The following instruments are used:
1. The Prodromal Questionnaire (authorized Dutch translation by M. van der Gaag, R. Klaassen, L. Wunderink)  will be used to screen patients for psychosis-proneness in the general help-seeking population and is a 92-item self-reporting lifetime questionnaire, rated on a two point scale ('agree' and 'disagree'.) Of these items, 45 apply in the case of possible sub-clinical positive symptoms. When maximizing the true positive cases, Loewy et al. found a cut-off score of eight symptoms at the positive subscale predictive for an ARMS on the SIPS (a CAARMS look-alike) with a sensitivity of 90% and a specificity of 48% in a population that was referred because of suspected prodromal state . Since we expect more false positives in the general help-seeking population, the cut-off score for the EDIE-screening is higher . Patients are invited for a further structured clinical interview when the total score of positive symptoms exceeds 18.
2. The Comprehensive Assessment of At Risk Mental State (CAARMS , authorised Dutch version by M. van der Gaag, J. van der Werf, L. Wunderink, A. Malda, R. Klaassen)  is a semi-structured interview that assesses sub-clinical psychotic symptoms in the last year before assessment. Both intensity and frequency of the symptoms are assessed, in order to dimensionally distinguish between not at risk, ARMS and psychosis (see table 1, 2, 3). The CAARMS has a good reliability. High scores on the CAARMS are predictive for transition into psychosis with a relative risk of 12.44 (95% CI = 1.5 - 103.41, p = 0.0025). An ARMS predicts psychosis onset within one year with good sensitivity (86%), specificity (91%), positive predictive value (80%) and negative predictive value (94%) [9, 25].
3. Drug and alcohol use are assessed with the Composite International Diagnostic Interview (CIDI) . The CIDI is a comprehensive, fully standardized instrument for assessing mental disorders according to the definitions and criteria of ICD-10 and DSM-IV. Good reliability and validity of the CIDI have been reported with all Kappa coefficients above .5 for reliability and above .7 for validity .
4. Semantic verbal fluency is assessed with a subtest of the Groninger Intelligence Test, a test that is part of a Dutch set of intelligence tests comparable to the Wechsler . Participants have to name as many animals as possible in one minute. Schizophrenia patients do have more difficulties with tests like this compared to depressed patients and healthy controls, due to cognitive problems  F(2,63) = 3.8 p < .05. A poor result on this test could be a predictor for schizophrenia .
5. Depression is assessed with the Dutch translation of the Beck Depression Inventory second edition (BDI-II-NL) . The BDI is a 21-item self-report questionnaire, which assesses the presence and severity of depressive symptoms. The score ranges from 0 - 63; a high score reflects more severe depression. The BDI-II is positively correlated to the Hamilton Depression Scale (Pearson r = .71). Also the test-retest reliability and the internal consistency show high rates (Pearson r = .93 and α = .91 respectively)
6. The Calgary Depression Scale (CDS)  is an 8-item interview that assesses depressive symptoms independent of the negative symptoms of schizophrenia with a goodness-of-fit index of 0.89 and a root square residual of 0.07. The internal reliability was good (α 0.85). The CDS shows weak statistically significant associations with the negative symptoms on the PANSS (0.33)
7. The Social Interaction Anxiety Scale (SIAS)  is a 20-item self-report questionnaire for social anxiety. Total scores range from 0 to 80. A high score on the SIAS reflects more severe social phobia. The SIAS discriminates significantly (p < .001) between social anxiety and other anxieties and healthy controls. High internal reliability (α ranges from .88 - .94) and test-retest reliability (α = .92) is reported for all scales.
8. Ethnic identity is assessed with the Dutch version of the ICSEY (International Comparative Study of Ethno Cultural Youth) Scale of Ethnic and National Identity [35, 36]. This is a 10-item self-report questionnaire, which assesses ethnic and national affirmation, sense of belonging and feelings about being a group member. Each item is rated on a 5-point scale, ranging from 'strongly disagree' (1) to 'strongly agree' (5). No information about reliability and validity is reported.
9. The Personal Beliefs about Illness Questionnaire-Revised (PBIQ-R)  assesses the subjective appraisal of the illness. It is a self-report questionnaire with five subscales: 1) loss, 2) humiliation, 3) shame, 4) attribution of behaviour to self or to illness and 5) entrapment in psychosis.
10. The Euroqol-5D  assesses quality of life. It is a self-report questionnaire and measures general health-related quality of life. The list contains five dimensions (mobility, self-care, usual activities, pain and anxiety/depression). Each item score ranges from no to extreme problem level. Good reliability and validity are reported for use within a schizophrenic population .
11. Manchester Short Assessment of Quality of Life (MANSA) . The MANSA was developed as a slightly modified instrument for assessing quality of life and satisfaction with specific life domains. The self-report questionnaire contains 16 items, which are rated on a 6-point scale. High face and construct validity was reported for assessing quality of life (coefficients above .82 for all domains). The measured quality of life isn't specifically illness or symptom related and therefore could be used for persons with several mental illnesses .
12. Genetic material will be derived from blood or saliva.
13. The Positive And Negative Syndrome Scale (PANSS)  is a 30 item structured interview that was developed for the assessment of positive (7 items) and negative (7 items) symptoms as well as general psychopathology (16 items) over the past two weeks. The PANSS uses 7-point Likert type scales. A study with 101 Schizophrenia patients  found the three scales to be normally distributed and found evidence of reliability and stability for the positive and negative scales (α .73 and .83, p < .001). The general psychopathology scale has a high internal consistency (α .79, p < .001)
14. The Psychosis Rating Scale (PSYRATS) consists of two subscales that assess auditory hallucinations (11 items) and delusions (6 items). Inter-rater reliability is good, with coefficients in the range of .79 to 1.00. Validity was checked by comparing the PSYRATS with the Psychiatric Assessment (KGV) scale and the PANSS. Significant relationships were found for hallucinations and delusional disruption reported at the KGV, PANSS and PSYRATS .
15. The Dutch version of the Schedules for Clinical Assessment in Neuropsychiatry (SCAN 2.1)  will be used to assess the DSM-IV disorder status at baseline and when a transition to psychosis occurs. The SCAN 2.1 is a semi-structured, diagnostic interview for DSM-IV and ICD-10 designed by the World Health organization and translated into Dutch by Giel and Nienhuis . This interview assesses all kind of symptoms belonging to the most common Axis I disorders, like mood disorders, anxiety disorders, eating disorders, psychotic disorders and cognitive decline. The reliability of this instrument is qualified as moderate to substantial. Diagnosis and non-diagnosis were recognised with a sensitivity of 86% percent and a specificity of 99%. Test-retest reliability was significant for diagnosis (k = .64).
16. Social Demographic Questionnaire (SDQ):
The sociodemographic questionnaire is developed by the researchers to assess socio-demographic factors in our study that may play a role in the development of a psychosis based on previous research and know risk factors for schizophrenia. The items are grouped by type:
1. General: e.g. Current residence; Birth; Relationship status; Household; Previous residences.
2. Education: e.g. number of years full-time; training completed; duplication number; highest level achieved; highest level completed; total numbers of education years; special education; cito-score.
3. Current situation: e.g. currently education; Number of months been successful in training last year; Paid job; Number of months been working successfully last year.
4. DSM-IV: e.g. have you ever received a psychiatric diagnosis? If yes, what diagnosis?
5. Medication: e.g. have you ever received medication? If yes, what medications?
6. Bullying: e.g. Have you been bullied in the past; at what age it started and stopped; Seriousness of harassment?
7. Family data (Hetero-history): e.g. General; Educational history of family.
8. Family history: e.g. familial psychiatric disorders; what degree of family.
9. Pregnancy: e.g. drugs, alcohol, smoking, anaemia during pregnancy; unwanted pregnancy; duration; age of mother at birth; APGAR-score; Birth weight; breastfeeding in baby time.
10. Head injury: e.g. involving injuries?
To determine baseline balance, several variables that are proven to be risk factors will be assessed, such as severity of ARMS-symptoms , heredity , growing up in a big city, stress and unhealthy behaviour of mother during pregnancy , migration and feelings of discrimination [35, 36] and alcohol and drug use, including cannabis use . If the baseline balance is disturbed, the analyses will be corrected for risk factors for transition to psychosis.
Missing data will be imputed by EM-algorithm. EM-algorithm recovers the complete factor loadings considerably better than simple imputation techniques [47, 48].
Group differences will be analysed by unadjusted chi-square. Pearson's chi-square tests will be performed to analyse gender differences between the transition group and the non-transition group, Comparisons will be analysed on intention-to-treat, using the Statistical Package for the Social Sciences (SPSS for windows, version 17). Multiple logistic regression analyses will be used to explore which factors predict transition to psychosis within the UHR-group. To examine the effects of CBT on the positive symptoms reported on the first four scales of the CAARMS, analysis of co-variance will be used.
Survival analyses will be used to measure time to transition and risk factors for developing a first episode psychosis. We will conduct Kaplan-Meier curves to explore the cumulative probability of developing psychosis with inclusion in the study as entry point and last follow-up assessment after 18 months as the end point.
We will use the Mann-Whitney test to examine if people who made a transition scored significantly higher on the CAARMS and the SOFAS compared to patients who did not make a transition. We considered p-values less than 0.05 to be statistically significant.