improving detection of first-episode psychosis by mental health-care services using a self-report...

Original Article

Improving detection of first-episode psychosisby mental health-care services using a

self-report questionnaireeip_147 289..295

Nynke Boonstra,1,2,3 Lex Wunderink,1,3 Sjoerd Sytema3 and Durk Wiersma3

1Friesland Mental Health Services,Department of Education and Research,2NHL University of Applied Science,Institute of Healthcare and WelfareStudies, Leeuwarden, and 3UniversityCenter of Psychiatry, University MedicalCenter Groningen, Groningen University,Groningen, The Netherlands

Corresponding author: Dr NynkeBoonstra, Friesland Mental HealthServices, PO Box 932, 8901 BSLeeuwarden, The Netherlands. Email:[email protected]

Received 9 December 2008; accepted 3July 2009

Abstract

Objective: To examine the utilityof the Community Assessment ofPsychic Experiences (CAPE)-42, a self-report questionnaire, to improvedetection of first-episode psychosisin new referrals to mental healthservices.

Method: At first contact with mentalhealth-care services patients wereasked to complete the CAPE-42 andwere then routinely diagnosed by aclinician. Standard diagnoses wereobtained by means of the mini-Schedule for Clinical Assessment inNeuropsychiatry.

Results: Of the 246 included patients,26 (10.6%) were diagnosed with

psychosis according to the mini-Schedule for Clinical Assessment inNeuropsychiatry. Only 10 of themwere recognized by clinical routine,and 16 psychotic patients were notproperly identified. Using an optimalcut-off of 50 on the frequency ordistress dimension of the positivesubscale of the CAPE-42 detected 14of these misdiagnosed patients. Thesensitivity of the CAPE-42 at this cut-off point was 77.5 and the specificity70.5.

Conclusion: Systematic screening ofpatients using a self-report ques-tionnaire for psychotic symptomsimproves routine detection of psy-chotic patients when they first comeinto contact with mental healthservices.

Key words: duration of untreated psychosis, early detection, earlyintervention, schizophrenia.

INTRODUCTION

Early treatment in first-episode psychosis is one ofthe few available points of intervention to improveprognosis. Duration of untreated psychosis (DUP) isdefined as the time interval between the onset of thefirst psychotic symptoms and initiation of adequatetreatment.1 Short DUP has been associated with anearlier and better level of remission,2–4 a betterchance of recovery,5 lower relapse rates,6,7 less cog-nitive deterioration,8 less positive9–12 and negativesymptoms,3,13 and better social functioning.14,15 Theresults of these studies are not directly comparablebecause of differences in adjusting to conceivabledeterminants of DUP, measurement of outcomevariables and the fact that the definition of DUPhas been operationalized differently. Despite these

difficulties in the comparing of results, all evidencepoints in the same direction: a longer DUP is asso-ciated with poorer outcomes. DUP seems to be anindependent predictor of illness and treatment out-come,16 though the causal nature of this relationshiphas yet to be established. The most importanthypothesis explaining the deleterious effects of alonger DUP is that emergent and active psychosiscauses progressive grey matter loss, mainly duringthe critical period around the first episode.17,18 Animportant prerequisite for early treatment is earlydetection. Over the past 10 years specific programsfor early detection and treatment of first-episodepsychosis have been developed with the aim ofreducing DUP. These programs – which are probablyworthwhile even though there is no firm evidence tosupport their efficacy19 – are primarily focused on

Early Intervention in Psychiatry 2009; 3: 289–295 doi:10.1111/j.1751-7893.2009.00147.x

© 2009 The AuthorsJournal compilation © 2009 Blackwell Publishing Asia Pty Ltd

289

reducing delay in help-seeking and delay in referralto health services. For example, the ScandinavianTreatment and Intervention Psychosis Study studyrevealed that intensive education of the generalpublic, schools and the primary health-care servicesto help them recognize psychotic symptoms was thekey to reducing delay in help seeking and delay inreferral by primary care.20,21 Norman et al. describedtwo components representing DUP; firstly the delaycaused by the patients contacting health profes-sionals, and secondly, the delay caused by servicesafter the first contact has already been made.1 Bothof these delay components appear to be of equalimportance. Patients who had their first contactwith services before the first onset of psychosis hada substantial longer service delay component. Thisalso holds true for those patients who had their firstcontact around their first onset of psychosis.1

Brunet et al.22 even described three components ofDUP as: delay in help-seeking, delay in referral anddelay in recognition by mental health-care services.The delay caused by mental health services wasfound to account for 35% of overall DUP in a studyof 80 participants from the inner city of Birming-ham.22 Both Norman et al. and Brunet et al. arguethat efforts to minimize effects of DUP should beprimarily targeted at a reduction of the servicedelay. Very recently, a large scale multi-centre studyof pathways to care also demonstrated that a sub-stantial proportion of DUP was caused by delayafter first contact by mental health-care services.23

The evidence suggests that mental health-careservices do not properly recognize psychosis. Wehypothesized: (i) that a significant number ofpatients with a first-episode psychosis who arenewly referred to mental health-care services arenot recognized by routine clinical procedures;and (ii) that use of a self-report questionnaire, theCommunity Assessment of Psychic Experiences(CAPE)-42, will improve recognition of first-episodepsychosis.

MATERIAL AND METHODS

Subjects

Patients with a first-ever contact with mentalhealth-care services in part of the Dutch province ofFriesland (catchment area of 300 000 inhabitants)during a period of 18 months (November 2006–April2008) were considered for inclusion in the presentstudy. Patients were excluded if they: (i) were notbetween the ages of 18 and 65; (ii) were unable tounderstand and/or speak Dutch; (iii) were mentally

retarded; (iv) did not enter treatment within2 months; or (v) were clinically diagnosed with aDiagnostic and Statistical Manual of Mental Disor-ders (DSM)-IV V-code for psychosocial problems.The characteristics of the sample are shown inTable 1. During the selected inclusion period of18 months 1329 first contact patients were referredto mental health-care services. Of them, 372 did notmeet inclusion criteria: seven patients were unableto speak Dutch; seven patients were excludedbecause of mental retardation; 23 patients wereyounger than 18 or older than 65 years; 257 did notenter treatment after their first visit and therefore noinformation was collected; and 78 patients wereclinically diagnosed with a DSM-IV V-code for psy-chosocial problems only. A random sample of 350patients was drawn from the 957 patients who metinclusion criteria, and they were invited to partici-pate in the study and provided informed consent.Data for 104 patients were incomplete: clinical diag-noses of 27 patients were not reported; 40 patientsdid not complete the CAPE-42;24 and 37 patientsrefused to be interviewed with the mini-Schedulefor Clinical Assessment in Neuropsychiatry (mini-SCAN).25 246 patients (70%) had complete data.These patients did not differ significantly on gender,age or baseline Global Assessment of Functioningscores from the total eligible population of 957patients or from the 104 patients from whom datawere incomplete.

Assessment

Figure 1 shows the consort flow chart. Patients wereasked to complete the self-report questionnaire

TABLE 1. Patient characteristics (n = 246)

n (%) Mean (SD)

GenderMale 118 (48.0)Female 128 (52.0)

Age (mean � SD) 37.5 (12.7)Male (mean � SD) 37.6 (12.2)Female (mean � SD) 37.3 (12.2)

GAF baseline score (mean � SD) 60.3 (8.4)Employment

Yes 148 (60.2)No 95 (38.6)Unknown 3 (1.2)Working hours/week (mean � SD) 148 27.9 (14.0)

Living situationLiving with a partner/family 115 (46.7)Living alone 130 (52.8)Unknown 1 (0.4)

GAF, Global Assessment of Functioning; SD, standard deviation.

Improving detection of psychosis

290 © 2009 The AuthorsJournal compilation © 2009 Blackwell Publishing Asia Pty Ltd

CAPE-42.24 In addition to the clinical diagnosticroutine outlined by the DSM-IV,26 patients werediagnostically assessed by a research psychiatristor a psychologist using the mini-SCAN. Weselected this instrument as the ‘gold standard’ fordiagnosis.25 Blinding procedures were put in placefor all assessments. Clinicians were blinded frommini-SCAN results, researchers were kept blindfrom clinicians’ diagnoses. Both clinicians andresearchers were kept blind from CAPE-42 results.

Instruments

The CAPE-42

The CAPE-42 is a 42 item self-report questionnairemeasuring positive and negative psychoticsymptoms and depressive symptoms on a two-dimensional scale. The first dimension measuresthe frequency of symptoms on a four-point scale of‘never’ = 1, ‘sometimes’ = 2, ‘often’ = 3 and ‘nearlyalways’ = 4, and the second dimension measuresthe degree of distress caused by the experience: ‘notdistressed’ = 1, ‘a bit distressed’ = 2, ‘quite dis-tressed’ = 3 and ‘very distressed’ = 4. The total scoreranges from 42 to 168 on both dimensions. The posi-tive subscale counts 20 items (range 20–80 on bothdimensions), the negative subscale 14 items (range14–56 on both dimensions) and the depressive sub-scale 8 items (range 8–32 on both dimensions).24,27–29

The CAPE-42 has been designed to assess lifetimepsychotic experiences in the general population.

Mini-SCAN

Patients were diagnostically assessed with the mini-SCAN, a short version of the validated Schedules forClinical Assessment in Neuropsychiatry (SCAN 2.1)

using the same algorithm.30 The mini-SCAN wasused as the ‘gold standard’ for diagnosis in thepresent study to assess the sensitivity and specificityof the CAPE-42. Like its predecessor, the mini-SCANis a semi-structured diagnostic interview used toestablish an axis-1 DSM-IV diagnosis.26 Themini-SCAN has an advantage in offering a moreuniformly defined interactive computerized inter-viewing schedule and a brief time of administrationof 30 to 60 min.

Training and reliability

The diagnostic mini-SCAN interviews were admin-istered by research psychiatrists and psychologistswho are familiar with psychopathology classifica-tion systems and were formally trained by the DutchWHO Centre for Training and Research at the Uni-versity Medical Center Groningen, Department ofPsychiatry. Interviewers were trained in administra-tion techniques and software operating instructionsof the software program. Reliability was enhancedby rating videotaped interviews, followed by groupreviews of the ratings. Inter-rater reliability for themini-SCAN was established by a pairwise compari-son of three raters, all rating the same 15 randomlyselected subjects. The mean of the pairwise com-parison unweighted kappa score on the diagnosticcategory appeared to be high (kappa = 0.87).

Statistical analysis

We used receiver operator characteristics (ROCs) tocalculate the maximum area under the curve (AUC)defining the best scale and the optimal cut-off levelon that scale of CAPE-42. The AUC ranges from 0.5(the discriminatory ability of a test is no better than

n = 1329 First referrals to mental health-care services during the inclusion

period

n = 957 Meting inclusion criteria n = 372 Not meeting inclusion criteria because of age, language, intellectual retardation, only psychosocial problems or not entering treatment

n = 350 Random sample

n = 104 Incomplete data set n = 246 Sample with complete data set

FIGURE 1. Flow chart.

N. Boonstra et al.


291

chance) to 1.0 (perfect discriminatory ability). SPSS(version 15.0; SPSS Inc., Chicago, IL, USA) was usedto analyse the data.

RESULTS

Patient sample

Twenty-six (11%) of the 246 included patients had apsychotic disorder according to the mini-SCAN.Seventy-three percent of them were males (n = 19)with a mean age of 33.7 years (standard deviation(SD) = 12.6); women had a mean age of 38.4 years(SD = 13.5). Only 10 of these 26 patients were diag-nosed with a psychotic disorder by the clinician, theother 16 patients were assigned to other diagnoses:mood disorder (n = 7), anxiety disorder (n = 1),adjustment disorder (n = 5), substance related

disorder (n = 2) and impulse control disorder(n = 1). An overview of mini-SCAN diagnoses versusclinical diagnosis is shown in Table 2. Patientsdetected as having a psychotic disorder by the mini-SCAN did not significantly differ from the undetec-ted patients on gender, age or baseline GlobalAssessment of Functioning scores. Using mini-SCAN as the ‘gold standard’ for psychiatric diagnosisof psychosis, ROC analyses were conducted on thepositive symptom subscale scores from the fre-quency and distress dimensions of the CAPE-42(Fig. 2). The AUC score for the positive subscalefrom the frequency dimension was.76 and the AUCfor the positive subscale from the distress dimen-sion was.66 (P < 0.001).

As the aim of the present study was to improvedetection of first-episode psychotic patients witha self-report questionnaire, sensitivity should be

TABLE 2. Diagnostic distribution; mini-Schedule for Clinical Assessment in Neuropsychiatry (mini-SCAN) diagnosis of psychosis versusclinical diagnosis (n = 26)

Clinical diagnosis

Psychotic disorder No psychotic disorder

Mini-SCANdiagnosis

295.9 297.1 298.8 298.9 Mooddisorder

Substance-induceddisorder

Adjustmentdisorder

Anxietydisorder

Impulse-controldisorder

295 0 0 0 3 1 0 0 0 0297.1 0 2 0 0 0 0 0 0 0298.8 1 0 1 1 1 1 1 0 0298.9 0 0 1 1 5 1 4 1 1

1 - Specificity1,00,80,60,40,20,0

Sen

siti

vity

1,0

0,8

0,6

0,4

0,2

0,0

CAPE positive distress

CAPE positivefrequency

Source of the curve

ROC curve

FIGURE 2. Receiver operator characteris-tic (ROC) curve for Community Assess-ment of Psychic Experiences (CAPE)positive subscale frequency and distressdimensions for the mini-Schedule forClinical Assessment in Neuropsychiatrypsychosis. Diagonal segments are pro-duced by ties.



maximized, with as few false positives as possible.Sensitivity, specificity, positive predictive value(PPV) and negative predictive value (NPV) ofthe CAPE positive subscale frequency or distressdimension at various cut-off points are presented inTable 3. A cut-off score of 50 on the CAPE-42 posi-tive subscale of the frequency dimension or 50 onthe positive subscale of the distress dimension witha sensitivity of 77% and a specificity of 70.5% appearto be optimal.

The proportion of patients with CAPE-42 scoresabove 50 on the positive subscale frequency or dis-tress who are correctly diagnosed with a psychoticdisorder (PPV) was 23.5% and the proportion ofpatients with negative CAPE-42 scores on the posi-tive subscale frequency or distress who are correctlydiagnosed as non-psychotic (NPV) was 96%. Four ofthe six patients not detected by CAPE-42 with thesecut-off scores were identified by clinical routine.Results are shown in Table 4. The clinicians agreedto change the clinical diagnosis into the mini-SCAN

diagnosis of a psychotic disorder for all 16 patientswho were not properly clinically diagnosed with apsychosis initially.

DISCUSSION

We found that in routine clinical practice patientswere not appropriately recognized leaving a sub-stantial number of psychotic patients undetected.We showed that implementation of a self-reportquestionnaire, the CAPE-42, improved recognitionof first-episode psychotic patients.

The finding that first-episode psychotic patientsare undetected is in accordance with recently pub-lished data reporting a delay in recognition andinitiation of adequate treatment of first-episodepsychotic patients caused by mental health serviceswas reported.23 These findings consistently stressthe importance of timely recognition of first-episode psychotic patients by mental health-care

TABLE 3. Different cut-off points on CAPE-42 positive subscale (n = 246)

Cut off point on CAPEpositive subscalefrequency or distressdimension

Sensitivity (%) Specificity (%) PPV (%) NPV (%)

41 100 5.9 11.2 10043 96.2 16.4 12.0 97.345 84.6 33.6 13.1 94.948 76.9 57.3 17.5 95.550 76.9 70.5 23.5 96.352 65.4 84.1 32.7 95.454 42.3 87.7 28.9 92.8

CAPE, Community Assessment of Psychic Experiences; NPV, negative predictive value; PPV, positivepredictive value.

TABLE 4. Diagnostic partition of patients (n = 246)

Clinical diagnosis Total

Psychotic disorder No psychotic disorder

Mini-SCANpsychotic disorder

Mini-SCAN nopsychotic disorder

Mini-SCAN psychoticdisorder

Mini-SCAN nopsychotic disorder

CAPE-42 score �50 on positivesubscale frequency or distressdimension

6 0 14 65 85

CAPE-42 score <50 on positivesubscale frequency or distressdimension

4 0 2 155 161

Total 10 0 16 220 246

CAPE, Community Assessment of Psychic Experiences; mini-SCAN, mini-Schedule for Clinical Assessment in Neuropsychiatry.

N. Boonstra et al.


293

services. In the present study 16 psychotic patients(62% of all psychotic patients) were initially missedby the clinician. A first possible explanation mightbe that many psychotic patients do not presenttheir psychotic symptoms overtly and clinicians donot thoroughly investigate the possibility of a psy-chosis as a routine procedure, for example in a dif-ferential diagnosis. In a recent study we found thatabout half of the patients presenting at least twospecific psychotic symptoms were diagnosed witha non-psychotic disorder and did not receiveadequate treatment in accordance with these psy-chotic symptoms.31 A second explanation might bethe lack of revision of diagnostic categorizationafter diagnosis has once been set. This has alsobeen suggested by Norman et al.1 Systematic initialscreening and routine outcome assessment proce-dures during the course of treatment could help toprevent these flaws in clinical diagnosis. In thepresent study we used the CAPE-42 as a systematicinitial screener. The next step is to define the mostuseful cut-off score: the lower the cut-off, the morepatients will be detected, but the more diagnosticinterviews will have to be done. We decided that aCAPE-42 cut-off score of 50 on positive subscalefrequency or distress dimension as add on to clini-cal diagnosis was the optimal threshold to furtherscreening with mini-SCAN. But other policymakers could make other choices depending ontheir aims. For example, lowering the cut-off scoreto 41 would include only two extra psychoticpatients in our sample, at the expense of 148 extramini-SCAN interviews (233 instead of 85). At aCAPE-42 cut-off score of 50 on the positive sub-scale frequency or distress dimension, of 26 psy-chotic patients, 20 were detected. Of the remainingsix patients four were clinically diagnosed with apsychotic disorder.

The six undetected patients all had relatively lowscores on the positive dimension, and also lowerscores on the negative subscale frequency and dis-tress dimension compared with the 20 patients whowere detected by CAPE-42. For the distress dimen-sion this difference was significant (P = 0.007).

The two patients who remained undetected hadhigher scores on the negative subscale frequencyand distress dimension, even though this differenceis not significant. These patients might have beenassigned to a diagnosis of non-psychotic disorderbecause of their more prominent negative symp-toms and relative lack of positive symptoms.

The incidence of first-episode psychosis isrelatively low. As a result, relatively large samplesof patients would have to be screened to identifynew cases of psychosis. However, the advantage

of screening a population of new referrals withpsychiatric problems is considerable; given the riskassociated with long treatment delay, the invest-ment seems to be a justified. This study demon-strates once again that the delay within mentalhealth-care services should not be underestimated.Psychotic patients appear to be seriously underde-tected. More systematic efforts should be investedin detecting psychotic patients at their first contactswith mental health care instead of emphasizing theimportance of detecting psychotic patients hiddenin the general population. Systematic screening by aself-report questionnaire, in addition to clinicaldiagnosis, seems to be an important tool to reducethe DUP because of mental health-care servicedelay. The CAPE-42 is a self-report questionnaireand therefore based on subjective reports. The reli-ability of self-reported psychotic symptoms can bedisputed. However, previous studies show that self-report can be used to assess severity of psychosis inclinical and research settings.32 Moreover, patientsappear to be more willing to report psychotic symp-toms and experiences using self-report question-naires than in a personal interview.33 A possiblelimitation of this study is the exclusion of patientswith only psychosocial problems, and of patientswho were not accepted for treatment within2 months, which may also explain the relativelylarge proportion of psychotic patients in our sample(i.e. 26/246). We also included only patients whowere able to speak and understand Dutch; findingsare not representative for the non-Dutch-speakingpatients. However the proportion of non-Dutchspeaking patients in the areas studied is very low.

It was shown recently by a number of studies thata significant proportion of psychotic patientsremain undetected by mental health services. To thebest of our knowledge this study is one of the first toexamine systematic screening using a self-reportquestionnaire for psychotic symptoms that showssignificant improvement in routine detection ofpsychotic patients at first contact with mentalhealth services.

ACKNOWLEDGEMENT

This study was supported by a grant from theProvince of Friesland.

REFERENCES

1. Norman RM, Malla AK, Verdi MB, Hassall LD, Fazekas C.Understanding delay in treatment for first-episode psychosis.Psychol Med 2004; 34: 255–66.



2. Malla AK, Norman RM, Manchanda R et al. One yearoutcome in first episode psychosis: influence of DUP andother predictors. Schizophr Res 2002; 54: 231–42.

3. Simonsen E, Friis S, Haahr U et al. Clinical epidemiologicfirst-episode psychosis: 1-year outcome and predictors. ActaPsychiatr Scand 2007; 116: 54–61.

4. Wunderink A, Nienhuis FJ, Sytema S, Wiersma D. Treatmentdelay and response rate in first episode psychosis. Acta Psy-chiatr Scand 2006; 113: 332–9.

5. Wunderink L, Sytema S, Nienhuis FJ, Wiersma D. Clinicalrecovery in first-episode psychosis. Schizophr Bull 2009; 35(2): 362–9; doi: 10.1093/schbul/sbn143.

6. Altamura AC, Bassetti R, Sassella F, Salvadori D, Mundo E.Duration of untreated psychosis as a predictor of outcome infirst-episode schizophrenia: a retrospective study. SchizophrRes 2001; 52: 29–36.

7. de Haan L, Linszen DH, Lenior ME, de Win ED, Gorsira R.Duration of untreated psychosis and outcome of schizophre-nia: delay in intensive psychosocial treatment versus delay intreatment with antipsychotic medication. Schizophr Bull2003; 29: 341–8.

8. Amminger GP, Edwards J, Brewer WJ, Harrigan S, McGorryPD. Duration of untreated psychosis and cognitive deterio-ration in first-episode schizophrenia. Schizophr Res 2002; 54:223–30.

9. Black K, Peters L, Rui Q, Milliken H, Whitehorn D, Kopala LC.Duration of untreated psychosis predicts treatment outcomein an early psychosis program. Schizophr Res 2001; 47: 215–22.

10. Bottlender R, Sato T, Jager M et al. The impact of the dura-tion of untreated psychosis prior to first psychiatric admis-sion on the 15-year outcome in schizophrenia. Schizophr Res2003; 62: 37–44.

11. Larsen TK, Friis S, Haahr U et al. Early detection and interven-tion in first-episode schizophrenia: a critical review. ActaPsychiatr Scand 2001; 103: 323–34.

12. Norman RM, Malla AK, Manchanda R. Early premorbidadjustment as a moderator of the impact of duration ofuntreated psychosis. Schizophr Res 2007; 95: 111–14.

13. Perkins D, Lieberman J, Gu H et al. Predictors of antipsychotictreatment response in patients with first-episode schizophre-nia, schizoaffective and schizophreniform disorders. Br JPsychiatry 2004; 185: 18–24.

14. Drake RJ, Haley CJ, Akhtar S, Lewis SW. Causes and conse-quences of duration of untreated psychosis in schizophrenia.Br J Psychiatry 2000; 177: 511–15.

15. Harris MG, Henry LP, Harrigan SM et al. The relationshipbetween duration of untreated psychosis and outcome: aneight-year prospective study. Schizophr Res 2005; 79: 85–93.

16. Norman RM, Lewis SW, Marshall M. Duration of untreatedpsychosis and its relationship to clinical outcome. Br J Psy-chiatry Suppl 2005; 48: s19–23.

17. DeLisi LE. The concept of progressive brain change in schizo-phrenia: implications for understanding schizophrenia.Schizophr Bull 2008; 34: 312–21.

18. Wood SJ, Pantelis C, Velakoulis D, Yucel M, Fornito A,McGorry PD. Progressive changes in the development towardschizophrenia: studies in subjects at increased symptomaticrisk. Schizophr Bull 2008; 34: 322–9.

19. Marshall M, Rathbone J. Early intervention for psychosis.Cochrane Database Syst Rev 2006; (4): CD004718.

20. Joa I, Johannessen JO, Auestad B et al. The key to reducingduration of untreated first psychosis: information cam-paigns. Schizophr Bull 2008; 34: 466–72.

21. Johannessen JO, McGlashan TH, Larsen TK et al. Early detec-tion strategies for untreated first-episode psychosis.Schizophr Res 2001; 51: 39–46.

22. Brunet K, Birchwood M, Lester H, Thornhill K. Delays inmental health services and duration of untreated psychosis.Psychiatr Bull 2007; 31: 408–10.

23. Marshall M, Husain N, Warwurton J et al. Delays in thepathway to care for a first episode of psychosis: an analysisof the components of duration of untreated psychosisfrom the Nation Eden Study. Early Interv Psychiatry 2008;2 (Suppl. 1): A18.

24. Stefanis NC, Hanssen M, Smirnis NK et al. Evidence that threedimensions of psychosis have a distribution in the generalpopulation. Psychol Med 2002; 32: 347–58.

25. Nienhuis FJ, Giel R. mini- Schedules for Clinical Assessment inNeuropsychiatry. Amsterdam: Harcourt, 2000.

26. Practice guideline for the treatment of patients with schizo-phrenia. American Psychiatric Association. Am J Psychiatry1997; 154 (Suppl. 4): 1–63.

27. Hanssen M, Peeters F, Krabbendam L, Radstake S, Verdoux H,Van Os J. How psychotic are individuals with non-psychoticdisorders? Soc Psychiatry Psychiatr Epidemiol 2003; 38: 149–54.

28. Hanssen MS, Bijl RV, Vollebergh W, Van Os J. Self-reportedpsychotic experiences in the general population: a validscreening tool for DSM-III-R psychotic disorders? Acta Psy-chiatr Scand 2003; 107: 369–77.

29. Konings M, Bak M, Hanssen M, Van Os J, Krabbendam L.Validity and reliability of the CAPE: a self-report instrumentfor the measurement of psychotic experiences in thegeneral population. Acta Psychiatr Scand 2006; 114: 55–61.

30. Wing JK, Babor T, Brugha T et al. SCAN. Schedules for clinicalassessment in neuropsychiatry. Arch Gen Psychiatry 1990;47: 589–93.

31. Boonstra N, Wunderink L, Sytema S, Wiersma D. Detection ofpsychosis by mental health care services; a naturalisticcohort study. Clin Pract Epidemol Ment Health 2008; 4:29.

32. Niv N, Cohen AN, Mintz J, Ventura J, Young AS. The validityof using patient self-report to assess psychotic symptoms inschizophrenia. Schizophr Res 2007; 90: 245–50.

33. Hamera EK, Schneider JK, Potocky M, Casebeer MA. Validityof self-administered symptom scales in clients with schizo-phrenia and schizoaffective disorders. Schizophr Res 1996;19: 213–19.

N. Boonstra et al.


295

improving detection of first-episode psychosis by mental health-care services using a self-report...

Documents