SCHRES-06872; No of Pages 6

Schizophrenia Research xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Schizophrenia Research

j ou rna l h omepag e: w w w . e l se v i e r . c o m / l o ca t e / s c h r e s

Hallucinations in adolescents and risk for mental disorders and suicidal

behaviour in adulthood: Prospective evidence from the MUSP birth cohort study
Melissa Connell a, Kim Betts b, John J. McGrath c,d, Rosa Alati b,g, Jake Najman b,f, Alexandra Clavarino b,h,
Abdullah Mamun b, Gail Williams b, James G. Scott a,d,e,⁎
a
The University of Queensland Centre for Clinical Research, Herston, QLD 4029, Australia
b
School of Public Health, University of Queensland, Herston, QLD 4029, Australia
c
Queensland Brain Institute, University of Queensland, St Lucia, QLD 4076, Australia
d
Queensland Centre for Mental Health Research, The Park Centre for Mental Health, Wacol, QLD 4076, Australia
e
Royal Brisbane and Women's Hospital, Herston, QLD 4029, Australia
f
School of Social Science, University of Queensland, St Lucia. QLD 4076, Australia
g
Centre for Youth Substance Abuse Research, University of Queensland, Royal Brisbane and Women's Hospital, Herston, QLD 4006, Australia
h
School of Pharmacy, The University of Queensland, 20 Cornwall Street, Woolloongabba, 4102, Queensland, Australia

a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 February 2016
Background: Hallucinations, once equated with serious mental disorders, are common in adolescents. Given the high
Received in revised form 7 May 2016 prevalence of hallucinations, it is important to determine if they are associated with adverse mental health outcomes in
Accepted 9 June 2016 adulthood. This study compared the mental health outcomes of participants (aged 30–33 years) in the Mater-University of
Available online xxxx Queensland Study of Pregnancy (MUSP) who reported hallucinations at (a) 14 years only and (b) 14 and 21 years versus
cohort members without hallucinations.
Keywords: Method: Participants (n = 333) were aged between 30 and 33 years and (a) reported hallucinations on the Youth Self-Report
Hallucinations Birth Questionnaire at 14 and/or the Young Adult Self-Report Questionnaire at 21 years and (b) controls (n = 321) who did not
cohort Mental report hallucinations. Lifetime diagnoses of mental disorders were ascertained by the Structured Clinical Interview for DSM
disorders Psychotic
Disorders (DSM IV-TR) administered by clinical psychologists. Suicidal be- haviour was measured by self report.
disorders Suicide
Results: Hallucinations at 14 years only were not associated with an increased risk of mental disorders in adult- hood.
Hallucinations reported at both 14 and 21 years were associated with lifetime diagnoses of psychotic dis- orders (OR, 8.84;
95% CI: 1.61–48.43 and substance use disorders (OR, 2.34; 95% CI: 1.36–4.07) and also strongly associated with lifetime
suicide attempts (OR, 7.11; 95% CI: 2.68–18.83).
Conclusions: Most adolescents who experience hallucinations do not have an increased rate of mental disorder in adulthood;
however, those with hallucinations that are experienced at more than one point in time are at in- creased risk of suicidal
behaviour and both psychotic and non-psychotic psychopathology.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction
Hallucinations, cardinal symptoms of psychotic disorders, also occur in other
mental disorders and in otherwise well individuals. For exam- ple, a large
community-based study reported the lifetime prevalence of hallucinations as
5.2% (McGrath et al., 2015) whilst studies of adoles- cents have found a higher
prevalence of almost 15% (Kelleher et al.,
* Corresponding author at: The University of Queensland Centre for Clinical Research, Herston,
QLD 4029, Australia.
E-mail address: james.scott@health.qld.gov.au (J.G. Scott).
2012a). Studies exploring the outcomes of hallucinations in adolescents have
found an increased risk of subsequent psychotic disorder (Poulton et al., 2000;
Welham et al., 2009) and other psychopathology (Dhossche et al., 2002; Fisher
et al., 2013; Kelleher et al., 2012b). However, these studies have their limitations
and the utility of hallucinations as a pre- dictor of future mental health disorders
is in debate.
Five population studies have examined adult mental health out- comes of
children and adolescents who hallucinate. Two studies based on the Dunedin
birth cohort assessed diagnostic outcomes of partici- pants who experienced
psychotic symptoms measured at 11 years (based on the Diagnostic Interview
Schedule for Children) (Fisher et

http://dx.doi.org/10.1016/j.schres.2016.06.009 0920-
9964/© 2016 Elsevier B.V. All rights reserved.

Please cite this article as: Connell, M., et al., Hallucinations in adolescents and risk for mental disorders and suicidal behaviour in adulthood: Prospective evidence
from the MUS..., Schizophr. Res. (2016), http://dx.doi.org/10.1016/j.schres.2016.06.009
2 M. Connell et al. / Schizophrenia Research xxx (2016) xxx–xxx
al., 2013; Poulton et al., 2000). Participants who were interviewed by child
psychiatrists and identified as having psychotic symptoms had a 16-fold
greater risk of adult schizophreniform disorder (odds ratio [OR], 16.4; 95%
confidence interval [CI], 3.9–67.8) at 26 years (Poulton et al., 2000). In a
subsequent study of the same cohort, outcomes were extended to 38 years,
where it was found that hallucinations were asso- ciated with an increased
likelihood of schizophrenia (relative risk [RR], 7.24; 95% CI, 2.17–24.13) and
posttraumatic stress disorder (RR, 3.03; 95% CI, 1.33–6.89) (Fisher et al.,
2013), suggesting childhood psychotic symptoms may be a useful predictor of
future mental health problems. In a longitudinal study of youth aged 14–17
years, Dominguez and col- leagues measured psychotic experiences at three
time points over an 8- year period. They found that those who reported
hallucinations and other psychotic experiences at only one time did not have an
increased risk of subsequent psychosis (compared with those who never
halluci- nated). However, adolescents with hallucinations at all three time
points had 10 times the odds of future psychosis (OR, 9.9; 95% CI, 2.5– 39.8)
(Dominguez et al., 2011).
Two studies have reported the diagnostic outcomes of adolescents who
experienced auditory or visual hallucinations measured by the Youth Self-
Report (Dhossche et al., 2002; Welham et al., 2009). Welham and colleagues
reported that participants of the Mater Hospital Univer- sity of Queensland Study
of Pregnancy (MUSP) who experienced hallu- cinations at 14 years were at
increased risk of non-affective psychosis at age 21 (OR, 5.09; 95% CI, 2.18–11.8
[males]; 2.27; 1.01–5.12 [females])
(Welham et al., 2009). However, this study did not examine non-psy- chotic
disorders as an outcome. By contrast, Dhossche and colleagues assessed both
psychotic and non-psychotic outcomes in adulthood of hallucinations in a
community sample of adolescents (Dhossche et al., 2002). Those reporting
hallucinations were at increased risk of a depres- sive disorder (OR, 3.0; 95% CI,
1.1–8.8) or substance use disorder (OR, 7.8; 95% CI, 2.5–24.6) at 8-year follow-
up but none had transitioned to psychotic disorder.
Collectively, these studies suggest children and adolescents who hal- lucinate
are at increased risk of psychotic and non-psychotic disorders in adulthood.
However, all studies assessed diagnostic outcomes using lay interviewers rather
than clinicians, few examined a range of both psychotic and non-psychotic
outcomes, and only one (Dominguez et al., 2011) examined hallucinations in
adolescents at more than one time point showing that persistence of
hallucinations was important in predicting future psychosis. There is a need for
studies that examine a range of outcomes associated with hallucinations, which
are mea- sured at more than one time point, and that follow participants into
adulthood after they have transitioned through the period of highest risk for
psychosis (Dhossche et al., 2002; Welham et al., 2009).
Another outcome associated with hallucinations in adolescents and adults is
suicidality. A cross-sectional study of a large representative sample of adults
showed those respondents with psychotic experiences (PE) were more likely to
report current suicidal ideation (OR, 5.24; 95% CI, 2.85–9.62) and suicide
attempts (OR, 9.48; 95% CI, 3.98–22.62) (DeVylder et al., 2015a).
Prospective studies have also reported that PE in adolescents elevate the risk for
suicidal behaviours at follow-up. A prospective cohort study of school-based
adolescents with baseline psychopathology and psychotic symptoms found
14% had reported a suicide attempt by 3 months (OR, 17.91; 95% CI, 3.61–
88.82) and 34%
reported a suicide attempt by 12 months (OR, 32.67; 95% CI, 10.42– 102.41)
(Kelleher et al., 2013). Another longitudinal cohort study of ad- olescents found
that PE at one time point only without psychological distress was not
associated with an increased risk of suicidality at 12 months (Martin et al.,
2015). However, persistent psychotic experi- ences (i.e. those present at both
baseline and one year follow-up) and those PE accompanied by psychological
distress were associated with increased odds of suicide attempts (OR, 4.63:
95% CI, 1.21–17.72 and 12.81; 4.02–40.88, respectively). Both studies have
follow-up times lim- ited to 12 months and did not report associations with
adult suicidal behaviour.
Please cite this article as: Connell, M., et al., Hallucinations in adolescents and risk for mental disorders and suicidal behaviour in adulthood: Prospective evidence
from the MUS..., Schizophr. Res. (2016), http://dx.doi.org/10.1016/j.schres.2016.06.009
The current study examined a broad range of mental health out- comes in
adulthood for those who reported experiencing hallucinations at (i) 14 years and
(ii) both 14 and 21 years. Based on previous findings (van Os et al., 2009), it
was hypothesised that those experiencing hallu- cinations at 14 years would be at
increased risk of adult psychotic and non-psychotic disorders as well as
suicidality. Additionally, those ado- lescents who experienced hallucinations at
both 14 and 21 years would have the highest risk of psychotic and non-
psychotic disorder and suicidality at 30–33 years.
2. Method
2.1. Participants
Participants were from the Mater-University of Queensland Study of
Pregnancy (MUSP), a prospective birth cohort study of mothers and their
offspring who received antenatal care at the Mater Misericordiae Mothers'
Hospital, a major public hospital in Brisbane, Australia, be- tween 1981 and
1984. Baseline data were collected on 7223 mothers and their singleton live-
birth offspring who have since been prospec- tively followed-up over thirty
years. Further information describing the MUSP cohort study can be found
elsewhere (Najman et al., 2005). At the 14- and 21-year data collections, the
Youth Self-Report (Achenbach, 1991) and the Young Adult Self-Report
(Achenbach, 1997) were used to identify those offspring experiencing auditory
and visual hallucinations (Fig. 1). Only participants who provided data at both
14- and 21-year follow-ups were eligible for recruitment. In total, 3535
participants completed the hallucination questions at both time points, of which
822 (23.3%) endorsed experiencing auditory or vi- sual hallucinations at 14
and/or 21 years. This group consisted of 455 participants (12.9%) with
hallucinations only at 14 years old, 227 (6.4%) with hallucinations only at 21
years old and 140 (4%) with hallu- cinations at both 14 and 21 years old.
2.2. Sampling protocol
The target sample included all 822 MUSP participants who endorsed
experiencing hallucinations at ages 14 and/or 21 and 490 randomly se- lected
MUSP participants who did not report hallucinations at either time point. The
intended ratio of participants endorsing hallucinations to those not endorsing
hallucinations (roughly 2:1), was chosen to oversample individuals who were at
a potentially greater risk of having a psychotic disorder by age 30. Four
hundred and forty-five (54%) of those who had experienced hallucinations
(56.7% male, M age = 31.6, SD = 0.88) and 321 (65%) of those who had not
experienced hallucina- tions were interviewed (42.2% male, M age = 31.1, SD
= 0.95). This re- sulted in a total sample of 766 participants. The current study
restricted cases to (a) 250 (54.9%) participants who reported hallucinations at age
14 alone and (b) 83 (59.2%) participants who reported hallucinations at age 14
and age 21. One hundred and twelve cohort members reported hallucinations
at 21 years only but were not included in the present study as it could not be
determined if they had already developed a psy- chotic disorder prior to this age
(we intend to follow these individuals in future studies). Methods to assess
potential bias resulting from the re- sponse rates are outlined below.
2.3. Interview procedure
Interviews were conducted face to face when possible, and phone
interviews were used when participants were unable to attend the study
location. Interviews involved completion of self-report question- naires and
participation in a semi-structured interview. Interviewers were ‘blind’ to age 14
and age 21 hallucination group status.
 M. Connell et al. / Schizophrenia Research xxx (2016) xxx–xxx
Fig. 1. Sampling procedure used in participant selection.
2.4. Structured clinical interview for DSM-IV Axis I disorders (SCID I)
The SCID I is a semi-structured interview used for making DSM-IV Axis I
diagnoses and covers key diagnostic groups including mood disor- ders,
psychotic disorders, anxiety disorders, substance use disorders, somatoform
disorders, eating disorders and adjustment disorders (First et al., 2002). It
requires administration by a clinician or trained mental health professional
familiar with DSM classification and diagno- sis. It assesses both lifetime and
current diagnoses and can be used to provide symptom ratings as either absent,
subthreshold or threshold.
The research version of the SCID was administered using netSCID, a
computerized program designed in collaboration with the SCID authors and found
to improve accuracy and reliability of administration. Inter- views were
administered by clinical psychologists and clinical psychol- ogy registrars with
experience and training in interviewing those with psychotic disorders. Raters
undertook training in administration of the SCID and were required to meet
criteria of 95% diagnostic agreement with SCID training interviews before
interviewing the MUSP partici- pants. Interviewers were required to participate in
regular group meet- ings with J.S. and M.C. in which ratings were reviewed
and complex presentations discussed. All interviews were audio recorded and
ran- dom audits were undertaken by M.C. to ensure adherence to the proto- col
and accuracy of ratings made.
3
2.5. Suicidality
To avoid the limitations of single item measures of suicidality, suicid- al
ideation, suicide plans and suicide attempts over the person's lifetime were
measured using three items which asked if the person had ever se- riously
thought about committing suicide, had they ever made a plan and had they
ever had an attempt.
2.6. Outcome variables
For the main analyses, we used the major categories of Axis 1 diag- noses
from the SCID, including any DSM-IV mood (i.e. major depressive disorder,
bipolar disorders and substance-induced mood disorders), anxiety (i.e. panic
disorder, agoraphobia, posttraumatic stress disorder, generalized anxiety disorder,
obsessive compulsive disorder, social pho- bia and specific phobia), psychotic
(i.e. schizophrenia, schizoaffective disorder, psychotic disorder NOS, substance
induced psychotic disorder and brief psychotic disorder), substance use (i.e.
disorders of abuse and dependence including alcohol, cannabis, cocaine, opiate,
hallucinogen, amphetamine, sedative and polysubstance dependence) and eating
dis- order (i.e. anorexia nervosa, bulimia nervosa and eating disorder NOS). A
secondary analysis was also conducted based on selected mental dis- orders
(major depressive disorder [MDD], social phobia, specific phobia,
4 M. Connell et al. / Schizophrenia Research xxx (2016) xxx–xxx

posttraumatic stress disorder [PTSD], generalized anxiety disorder [GAD],

Table 2
panic disorder, obsessive compulsive disorder [OCD], alcohol abuse and/or Multivariable associations between hallucinations at age 14 only and hallucinations at both 14 and
dependence [combined as alcohol use disorder], cannabis 21 with DSM-IV mental disorder categories, adjusted for potential confoundersa [expressed in
OR with 95% confidence intervals (CI)] (n = 650).
abuse and/or dependence [combined as cannabis use disorder] and
polysubstance dependence disorders). Suicidality was measured using three Hallucinations
categorical variables of suicidal thoughts, suicidal plans and suicide None Age 14 only Ages 14 and 21
attempts. Disorder category OR (95% CI) OR (95% CI) OR (95% CI)
Mood 1.00 1.04 (0.74–1.48) 1.47 (0.89–2.43)
2.7. Predictor variables Anxiety 1.00 0.92 (0.63–1.33) 1.52 (0.91–2.54)
Psychotic 1.00 1.09 (0.16–7.60) 8.84 (1.61–48.43)
Substance use 1.00 1.08 (0.71–1.62) 2.34 (1.36–4.07)
Hallucinations were measured at age 14 using the YSR (Achenbach, 1991) Eating 1.00 1.23 (0.35–4.30) 3.60 (0.99–13.07)
and at age 21 using the YASR. (Achenbach, 1997) Hallucinations were assessed
a
Adjusted confounding factors were age, sex, maternal education at baseline and sub- stance use at
at 14 and 21 years with the YSR/YASR questionnaires which included two
14 years.
questions: (i) “I hear sounds or voices which other people think aren't there”;
(ii) “I see things that other people think aren't there”. Respondents answered
with very true or often true, somewhat or sometimes true, or not true, 3. Results
which were then reclassified into three discrete hallucinations categories: (1)
did not experience hal- lucinations, (2) experienced hallucinations sometimes After adjusting for age, sex, maternal education at baseline and sub- stance
or often at age 14, (3) experienced hallucinations sometimes or often at ages use at 14 years, hallucinations at 14 years only were not associat- ed with any
14 and 21.The reliability of self-report measures for assessing hallucinations in groups of mental disorder in adulthood (Table 2). Those participants who
adolescents has been established (Kelleher et al., 2011). Of all psy- chotic reported hallucinations at both 14 and 21 years had increased odds of being
experiences, self-report assessment of hearing voices or sounds that others diagnosed with a psychotic disorder (OR, 8.84; 95% CI, 1.61–48.43) or
don't hear was found to be the strongest predictor of a clin- ically verifiable substance use disorder (OR, 2.34; 95% CI,
history of auditory hallucinations (positive predictive value [PPV] = 71.4% 1.36–4.07) at 30–33 years (See Tables 1 and 2).
and negative predictive value [NPV] = 90.4%), with visual hallucinations also Pertaining to specific mental disorders, there were no associations with
showing moderately successful predic- tive power (PPV = 81.8% and NPV = hallucinations at 14 years. Those with hallucinations at 14 and 21 years had
80%) (Kelleher et al., 2011). Hallu- cination items from the YSR have been increased risk of alcohol use disorder (OR, 2.41; 95% CI, 1.35–4.30), cannabis
used in previous longitudinal studies (Dhossche et al., 2002; Welham et al., use disorder (OR, 2.71; 95% CI, 1.36–5.41), panic disorder (OR, 3.30; 95% CI,
2009). 1.35–8.08), polysubstance dependence (OR, 3.83; 95% CI, 1.13–13.04) and
specific phobia (OR, 2.43; 95% CI, 1.11–
2.8. Statistical analyses 5.29) (Supplementary Table 1). There was no association between hal-
lucinations at both 14 and 21 years and major depressive disorder, gen- eralized
Chi-square tests were used to examine the associations between the different anxiety disorder, obsessive compulsive disorder, posttraumatic stress disorder
categories of hallucinations with the major groups of mental disorders at age and social phobia.
30. Univariate and multivariate logistic regressions were then used to examine Those reporting hallucinations in both adolescence and young adult- hood
the associations between hallucinations ad- justed for potential covariates with were more likely to report lifetime suicidal ideas and behaviours at 30–33
the mental disorder groups and with suicidal behaviour. Covariates used in the years. Suicide attempts were almost three times more likely to occur in those
adjusted models were partic- ipant age, sex, substance use at age 14 and maternal who had hallucinations at 14 years only (OR, 2.63; 95% CI, 1.04–6.63) and
education at base- line. Secondary analyses were then conducted examining seven-fold more likely in those with hallucina- tions at both follow-ups (OR,
the associations between hallucinations with specific mental disorders. The 7.11; 95% CI, 2.68–18.83) (Table 3).
final sample in all analyses consisted of 650 participants with values for all Lastly, the attrition analyses showed that among participants who reported
variables of interest. hallucinations, those included in the sample were more likely to be female and
their mothers were more likely to have completed ter- tiary education
2.9. Response rates analyses (Supplementary Table 2). Among participants who did report hallucinations,
those included in the study were more likely to be female, have a higher
Two separate attrition analyses were conducted using multivariate logistic family income and an older maternal age at birth.
regression to compare differences among baseline measures (i.e., pre-
pregnancy) between (1) MUSP participants who reported hal- lucinations at 14 4. Discussion
and/or 21 years of age who participated in the current study versus those who did
not participate and (2) MUSP participants who did not report hallucinations at Adolescents who reported experiencing hallucinations at both 14 and 21
14 or 21 years of age (controls) and participated in the current study versus years were at increased risk of both adult psychopathology and lifetime suicidal
those who did not partici- pate These analyses were further limited to participants ideas and behaviours. Those who reported experienc- ing hallucinations at 14
with all data on baseline variables of interest. years only did not have an increased risk of mental disorders in the main
analysis although they were more likely to have attempted suicide.
Together, these findings suggest that

Table 1
Univariable associations between hallucinations at age 14 only and hallucinations at both 14 and 21 with lifetime DSM-IV mental disorder categories (n = 650).

Hallucinations, %(n)

Disorder category None Age 14 only Ages 14 and 21 Chi-square (df) P value

Mood 42.3 (134) 48.8 (112) 53.0 (44) 3.08 (2) 0.22
Anxiety 33.1 (105) 33.6 (84) 47.0 (39) 5.94 (2) 0.05
Psychotic 0.6 (2) 1.2 (3) 07.2 (6) 17.80 (2) b0.001
Substance use 23.7 (75) 30.0 (75) 44.6 (37) 14.35 (2) 0.001
 M. Connell et al. / Schizophrenia Research xxx (2016) xxx–xxx
Table 3
Multivariable associations between hallucinations at age 14 only and hallucinations at both 14 and 21 with suicidal behaviours at age 30, Adjusted for potential confoundersa [expressed in OR with 95%
confidence intervals (CI)] (n = 644).
Hallucinations
None
Suicidal behaviour % yes (n) OR (95% CI)
Ideation 16.8 (53) 1.00
Plans 6.0 (19) 1.00
Attempts 2.2 (7) 1.00
a
Adjusted confounding factors were age, sex, maternal education at baseline and substance use at 14 years.
adolescents whose hallucinations remit before adulthood are at no greater risk
of future mental disorders than individuals who do not ex- perience
hallucinations. However, those who experience hallucinations in adolescence and
adulthood are more likely to develop a mental disor- der in adulthood.
Those with hallucinations present at 14 and 21 years had an almost nine-fold
risk of developing a psychotic disorder and a three-fold risk of a substance use
disorder. When examining for specific diagnostic out- comes, hallucinations
experienced in both adolescence and adulthood were associated with a wide
range of non-psychotic illness as assessed by clinicians including alcohol and
cannabis use disorders, panic disor- der, polysubstance dependence and specific
phobia.
This study supports and extends previous research. Dominguez and
colleagues (Dominguez et al., 2011) previously reported that hallucina- tions
that were present on more than one follow-up were associated with later
psychotic disorders. Other studies have shown that hallucina- tions in
adolescents are associated with both psychotic and non-psy- chotic disorders
(Fisher et al., 2013). There is now consistent evidence of an association
between hallucinations and suicidal behaviour (DeVylder et al., 2015b;
Kelleher et al., 2013; Link et al., 2015; Martin et al., 2015). The risk of suicidal
ideation and plans and of suicide at- tempt was greater in those who had
hallucinations at both 14 and 21 years compared with those who reported
hallucinations at 14 years only. Persistent hallucinations in adolescents have been
shown to be as- sociated with suicidality in a study where the follow-up was
limited to 12 months (Martin et al., 2015). This study is consistent with the
find- ings of Fisher and colleagues (Fisher et al., 2013) who reported psychot- ic
experiences in childhood were associated with suicide attempts at 38 years.
These findings demonstrate that hallucinations occurring at more than one
point in time are predictive of a range of adult mental disor- ders (Kelleher et
al., 2012b). The relationship found with anxiety and substance use disorders
suggests that hallucinations in adolescents may be indicative of a shared
vulnerability for broader psychopatholo- gy. It may be that affective
dysregulation is the core deficit that under- lies both the experience of
hallucinations in adolescents and the risk of developing a mental disorder in
adulthood. Affective dysregulation is a key feature of affective, anxiety,
suicidality (Cisler et al., 2010) and substance use (Cheetham et al., 2010)
difficulties and has been pro- posed as forming a reciprocal relationship with
reality distortion (Van Rossum et al., 2011; Wigman et al., 2012). Further
research is needed to establish whether psychosis and affective dysregulation
are part of the same continuum or constitute separate but interacting continua
(Taylor et al., 2014; van Os, 2014).
4.1. Clinical implications
In this study, hallucinations that are experienced in early adoles- cence
and remit were found not to be associated with increased risk for adult
psychopathology. However, those whose hallucinations were present in both
adolescence and young adulthood had an increased risk of mental disorders
and suicidal ideation and behaviours at 30 years. It is important to take into
account the natural history of hal- lucinations in young people to avoid
unnecessary interventions based
5
Age 14 only Ages 14 and 21
% yes (n) OR (95% CI) % yes (n) OR (95% CI)
24.8 (61) 1.44 (0.93–2.23) 39.0 (32) 2.89 (1.66–5.02)
11.0 (27) 1.30 (0.67–2.52) 20.7 (17) 2.89 (1.37–6.09)
7.7 (19) 2.63 (1.04–6.63) 18.3 (15) 7.11 (2.68–18.83)
on the assumption that these experiences are forerunners of psychosis. However,
previous studies have shown that hallucinations in adoles- cents are associated
with serious psychopathology in the short term, in particular, increased risk of
suicidal behaviours (Kelleher et al., 2013; Martin et al., 2015) which are also
linked to distress and affective dysregulation (Yen et al., 2013). Interventions
aimed at improving dis- tress tolerance and affect regulation may ameliorate
psychopathological processes which perpetuate these difficulties and result in
mental disor- der diagnoses in adulthood (Taylor et al., 2014). Adolescents and
their families can be reassured that most who experience hallucinations will
not develop psychosis or other mental disorders later in life.
4.2. Limitations
The use of lifetime assessments of mental disorders and suicidality made it
difficult to confidently ascribe the temporal sequence of the onset of the
disorders assessed at age 30, and it is feasible that onset may have preceded
the assessments at 14 and 21 years. However, it would be unlikely that a
clinical psychotic disorder would be present prior to 14 years and there is a
tendency for respondents to favour re- cent experiences in reports of lifetime
symptoms and behaviours (Kelleher et al., 2012c). There are additional
confounding factors which may have been associated with outcomes that were
not mea- sured such as experiences of trauma and adversity. Amongst MUSP par-
ticipants experiencing hallucinations at age 14 and/or 21, the response rate was
54% which may have impacted the generalisability of our find- ings. However,
the attrition analysis found that aside from gender and maternal education, those
included in the sample did not differ by base- line variables, and also did not
differ by the severity of the reported hal- lucinations. However, those with
mental disorders in adulthood are more likely to drop out of the MUSP study,
thus our estimates may be conservative.
Finally, whilst a strength of this study was the measurement of hal-
lucinations at two time points encompassing the transition from adoles- cence to
young adulthood, these were seven years apart, which precludes assumptions
as to whether the hallucinations were persistent or recurrent. Furthermore, there
was no information regarding the fre- quency or emotional valence of the
content of the hallucinations, two characteristics reported of hallucinations in
need of treatment (Daalman et al., 2011; Johns et al., 2014). Ideally, more
measurements of hallucinations at closer time points would better inform the
course of hallucinations in adolescents and their associated adult outcomes.
5. Conclusions
Hallucinations in adolescents are common. When mental health out- comes
were assessed by clinicians using a standardised diagnostic clin- ical instrument,
those experiencing hallucinations at both 14 and 21 years had increased odds of
adult mental disorders and suicidal be- haviours. In most adolescents, the
hallucinations resolved and were not associated with increased odds of adult
psychopathology.
Conflict of interest
All authors declare no conflicts of interest.
6 M. Connell et al. / Schizophrenia Research xxx (2016) xxx–xxx
Funding/Support
This study was funded by the National Health and Medical Research Council
(NHMRC #1046216). JGS is supported by a National Health and Medical
Research Council Practitioner Fellowship Grant #1105807. AAM was funded
by the NHMRC CDF Level 2 (ID 1026598). John McGrath received John Cade
Fellowship #1056929 from the National Health and Medical Research Council.
Funders' roles
The sponsors of the study (the NHMRC) had no role in the design and
conduct of the study; collection, management, analysis and inter- pretation of
the data; and preparation, review or approval of the manuscript.
Contributors' statement
JGS and JMM planned the study. MC and JGS supervised the data col- lection.
JGS, MC and KB planned the statistical analysis which was con- ducted by KB.
MC and JGS wrote the initial drafts of the manuscript. All authors were involved
in later revisions and all authors approved the final manuscript.
Access to data
MC, JS and KB had full access to the data used in the study and take
responsibility for the integrity of the data and the accuracy of the data analysis.
Acknowledgements
The authors thank the MUSP team, MUSP participants, the Mater Misericordiae Hos- pital, and
the Schools of Social Science, Population Health and Medicine (University of Queensland).
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.schres.2016.06.009.
References
Achenbach, T.M., 1991. Manual for the Youth Self Report and 1991 Profile. University of Vermont
Department of Psychiatry, Burlington, VT.
Achenbach, T.M., 1997. Manual for the Young Adult Self-Report and Young Adult Behavior Checklist.
University of Vermont, Department of Psychiatry, Burlington, VT.
Cheetham, A., Allen, N.B., Yücel, M., Lubman, D.I., 2010. The role of affective dysregulation in drug
addiction. Clin. Psychol. Rev. 30 (6), 621–634.
Cisler, J.M., Olatunji, B.O., Feldner, M.T., Forsyth, J.P., 2010. Emotion regulation and the anxiety
disorders: an integrative review. J. Psychopathol. Behav. Assess. 32 (1), 68–82.
Daalman, K., Boks, M.P., Diederen, K.M., de Weijer, A.D., Blom, J.D., Kahn, R.S., Sommer, I.E., 2011.
The same or different? A phenomenological comparison of auditory verbal hal- lucinations in
healthy and psychotic individuals. J. Clin. Psychiatry 72 (3), 320–325.
DeVylder, J.E., Lukens, E.P., Link, B.G., Lieberman, J.A., 2015a. Suicidal ideation and suicide attempts
among adults with psychotic experiences: data from the Collaborative Psy- chiatric Epidemiology
Surveys. JAMA Psychiatry 72 (3), 219.
DeVylder, J.E., Thompson, E., Reeves, G., Schiffman, J., 2015b. Psychotic experiences as in- dicators
of suicidal ideation in a non-clinical college sample. Psychiatry Res. 226
(2– 3), 489–493.
Dhossche, D., Ferdinand, R., Van Der Ende, J., Hofstra, M.B., Verhulst, F., 2002. Diagnostic outcome
of self-reported hallucinations in a community sample of adolescents. Psychol. Med. 32 (4),
619–627.
Dominguez, M.D.G., Wichers, M., Lieb, R., Wittchen, H.-U., Van Os, J., 2011. Evidence that onset of
clinical psychosis is an outcome of progressively more persistent subclinical psychotic
experiences: an 8-year cohort study. Schizophr. Bull. 37 (1), 84–93.
First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 2002. Structured Clinical Interview for DSM-
IV-TR Axis I Disorders, Research Version, Non-patient Edition (SCID-I/NP) Revision: January
2010 ed. Biometrics Research. New York State Psychiatric Institute, New York.
Fisher, H.L., Caspi, A., Poulton, R., Meier, M.H., Houts, R., Harrington, H., Arseneault, L., Moffitt,
T.E., 2013. Specificity of childhood psychotic symptoms for predicting schizo- phrenia by 38 years of
age: a birth cohort study. Psychol. Med. 43 (10), 2077–2086. Johns, L.C., Kompus, K., Connell, M.,
Humpston, C., Lincoln, T.M., Longden, E., Preti, A., Alderson-Day, B., Badcock, J.C., Cella, M.,
Fernyhough, C., McCarthy-Jones, S., Peters, E., Raballo, A., Scott, J., Siddi, S., Sommer, I.E., Laroi, F.,
2014. Auditory verbal hallucina- tions in persons with and without a need for care. Schizophr. Bull.
40 (Suppl. 4),
S255–S264.
Kelleher, I., Harley, M., Murtagh, A., Cannon, M., 2011. Are screening instruments valid for psychotic-
like experiences? A validation study of screening questions for psychotic- like experiences
using in-depth clinical interview. Schizophr. Bull. 37 (2), 362–369.
Kelleher, I., Connor, D., Clarke, M.C., Devlin, N., Harley, M., Cannon, M., 2012a. Prevalence of
psychotic symptoms in childhood and adolescence: a systematic review and meta- analysis of
population-based studies. Psychol. Med. 42 (9), 1–7.
Kelleher, I., Keeley, H., Corcoran, P., Lynch, F., Fitzpatrick, C., Devlin, N., Molloy, C., Roddy, S.,
Clarke, M.C., Harley, M., Arseneault, L., Wasserman, C., Carli, V., Sarchiapone, M., Hoven, C.,
Wasserman, D., Cannon, M., 2012b. Clinicopathological significance of psy- chotic experiences in
non-psychotic young people: evidence from four population- based studies. Br. J. Psychiatry 201
(1), 26–32.
Kelleher, I., Lynch, F., Harley, M., Molloy, C., Roddy, S., Fitzpatrick, C., Cannon, M., 2012c.
Psychotic symptoms in Adolescence Index Risk for Suicidal Behavior: findings from 2
population-based case-control clinical interview studies. Arch. Gen. Psychiatry 69 (12), 1277–
1283.
Kelleher, I., Corcoran, P., Keeley, H., Johanna, T.W.W., Devlin, N., Ramsay, H., Wasserman, C., Carli,
V., Sarchiapone, M., Hoven, C., Wasserman, D., Cannon, M., 2013. Psychotic symptoms and
population risk for suicide attempt: a prospective cohort study. JAMA Psychiatry 70 (9), 940.
Link, B.G., Lieberman, J.A., Lukens, E.P., DeVylder, J.E., 2015. Suicidal ideation and suicide attempts
among adults with psychotic experiences: data from the Collaborative Psy- chiatric Epidemiology
Surveys. JAMA Psychiatry 72 (3), 219–225.
Martin, G., Thomas, H., Andrews, T., Hasking, P., Scott, J.G., 2015. Psychotic experiences and
psychological distress predict contemporaneous and future non-suicidal self-in- jury and suicide
attempts in a sample of Australian school-based adolescents. Psychol. Med. 45 (2), 429–437.
McGrath, J.J., Saha, S., Al-Hamzawi, A., et al., 2015. Psychotic experiences in the general population: a
cross-national analysis based on 31 261 respondents from 18 countries. JAMA Psychiatry.
Najman, J.M., Bor, W., O'Callaghan, M., Williams, G.M., Aird, R., Shuttlewood, G., 2005. Co- hort
profile: the Mater-University of Queensland Study of Pregnancy (MUSP). Int. J. Epidemiol. 34
(5), 992–997.
Poulton, R., Caspi, A., Moffitt, T.E., Cannon, M., Murray, R., Harrington, H., 2000. Children's Self-
Reported Psychotic Symptoms and Adult Schizophreniform Disorder: a 15-year longitudinal
study. Arch. Gen. Psychiatry 57 (11), 1053–1058.
Taylor, H.E., Stewart, S.L.K., Dunn, G., Parker, S., Bentall, R.P., Birchwood, M., Morrison, A.P., 2014.
Psychopathology and affect dysregulation across the continuum of psychosis: a multiple
comparison group study. Early Interv. Psychiatry 8 (3), 221–228.
van Os, J., 2014. The many continua of psychosis. JAMA Psychiatry 71 (9), 985–986. van Os, J.,
Linscott, R.J., Myin-Germeys, I., Delespaul, P., Krabbendam, L., 2009. A systemat-
ic review and meta-analysis of the psychosis continuum: evidence for a psychosis
proneness–persistence–impairment model of psychotic disorder. Psychol. Med. 39 (2), 179–
195.
Van Rossum, I., Dominguez, M.-D.-G., Lieb, R., Wittchen, H.-U., Van Os, J., 2011. Affective
dysregulation and reality distortion: a 10-year prospective study of their association and clinical
relevance. Schizophr. Bull. 37 (3), 561–571.
Welham, J., Scott, J., Williams, G., Najman, J., Bor, W., O'Callaghan, M., McGrath, J., 2009. Emotional
and behavioural antecedents of young adults who screen positive for non-affective psychosis: a 21-
year birth cohort study. Psychol. Med. 39 (4), 625–634. Wigman, J.T.W., van Nierop, M.,
Vollebergh, W.A.M., Lieb, R., Beesdo-Baum, K., Wittchen, H.-U., van Os, J., 2012. Evidence that
psychotic symptoms are prevalent in disorders of anxiety and depression, impacting on illness onset,
risk, and severity-implications for
diagnosis and ultra-high risk research. Schizophr. Bull. 38 (2), 247–257.
Yen, S., Weinstock, L.M., Andover, M.S., Sheets, E.S., Selby, E.A., Spirito, A., 2013. Prospec- tive
predictors of adolescent suicidality: 6-month post-hospitalization follow-up. Psychol. Med. 43
(5), 983–993.