Introduction
Hallucinations and delusions, the classic symptoms of psychosis, are more prevalent in the general population than previously considered (Freeman et al. Reference Freeman, Garety, Kuipers, Fowler and Bebbington2002; Scott et al. Reference Scott, Martin, Bor, Sawyer, Clark and McGrath2009; van Os et al. Reference van Os, Linscott, Myin-Germeys, Delespaul and Krabbendam2009; Bartels-Velthuis et al. Reference Bartels-Velthuis, van de Willige, Jenner, van Os and Wiersma2011; Mackie et al. Reference Mackie, Castellanos-Ryan and Conrod2011; Laurens et al. Reference Laurens, Hobbs, Sunderland, Green and Mould2012; van Os & Murray, Reference van Os and Murray2013). These symptoms may be frankly psychotic in nature but more commonly occur in an attenuated form – that is, experiences that are hallucinatory or delusional in nature but where reality testing remains intact. They are commonly referred to as ‘psychotic experiences’ or ‘psychotic-like experiences’. A meta-analysis of community-based studies of psychotic experiences showed a median population prevalence of 17% in children aged 9 to 12 years and 7.5% in adolescents aged 13 to 18 years (Kelleher et al. Reference Kelleher, Connor, Clarke, Devlin, Harley and Cannon2012a ). Initial research focused largely on the finding that psychotic experiences in youth predicted an increased risk for psychotic disorder in adulthood (Poulton et al. Reference Poulton, Caspi, Moffitt, Cannon, Murray and Harrington2000), a finding that has been replicated a number of times (Welham et al. Reference Welham, Scott, Williams, Najman, Bor, O'Callaghan and McGrath2009; Kaymaz et al. Reference Kaymaz, Drukker, Lieb, Wittchen, Werbeloff, Weiser, Lataster and van Os2012). More recent community research, however, has found that psychotic experiences are associated with a broad spectrum of (non-psychotic) symptoms of psychopathology (Scott et al. Reference Scott, Nurcombe, Sheridan and McFarland2007b ; Yung et al. Reference Yung, Buckby, Cosgrave, Killackey, Baker, Cotton and McGorry2007; Varghese et al. Reference Varghese, Scott, Welham, Bor, Najman, O'Callaghan, Williams and McGrath2011; Kelleher et al. Reference Kelleher, Keeley, Corcoran, Lynch, Fitzpatrick, Devlin, Molloy, Roddy, Clarke, Harley, Arseneault, Wasserman, Carli, Sarchiapone, Hoven, Wasserman and Cannon2012b ; Werbeloff et al. Reference Werbeloff, Drukker, Dohrenwend, Levav, Yoffe, van Os, Davidson and Weiser2012; Wigman et al. Reference Wigman, van Nierop, Vollebergh, Lieb, Beesdo-Baum, Wittchen and van Os2012; Fisher et al. Reference Fisher, Caspi, Poulton, Meier, Houts, Harrington, Arseneault and Moffitt2013), and with suicidal thoughts and behaviour (Saha et al. Reference Saha, Scott, Johnston, Slade, Varghese, Carter and McGrath2011; Kelleher et al. Reference Kelleher, Lynch, Harley, Molloy, Roddy, Fitzpatrick and Cannon2012c , Reference Kelleher, Corcoran, Keeley, Wigman, Devlin, Ramsay, Wasserman, Carli, Sarchiapone, Hoven, Wasserman and Cannon2013a ). In a recent multicentre community-based study, we found that the majority of young people who reported psychotic experiences met criteria for at least one (non-psychotic) Axis I psychiatric disorder (Kelleher et al. Reference Kelleher, Keeley, Corcoran, Lynch, Fitzpatrick, Devlin, Molloy, Roddy, Clarke, Harley, Arseneault, Wasserman, Carli, Sarchiapone, Hoven, Wasserman and Cannon2012b ). Psychotic experiences were associated with a range of disorders but, interestingly, were particularly predictive of multimorbidity, that is, the presence of more than one psychiatric diagnosis. Furthermore, young people with Axis I disorders who reported psychotic experiences were found to have a greatly increased odds of suicidal behaviour even compared with individuals with the same diagnosis who did not report psychotic experiences (Kelleher et al. Reference Kelleher, Lynch, Harley, Molloy, Roddy, Fitzpatrick and Cannon2012c ). In fact, in a community-based population study, more than one-third of young people with psychopathology who reported psychotic experiences at baseline had at least one suicide attempt by 12-month follow up (Kelleher et al. Reference Kelleher, Corcoran, Keeley, Wigman, Devlin, Ramsay, Wasserman, Carli, Sarchiapone, Hoven, Wasserman and Cannon2013a ), demonstrating that these experiences are an important marker of risk for suicidality. Data on psychotic experiences in clinical populations, however, are lacking.
In the clinic setting, research on attenuated psychotic experiences has focused mainly on ‘at risk mental states’ (ARMS) for psychosis (Yung et al. Reference Yung, McGorry, McFarlane, Jackson, Patton and Rakkar1996; McGlashan et al. Reference McGlashan, Miller and Woods2001; Addington et al. Reference Addington, Cornblatt, Cadenhead, Cannon, McGlashan, Perkins, Seidman, Tsuang, Walker, Woods and Heinssen2011). It is important to note, however, that ARMS account for only a proportion of individuals with psychotic experiences. In community-based research, for example, we found that only one in three to one in 25 adolescents who reported psychotic experiences met formal ARMS criteria (depending on the specific ARMS criteria used) (Kelleher et al. Reference Kelleher, Murtagh, Molloy, Roddy, Clarke, Harley and Cannon2012d ). Similarly, Schimmelmann et al. (Reference Schimmelmann, Michel, Schaffner and Schultze-Lutter2011) reported, in a predominantly adult sample, that just one in 12 who reported psychotic experiences met formal ARMS criteria. Despite the high prevalence of psychotic experiences compared with ARMS, there has been relatively little clinic-based research investigating the clinicopathological significance of psychotic experiences in general. Chambers et al. (Reference Chambers, Puig-Antich, Tabrizi and Davies1982) reported that 40% of a clinical sample of children diagnosed with major depressive disorder (MDD) also experienced psychotic experiences, demonstrating that psychotic experiences are common in the clinic. What, then, is the clinical significance when a young person reports psychotic experiences to a clinician? The current study aimed to systematically address this question from a number of clinical perspectives. First, we wished to investigate the prevalence of psychotic experiences (attenuated or frank) in a general child and adolescent mental health clinic sample and to examine the association between these symptoms and global functioning. Second, we wished to examine the association between psychotic experiences and psychopathology in general and multimorbid psychopathology in particular. Third, we wished to investigate the relationship between psychotic experiences and suicidal behaviour and test if young people with psychiatric illness who report psychotic experiences have a high risk of suicidal behaviour compared with young people with psychiatric illness who do not report psychotic experiences. Fourth, we wished to investigate, using multivariate modelling, if multimorbid psychopathology could explain a relationship between psychotic experiences and suicidal behaviour.
Method
Setting
The study was carried out in a large child and adolescent mental health out-patient service (CAMHS) in the Republic of Ireland. CAMHS in the Republic of Ireland are government funded and are organized on a geographical area basis. They are available at no cost to all those under 16 years of age who live in their catchment area. The catchment area of the service has a population of approximately 380 000 of whom approximately 73 000 are under the age of 16 years. It is divided into five sectors, each served by a multidisciplinary team. The study involved two of these multidisciplinary teams. The area served by these two teams includes pockets of severe inner-city deprivation, large suburban housing estates and more affluent areas of private housing.
Participants and procedures
This was a ‘clinical case–clinical control’ study, with the patients of interest (those with psychotic experiences) being compared with patients in the same clinic who did not have psychotic experiences (as opposed to being compared with healthy controls). The participants were 108 adolescents with at least one current Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) psychiatric disorder in the 12–16 years age range, newly referred to the service between 2008 and 2009. Each newly referred adolescent was discussed at the weekly clinical team meeting, and provided a clinician could see them within a 4-week period of their research assessment, they were deemed suitable for the study. Their parent was then contacted by telephone, informed of the study, asked to discuss it with their son/daughter, and invited to participate in the research assessment, which was carried out in the adolescent's home or in the clinic. The requirement that a clinician be available to see them within a 4-week period was part of the study protocol for ethical reasons, as it was considered not to be in the best interests of adolescents and their families to expose them to a detailed research assessment, unless a clinical service could be guaranteed within a reasonable period. The study was approved by the Mater University Hospital's Research Ethics Committee. During the study period 274 adolescents in the 12–16 years age range were referred to the study teams. A total of 112 adolescents were not approached to take part in the study because it was not possible to offer a clinical service within a 4-week period of their research assessment, leaving a possible sample of 162 adolescents. A total of 20 adolescents or their parents refused to participate, with 142 (88%) agreeing to take part. Following clinical assessment, 34 of these adolescents did not have a current diagnosable psychiatric disorder, leaving a total sample of 108 participants on whom the analyses were conducted. In all, 10% were randomly selected to test for inter-rater reliability of the Kiddie Schedule for Schizophrenia and Affective Disorders Present and Lifetime Versions (K-SADS-PL). This involved a second interviewer (a psychiatric registrar) sitting in and independently rating the selected interviews. Inter-rater reliability for the K-SADS-PL was >90%.
Measures
The interview instrument used was the K-SADS-PL (Kaufman et al. Reference Kaufman, Birmaher, Brent, Rao and Ryan1996). Children and parents were interviewed separately, both answering the same questions about the child. Interviews were conducted by one psychiatrist and one psychologist trained in the use of the interview schedules. The exposure measure for the current study was a positive report at clinical interview of psychotic experiences. Outcome measure 1 was a DSM-IV diagnosable lifetime affective (MDD and bipolar affective disorder), anxiety (panic disorder, separation anxiety disorder, social phobia, specific phobia, generalized anxiety disorder, obsessive compulsive disorder, post-traumatic stress disorder) or behavioural disorder (attention deficit/hyperactivity disorder, oppositional defiant disorder, conduct disorder). Outcome measure 2 was socio-occupational functioning as measured by the Children's Global Assessment Scale (CGAS). Outcome measure 3 was a history of suicidal behaviour. See Table 1 for details of measures. The methods used by the K-SADS-PL to assess psychotic experiences in children are described in detail elsewhere. Briefly, to rate a hallucination as positive, the experience must occur in clear consciousness (i.e. not hypnopompic or hypnagogic, or associated with drug/alcohol use or physical illness). Hallucinations are distinguished from hallucination-like phenomena, such as illusions, elaborate fantasies and imaginary friends. Three categories of hallucinations are considered. First, non-diagnostic hallucinatory experiences are symptoms which are considered to have little pathological significance. These include hearing one's name being called without any other verbal hallucinations and hearing non-verbal noises, such as footsteps or creaking floorboards. These were not classified as psychotic experiences in the current study. The second category, diagnostic auditory hallucinations, includes experiences of hearing one or more voices saying at least one word other than one's own name (with or without insight that the voice was a product of one's own mind – i.e. attenuated or frank symptoms). These experiences were classified as psychotic experiences in the current study. The third category, non-auditory hallucinations, involves hallucinatory experiences affecting the other senses, predominantly visual hallucinations (attenuated or frank). These experiences were also classified as psychotic experiences in the current study. A range of delusional thoughts are also assessed in the K-SADS-PL, including delusions of reference, control or influence, persecution, grandiosity and nihilism. Delusions made up the other major category classified as psychotic experiences in the current study. Magical thinking was not classified as a psychotic experience. The interviewers recorded notes of any potential psychotic phenomena during the interview. A clinical consensus meeting was held following the interviews (with I.K., A.M. and M.C.) to classify these phenomena as psychotic experiences (or not), blind to diagnoses and all other information on the participants.
Description of the study measures
K-SADS-PL, Kiddie Schedule for Schizophrenia and Affective Disorders Present and Lifetime Versions; DSM, Diagnostic and Statistical Manual of Mental Disorders; CGAS, Children's Global Assessment Scale.
Statistical analyses
Statistical analyses were conducted using Stata version 11.0 for Windows (StataCorp LP, USA). First, we calculated the prevalence of current Axis I disorders and of psychotic experiences in the sample. We then calculated the percentage prevalence of psychotic experiences for each of the three major Axis I disorder domains (affective, anxiety and behavioural disorders). Next, we calculated the percentage prevalence of psychotic experiences across increasing levels of current Axis I disorders (one, two and three or more diagnoses) and, in order to test for a linear increase (or decrease) in diagnoses with the prevalence of psychotic experiences, we used the Stata command nptrend, which is an extension of the Wilcoxon rank-sum test and performs a non-parametric test for trend across ordered groups. We then used ordered logistic regression to look at the relationship between psychotic experiences and socio-occupational functioning on the CGAS. Then, in order to test whether multimorbidity might be responsible for the relationship between psychotic experiences and socio-occupational functioning, we conducted a multivariate regression analysis including both psychotic experiences and multimorbidity in the same model.
In terms of suicidal behaviour, we calculated the prevalence of suicidal ideation, plans and attempts and used logistic regression to assess the relationship with psychotic experiences. We then conducted a series of stratified analyses to investigate if affective, anxiety and behavioural disorders interacted with psychotic experiences in terms of risk for suicidal behaviour. Using logistic regression, we assessed the odds of suicide plans and attempts in individuals with affective disorders who reported psychotic experiences compared with those with affective disorders who did not report psychotic experiences. We repeated this for anxiety and behavioural disorders. Finally, in order to test whether the relationship between psychotic experiences and suicidal behaviour was explained by multimorbidity, we conducted a multivariate regression analysis, including the number of disorders in the model. We used χ 2 to test associations between sex and the outcome and exposure variables, with sex included as a covariate in regression analyses where there was a significant association.
Results
Study diagnoses
Current DSM-IV affective, anxiety and behavioural diagnoses were as follows: affective disorders – MDD (34%) (zero cases of bipolar affective disorder); anxiety disorders – generalized anxiety disorder (34%), separation anxiety disorder (23%), social phobia (21%), specific phobia (8%), obsessive compulsive disorder (12%), post-traumatic stress disorder (8%) and panic disorder (2%); behavioural disorders – oppositional defiant disorder (37%), attention deficit/hyperactivity disorder (36%) and conduct disorder (11%). Two patients were diagnosed with a psychotic disorder. Additional diagnoses are listed in online Supplementary Table S1. Multimorbidity (i.e. more than one diagnosis) was common: 37% (n = 40) had one current disorder, 31% (n = 33) had two current disorders and 32% (n = 34) had three or more disorders.
Psychotic experiences, psychopathology and socio-occupational functioning
Psychotic experiences were reported by 46% (n = 52) of the total clinical sample, including 68% of individuals with an affective disorder, 60% of individuals with an anxiety disorder and 41% of individuals with a behavioural disorder. Sex did not show a statistically significant association with psychotic experiences (χ 2 = 3.3, p > 0.05) or with multimorbidity (χ 2 = 1.1, p > 0.05) and so was not included as a covariate in the analyses. Of the sample with psychotic experiences, 50% (n = 26) reported auditory hallucinatory experiences, most commonly hearing voices talking when no one was present (73%). The commonest auditory vocal hallucinatory experiences included hearing voices commenting on behaviour (53%) and hearing voices giving commands (26%). In terms of visual hallucinatory experiences, 29% (n = 15) reported seeing people or faces where none really existed. In terms of delusional-type experiences, the most common experiences included beliefs about being followed or spied upon (63%, n = 33), beliefs that they had interacted with ghosts (39%, n = 19) and thought withdrawal or broadcast (21%, n = 11).
Psychotic experiences were associated with increased risk of multimorbidity: 78% of young people who reported psychotic experiences had two or more current disorders compared with 49% of young people without psychotic experiences, while 46% of young people who reported psychotic experiences had three or more disorders, compared with 19% of young people without psychotic experiences [odds ratio (OR) 2.35, 95% confidence interval (CI) 1.43–3.87; test for linear trend: Z = 3.44, p = 0.001].
Patients who reported psychotic experiences demonstrated significantly poorer socio-occupational functioning than patients who did not report psychotic experiences (OR 0.31, 95% CI 0.15–0.62). Poorer functioning was also associated with increasing numbers of diagnosable disorders (i.e. multimorbidity) (OR 0.35, 95% CI 0.22–0.55). In order to test whether poorer functioning in individuals with psychotic experiences was a result of the relationship between psychotic experiences and multimorbidity, we conducted a multivariate regression analysis including both psychotic experiences and multimorbidity. Multimorbidity did not explain the relationship between psychotic experiences and poorer functioning, with both variables demonstrating independent main effects on functioning (psychotic experiences: OR 0.45, 95% CI 0.22–0.95; increasing levels of multimorbidity: OR 0.64, 95% CI 0.50–0.88).
Suicidal behaviour
Of the total sample, 12% (n = 14) reported isolated suicidal ideation (without suicide plans or acts), 34% (n = 37) had a history of specific suicide plans and 27% (n = 30) had a history of suicide attempt. There was a statistically significant association between sex and suicide plans (χ 2 = 12.98, p < 0.001) and attempts (χ 2 = 14.79, p < 0.001); therefore, sex was included as a covariate in the analyses on suicidal behaviour. Psychotic experiences were not associated with isolated suicidal ideation (OR 1.35, 95% CI 0.43–4.26) but were associated with a 3-fold increased odds of suicide plans (OR 3.35, 95% CI 1.39–8.08) and attempts (OR 2.70, 95% CI 1.06–6.89). Multimorbidity did not explain the relationship between psychotic experiences and suicide plans (adjusted OR 3.25, 95% CI 1.27–8.33) or suicide attempts (adjusted OR 3.13, 95% CI 1.11–8.79).
The relationship between psychotic experiences and affective, anxiety and behavioural disorders is shown in Table 2. In order to examine for potential interactions between affective, anxiety and behavioural disorders and psychotic experiences in terms of risk for suicidal behaviour, we conducted a series of stratified analyses (see Table 3). Of the patients with a diagnosis of MDD, 18 had a suicide attempt, 16 of whom (89%) reported psychotic experiences. Of the patients with an anxiety disorder, 15 had a suicide attempt, 14 of whom (93%) reported psychotic experiences. Of the patients with a behavioural disorder, 16 had a suicide attempt, 10 of whom (63%) reported psychotic experiences.
Association between major diagnoses and suicidal behaviour
OR, Odds ratio; CI, confidence interval.
Stratification of disorder groups into those with and without psychotic experiences and association with suicide attempts
OR, Odds ratio; CI, confidence interval.
Discussion
Using a child and adolescent mental health clinic sample we found a number of clinically important results. First, psychotic experiences were relatively common and were prevalent in a wide range of non-psychotic psychiatric disorders. Second, psychotic experiences indexed particularly high risk for multimorbidity, that is, having multiple DSM diagnoses. Third, young people with psychopathology who also reported psychotic experiences demonstrated significantly poorer socio-occupational functioning compared with young people with psychopathology who did not report psychotic experiences, even when controlling for levels of multimorbidity. Fourth, psychotic experiences were a marker of high risk for severe suicidal behaviour, that is, suicide plans and attempts, even when controlling for levels of multimorbidity.
The prevalence of psychotic experiences reported in the current study is similar to prevalence findings of 48% reported by Chambers et al. (Reference Chambers, Puig-Antich, Tabrizi and Davies1982) and 40% reported by Altman et al. (Reference Altman, Collins and Mundy1997) in child and adolescent mental health services. Ulloa et al. (Reference Ulloa, Birmaher, Axelson, Williamson, Brent, Ryan, Bridge and Baugher2000) reported a prevalence of approximately 9% for ‘definite’ or ‘probable’ psychotic experiences in a clinical sample of young people. The reason for the lower prevalence in the Ulloa et al. (Reference Ulloa, Birmaher, Axelson, Williamson, Brent, Ryan, Bridge and Baugher2000) sample is not clear, although it is important to note that the age range of participants in their study was up to 21 years, since we have previously shown that, in community samples at least, psychotic experiences are less common in late compared with early childhood and adolescence (Kelleher et al. Reference Kelleher, Connor, Clarke, Devlin, Harley and Cannon2012a ).
Psychotic experiences were common across a range of disorders but demonstrated a particularly strong relationship with multimorbid psychopathology, with prevalence of psychotic experiences increasing in a dose–response manner with the number of diagnosable disorders – 28% of patients with one disorder reported psychotic experiences, compared with 48% of patients with two disorders and 68% of patients with three or more disorders. This is in line with community-based research, which also showed a dose–response relationship between the prevalence of psychotic experiences and number of diagnosable disorders (Kelleher et al. Reference Kelleher, Keeley, Corcoran, Lynch, Fitzpatrick, Devlin, Molloy, Roddy, Clarke, Harley, Arseneault, Wasserman, Carli, Sarchiapone, Hoven, Wasserman and Cannon2012b ). Clinical need is at a premium in these cases as patients with multimorbidity are at very high risk for a wide range of poor outcomes, even when compared with other patients with psychopathology. Angst et al. (Reference Angst, Sellaro and Ries Merikangas2002), for example, showed that multimorbidity predicts greater work impairment, poorer social functioning, more relationship breakdown and higher levels of subjective distress. This was evident in the current study, with multimorbidity predicting poorer socio-occupational functioning compared with single-disorder psychopathology. Interestingly, however, psychotic experiences were an even stronger predictor of poor socio-occupational functioning, and multivariate analysis demonstrated that this relationship was not explained by the effect of multimorbidity. Psychotic experiences reported in the clinic, then, should alert the clinician to high risk for multimorbid psychopathology (patients with psychotic experiences in the current study had an average of three Axis I disorders) and poorer socio-occupational functioning (beyond that predicted by multimorbidity itself).
Psychotic experiences were associated with a significantly increased prevalence of suicide plans and attempts. There was no relationship, however, between psychotic experiences and isolated suicidal ideation (in the absence of suicide plans or attempts), indicating that these symptoms are a marker of risk for more severe suicidal behaviour. The prevalence of suicidal behaviour was relatively low in individuals with affective, anxiety or behavioural disorders who did not report psychotic experiences but high in those with psychotic experiences. These findings highlight that a report of psychotic experiences in a young person with psychopathology should alert clinicians to risk for suicidal behaviour, even relative to other patients with a disorder. The finding of a high prevalence of suicidal behaviour in the sample with MDDs who reported psychotic experiences demonstrates interesting overlap with established findings on the (narrower) diagnosis of MDD with psychotic features. Unlike MDD with psychotic features, however, which is considered to be an uncommon illness (of note, none of the current sample was diagnosed with MDD with psychotic features), psychotic experiences were common among young people in the current sample who had a diagnosis of MDD. Furthermore, it is important to recognize that most of the individuals who reported psychotic experiences in this sample had attenuated symptoms (i.e. they did not have true hallucinations or delusions, but, rather, attenuated symptoms) and, thus, diagnostically, could not be considered to have MDD with psychotic features. Our findings suggest, however, that there is a much broader spectrum of (attenuated) psychotic experiences associated with MDD and that, as with the narrower diagnosis of ‘MDD with psychotic features’, this group is at high risk for suicidal behaviour. It is also important to note that the relationship between suicidal behaviour and psychotic experiences was not limited to MDD with psychotic experiences; this relationship was also evident in anxiety and behavioural disorders with psychotic experiences, which do not have a ‘with psychotic features’ subcategory in the DSM.
There are a number of possible explanations for the relationship between psychotic experiences and suicidal behaviour. The strong relationship between psychotic experiences and multimorbidity may be part of this explanation; however, the fact that psychotic experiences remained a strong marker of risk for suicidal behaviour even when multimorbidity was included in the multivariate model means that multimorbidity can be, at best, only part of the mechanistic explanation linking psychotic experiences and suicidal behaviour. Other potential mechanisms might include shared risk factors for suicidal behaviour and psychotic experiences. Childhood traumatic experiences, such as physical and sexual abuse, predict a higher prevalence of suicidal behaviour (Dube et al. Reference Dube, Anda, Felitti, Chapman, Williamson and Giles2001; Evans et al. Reference Evans, Hawton and Rodham2005; Afifi et al. Reference Afifi, Enns, Cox, Asmundson, Stein and Sareen2008) and are also more prevalent among individuals who report psychotic experiences (Scott et al. Reference Scott, Chant, Andrews, Martin and McGrath2007a ; Kelleher et al. Reference Kelleher, Harley, Lynch, Arseneault, Fitzpatrick and Cannon2008, Reference Kelleher, Keeley, Corcoran, Ramsay, Wasserman, Carli, Sarchiapone, Hoven, Wasserman and Cannon2013b ; Freeman & Fowler, Reference Freeman and Fowler2009; Arseneault et al. Reference Arseneault, Cannon, Fisher, Polanczyk, Moffitt and Caspi2011; Bebbington et al. Reference Bebbington, Jonas, Kuipers, King, Cooper, Brugha, Meltzer, McManus and Jenkins2011; Fisher et al. Reference Fisher, Schreier, Zammit, Maughan, Munafò, Lewis and Wolke2012). There is also research to suggest that individuals who endorse psychotic experiences on questionnaires have stronger emotional reactivity to daily stressors (Lataster et al. Reference Lataster, Wichers, Jacobs, Mengelers, Derom, Thiery, Van Os and Myin-Germeys2009) and poorer coping skills (Lin et al. Reference Lin, Wigman, Nelson, Vollebergh, van Os, Baksheev, Ryan, Raaijmakers, Thompson and Yung2011) which may in turn put them at increased risk for suicidal behaviour. Further research will be necessary to understand the mechanisms underlying the relationship between psychotic experiences and both multimorbidity and suicidal behaviour.
Strengths and limitations
Strengths of the current study include the use of ‘gold standard’ assessment of psychiatric disorders (full semi-structured Axis I diagnostic interviews) and the use of highly trained professionals to conduct the interviews (one psychiatrist and one psychologist). While clinical service limitations meant that a significant proportion of patients referred to participating clinics could not be invited to participate in the study, consent to take part among eligible patients was high (88%). Because part of the study requirements meant that participation was limited to patients who would receive a prompt clinical service (within 4 weeks of referral), it is likely that participants in the current study had more severe illness than might be seen in a fully representative clinical sample (for example, an individual who reports suicidal behaviour is likely to receive a more prompt clinical service than one who does not). Thus, severe psychopathology (including multimorbidity and suicidal behaviour) and psychotic experiences may have been more prevalent in the current study than would be the case in the total sample of patients in a clinic. However, this ‘enrichment’ for psychopathology severity was beneficial in terms of testing the study hypotheses because it allowed us to test the relationship between the exposure and outcome measures with maximal statistical power. Our findings are all the more striking because, rather than comparing our cases with a sample of healthy community controls as is usually reported, we compared our cases with an unwell clinical sample (who did not, however, have psychotic experiences). The fact that large differences emerged in our cases compared with our clinical controls illustrates the significance of a report of psychotic experiences in the clinic. Because this was not a longitudinal study, the issue of temporality cannot be addressed, i.e. which arose first: psychotic experiences, psychopathology, or both together. However, this does not detract from the clinical importance of understanding the contemporaneous relationship between psychotic experiences and non-psychotic psychopathology. As with any in-depth clinical interview study, its strength is its weakness: it is not possible to conduct this type of in-depth diagnostic research with very large numbers of participants in the same way as can be done with questionnaire or lay interview studies. As a result, subgroup analyses involved smaller groups and, because of this, CIs are wide in some cases. However, despite this, the results demonstrated a clear dose–response effect in terms of the prevalence of psychotic experiences and increasing levels of multimorbidity and, in terms of suicidal behaviour, even the lower ranges of CIs for significant interactions between psychotic experiences and psychopathology were well in excess of 1.
Conclusions
The current paper posed the question: what does it mean when a young person reports psychotic experiences to a clinician? The results of this study suggest a number of answers. First, beyond thinking about the primary diagnosis for which the individual was referred, the clinician should consider that, based on our findings, most of these individuals will have no fewer than three Axis I diagnoses. These individuals are also likely to have poorer socio-occupational functioning, even compared with individuals with the same level of multimorbidity who do not have psychotic experiences. Of particular clinical importance, a report of psychotic experiences points to a very strong relationship with suicidal behaviour. From a research point of view, these findings also highlight the importance of including assessments of psychotic experiences in future studies of suicidal behaviour and in research on multimorbid psychopathology. The current study also highlights the fact that so-called ‘psychotic’ experiences are, in fact, symptoms that occur in a wide range of disorders and should not, when reported by adolescents, be mistaken to indicate incipient risk for psychotic disorder. It is also important to note that there is a lack of evidence that the presence of these symptoms indicates the need for treatment with antipsychotic medication; treatment should be targeted at the underlying diagnoses (e.g. MDD, generalized anxiety disorder, attention deficit hyperactivity disorder, etc.), though recognizing that this may be a more severe case than the same diagnosis in someone without psychotic experiences. Inherent to all of this, of course, is the assumption that all patients will be assessed for psychotic experiences; anecdotal evidence, unfortunately, suggests this is not the case in many youth mental health services. Our own clinical experience is that young people will rarely volunteer information on psychotic experiences in the context of a routine psychiatric assessment unless they are directly (and sensitively) asked about them. When they are asked, however, young people are usually willing to discuss these symptoms and are, in fact, often greatly relieved to share these experiences with a calm (unflinching) clinician who can explain that these symptoms are not uncommon and do not indicate that she or he is at the precipice of psychotic disorder. In light of the findings of the present study, greater recognition of the importance of routine assessment for (attenuated and frank) psychotic experiences by clinicians working with young people, and greater understanding of the pathological significance of these symptoms, are urgently needed.
Supplementary material
For supplementary material accompanying this paper visit http://dx.doi.org/10.1017/S0033291713002122.
Acknowledgements
This work was supported by the Brain and Behavior Research Foundation (NARSAD) and the Health Research Board (Ireland). I.K. was supported by an Interdisciplinary Capacity Enhancement Award from the Health Research Board Ireland (ICE/2012/11). J.T.W.W. was supported by a Stichting Koningsheide grant. The study sponsors had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; and preparation, review or approval of the manuscript.
Declaration of Interest
None.
