Psychotic-Like Experiences in Nonpsychotic Help-Seekers: Associations With Distress, Depression, and Disability

Abstract

Psychotic-like experiences (PLEs) increase the risk of schizophrenia and other psychotic disorders yet are common in the community. Some PLEs, such as those associated with depression, distress, and poor functioning, may confer increased risk. The aim of this study is to determine the prevalence of PLEs in a nonpsychotic clinical sample and to investigate whether any subtypes of PLEs are associated with the above factors. Consecutive referrals to a youth psychiatric service (N = 140) were assessed to measure PLEs, depression, and functioning. PLE data were factor analyzed, and the associations of psychotic subtypes and distress, depression, and disability were analyzed. Three subtypes of PLEs were identified: Bizarre Experiences, Persecutory Ideas, and Magical Thinking. Bizarre Experiences and Persecutory Ideas were associated with distress, depression, and poor functioning. Magical Thinking was not, unless accompanied by distress. Bizarre Experiences and Persecutory Ideas may be more malignant forms of psychotic symptoms, as they are associated with current disability, and may confer increased risk of development of full-blown psychotic disorder.

Introduction

Previous studies have shown that young people with psychotic-like experiences (PLEs) who are referred to a clinical service are at “ultra high risk” (UHR) of imminent development of a psychotic disorder such as schizophrenia. For example, we found that 35–40% of UHR individuals who presented to the Personal Assessment and Crisis Evaluation (PACE) Clinic with PLEs had onset of a psychotic disorder within 12 months.^1,2 Specific PACE UHR criteria are detailed in previous publications^1,3 but consist largely of PLEs associated with help-seeking. Other clinical services around the world using similar intake criteria have also found high rates of onset of psychotic disorder in their cohorts.^4,5 Thus, it seems possible to detect emerging psychotic disorder in people with PLEs. These individuals seeking help with PLEs, who subsequently develop full-blown psychotic disorder, could be conceptualized as being in the prodromal phase of the disorder.

Another finding from the PACE Clinic was that the process of change from prodrome to full-threshold psychotic disorder may be associated with brain changes. A subsample of UHR individuals from the PACE Clinic with PLEs underwent a magnetic resonance imaging (MRI) brain scan before and after onset of psychotic disorder. A reduction in gray matter in right temporal areas over time was found. No brain changes were found in those who had PLEs at baseline but who did not develop a psychotic disorder.⁶ Thus, it appears that at some point in transition from prodromal state to psychotic disorder, alterations in brain structure (and presumably function) occur.

However, despite the seemingly malignant course of PLEs found in these clinical populations, community studies suggest that PLEs are not uncommon and may not be associated with help-seeking or distress. For example, the Epidemiological Catchments Area study found a lifetime prevalence of hallucinations of 10% in men and 15% in women.⁷ The Baltimore Mental Health survey found that 10% of subjects reported paranoid symptoms,⁸ and a Dutch population study found that 17.5% of adult subjects had a lifetime prevalence of psychotic experiences.⁹ A birth cohort study from New Zealand found that over 20% of subjects assessed at age 26 had had at least one psychotic-like experience.¹⁰

Aside from the possibility that some self-reported PLEs in population studies are the result of misinterpretation, it seems that not all people with PLEs will go on to develop a psychotic disorder. We need to know whether some PLEs are more likely to progress to full-blown psychotic disorder than others and whether there are any associated features that make progression more likely. Existing literature indicates that appraisal and coping style in relation to PLEs may be a factor in determining outcome,^11,12 and PLEs associated with distress may influence help-seeking.^1,13,14 Depression and poor functioning have also been found to increase the risk of transition to full-blown psychotic disorder in UHR individuals at the PACE Clinic.¹ The role of these factors in influencing the transition from psychotic-like experience to full-blown psychotic disorder needs to be explored.

Thus, further investigation is required into different types of PLEs and into their association with distress, depression, and psychosocial functioning. To further investigate these issues, we studied psychotic phenomena in a nonpsychotic population.

Method

Aims

The aims of this study are as follows:

1. To assess the characteristics of psychotic-like experiences in a clinical population of nonpsychotic young people and to determine if there are different subtypes of PLEs in this population;

2. To assess the prevalence of PLEs in this population; and

3. To investigate the association between PLEs and level of distress, depression, and psychosocial functioning.

Hypotheses

We hypothesize:

1. That there would be different subtypes of PLE in the study population;

2. That, based on previous community⁹ and primary care¹⁵ research, PLEs would be common; and

3. That a subgroup of young people with PLEs would have associated distress, depression, and low functioning but that a subgroup would not.

Setting

ORYGEN Youth Health (OYH) is a public mental health program for young people between 15 and 24 years old who live in the north and west of metropolitan Melbourne, Australia. The catchment area covers approximately 900,000 people, about 200,000 of whom are aged between 15 and 24. The clinical service has 3 components: EPPIC (the Early Psychosis Prevention and Intervention Centre, a service for people with first-episode psychotic disorder), PACE (the Personal Assessment and Crisis Evaluation clinic, a service for individuals thought to be at imminent risk of psychotic disorder—that is, prodromal for psychosis), and Youthscope (a service for nonpsychotic individuals). Referrals to OYH are taken from a range of sources, including general practitioners (GPs) and other primary care services, school and university counseling services, drug and alcohol services, the justice system (ie, prisons and youth detention centers), and youth accommodation centers, as well as from families/caregivers and young people themselves. The procedure for entry into OYH is as follows: first, a brief telephone “triage” interview is undertaken with the referrer and the young person. Once basic criteria are met (ie, age and location), the young person is referred to 1 of the 3 clinical teams (EPPIC, PACE, or Youthscope) for a face-to-face interview to determine eligibility. All young people who are recognized as meeting the criteria for EPPIC and PACE at this interview are accepted. However, because of the high prevalence of nonpsychotic disorders such as depression, there are many more young people referred to Youthscope than can be accepted. Thus, criteria have been established for acceptance into Youthscope. These are based largely on presence of a diagnosis of a nonpsychotic mental disorder, such as depression, anxiety, and/or severe personality disorder. Degree of risk of suicide or self-harm, disability or functional impairment, and previous history of unsuccessful treatment within primary care services are also taken into account by the Youthscope clinician. Many young people not accepted into Youthscope also have high rates of depressive symptoms and poor functioning. Those not accepted also include young people with a primary diagnosis of an uncomplicated substance use disorder or a primary diagnosis of oppositional defiant disorder. The study group for this project was taken from those who were referred to, but not necessarily accepted into, Youthscope.

Participants

During the 6-month period of April to October 2003, young people consecutively referred to Youthscope were invited to participate in the study. Inclusion criteria were age between 15 and 24 years and residence in the western metropolitan region of Melbourne, Australia. Exclusion criteria included known organic cause for presentation, known intellectual disability (defined as IQ less than 70), and an inability to speak English. Research interviews took place independently of clinical decision making. Both referrals who were accepted into the Youthscope service and those who were not were approached to participate in the research. Thus, the sample consists of help-seeking people age 15–24 who sought help for a psychiatric problem from a specialized youth mental health service and were considered to be neither psychotic nor at immediate risk of developing psychotic disorder (ie, not considered prodromal for psychosis). All participants gave written informed consent, and the research was approved by the local ethics committee.

Instruments

The CAPE (Community Assessment of Psychiatric Experiences)¹⁶ positive symptoms scale was used to assess PLEs. This self-report scale measures the lifetime prevalence of PLEs on both a frequency scale (0 = never to 4 = nearly always) and a distress scale (1 = not distressed to 4 = very distressed).

Functioning was measured with either the Global Assessment of Functioning (GAF)¹⁷ (used for subjects 18–24 years old) or the Childhood Global Assessment Scale (CGAS) (used for subjects 15–17 years old), both of which give a single-item summary score of overall psychosocial functioning. These instruments both use scales that range from 0 to 100 but have different anchor points, with the CGAS including more examples of school situations than the GAF. However, the ratings are equivalent. A score of 70 or below indicates some difficulty with important areas of functioning; a score of 50 or below indicates serious impairment in functioning. In addition, the Revised Multidimensional Assessment of Functioning Scale (RMAFS) was used to gain a clearer picture of specific areas of disability. This is a 23-item self-report scale, which generates a Total Functioning score and 3 subscale scores: General Functioning (10 items, eg, “I feel like I'm working towards a goal”), Peer Relationships (6 items, eg, “I spend quite a lot of time with my friends”), and Family Functioning (7 items, eg, “I get on well with my parents”). RMAFS is scored on a 0 to 4 scale (0 = not applicable, 1 = not at all/rarely to 4 = always/almost always). Higher scores indicate better functioning. A manuscript describing the psychometric properties of this scale is in preparation.

Level of self-reported depressive symptomatology was assessed with the Mood and Anxiety Questionnaire: Anhedonic Depression (MASQ:AD) scale.¹⁸ MASQ:AD items comprise both loss of interest (eg, “felt withdrawn from others”) and high positive affect (eg, “felt hopeful about the future”). MASQ:AD scores range from 1 (not at all) to 5 (frequently).

The Structured Clinical Interview for DSM-IV (SCID-IV)¹⁹ was used to assess Axis I psychiatric diagnoses. Interrater assessments were conducted in approximately 15% of interviews to ensure agreement across raters. Kappa values for mood diagnoses were excellent (k = .89).

Data Analysis

Analyses were conducted using the Statistical Package for Social Sciences (SPSS), Version 11.5. Data were initially screened for missing values, for the assumptions of normality, linearity, and homogeneity, and for outliers. Eleven participants had more than 25% of the data missing and were subsequently removed from further analyses. This left 140 participants with valid data.

The correlation matrix among items was examined. There were many correlations greater than 0.3, indicating that factor analysis was a meaningful technique.

A principal components analysis was conducted on the CAPE frequency scores to determine the number of factors. The factor structure within this identified number of factors was assessed using a principal axis factoring technique, with direct oblimin rotation. Two tests were used to determine whether the data solution was factorable. First, the Kaiser-Meyer-Olkin measure of sampling adequacy tests whether the partial correlations among variables are small. Values of 0.60 or greater are required for good factor analyses.²⁰ The second test was Bartlett's test of sphericity, which assesses whether the factor model is appropriate. This oblique rotation was chosen as the CAPE factors are theoretically interrelated.

Once the factor structure was identified, correlations were conducted to assess the relationship between CAPE factors and measures of functioning and symptomatology. Chi-square and t-tests were run to compare factor scores with variables including age, sex, and functioning. Simple and multivariable linear regressions were undertaken, using both depression and functioning as dependent variables and PLEs as independent variables, to examine the association between these factors.

Results

Baseline Characteristics of the Sample

Of the 209 young people referred to Youthscope during the 6-month study period, 150 consented to take part (72% of those eligible). There were 63 males (42%) and 87 females (58%). Eighty-eight subjects (58.7%) were accepted for treatment at Youthscope. Participants in the research were significantly more likely to have been accepted into the service than refusers (58.3% of those accepted into the service took part in the research versus 41.7% of those not accepted into the service; p = .005). Study participants were significantly younger than refusers (mean age of participants = 17.67 versus mean age of refusers = 18.66; p = .020). There was no gender difference between participants and refusers.

Table 1 presents baseline data for the sample. The sample had high levels of depression as the presenting complaint, with a mean score of 77.5 on the MASQ:AD scale, compared with a mean of 54–56 in community samples.^21,22 There was a high prevalence of depressive disorders. Nearly half of the sample (n = 69, 46%) had a current mood disorder. Of those with mood disorders, the majority had Major Depressive Disorder (n = 56, or 81.2% of those with a current mood disorder); 8 (11.6%) had Dysthymia; and 5 (7.2%) had Mood Disorder NOS (not otherwise specified). Prevalence of other current Axis I disorders were as follows: 63 subjects (42%) had a current Anxiety Disorder, and 33 (22%) had a Substance Use Disorder, 11 (7%) had a current Eating Disorder, and 16 (11%) had a Disruptive Behavioral Disorder (numbers add up to over 100% as some subjects had more than 1 disorder). No subjects had Bipolar Disorder or Pervasive Developmental Disorder.

Table 1.

Baseline Characteristics of the Sample (N = 140)

	Median	Mean (SD)	Range
CGAS/GAF	51	52.90 (14.39)	11–88
RMAFS—Total	57	56.90 (14.18)	28–83
RMAFS—Family	19	18.74 (5.68)	5–28
RMAFS—Peer	16	16.06 (5.23)	0–24
RMAFS—General	22	22.10 (7.70)	9–39
MASQ:AD	77.50	75.81 (17.43)	36–110

Note: CGAS = Childhood Global Assessment Scale; GAF = Global Assessment of Functioning; MASQ:AD = Mood and Anxiety Questionnaire: Anhedonic Depression; RMAFS = Revised Multidimensional Assessment of Functioning Scale.

The sample had moderate to serious difficulty in functioning across a range of domains, as indicated by a mean GAF/CGAS score of 52.9 and mean RMAFS scores for Total, Family, Peer, and General Functioning of 56.90, 18.74, 16.06, and 22.10, respectively (see Table 1). This compares with mean RMAFS scores from a community sample of 14 to 15 year olds (n = 913): Total 72.22 (SD = 10.34), Family 22.51 (SD = 4.33), Peer 18.96 (SD = 3.59) and General 30.76 (SD = 5.26) (unpublished raw data).

PLEs in the Sample: Determination of Number of Factors

Both 2- and 3-factor solutions for the CAPE data were inspected, and it was determined that the 3-factor structure best represented the data and accounted for 52.44% of the explained variance. There were few variables with high cross-loadings in the 3-factor model, and all items loaded significantly (r > .30) on 1 of the factors. The identified structure of the 3-factor solution is presented in Table 2. Factor 1 was composed of 8 items that generally related to Bizarre Experiences. Factor 2 contained 5 items that related to Persecutory Ideas. Factor 3 consisted of 5 items that related to Magical Thinking. Factor scores were generated for each factor. There was no significant difference between males and females for any factor (Factor 1, t(138) = −.85, p > .05; Factor 2, t(138) = .77, p > .05); Factor 3, t(138) = −.17, p > .05). Correlations between factors were Bizarre:Persecutory −0.48, Bizarre:Magical 0.39, and Persecutory:Magical −0.40. Participants were divided according to age with those less than 18 years old (n = 82) compared with those aged 18 years or older (n = 58). There was no significant difference between the groups for any factor (Factor 1, t(138) = .20, p > .05; Factor 2, t(138) = .83, p > .05; Factor 3, t(138) = −1.58, p > .05). These results indicate the factor structure is stable across both age and sex in the present sample.

Table 2.

Factor Loadings for CAPE (Community Assessment of Psychiatric Experiences) Items With a 3-Factor Solution

Item No.	Item	F1	F2	F3
FACTOR 1—Bizarre Experiences
13	Do you ever feel as if the thoughts in your head are not your own?	.71
16	Do you ever feel as if you are under the control of some force or power other than yourself?	.70
17	Do you ever hear voices when you are alone?	.69
14	Have your thoughts ever been so vivid that you were worried other people would hear them?	.66
12	Do you ever feel as if the thoughts in your head are being taken away from you?	.59
18	Do you ever hear voices talking to each other when you are alone?	.55
15	Do you ever hear your own thoughts being echoed back to you?	.54
2	Do you ever feel as if things in magazines or on TV were written especially for you?	.39	.28
FACTOR 2—Persecutory Ideas
4	Do you ever feel as if you are being persecuted in some way?		.86
5	Do you ever feel as if there is a conspiracy against you?		.65
1	Do you ever feel as if people seem to drop hints about you or say things with a double meaning?		.59
3	Do you ever feel as if some people are not what they seem to be?	.31	.46	.29
11	Do you ever feel that people look at you oddly because of your appearance?	.39	.44
FACTOR 3—Magical Thinking
8	Do you believe in the power of witchcraft, voodoo or the occult?			.88
10	Do you think that people can communicate telepathically?			.53
7	Do you ever feel that you are a very special or unusual person?		.38	.42
6	Do you ever feel as if you are destined to be someone very important?			.32
9	Do you ever feel as if electrical devices such as computers can influence the way you think?			.29

Note: Only loadings > .25 are displayed.

Prevalence of PLEs

To determine the prevalence of psychotic-like experiences, the participants' responses were dichotomized. Responses were recoded to 0 (never) and 1 (at least sometimes). Almost all the sample (98.6%) (n = 138) reported experiencing at least 1 of the 18 PLEs at least “sometimes” in their lifetime. Persecutory Ideas was the most prevalent of the factors, with 97.9% of participants endorsing at least 1 of the 5 items at least sometimes, compared with 73.6% reporting Bizarre Experiences and 64.3% reporting Magical Thinking. Within each factor there was variation in prevalence. For example, within Factor 1 (Bizarre Experiences), lifetime prevalence ranged from 12.9% (for experience of hearing voices conversing with each other) to 73.6% (for feeling as if things in magazines or on television were written especially for the person). Within Factor 2 (Persecutory Ideas) prevalence ranged from 28.6% (for feeling there is a conspiracy against the subject) to 62.1% (for feeling as if some people are not who they seem to be). Within Factor 3 (Magical Thinking) the range was from 83.6% (feeling of being a very special or unusual person) to 47.1% (feeling that electrical devices can influence the way the subject thinks).

Distress Associated with PLEs

To determine whether differing levels of distress were associated with each CAPE subscale, correlations between frequency and distress were determined separately for each subscale. The highest correlation between distress and frequency was for Bizarre Experiences (0.89, p < .01), followed by Persecutory Ideas (0.83, p < .01) and then Magical Thinking (0.58, p < .01) . To investigate whether these correlations were significantly different from each other, correlations were first transformed using Fisher's Z transformation. These Z scores were then compared using the formula described in Hinkle, Wiersma, and Jurs²³ for the difference between 2 independent correlations. The correlation between the distress and frequency for Bizarre Experiences was significantly higher than the correlations between distress and frequency for Persecutory Ideas (z = 2.0, p = .045) and Magical Thinking (z = 6.26, p < .001). The correlation between distress and frequency for Persecutory Ideas was also significantly greater than the correlation for Magical Thinking (z = 4.26, p < .001).

Relationship Between PLEs and Depression

Frequency of overall PLEs, Bizarre Experiences, and Persecutory Ideas were significantly correlated with level of depression as measured by the MASQ:AD scale, but Magical Thinking was not (Table 3). Using a standard linear multiple regression, Persecutory Ideas were significantly associated with level of depression, but Bizarre Experiences were not. Magical Thinking was significantly negatively associated with depression. That is, as level of Magical Thinking increased, the level of depression was reduced. PLEs accounted for 28% of the variance of depression level (r² = 0.28, F_3,132 = 17.30, p < .000) (Table 4). Age and gender were not associated with level of depression.

Table 3.

Correlations Between CAPE Scores and Depressive Symptomatology

	CAPE Total	Bizarre Experiences	Persecutory Ideas	Magical Thinking
Depression MASQ:AD	.41*	.36*	.51*	.14

Note: CAPE = Community Assessment of Psychiatric Experiences; MASQ:AD = Mood and Anxiety Questionnaire: Anhedonic Depression.

^*p < .01.

Table 4.

Multiple Regression Showing Association Between Psychotic-Like Experiences (PLEs) (Independent Variable) and Depression Level (Dependent Variable)

	B	t	Significance (p)	CI	sr2
Bizarre Experiences	.45	1.05	.294	−.39, 1.28	.01
Persecutory Ideas	2.56	5.39	.000	1.62, 3.50	.16
Magical Thinking	−1.06	−2.10	.037	−2.06, −.06	.02

Note: Depression measured by Mood and Anxiety Questionnaire: Anhedonia Depression (MASQ:AD) scale. B = beta coefficient, the change in depression level per unit change in PLEs.

To further explore the relationship between PLE and depression, the data was split into those who had a current DSM-IV mood disorder (n = 69) versus those without a mood disorder (n = 81). Of those without a current mood disorder, 25 (31%) had a current Anxiety Disorder, 14 (17%) had a Substance Use Disorder, and 10 (12.5%) had a Disruptive Behavioral Disorder. Individuals with an Anxiety Disorder did not have significantly more PLEs than those without an Anxiety Disorder (with Anxiety Disorder: mean CAPE total score = 32.13, SD = 8.80; without Anxiety Disorder: mean CAPE total score = 29.46, SD = 8.41; p = .072). There was no significant difference between those with or without a current Substance Use Disorder (dependence or abuse) (respectively, mean total CAPE = 30.07, SD = 8.89 versus mean of 30.71, SD = 8.65; p = .717). Table 5 presents the mean PLE scores for those with and without mood disorders. Participants with a current mood disorder had significantly higher levels of Bizarre Experiences and Persecutory Ideation, as well as total CAPE score, than those without a Mood Disorder. However, there was no significant difference between groups for Magical Thinking.

Table 5.

Comparison of Mean Psychotic-Like Experience (PLE) Scores for Those With and Without Mood Disorders

	With Mood Disorder	Without Mood Disorder
	Mean (SD)	Mean (SD)	p
CAPE Total	33.19 (8.84)	28.35 (7.88)	.001
Bizarre Experiences	12.24 (4.17)	10.46 (3.38)	.006
Persecutory Ideas	11.98 (3.83)	9.58 (3.08)	.000
Magical Thinking	8.97 (2.84)	8.31 (3.08)	.195

Note: CAPE = Community Assessment of Psychiatric Experiences.

In addition to investigating levels of PLEs, we also sought to determine whether subjects with mood disorders experienced greater numbers of PLEs compared with those without a mood disorder. To do this, PLEs were recoded to “never” versus “at least sometimes.” The total scale and the subscales data were then summed to determine the number of items that participants had endorsed. Participants with mood disorders reported experiencing more items for Total CAPE (M = 9.34 vs 7.24 items, t(138) = −2.95, p = .0047), Persecutory Ideas (M = 3.84 vs 3.08, t(138) = −3.10, p = .002), and Bizarre Experiences (M = 3.00 vs 1.96, t(138) = −2.77, p = .006). There was no significant difference for Magical Ideation (M = 2.50 vs 2.19, t(138) = −1.18, p = .241).

Relationship Between PLEs and Functioning

Frequency of total PLEs (Total CAPE frequency score) and frequency of Bizarre Experiences and Persecutory Ideas were significantly correlated with lower overall functioning, but frequency of Magical Thinking was not. High score on Persecutory Ideas was significantly correlated with poor peer and family functioning. All three subscales were significantly associated with poor global functioning, if the symptoms were associated with distress (using distress scores from the CAPE) (Table 6).

Table 6.

Correlations Between Psychotic-Like Experiences (PLEs) and Functioning (CGAS/GAF Score)

	R2	Coefficient	p
Bizarre Experiences—Frequency	.15	−.27	.003
Persecutory Ideas—Frequency	.17	−.31	.001
Magical Thinking—Frequency	.09	−.11	.127
Bizarre Experiences—Distress	.15	−.27	.005
Persecutory Ideas—Distress	.16	−.30	.001
Magical Thinking—Distress	.11	−.20	.040

Note: CGAS = Children's Global Assessment Scale; GAF = Global Assessment of Functioning.

In a multiple regression analysis, increasing age (p = .001) and level of depression (p < .001) were also associated with poor functioning. After adjusting for age, Persecutory Ideas remained significantly associated with poor functioning (p = .012). However, once level of depression was controlled for, PLEs were no longer associated with poor functioning (Table 7).

Table 7.

Multiple Regression Examining Association Between Age, Sex, Psychotic-Like Experiences (PLEs), and Depression (Independent Variables) on Functioning (CGAS/GAF Score) (Dependent Variables)

	B	t	Significance	CI	sr2
Age	−1.48	−3.26	.001	−2.38, −.58	.07
Sex	−.71	−.29	.769	−5.49, 4.07	.001
MASQ:AD	−.35	−5.58	.000	−.47, −.22	.18
Bizarre Experiences	−.46	−1.30	.195	−1.17, .24	.01
Persecutory Ideas	−.26	−.58	.564	.621, −.29	.001
Magical Thinking	−.19	.42	.674	1.06, −.12	.001

Note: CGAS = Children's Global Assessment Scale; GAF = Global Assessment of Functioning; MASQ:AD = Mood and Anxiety Questionnaire: Anhedonic Depression scale. B = beta coefficient, the change in CGAS/GAF score per unit change in PLE.

Discussion

A paradox in psychosis research is the finding that psychotic-like experiences are common in the community,^7–9 yet in clinical settings established for management of high-risk individuals, PLEs are predictive of onset of frank psychotic disorder within a brief time frame.^2–5 An important area of research is to investigate whether there are any types of PLEs that are more likely than others to be associated with poor outcome and whether there are any associated features that make transition from PLEs to threshold psychotic disorder and need for care more likely. Distress, depression, and disability might be important associations to assess.

Prevalence of PLEs

The study group consisted of help-seeking young people with nonpsychotic disorders. This sample should be contrasted with the PACE sample, which consists of young people identified as being at imminent risk of, or prodromal for, psychotic disorder. Unlike the PACE sample, subjects in the current study group were not suspected of being “prodromal” by referrers or triage clinicians at OYH. These subjects did not present primarily because of their PLEs, and PLEs were found only by direct inquiry by the CAPE instrument during the current study. It is likely that some subjects would actually meet PACE criteria; however, as this study took place independently of clinical assessment, they were not referred to PACE. Although not meeting EPPIC or PACE criteria, the subjects were distressed, help-seeking, and symptomatic, with high levels of depression and other psychiatric symptoms and generally low functioning. Thus, they are clearly not a general population sample.

This apparently nonpsychotic group reported a high lifetime prevalence of PLEs. Persecutory Ideas were most common (97.9%), followed by Bizarre Experiences (73.6%) and Magical Thinking (64.3%). The finding of high prevalence of self-reported psychotic symptoms is consistent with previous research in both community samples^7–9 and primary care samples.^15,24 However, the prevalence of PLEs in this sample is higher than in these previously studied cohorts. One possible reason for this is that the current sample is younger than the community and primary care samples previously investigated. It has been shown that young people report PLEs more frequently than older people.²⁵ A second explanation is that the current sample is one of help-seekers in a mental health service with high rates of psychiatric disorder, particularly depression. Psychotic symptoms and PLEs are known to occur in association with major depression and have been shown to be a risk factor for the development of major depression in a primary care study.²⁶

Characteristics of PLEs

Bizarre Experiences and Persecutory Ideas were found to be associated with increased distress and level of depression, presence of mood disorder, and poor functioning. Magical Thinking was not found to be associated with any of these variables, unless accompanied by distress. This finding is not unexpected if the content of the PLEs is taken into account. Magical Thinking includes experiences such as belief in the occult and the power of voodoo and belief that one is a special person. These so-called PLEs can be normal phenomena for adolescents and young adults. Although appraisal of PLEs was not specifically studied, it seems likely that such experiences are not viewed negatively or as threatening. Their low distress ratings support this impression. Their lack of association with any maladaptive feature such as depression or poor functioning also suggests that the Magical Thinking experiences may be benign. However, if these beliefs were experienced as distressing, they were associated with disadvantage (depression or poor functioning).

Because Bizarre Experiences and Persecutory Ideas were associated with distress, depression, and poor functioning, they may also be associated with increased risk of development of frank psychotic disorder. This is supported by the finding that a person's reaction to and coping with unusual experiences are important in determining outcome, particularly whether PLEs evolve to actual psychiatric disorder.^27,28 It is likely, however, that the degree of risk for psychosis is lower than that of the PACE sample, as the current subjects do not present primarily because of their PLEs; the PLEs are either not being detected by clinicians or are not deemed significant enough to warrant referral to the specialized PACE Clinic. This hypothesis can be tested by follow-up of the sample. It is not possible to determine the exact nature of the association between depression and Bizarre Experiences and Persecutory Ideas from this current cross-sectional study. It may be that these PLEs cause distress and depression, or vice versa. Alternatively, they may be part of the depression syndrome for some people. Another theory is that PLEs may be nonspecific risk factors for mental disturbance and confer risk for both schizophrenia and mood disorders.²⁹ Follow-up of the sample can examine whether PLEs diminish in response to treatment for depression.

The association between PLEs and poor functioning may be entirely explained by the level of depression. In other words, depression is associated with both poor functioning and PLEs, and any apparent association between PLEs and poor functioning is spurious. However, high levels of depression were found in this sample of young help-seekers, and depression explained over 19% of the variance in functioning. Thus, any additional impact on functioning by PLEs might be hard to identify in this sample size.

The limitations of the study should be acknowledged. The sample size of 140 is small. Additionally, the data were collected from a specialized youth mental health service and will not be generalizable to the community population. The sample consists of help-seeking young people with apparently nonpsychotic disorders referred to a youth mental health service. Apart from a few private psychiatrists (who are generally not accessible to young people due to expense, long waiting periods, and other barriers³⁰), there are few services within this relatively indigent area for young people with psychiatric problems who are considered to need specialized help. Thus, those referred to Youthscope are likely to representative of those referred for psychiatric care from GPs and other sources. Data were collected on those who were referred to the service, not just those who were accepted into the service, hence eliminating one filter that would otherwise make the subjects less representative of their population. However, a study limitation is that participants in this study were more likely to be accepted into the clinical service for treatment than nonparticipants. Thus, they may not be representative of the overall population of young people referred to specialized services. Because those accepted into the service were more likely to be studied than those not accepted into the service, there may be a bias toward studying sicker and more disabled individuals, and the prevalence of PLEs may be overestimated. A further limitation is that PLEs were assessed by a self-report measure developed for community samples rather than by an interview. Some subjects may have misinterpreted questions and endorsed PLEs incorrectly.

Despite these limitations, some tentative clinical implications can be drawn. First, in help-seeking young people PLEs are likely to be common and should be inquired about. Second, the level of distress associated with PLEs should be noted, and individuals with distressing PLEs should be offered support and follow-up, as they may be at risk of psychotic disorder. Finally, persecutory ideas and bizarre experiences in particular should be noted and monitored, as they are likely to be associated with distress, depression, and poor functioning. Thus, they may be the focus of clinical attention in their own right. They may also confer an increased risk of development of a psychotic disorder over a brief time period, such as 12 months.

In summary, it appears that different subtypes of PLEs can be identified in this clinical but nonpsychotic sample. Bizarre Experiences and Persecutory Ideation seem to be maladaptive, as they are associated with depression, distress, and disability and may therefore confer heightened risk of the development of psychotic disorder. However, the direction of causality and the magnitude of the risk are not clear. Further longitudinal research is needed and is currently underway.