed to and approved the final manuscript.

Conflict of interest statementA/Prof. Leonardo F. Fontenelle has received travel expense

reimbursement from Solvay and has acted as a consultant toLundbeck. Prof. Chris Pantelis has acted as a consultant for Pfizerand received research grants and acted as a consultant for Janssen-Cilag, Eli Lilly, Hospira (Mayne), and Astra Zeneca. All other authorsdeclare that they have no conflicts of interest

Acknowledements

A/Prof. Leonardo F. Fontenelle was supported by an EndeavourResearch Fellowship (postdoctoral research) from the Departmentof Education, Employment and Workplace Relations (Australia),and Bolsa de Produtividade em Pesquisa, Conselho Nacional deDesenvolvimento Científico e Tecnológico (Brazil); Dr. Ashleigh Linwas supported by an NHMRC program grant; Prof. Christos Panteliswas supported by an NHMRC Senior Principal Research Fellowshipand NHMRC Program Grants; A/Prof. Stephen J. Wood was sup-ported by an NHMRC Clinical Career Development Award anda National Alliance for Research on Schizophrenia and Depression(NARSAD) Young Investigator Award; Dr. Barnaby Nelson wassupported by a Ronald Griffith Fellowship and a NARSAD YoungInvestigator Award; Prof. Alison R. Yung was supported by anNHMRC Senior Research Fellowship.

References

Andreasen NC. Negative symptoms in schizophrenia. Definition and reliability.Archives of General Psychiatry 1982;39:784e8.

Berman I, Pappas D. Bermam S. Obsessive-compulsive symptoms in schizophrenia:are they manifestations of a distinct subclass of schizophrenia disorders. CNSSpectrum 1997;2:45e8.

Bottas A, Cooke RG, Richter MA. Comorbidity and pathophysiology of obsessive-compulsive disorder in schizophrenia: is there evidence for a schizo-obsessivesubtype of schizophrenia? Journal of Psychiatry and Neuroscience2005;30:187e93.

Byerly M, Goodman W, Acholonu W, Bugno R, Rush AJ. Obsessive compulsivesymptoms in schizophrenia: frequency and clinical features. SchizophreniaResearch 2005;76:309e16.

L.F. Fontenelle et al. / Journal of Psychiatric Research 45 (2011) 1140e1145 1145

Craig T, Hwang MY, Bromet EJ. Obsessive-compulsive and panic symptoms inpatients with first-admission psychosis. American Journal of Psychiatry2002;159:592e8.

Cunill R, Castells X, Simeon D. Relationships between obsessive-compulsivesymptomatology and severity of psychosis in schizophrenia: a systematicreview and meta-analysis. Journal of Clinical Psychiatry 2009;70:70e82.

de Haan L, Dudek-Hodge C, Verhoeven Y, Denys D. Prevalence of psychotic disordersin patients with obsessive-compulsive disorder. CNS Spectrum 2009;14:415e7.

de Haan L, Hoogenboom B, Beuk N, van Amelsvoort T, Linszen D. Obsessive-compulsive symptoms and positive, negative, and depressive symptoms inpatients with recent-onset schizophrenic disorders. Canadian Journal ofPsychiatry 2005;50:519e24.

Faravelli C, Abrardi L, Bartolozzi D, Cecchi C, Cosci F, D’Adamo D, et al. The SestoFiorentino study: background, methods and preliminary results. Lifetime prev-alence of psychiatric disorders in an Italian community sample using clinicalinterviewers. Psychotherapy and Psychosomatics 2004;73:216e25.

First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical Interview forDSM-IV-TR Axis I disorders. Research Version, Patient Edition. (SCID-I/P). NewYork: Biometrics Research, New York State Psychiatric Institute; 2002.

Gittleson NL. Depressive psychosis in the obsessional neurotic. British Journal ofPsychiatry 1966;112:883e7.

Guillem F, Satterthwaite J, Pampoulova T, Stip E. Relationship between psychoticand obsessive compulsive symptoms in schizophrenia. Schizophrenia Research2009;115:358e62.

Hamilton M. The assessment of anxiety states by rating. British Journal of MedicalPsychology 1959;32:50e5.

Hamilton M. A rating scale for depression. Journal of Neurology, Neurosurgery andPsychiatry 1960;23:56e62.

Heinrichs DW, Hanlon TE, Carpenter Jr WT. The Quality of life scale: an instrumentfor rating the schizophrenic deficit syndrome. Schizophrenia Bulletin1984;10:388e98.

Henry LP, Harris MG, Amminger GP, Yuen HP, Harrigan SM, Lambert M, et al. Earlypsychosis prevention and intervention centre long-term follow-up study offirst-episode psychosis: methodology and baseline characteristics. Early Inter-vention in Psychiatry 2007;1:49e60.

Insel TR, Akiskal HS. Obsessive-compulsive disorder with psychotic features:a phenomenologic analysis. American Journal of Psychiatry 1986;143:1527e33.

Kringlen E. Obsessional neurotics: a long-term follow-Up. British Journal ofPsychiatry 1965;111:709e22.

Lo WH. A follow-up study of obsessional neurotics in Hong Kong Chinese. BritishJournal of Psychiatry 1967;113:823e32.

Lykouras L, Alevizos B, Michalopoulou P, Rabavilas A. Obsessive-compulsivesymptoms induced by atypical antipsychotics. A review of the reported cases.

Progress in Neuropsychopharmacology and Biological Psychiatry2003;27:333e46.

Lysaker PH, Whitney KA, Davis LW. Associations of executive function withconcurrent and prospective reports of obsessive-compulsive symptoms inschizophrenia. Journal of Neuropsychiatry and Clinical Neurosciences2009;21:38e42.

Matsunaga H, Kiriike N, Matsui T, Oya K, Iwasaki Y, Koshimune K, et al. Obsessive-compulsive disorder with poor insight. Comprehensive Psychiatry2002;43:150e7.

McGorry PD, Goodwin RJ, Stuart GW. The development, use, and reliability of thebrief psychiatric rating scale (nursing modification)ean assessment procedurefor the nursing team in clinical and research settings. Comprehensive Psychi-atry 1988;29:575e87.

Nelson E, Rice J. Stability of diagnosis of obsessive-compulsive disorder in theepidemiologic catchment area study. American Journal of Psychiatry1997;154:826e31.

Niendam TA, Berzak J, Cannon TD, Bearden CE. Obsessive compulsive symptoms inthe psychosis prodrome: correlates of clinical and functional outcome.Schizophrenia Research 2009;108:170e5.

Pollitt J. Natural history of obsessional states; a study of 150 cases. British MedicalJournal 1957;1:194e8.

Poyurovsky M, Fuchs C, Weizman A. Obsessive-compulsive disorder in patientswith first-episode schizophrenia. American Journal of Psychiatry1999;156:1998e2000.

Poyurovsky M, Hramenkov S, Isakov V, Rauchverger B, Modai I, Schneidman M,et al. Obsessive-compulsive disorder in hospitalized patients with chronicschizophrenia. Psychiatry Research 2001;102:49e57.

Sevincok L, Akoglu A, Kokcu F. Suicidality in schizophrenic patients with andwithout obsessive-compulsive disorder. Schizophrenia Research2007;90:198e202.

Takahashi T, Suzuki M, Velakoulis D, Lorenzetti V, Soulsby B, Zhou SY, et al.Increased pituitary volume in schizophrenia spectrum disorders. SchizophreniaResearch 2009;108:114e21.

Tibbo P, Kroetsch M, Chue P, Warneke L. Obsessive-compulsive disorder inschizophrenia. Journal of Psychiatric Research 2000;34:139e46.

Wechsler D. Wechsler abbreviated scale of intelligence. San Antonio, Texas: ThePsychological Corporation; 1999. San Antonio.

Weschler D. Manual for the Wechsler adult intelligence scaledrevised. New York:Psychological Corporation; 1981.

Yung AR, Yuen HP, McGorry PD, Phillips LJ, Kelly D, Dell’Olio M, et al.Mapping the onset of psychosis: the comprehensive assessment of at-risk mental states. Australian and New Zealand Journal of Psychiatry2005;39:964e71.