Symptoms associated with psychosis risk in an adolescent birth cohort: improving questionnaire utility with a multidimensional approach

Aim: Specialized self‐report questionnaires have been developed for detection of symptoms indicative of psychosis risk. The identification of at‐risk individuals is typically based on sum scores, which assume equal severity and discriminability of all symptoms, and a single dimension of illness. Our aim was to test whether separable dimensions of risk could be identified in the general population. Methods: We explored the latent structure of one such questionnaire using full‐information item factor analysis, deriving exploratory models from the PROD‐Screen questionnaire responses of the adolescent generalpopulation based on the Northern Finland 1986 Birth Cohort (n = 6611). Results: A three‐dimensional factor structure of positive, negative and general symptoms emerged. The factor structure, the appropriateness of the statistical model and the application of the results to the detection of heightened psychosis risk are discussed. Conclusions: In explicitly taking into account the multidimensionality and varying symptom severity of the included items, the current model provides an improvement in questionnaire‐based assessment of psychosis risk.     

Young people at risk for psychosis: case finding and sample characteristics of the Oulu Brain and Mind Study

Aim: Set within the general population‐based Northern Finland Birth Cohort 1986, the Oulu Brain and Mind Study aims to explore the causes and pathogenesis of psychotic illness by following young people at risk for psychosis due to having a first‐degree relative with psychotic illness or due to having experienced psychotic‐like symptoms themselves. We report the study methods and explore the relationship between these definitions of high risk for psychosis and operational criteria for a prodromal psychosis syndrome based on interview. Methods: Prospectively collected data from earlier follow‐ups of this cohort were combined with health register data to categorize subjects as those with familial risk (n = 272), symptomatic risk (n = 117), psychosis (n = 78), attention deficit hyperactivity disorder (ADHD) (n = 103) and a sample of controls (n = 193) drawn randomly from the remaining cohort. The Structured Interview for Prodromal Syndromes (SIPS) was applied to all, 295 participants together with questionnaires measuring psychosis vulnerability and schizotypal traits. Results: There were 29 (10%) current prodromal cases. Criteria for the current prodromal syndrome were fulfilled by 12% of the familial risk group and 19% of the symptomatic risk group, compared with 5% of the ADHD group and 4% of controls. Conclusion: We successfully detected young people with a prodromal psychosis syndrome although relatively few subjects deemed to be at high risk met the full operational criteria according to the SIPS interview. Combining methods from familial, clinical and psychometric high‐risk approaches provides a tractable method for studying risk of psychosis in the general population.     

Prevalence of suicide risk factors in people at ultra-high risk of developing psychosis: a service audit

Background: Little data is available on the prevalence of suicide risk factors in people at ultra‐high risk (UHR) of developing psychosis. Aim: The aim of this study was to provide an estimate of the cross‐sectional prevalence of possible suicide risk factors in those attending a routine clinical service for people at UHR of developing psychosis. Methods: For all patients in treatment (n = 34) over a 4‐week period, levels of suicidal ideation and depression upon entry to the service were assessed by auditing intake scores on the Beck Depression Inventory, second edition. Level of engagement with services, social isolation, substance and alcohol misuse, ready access to means, current suicidal ideation, previous suicide attempts, current or previous self‐harm, expressions of concern from others, depression, agitation, hopelessness, worthlessness, suspiciousness and fears of mental disintegration were all assessed by case note review and interview with the treating clinician. Results: There was a high prevalence of at least mild suicidal ideation (58.8%, n = 20) and severe depressed mood (47%, n = 16) in this client group at point of entry to the service. Seven people (20.6%) had engaged in serious self‐harm (including attempted suicide) during the time they were in contact with the service. Forty‐seven per cent (n = 16) reported at least 27 suicide attempts between them; the mean number of attempts being 1.69 (standard deviation = 1.08). Conclusion: Suicide risk was high in this small sample of people at UHR of developing psychosis. Controlled research with larger samples and better methodology is urgently required to inform legal, ethical and scientific debates surrounding this group.     

Maladaptive schemas as a mediator between social defeat and positive symptoms in young people at clinical high risk for psychosis

Aim: Social defeat may be the mechanism that links past social adversities with the development of psychosis. In depression research, it is accepted that adverse early social experiences can lead to enduring cognitive vulnerabilities, characterized by negative schemas about the self and others. The aim of this study was to examine whether negative beliefs about the self and others link social defeat to early signs of psychosis. Methods: Data from a sample of individuals at high risk for developing psychosis (n = 38) were assessed using measures of social defeat and schemas. Results: High levels of social defeat and negative evaluations of the self and others were displayed. Negative beliefs mediated the relationship between social defeat and early symptoms, offering some support for the notion that maladaptive self‐beliefs play a role in the onset of psychosis. Conclusions: These results have implications for prevention because these maladaptive schemas are malleable factors for which we have effective psychological interventions.     

Reasons for using substances in adolescents with and without psychosis

Aim: To increase our understanding of substance use among adolescents with psychosis by comparing their reasons for use with those of typically developing adolescents. Methods: Participants were 35 outpatient adolescents, recently admitted for a first episode of psychosis to an early psychosis programme, and 35 typically developing adolescents. Measures used included the Personal Experience Screening Questionnaire and Reasons for Use Scale. Results: Adolescents with psychosis used significantly more substances than adolescents without psychosis (P < 0.01). Reasons for use in the two groups were similar. Conclusions: Although adolescents with psychosis report using substances for the same reasons as other adolescents, they may also use to self‐medicate secondary morbidity associated with psychosis. These results have a number of implications for prevention and treatment.     

Personality disorders and accentuations in at‐risk persons with and without conversion to first‐episode psychosis

Aim: The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition Cluster A personality disorders (PDs), particularly schizotypal PD, are considered a part of the schizophrenia spectrum and a risk factor of psychosis. The role of PDs and personality accentuations (PAs) in predicting conversion to psychosis was studied in patients symptomatically considered at risk, assuming a major role of the schizotypal subtype. Methods: PDs and PAs, assessed at baseline with a self‐report questionnaire, were compared between risk‐, gender‐ and age‐matched at‐risk patients with (n = 50) and without conversion to psychosis (n = 50). Results: Overall, Cluster A‐PDs were the least frequent cluster (14%), and schizotypal PD was rare (7%). Yet, PDs in general were frequent (46%), especially Cluster B‐ (31%) and C‐PDs (23%). Groups did not differ in frequencies of PDs, yet converters tended to have a higher expression of schizoid (P = 0.057) and Cluster A‐PAs (P = 0.027). In regression analyses, schizoid PA was selected as sole but weak predictor of conversion (OR = 1.685; 95% CIs: 1.134/2.504). Conclusions: Unexpectedly, schizotypal PD was infrequent and did not predict conversion. Conversion was best predicted by schizoid PA, indicating more severe, persistent social deficits already at baseline in later converters. This corresponds to premorbid social deficits reported for genetic high‐risk children and low social functioning in at‐risk patients later converting to psychosis. Further, PDs occurred frequently in at‐risk patients irrespective of conversion. As psychopathology and personality relate closely to one another, this result highlights that, beyond the current narrow focus on schizotypal PD, personality‐related factors should be considered more widely in the prevention of psychosis.     

Distinct and opposite profiles of connectivity during self‐reference task and rest in youth at clinical high risk for psychosis

Self‐reference is impaired in psychotic disorders such as schizophrenia, associated with disability, and closely related to characteristic patterns of aberrant brain connectivity. However, at present, it is unclear whether self‐reference is impacted in pathogenesis of the disorder. Alterations in connectivity during a self‐reference task or resting‐state in the psychosis risk (i.e., prodromal) period may yield important clues for biomarker development, as well as for novel treatment targets. This study examined a task‐based and resting‐state functional magnetic resonance imaging in individuals at clinical high risk (CHR) for psychosis (n = 22) and healthy control unaffected peers (n = 20). The self‐reference task comprised three task conditions where subjects were asked if an adjective was relevant to themselves (self), a designated other individual (other), or to evaluate the word's spelling (letter). Connectivity analyses examined medial prefrontal cortex (mPFC) and posterior cingulate cortex (PCC), regions commonly found in conjunction analyses of self‐reference, during both the self‐reference task and rest. In task connectivity analyses, CHR individuals exhibited decreased mPFC–PCC connectivity when compared to controls. In resting‐state analyses, CHR participants showed greater mPFC–PCC connectivity. Taken together, results suggest that psychosis‐like alterations in mPFC–PCC connectivity is present prior to psychosis onset across both task and rest.     

Associations between number of different type of care meetings with social network and improvement in mental well-being in adolescents at risk of first-episode psychosis

Aim: As research in the care of people at risk of developing first‐episode psychosis has mostly focused on cognitive behavioural therapy and antipsychotic medication, little is known about associations between changes in mental well‐being and effect of people participating in the care. Methods: Multiprofessional early intervention team met with adolescents who are at risk of psychosis, including coworkers and their families. Correlations were calculated between change scores in prepsychotic, functioning ability, quality of life (QoL), depression and anxiety scales, and number of family, coworker and adolescent‐participating social network meetings, and total number of social network meetings during the care. Results: Larger change scores in functioning ability were positively associated with the number of social network meetings with participating coworker (P = 0.041), but not with other types of participant meetings. Larger change scores in prepsychotic symptoms were positively associated with the number of meetings where the adolescent was participating (P = 0.001), the number of network meetings where the coworker was participating (P = 0.007) and the number of all meetings (P = 0.001). The number of any other type of meetings did not associate with change scores in QoL, depression and anxiety. Conclusion: According to the present results, adolescents at risk of psychosis seem to benefit from the inclusion of coworkers from the adolescents' natural surroundings in care; this could help to increase functioning ability. Different combinations of meetings, such as larger number of total meetings, larger number of meetings with the adolescent and larger number of meetings with coworkers from the adolescents' natural surroundings, seem to associate with stronger decrease in pre‐psychotic symptoms.     

The Basel Screening Instrument for Psychosis (BSIP): development, structure, reliability and validity

BACKGROUND: Early detection of psychosis is of growing clinical importance. So far there is, however, no screening instrument for detecting individuals with beginning psychosis in the atypical early stages of the disease with sufficient validity. We have therefore developed the Basel Screening Instrument for Psychosis (BSIP) and tested its feasibility, interrater-reliability and validity. AIM: Aim of this paper is to describe the development and structure of the instrument, as well as to report the results of the studies on reliability and validity. METHOD: The instrument was developed based on a comprehensive search of literature on the most important risk factors and early signs of schizophrenic psychoses. The interraterreliability study was conducted on 24 psychiatric cases. Validity was tested based on 206 individuals referred to our early detection clinic from 3/1/2000 until 2/28/2003. RESULTS: We identified seven categories of relevance for early detection of psychosis and used them to construct a semistructured interview. Interrater-reliability for high risk individuals was high (Kappa .87). Predictive validity was comparable to other, more comprehensive instruments: 16 (32 %) of 50 individuals classified as being at risk for psychosis by the BSIP have in fact developed frank psychosis within an follow-up period of two to five years. CONCLUSIONS: The BSIP is the first screening instrument for the early detection of psychosis which has been validated based on transition to psychosis. The BSIP is easy to use by experienced psychiatrists and has a very good interrater-reliability and predictive validity.     

Detection of people at risk of developing a first psychosis: comparison of two recruitment strategies

Rietdijk J, Klaassen R, Ising H, Dragt S, Nieman DH, van de Kamp J, Cuijpers P, Linszen D, van der Gaag M. Detection of people at risk of developing a first psychosis: comparison of two recruitment strategies. Objective: Better recruitment strategies are needed to improve the identification of people at ultra‐high risk of developing psychosis. This study explores the effectiveness of two recruitment strategies: a screening method in a consecutive help‐seeking population entering secondary mental health services for non‐psychotic problems vs. a population referred to the diagnostic center of an early‐psychosis clinic. Method: From February 2008 to February 2010, all general practitioner and self‐referrals (aged 18–35 years) to the secondary mental healthcare service in The Hague and Zoetermeer were screened with the Prodromal Questionnaire; patients who scored above the cutoff of 18 and had a decline in social functioning were assessed using the Comprehensive Assessment of At‐Risk Mental States (CAARMS). All referrals (aged 14–35 years) to the diagnostic center in Amsterdam were also assessed with the CAARMS. Results: The screening detected a three‐fold higher prevalence of at‐risk mental states: these subjects were older and more often female. manova showed significantly higher scores for the screened population on depression, social anxiety, distress with positive symptoms, and a higher rate of transition to psychosis within 12 months. Conclusion: The screening method detects more patients with at‐risk mental states than the referral method. The latter method is biased to young male patients in an earlier prodromal stage and a lower transition rate.     

Norsk Psykiatrisk Forening: Årsmøste

Background: While functioning ability, quality of life (QoL) and depression are widely studied phenomena in schizophrenia, little is known about functioning ability, QoL and depression, especially among adolescents at high risk of developing first-episode psychosis. Aim: To investigate associations between high risk of developing psychosis and functioning ability, depression and QoL among adolescents. Methods: The data was collected by an early intervention team in Espoo, Finland, between 1 January 2007 and 31 May 2008. Subjects at high risk of developing psychosis were compared with subjects not at high risk in terms of functioning ability (GAF), QoL (16D) and depressive symptoms (RBDI) in a cross-sectional setting. The study was conducted with 80 adolescents (mean age 14.7, range 12–18 years). Results: Those at high risk of developing psychosis had significantly lower and poorer scores in functioning ability (53.4 vs. 58.4, P=0.006), had higher and poorer scores in QoL (10.81 vs. 7.05, P=0.002) and higher and poorer scores in depression (8.95 vs. 4.76, P=0.001) than those who did not meet the criteria of being at high risk of developing psychosis. Poorer functioning ability independently explained being at high risk of developing psychosis at a statistically significant level (P=0.021) in a logistic regression analysis after age, gender, depression and QoL were adjusted for. Conclusions: Poor functioning ability seems to be associated with high risk of developing psychosis among adolescents.     

Intervention to improve level of overall functioning and mental condition of adolescents at high risk of developing first-episode psychosis in Finland

Aim: Being at high risk of developing psychosis has been suggested to be a result of a combination of acute life stressors and trait‐like vulnerability to psychosis. Reducing levels of stress could support overall functioning and mental condition in those at risk. Methods: The Jorvi Early Psychosis Recognition and Intervention (JERI) project at Helsinki University Central Hospital, Jorvi Hospital, Finland, is an early intervention team for adolescents at risk of developing first‐episode psychosis. The project is based on the idea of multiprofessional, community, home, family and network‐oriented, stress‐reducing, overall functioning‐supporting, low‐threshold care. The JERI team meets multiprofessionally with adolescents in their natural surroundings, for example, at school or at home, together with their parents, network and community co‐worker, who has originally contacted the JERI team because of unclear mental health problems. Subjects were assessed with the PROD‐prodromal screen to identify those at risk of developing first‐episode psychosis. Results: Statistically significant difference between baseline and follow‐up measures was found in at risk subjects (n = 28) in scales of overall functioning (P = 0.000), depression (P = 0.001), anxiety (P = 0.001), quality of life (QOL) and pre‐psychotic symptoms. Conclusions: JERI‐type intervention may improve level of overall functioning and support mental condition in adolescents at risk of developing first‐episode psychosis, even though further study with larger numbers of subjects, with a control group and with a longer follow‐up time, is needed.     

The Frankfurt Complaint Questionnaire for self‐assessment of basic symptoms in the early detection of psychosis—Factor structure,reliability, and predictive validity

Objectives

Patients with schizophrenia often experience subtle disturbances in several domains of information processing—so‐called basic symptoms (BS). BS are already present before onset of frank psychosis and can be assessed by interviews but also by the self‐administered Frankfurt Complaint Questionnaire (FCQ). We investigated the factor structure, reliability, and predictive validity for transition to psychosis of the FCQ, comparing previously proposed factor solutions containing 1, 2, 4, and 10 factors.

Methods

Confirmatory factor analysis was used in a sample of 117 at‐risk mental state and 92 first‐episode psychosis participants of the Basel FePsy (early detection of psychosis) study.

Results

Although all factor models fitted to the data, the 2‐ or 4‐factor solutions performed best among the models that used at least half of the FCQ items, suggesting the covariance between FCQ items is best explained by 2 to 4 underlying factors. No FCQ‐scale predicted transition to psychosis.

Conclusion

We could confirm a 2‐ to 4‐factor structure of the FCQ in a sample of at‐risk mental state and first‐episode psychosis patients using confirmatory factor analysis. Contrary to interview‐assessed cognitive–perceptive BS, self‐assessed BS do not seem to improve prediction of psychosis. This result reinforces reports of poor correspondence between interview‐ and questionnaire‐assessed BS.

     

Early detection and intervention evaluation for people at high-risk of psychosis-2 (EDIE-2): trial rationale, design and baseline characteristics

Aims: Much research has begun to focus on the identification of people who are at high risk of developing psychosis, and clinical services have been initiated for this population. However, only a small number of studies have reported on the efficacy of interventions for preventing or delaying the onset of psychosis. The results of prior work suggest that cognitive therapy (CT) may be an effective, well‐tolerated treatment. We report on the rationale and design for a large‐scale, multi‐site randomized, controlled trial of CT for people who are assessed to be at high risk of psychosis because of either state or state‐plus‐trait risk factors. Methods: The study employs a single‐blind design in which all participants receive frequent mental‐state monitoring, which will efficiently detect transition to psychosis, and half are randomized to weekly sessions of CT for up to 6 months. Participants will be followed‐up for a minimum of 12 months and to a maximum of 2 years. Results: We report the characteristics of the final sample at baseline (n = 288). Conclusions: Our study aimed to expand the currently limited evidence base for best practice in interventions for individuals at high risk of psychosis.     

The assessment of positive dimension of the psychosis phenotype in college students

The Wisconsin Schizotypy Scales are one of the most used measuring instruments for the assessment of psychometric risk for psychosis. The main goal of the present study was to analyze the reliability of the scores and to provide new sources of validity evidence for the brief version of the Magical Ideation Scale (MIS-B) and the Perceptual Aberration Scale (PAS-B). The final sample was comprised of a total of 1349 university students divided into two subsamples (n1 = 710; M = 19.8 years; n2 = 639; M = 21.2 years). Results show that both measurement instruments have adequate psychometric properties under Classical Test Theory and Item Response Theory. Internal structure analysis of MIS-B and PAS-B, through exploratory and confirmatory factor analysis, yielded an essentially one-dimensional solution. Cronbach’s alpha coefficient for the total score of MIS-B ranged between 0.86 and 0.87, whereas for the PAS-B it ranged between 0.78 and 0.89. A total of 5 items showed a differential functioning for sex. The results indicate that the MIS-B and PAS-B are brief measurement instruments with adequate psychometric properties for the assessment of the positive dimension of the psychosis phenotype and could be used as screening tools in the detection of individuals at risk for psychosis in the general population.     

First self‐perceived signs and symptoms in emerging psychosis compared with depression

Aim: To investigate differences between the early symptoms of schizophrenia and depressive disorders. Methods: Sixty‐one individuals with an at‐risk mental state (ARMS), 17 of whom later made the transition to psychosis, 37 patients with a first episode of psychosis and 16 controls with depressive disorders were interviewed about first self‐perceived signs and symptoms. Results: In ARMS and first episode of psychosis, on average, first self‐perceived signs or symptoms had occurred about 5–6 years before the interview. In ARMS, including transition to psychosis, ‘loss of energy’ and ‘difficulties concentrating’ were the most frequently recalled first signs. There was much overlap for the four most frequently mentioned symptoms in the three groups. As compared with ARMS, controls with depressive disorders significantly more often recalled ‘depression’ and ‘social isolation’ as the very first signs of disease. Conclusions: Clinicians should consider the development of self‐recalled first signs over time carefully when assessing suspected early prodromal stages of schizophrenia and beginning depressive disorder.     

Markers of vulnerability to obsessive-compulsive disorder in an ultra-high risk sample of patients who developed psychosis

Aims: The study aims to identify markers of vulnerability to obsessive–compulsive disorder (OCD) in an ultra‐high risk sample of patients who developed psychosis. Methods: Three hundred and eleven patients at ultra‐high risk for psychosis were examined at baseline and after a mean of 7.4 years follow‐up. Patients who developed psychosis with OCD (PSY + OCD; n = 13) and psychosis without OCD (PSY − OCD; n = 45) were compared in terms of socio‐demographic and clinical features. Results: PSY + OCD patients displayed greater severity of depression before and after conversion to PSY + OCD, and increased rates of depressive disorders before exhibiting PSY + OCD. However, they only displayed greater severity of anxiety and increased rates of non‐OCD anxiety disorders after psychosis. Further, PSY + OCD patients were more likely to report a positive family history for anxiety disorders than PSY − OCD. Conclusion: Although depression and a family history of anxiety disorder may act as vulnerability markers for OCD in psychosis, the resulting anxiety may be a correlate or a consequence of PSY + OCD.     

Factor structure and psychometric properties of the Modified Scale for Suicidal Ideation among suicidal youth

Background: Although suicidal behaviors occur at a high rate in adolescence, relatively few interview‐based measures are available to assess suicidal ideation among youth. Existing interview measures are limited by a paucity of empirical study, a failure to conform to standard suicide nomenclature, or a lengthy administration time. This study presents data on the psychometric properties and factor structure of the brief, layperson‐administered Modified Scale for Suicidal Ideation (MSSI) among suicidal youth. Methods: The MSSI was administered to an inpatient sample of 102 suicidal youth aged 13–17 years. Additional interview and self‐report measures were administered to examine the convergent validity of the MSSI. Results: Consistent with previous findings among suicidal adults, the MSSI displayed good internal consistency and expected patterns of convergent validity. Principal component analysis revealed a bidimensional structure, with factors corresponding to (1) Desire and Ideation and (2) Plans and Preparations. Each factor displayed acceptable internal consistency and expected patterns of convergent validity via associations with hopelessness, depressive symptoms, impulsivity, and a self‐report measure of suicidal behaviors. The Plans and Preparations factor significantly associated with the presence of a current suicide attempt and with greater suicide intent among attempters, whereas the Desire and Ideation factor did not. Conclusions: The MSSI appears to be a reliable and valid instrument to assess suicidal ideation among distressed youth. Clinicians are encouraged to pay particular attention to responses on the Plans and Preparations factor given its stronger association with suicide attempt and more serious suicide intent. Depression and Anxiety, 2009. © 2009 Wiley‐Liss, Inc.     

Potential use of Internet‐based screening for anxiety disorders: a pilot study

Background: The Internet is a widely used resource for obtaining health information. Internet users are able to obtain anonymous information on diagnoses and treatment, seek confirmatory information, and are able to self‐diagnose. We posted a self‐report diagnostic screening questionnaire for DSM‐IV anxiety and mood disorders (MACSCREEN) on our clinic website. Method: Three hundred and two individuals completed the MACSREEN. For those who qualified for a DSM‐IV disorder, self‐report symptom severity measures were completed for the specified disorder: Quick Inventory of Depressive Symptomatology, self‐report, Social Phobia Inventory, GAD‐7, Davidson Trauma Scale, Panic and Agoraphobia Scale, and Yale/Brown Obsessive Compulsive Scale, self‐report. Cutoff scores for each self‐report measure were used to evaluate clinically significant symptom severity. Respondents were also asked to complete a series of questions regarding their use of the Internet for health information. Results: The mean age of the MACSCREEN sample was 35.2 years (±13.9), where the majority (67.2%) were female. The most frequently diagnosed conditions were social phobia (51.0%), major depressive disorder (32.4%), and generalized anxiety disorder (25.5%). Sixty‐five percent of the sample met criteria for at least one disorder. Most respondents reported completing the MACSCREEN, as they were concerned they had an anxiety problem (62.3%). The majority of respondents reported seeking health information concerning specific symptoms they were experiencing (54.6%) and were planning to use the information to seek further assessment (60.3%). Conclusion: Individuals with clinically significant disorder appear to be using the Internet to self‐diagnose and seek additional information. Depression and Anxiety, 2010. © 2010 Wiley‐Liss, Inc.     

Domain‐specific cognitive impairment in non‐demented Parkinson's disease psychosis

Introduction

In Parkinson's disease (PD), psychosis is associated with cognitive impairment that may be more profound in particular cognitive domains. Our goal was to determine whether psychosis in non‐demented PD participants is associated with domain‐specific cognitive impairment on the Mini‐Mental State Exam (MMSE).

Methods

The Morris K. Udall Parkinson's Disease Research Center of Excellence Longitudinal Study at Johns Hopkins is a prospective study that was initiated in 1998. Clinical assessments are conducted at two‐year intervals at the Johns Hopkins Hospital. We analyzed data from 137 enrolled participants with idiopathic PD. Psychosis diagnoses were established by psychiatrist interview per DSM‐IV criteria. An incident dementia diagnosis resulted in exclusion from analysis for that evaluation and any future evaluations in that participant. We used logistic regression with generalized estimated equations (GEE) to model the time‐varying relationship between MMSE subscale scores and psychosis, adjusting for potential confounding variables identified through univariable analysis.

Results

Thirty‐one unique psychosis cases were recorded among non‐demented participants. Fifty total evaluations with psychosis present were analyzed. In multivariable regressions, psychosis was associated with lower scores on the orientation (relative odds ratio, rOR: 0.73; 95% CI: 0.58–0.93; p = 0.011), language (rOR: 0.64; 95% CI: 0.48–0.86; p = 0.003), and intersecting pentagon (rOR: 0.43; 95% CI: 0.20–0.92 p = 0.030) subscales of the MMSE.

Conclusions

In PD, executive dysfunction, disorientation, and impaired language comprehension may be associated with psychosis. Our findings suggest that the corresponding MMSE subscales may be useful in identifying participants with a higher likelihood of developing psychosis. Copyright © 2017 John Wiley & Sons, Ltd.