Schizophrenia Research 182 (2017) 4–12

Contents lists available at ScienceDirect

Schizophrenia Research

journal homepage: www.elsevier.com/locate/schres

Experimentally induced psychosocial stress in schizophrenia spectrum

disorders: A systematic review
Claudia Lange ⁎,1, Lorenz Deutschenbaur 1, Stefan Borgwardt, Undine E. Lang, Marc Walter, Christian G. Huber
Department of Psychiatry, University of Basel, Basel, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: Background: There is evidence that exposure to social stress plays a crucial role in the onset and relapse of schizo-
Received 14 June 2016 phrenia; however, the reaction of patients with schizophrenia spectrum disorder (SSD) to experimentally in-
Received in revised form 22 September 2016 duced social stress is not yet fully understood.
Accepted 4 October 2016 Method: Original research published between January 1993 and August 2015 was included in this systematic lit-
Available online 10 October 2016
erature research. Social stress paradigms, reporting subjective responses to stress measures, plasma or saliva cor-
tisol, or heart rate (HR) in patients with SSD were included. 1528 articles were screened, 11 papers (390 patients)
Keywords:
Psychosocial stress
were included.
Schizophrenia spectrum disorder Results: Three main findings were attained concerning chronically ill patients: (1) overall similar subjective re-
HPA axis sponses to stress ratings between SDD patients and controls, (2) no group differences in cortisol response to psy-
Heart rate chosocial stress and (3) an increase in HR after the stress exposure was seen in patients and controls. The study
examining first-episode patients found higher subjective responses to stress and lower stress-induced cortisol
levels.
Conclusion: The results indicate that first-onset medication free patients may show differences in subjective re-
sponses to stress measures and cortisol release while chronically ill patients display no differences in subjective
and cortisol response. This may be the correlate of a pathophysiological dysfunction of the hypothalamic-pitui-
tary-adrenal axis prior or at the onset of SSD and a subsequent change in dysregulation during the course of
the illness. Given the paucity of studies investigating psychosocial stress in SSD and the pathophysiological rele-
vance of psychosocial stress for the illness, there is need for further research. (PROSPERO registration number:
CRD42015026525)
© 2016 Elsevier B.V. All rights reserved.

1. Introduction (Mizrahi, 2015): The neural diathesis-stress model of Walker and

Schizophrenia spectrum disorders (SSD) are major mental illnesses
that are characterized by positive and negative symptoms, as well as
cognitive impairment (Dinzeo et al., 2008). The onset and the course
of SSD are regulated by an interplay of biological factors, vulnerability,
and psychosocial influences (Meltzer et al., 2001; Norman and Malla,
1993).
There is evidence that patients with SSD are exposed to higher rates
of psychosocial stress: For example, it has been reported that patients
with SSD often live a stressful social life with limited social support
and critical family environment (Lukoff et al., 1984; van Winkel et al.,
2008).
In SSD, stress is not only a consequence of the illness (Rusch et al.,
2015), but also an important factor modulating the course of the disease
⁎ Corresponding author at: Department of Psychiatry, University of Basel, Wilhelm
Klein-Strasse 27, CH-4012 Basel, Switzerland.
E-mail address: claudia.lange@upkbs.ch (C. Lange).
1
Equal contribution.
Diforio (1997) (Walker and Diforio, 1997) assumes that stress is a trig-
ger factor for transition to psychosis in patients with a pre-existing vul-
nerability. In line with this hypothesis, a meta-analysis by Beards et al.
found about 3-fold increased odds of adverse life events in the period
prior to psychosis onset (Beards et al., 2013). Psychosocial stress can
also be regarded as a cause triggering recurrent psychotic episodes
(Mizrahi, 2015). Overall, there seem to be a coincidence of adverse life
events and the onset or exacerbation of psychotic symptoms (for review
see van Winkel et al., 2008; Dinzeo et al., 2004).
In addition, psychosocial stressors such as childhood trauma appear
to play a significant role for the development of a vulnerability for SSD:
A recent meta-analysis found that childhood trauma scores are substan-
tially higher in patients with ultra high risk for psychosis compared with
healthy controls (Kraan et al., 2015). Furthermore, the stress of mothers
during pregnancy – for example the death of the father before the
child's birth (Huttunen and Niskanen, 1978), or exposure to war during
pregnancy (Maslina et al., 2008) – can also be considered as a risk factor:
research found that prenatal maternal stress is associated with the de-
velopment of SSD in the children.

http://dx.doi.org/10.1016/j.schres.2016.10.008
0920-9964/© 2016 Elsevier B.V. All rights reserved.
 C. Lange et al. / Schizophrenia Research 182 (2017) 4–12
In addition, patients with schizophrenia report an increased feeling
of uncontrollability in these stressful events (Horan et al., 2005). Uncon-
trollability is one of the main factors in eliciting a physiological stress re-
sponse (Dickerson and Kemeny, 2004). In summary, current literature
suggests that psychosocial stress constitutes an important factor in the
clinical course of SSD, which can facilitate transition to psychosis, trigger
the onset of a psychotic episode, and can aggravate psychotic symptoms
(Mizrahi, 2015).
Psychosocial stress reactions have been intensively investigated in
experimental settings over the past decades. Inducing stress in an ex-
perimental setting is a good way to imitate real life stressors (Allen et
al., 2014). These tests serve to examine the stress response more direct-
ly, leading to subjective responses to stress, constituting a potent
challenge of the hypothalamic-pituitary-adrenal axis, and eliciting
vegetative stress reactions (Kirschbaum and Hellhammer, 1999). The
biomarkers mostly investigated in stress research are cortisol and
heart rate (HR). The hypothalamic-pituitary-adrenal axis (HPA axis) is
commonly studied as the physiological system that acts in response to
stressors of everyday life. It regulates the release of the stress hormone
cortisol (Kirschbaum and Hellhammer, 1999). Psychosocial stress
commonly leads to an increased HR and to the activation of the HPA
axis, inducing the release of significantly more cortisol compared to a
resting situation.
HPA axis abnormalities can be involved in the development of
psychiatric illnesses and are often seen in psychiatric patients. Panic dis-
order (Petrowski et al., 2012; Schreiber et al., 1996), substance use dis-
order (Walter et al., 2008; Walter et al., 2011), and depression (Claes,
2009; Holsboer et al., 1992; Horstmann and Binder, 2011) are associat-
ed with a dysregulated HPA axis function. As detailed above, neuroen-
docrinological studies have shown that a dysfunction of the HPA axis
might also play a role in the pathophysiology of schizophrenia
(Mizrahi, 2015; Walker and Diforio, 1997; Altamura et al., 1999; Guest
et al., 2011). To date, there is a narrative review examining the role of
stress in general and the HPA axis for the course of schizophrenia
(Walker et al., 2008) and covering the literature until 2007, a recent sys-
tematic review of the activity of the hypothalamic-pituitary-adrenal
axis only pertaining to first-episode psychosis (Borges et al., 2013),
and a meta-analysis comparing HPA axis activation in schizophrenia
and depression (Ciufolini et al., 2014) including only a limited subset
of three studies on SSD. However, there is no current systematic review
focused on experimentally induced psychosocial stress as an important
parameter for the development and course of SSD, examining the sub-
jective responses to stress and vegetative stress response in addition
to HPA axis functioning, and covering the broad diagnostic range of
schizophrenia-spectrum disorders.
2. Aim/Objective
This systematic review aims to give an overview on all studies pub-
lished from January 1993 to August 2015 using psychosocial stress tasks
to examine the emotional response, as well as the two most frequently
employed biomarkers of acute stress response, namely cortisol and
heart rate, in patients with schizophrenia spectrum disorders.
3. Method
For this review, we performed a computer-based systematic litera-
ture search using PubMed and Web of Knowledge. A combination of
keywords, one part pertaining to experimental induction of psychoso-
cial stress (“cortisol”, “stress”, or “HPA axis”) and one pertaining to the
group of patients with schizophrenia spectrum disorder (“schizophre-
nia”, “schizoaffective”, “psychotic” or “psychosis”) was employed
using explosion of search terms. Our literature search included original
work published from January 1993 until August 2015. The start date
was chosen as the development of the Trier Social Stress Test (TSST)
in 1993 preceded a substantial increase of research using standardized
5
experimental protocols to induce psychosocial stress. Potentially rele-
vant articles were identified, retrieved, and assessed for possible inclu-
sion in this review by author CL. The reference lists of all retrieved
articles were additionally screened for further applicable original papers
(CL). Review articles were not considered. The screening of the articles
was performed in a three-step procedure: The first step determined
whether an article might be appropriate for this review based on the
article's title and abstract. In the second step, full-text articles were ex-
amined to decide whether they should be included based on the in- and
exclusion criteria (see below). Lastly, the selected studies were screened
for potential sample overlap.
3.1. In- and exclusion criteria
Psychosocial stress tasks had to include an uncontrollable (e.g. time
pressure, unsolvable task) and/or an evaluating (e.g. speech in front of a
panel) characteristic. Tasks investigating chronic stress exposure (e.g.
caregiving) and naturalistic observations (e.g. class examinations)
were not included. To be included, studies had to examine at least one
of the following outcome parameters: subjective responses to stress,
HPA axis activity in response to psychosocial stress, or heart rate. Corti-
sol levels had to be analyzed with a standard biological assay; only re-
ports on plasma cortisol or free/saliva cortisol were used. Studies
examining HPA axis response to biological/pharmacological challenges
(e.g. DEX/CRH test) or only reporting adrenocorticotropic hormones
(ACTH) as measures of HPA axis response were excluded.
Patients had to have a DSM-IV or ICD-10 diagnosis of a schizophrenia
spectrum disorder (i.e., paranoid, disorganized, catatonic, undifferenti-
ated and residual schizophrenia; schizophreniform disorder; brief
psychotic disorder; delusional disorder; shared psychotic disorder;
schizoaffective disorder; excluding drug induced psychotic disorder).
Furthermore, studies investigating only individuals under the age of
18 were not included. We included only studies written in English.
Out of 1528 screened papers, 13 papers matched inclusion criteria
based on the article's title and abstract (cf. Fig. 1). Of these, 2 papers
were excluded after evaluation of the full-text articles: one because pa-
tients had not been diagnosed with schizophrenia spectrum disorder
(Smith and Lenzenweger, 2012), and one because the study did not
use a psychosocial stress task (Dinzeo et al., 2004). Overall, 11 papers
were included in the qualitative review.
To assess the methodological quality of the current review, we used
the PRISMA 2009 checklist (Moher et al., 2009). While seven items of
the checklist pertained to meta-analyses and were not applicable, 17
out of the remaining 20 items were fulfilled (items 1–11, 17, 18, and
24–27), indicating an overall good methodological quality. The study pro-
tocol was registered in the International Prospective Register of System-
atic Reviews (PROSPERO, n.d; registration number: CRD42015026525)
prior to the completion of data extraction.
4. Results
The main results including level of significance can be found in Table 1.
We included 11 studies examining 420 patients with SSD. Sample over-
lap was found between Brenner et al., 2009 and Brenner et al., 2011
(Brenner et al., 2009; Brenner et al., 2011), both reporting findings from
the same 30 patients and 29 controls. Overall, 7 studies investigated subjec-
tive responses to stress, 8 studies investigated cortisol and 7 studies inves-
tigated heart rate. Only one study examined first-onset medication naïve
SSDpatients(vanVenrooijetal.,2012),all otherstudiesincludedchronical-
ly ill SSD patients.
4.1. Subjective responses to psychosocial stress
To our knowledge, only one study has investigated the subjective re-
sponse to psychosocial stress in first episode, medication naïve patients
6 C. Lange et al. / Schizophrenia Research 182 (2017) 4–12
Fig. 1. Flowchart of the studies considered and included (PRISMA Flow Diagram).
with SSD. The public speaking task study by van Venrooij (2012)
showed that Spielberger anxiety scores were significantly higher in
the patient group (p = 0.002). Additionally, patients experienced less
control during the task (measured via 10-point visual analog scales
(VAS), p = 0.035). Following the task, patients felt more nervous (4
points vs. 2 points in the state trait anxiety inventory (STAI)) (van
Venrooij et al., 2012). Studies focusing on the subjective emotional re-
sponse in medicated, chronically ill SSD outpatients found that the pa-
tients and the healthy control group both experienced a similar
elevation in negative mood (Nugent et al., 2014) and a similar decrease
in positive mood (based on BPRS scores, p N 0.05) (Horan and
Blanchard, 2003). While Brenner et al. (2009) found similar subjective
responses to stress ratings between the groups (measures via VAS; 6.5
in controls and 6.6 in schizophrenia patients) following the psychosocial
stress exposure, Lincoln et al. (2015) found significantly more stressed
patients than controls (measured via 10 cm long VAS scale, p ≤ 0.01),
but no group × time effect (p = 0.18). Similar results have been found
by Jansen el al. (higher STAI scores for patients: p = 0.014 resp.
p b 0.05 but no time × group interaction p = 0.255) (Jansen et al.,
1998; Jansen et al., 2000).
4.2. HPA axis response
So far, only one study has examined the endocrinological response
to psychosocial stress in first-episode, medication naïve patients with
SSD. The cortisol level of these patients was significantly lower (p =
0.042) compared to healthy controls, possibly revealing an already im-
paired HPA axis in first-episode patients (van Venrooij et al., 2012).
Experimental studies examining psychosocial stress in patients with
chronic SSD revealed differing endocrinological results: Early studies
from one group using psychosocial stressors in patients with SSD
presented findings suggesting a blunted cortisol response (p b 0.01)
(Jansen et al., 1998; Jansen et al., 2000). Brenner et al. (2009) applied
a modified version of the Trier Social Stress Test (TSST) where the
topic of the patient's speech was changed and no recording equipment
was used. The authors only found a trend for a blunted cortisol response
in patients with SSD compared to controls (measured via area under the
curve (AUC), p = 0.062). In a later publication reevaluating these results
the authors concluded that there were no significant group differences
regarding baseline cortisol (p = 0.90) as well as AUCi (controls: 5.0
(±7.4), patients: 2.3 (±8.9); p = 0.21) (Brenner et al., 2011). This is
in line with Lincoln et al. (2015) and Steen et al. (2011), who also did
not find significant group differences (Lincoln et al., 2015; Steen et al.,
2011). It should be mentioned that the Steen and Lincoln studies were
not classical TSST studies. Yet, the factors of uncontrollability and/or so-
cial evaluation were still given and the experimental settings were
therefore considered as psychosocial stress test in this review. Only
one more recent study has found endocrinological differences between
chronically ill SSD patients and controls (p = 0.01) with lower levels for
the patients (Rubio et al., 2015), yet this study was also no classical TSST
but rather a socially evaluated cold-pressor test study, lacking the one-
on-one interaction with the participants.
4.3. Heart rate
In general, while some studies have found no differences between
patients and healthy controls for heart rate (31, 39), there is broad evi-
dence for a significantly increased HR in SSD patients (Dinzeo et al.,
2008; van Venrooij et al., 2012; Jansen et al., 1998; Rubio et al., 2015)
over all phases of illness irrespective of time of testing. Considering
the autonomic nervous system stress reaction it has been shown that
first episode, medication naïve patients with SSD (p = 0.004) (33) as
 C. Lange et al. / Schizophrenia Research 182 (2017) 4–12
well as patients with chronic SSD displayed a significantly higher heart
rate after exposure to psychosocial stress, indicating a physiological
stress reaction (all p's b 0.05) (Brenner et al., 2009; Jansen et al., 2000;
Rubio et al., 2015). Yet, some findings suggest the absence of group
differences regarding the ANS stress reaction in SSD patients
and controls (Brenner et al., 2009; van Venrooij et al., 2012; Jansen
et al., 1998; Jansen et al., 2000; Rubio et al., 2015). Only one study
by Rubio et al. (2015) also reporting a blunted cortisol response
(see above) found significantly higher HR in SSD patients than in
controls in response to stress (p = 0.01), presumably due to the use
of a different stress paradigm with the absence of a 1:1 interaction
(Rubio et al., 2015).
5. Discussion
We conducted a systematic review on original research covering ex-
perimentally induced psychosocial stress in schizophrenia spectrum
disorders, examining publications from 1993 to August 2015. Although
the stress response in patients with SSD is an increasingly active
research area, only 11 studies have investigated the subjective psycho-
social stress response, cortisol levels, or heart rate in response to exper-
imentally induced social stress since 1993.
The existing reviews and meta-analyses pertaining to our topic suf-
fer from several limitations, employing no systematic literature search
(Walker et al., 2008), not covering the current literature (Walker et al.,
2008), only pertaining to patients with schizophrenia (Walker et al.,
2008; Ciufolini et al., 2014), only covering patients with first-episode
psychosis (Borges et al., 2013), employing narrow in- and exclusion
criteria (Ciufolini et al., 2014), and only reporting HPA axis related re-
sults (Walker et al., 2008; Borges et al., 2013; Ciufolini et al., 2014). No
review was focused on experimentally induced social stress. The pres-
ent systematic review addresses these issues: literature on first-episode
as well as chronically ill patients with SSD and published until August
2015 was examined. In addition to HPA axis functioning, subjective re-
sponses to stress response and findings on heart rate were described.
Furthermore, the methodological quality of our systematic review was
confirmed by fulfilling 17 out of 20 applicable criteria of the PRISMA
checklist, and the study was registered in the PROSPERO database.
5.1. Main findings
The primary focus of this review was to examine acute subjective,
endocrinological and ANS response to experimentally induced psycho-
social stress in patients with SSD.
Considering the subjective response to psychosocial stress, this re-
view shows that first-episode, medication naïve patients with SSD expe-
rience more anxiety and less control during the task than healthy
controls. At the end of the task, patients also felt more nervous indicat-
ing that they experienced stress (van Venrooij et al., 2012). Reviewing
the subjective emotional response in medicated, chronically ill schizo-
phrenia-spectrum patients we found homogeneous results: most stud-
ies report no differences between patients with schizophrenia and
healthy controls (Brenner et al., 2009; Nugent et al., 2014; Horan et
al., 2005). One study found that patients were significantly more
stressed than controls in general, but not in respect to the psychosocial
stress task (Lincoln et al., 2015). It is noticeable that first episode, med-
ication free patients seem to be more disposed to a subjective responses
to stress reaction compared to chronically ill patients. Yet, only one
study has investigated first-episode patients thus far, and further re-
search has to show if these results can be replicated.
A similar pattern of results is evident in the HPA axis data: Endocri-
nological studies regarding psychosocial stress found a lower cortisol
response in medication-free first episode patients (van Venrooij et al.,
2012). Taking into consideration that this particular patient group dis-
plays hypercortisolism in their cortisol day profile, it is plausible that
the HPA axis is not able to additionally adapt to the new, stressful
7
situation, therefore leading to a dysfunctional cortisol response (van
Venrooij et al., 2012). One possible explanation for the HPA axis
hypoactivity in psychosocial stress tasks in first-onset patients with
SSD could be that this particular patient group seems to experience
more anxiety and less control (van Venrooij et al., 2012). Heightened
ongoing feelings of fear and anxiety could result in elevated cortisol
levels, initiating a negative feedback inhibition of cortisol at the pituitary
gland and the hypothalamus and leading to reduced HPA axis activity
(for review: Shirayama et al., 2002; Wolf, 2003). This could explain a re-
duced reactivity of the HPA axis to psychosocial stress stimulation. In
addition, this assumption is supported by data of individuals with
ultra-high risk (UHR) for psychosis. Pruessner et al. (2011) found that
individuals with UHR for psychosis are more stressed compared to
first-episode psychosis patients (Pruessner et al., 2011). Additionally,
when confronted with the TSST, UHR patients release significantly less
cortisol than the healthy controls (Pruessner et al., 2013), a pattern
we have already seen in first-episode medication naïve schizophrenia
patients (van Venrooij et al., 2012). Taken together, these data suggest
that individuals with high risk for psychosis as well as first-episode pa-
tients may show a dysregulated HPA axis activity in response to stress
compared to healthy controls. However, these assumptions are based
on a limited number of observations, and future research is needed in
this area.
Endocrinological stress response data in chronically ill patients, on
the other hand, mostly fail to show group differences between the SSD
patients and healthy controls (Brenner et al., 2011; Lincoln et al.,
2015; Steen et al., 2011). Two studies, however, found a blunted cortisol
response (Jansen et al., 1998; Jansen et al., 2000). It is noticeable that
these findings come from only one group and emerge in the earlier pub-
lications (published until 2009), while most current studies (published
between 2011 and 2015) found no statistical differences between the
groups. This could be due to improved methodology and increased sam-
ple size in the studies conducted from 2011 onwards. The publications
with non-significant results included a higher number of participants,
increasing statistical power. Alternatively, this could be an effect of pub-
lication bias, assuming that earlier studies would have had a higher
chance to get published if they could show a blunted HPA axis response,
whereas newer studies would have a better chance to get published
even with non-significant findings, as the absence of group differences
became more accepted in the literature. Although one recent study re-
ported a blunted cortisol response in SSD (Rubio et al., 2015), this
should not be interpreted as conflicting finding, as it is presumably the
result of using a psychosocial stress paradigm lacking 1:1 interaction
and differing from the experimental settings used in the other included
studies (Rubio et al., 2015). This is in line with differing findings
concerning heart rate in the same study (see below).
Another biomarker that is frequently investigated in stress research
is heart rate (HR). The studies included in this review found a signifi-
cantly increased HR in SSD patients compared to healthy controls
(Dinzeo et al., 2008; van Venrooij et al., 2012; Jansen et al., 1998;
Jansen et al., 2000; Rubio et al., 2015). Furthermore, both patients and
controls showed a significant increase in HR after exposure to psychoso-
cial stress (Brenner et al., 2009; Jansen et al., 2000; Rubio et al., 2015). In
all but one study there was no evidence for group differences regarding
stress reaction of the autonomous nervous system in SSD patients and
controls (Brenner et al., 2009; van Venrooij et al., 2012; Jansen et al.,
1998; Jansen et al., 2000; Lincoln et al., 2015), and the study reporting
significant differences used a different experimental setting which can
be assumed to be responsible for the different findings (Rubio et al.,
2015). In summary, while patients tend to have higher HR in general
and respond to psychosocial stress with a HR increase, the autonomous
nervous system (ANS) response to social stress in first-episode patients
and chronically patients with SSD seems to be similar to controls. This
suggests that psychosocial stress tasks for SSD patients are suitable to
trigger at least an autonomic response, regardless of the chronicity of
the illness.
8 C. Lange et al. / Schizophrenia Research 182 (2017) 4–12

Table 1
Summary of studies investigating experimentally induced psychosocial stress in schizophrenia spectrum disorders.

First author, Year Sample characteristics Phase of illness Tasks, outcome measure Main results, outcome, and comments

Brenner et al., 2009 N = 30 patients • Chronically ill patients • Modified TSST • No group difference in subjective stress
rating
• Cortisol baseline in patients similar to
controls (p-value not reported)
− Gender: 24 m/6 f • Subjective stress (self-rated 10-point
• Lower increase in cortisol in patients in
− Age: 30.5 ± 7.3 scale)
stress reaction (t4 – t1) (p = 0.062)
• Cortisol
• No group difference in cortisol in stress
N = 29 controls • HR
reaction (t5-t1) (p = 0.48)
• No group differences in peak of stress
− Gender: 21 m/8 f reaction (p = 0.290)
Age: 29.3 ± 8.1 • Only trend for lower overall cortisol
(AUC) in patients (p = 0.062)
• Significant HR increase at stress measure
for both groups (p-value not reported)
• No HR differences between groups
(p-value not reported)

• Controls also displayed positive

symptoms
Brenner et al., 2011 N = 30 patients • Chronically ill patients • Modified TSST • No group differences regarding baseline
cortisol (p = 0.90) and auci (p = 0.21)

− Gender: 24 m/6 f
− Age: 30.5 ± 7.3 • Cortisol

N = 29 controls

− Gender: 21 m/8 f
Age: 29.3 ± 8.1
Dinzeo et al., 2008 N = 58 patients • No information regarding • Stress/challenge paradigm: unsolvable • Patients: higher heart rate than controls
chronicity reported anagram task in general (p b 0.001)
• HR
− Gender: 48 m/10 f
− Age: 41.4 ± 8.7
• Most patients had forensic history
N = 21 controls

− Gender: 14 m/7 f
Age: 39.4 ± 10.9
Horan and N = 36 patients • Chronically ill patients • Role play task • Similar elevation in negative mood
Blanchard, 2003 • Subjective stress (36-item self-report (p N 0.05) and decrease in positive
questionnaire; COPE) mood (p N 0.05) in both groups
− Only male participants
− Age: 39.0 ± 12.2

N = 15 controls

− Only male participants

Age: 36.1 ± 7.9
Jansen et al., 1998 N = 10 patients • In- and outpatients, no in- • Public speaking task • Higher STAI scores in patients for stress
formation regarding chro- task (p = 0.014)
nicity reported • Higher von Zerssen mood scores equal-
− Only male participants
ling more negative mood in patients
− Age: 27.1 ± 7.0
(p b 0.05) except after stress task
• Subjective stress (STAI & Von Zerssen
(p b 0.10), but no significant differences
N = 10 controls mood scale)
between scores prior to and after task
• Cortisol
for patients and controls
• HR
− Only male participants • Significant test × time × group effect
Age: 26.9 ± 5.8 regarding cortisol response (p = 0.033)
with a smaller cortisol response to psy-
chosocial stress in patients
• Patients had higher heart rate irrespec-
tive of time (p = 0.025)
• No significant test × time × group effect
in HR (p = 0.967)
Jansen et al., 2000 N = 18 patients • Outpatients, no informa- • Physical (bike) and psychosocial • STAI scores higher in patients (p b 0.05)
tion regarding chronicity (speech) task but no test × group effect (p = 0.53)
reported nor time × group effect (p = 0.255)
− Gender: 11 m/7 f
• Lower cortisol response to psychosocial
− Age: 27.7 ± 4.3
(p b 0.01) but not physical task (p =
0.87) in patients
N = 21 controls
• Overall significant increase in HR in both
groups (p-value not reported)
 C. Lange et al. / Schizophrenia Research 182 (2017) 4–12 9

Table 1 (continued)

First author, Year Sample characteristics Phase of illness Tasks, outcome measure Main results, outcome, and comments

− Gender: 13 m/8 f
− Age: 27.0 ± 5.4
Lincoln et al., 2015 N = 35 patients
− Gender: 20 m/15 f
− Age: 40.5 ± 12.5
N = 28 controls
− Gender: 17 m/11 f
Age: 34.9 ± 14.4
Nugent et al., 2014 N = 65 patients
− Gender: 44 m/11 f
− Age: 36.8 ± 12.1
• Acute and remitted
patients
• Outpatients, no informa-
tion regarding chronicity
reported
• Subjective stress (STAI)
• Cortisol
• HR
• Noise and social stress task
• Subjective stress (VAS)
• Cortisol
• HR
• Psychological distress challenge
• Subjective stress (PANAS)
• No group × test interaction (p-value not
reported) but patients had overall
higher heart rate (p b 0.001)
• Patients felt significantly more stressed
(p ≤ 0.01) but no group × time effect
(p = 0.18)
• No group × condition differences re-
garding cortisol (p = 0.68)
• No group × time interaction in cortisol
(p = 0.29)
• No group × time interaction in heart
rate (p = 0.32)
• No group differences regarding heart
rate (p = 0.15)
• Patients more likely to quit task early
(p = 0.005)
• No group differences in negative affect
in response to the task

N = 43 controls

− Gender: 22 m/21 f
Age: 38.5 ± 12.8
Rubio et al., 2015 N = 58 patients
− Only male participants
− Age: 36.3 ± 6.4
N = 28 controls
− Only male participants
Age: 36.6 ± 6.3
Steen et al., 2011 N = 70 patients
− Gender: 40 m/30 f
− Age male: 25.5 ± 11.0
− Age female: 29.5 ±
15.0
• Chronically ill patients
• No information regarding
chronicity reported
• Socially evaluated cold-pressor test • Increase of cortisol in both groups
(p ≤ 0.05)
• Patients had lower cortisol levels than
controls (p = 0.01) in the stress condi-
• Cortisol
tion
• HR
• Increase of heart rate in both groups
• Patients had higher heart rates than
controls as response to stress (p =
0.01)
• Mental challenge task • Generally no group difference in cortisol
• Cortisol reaction (p = 0.06)
• Blunted cortisol response in male pa-
tients compared to male controls (p =
0.037)

N = 98 controls
− Gender: 52 m/46 f
− Age male: 32.0 ± 15.0
Age female: 31.5 ± 14.0
van Venrooij et al., N = 10 patients
2012
− Only male participants
− Age: 23 (19–29)
N = 15 controls
− Only male participants
Age: 22 (20–25)
• First-episode, medication
naïve patients
• Patient group included also bipolar dis-
order patients
• Testing in morning
• Only one baseline and one post-test cor-
tisol sample
• Public speaking task • Spielberger anxiety scores higher in pa-
tients during the task (p = 0.002)
• No VAS differences before and during
task (p-value not reported)
• Subjective stress (PSS, STAI, VAS)
• Patients more nervous after task (p =
• Cortisol
0.001)
• HR
• Patients experienced less control during
the task (p = 0.035)
• Patients: significantly lower cortisol re-
sponse to stress task (p = 0.042)
• Patients: significantly higher heart rate
at baseline (p = 0.002) and irrespective
of time (p = 0.004)
• No group × time interaction concerning
HR (p = 0.245)

Age is given in years (mean ± standard deviation, or median and range). AUC = Area under the curve, AUCi = Area under the curve in respect to increase, COPE = measured coping style,
f = female, HR = Heart Rate, m = male, PANAS = Positive and Negative Affect Schedule, PSS = Perceived Stress Scale, STAI = Spielberger State Trait Anxiety Inventory, TSST = Trier Social
Stress Test, VAS = Visual Analog Scale.

5.2. Experimental induction of psychosocial stress
Subjective, HPA axis, and vegetative response may be influenced by
the experimental paradigm chosen to exert psychosocial stress. Some of
the studies included in the current review employed the Trier Social
Stress Test, which is a well-established paradigm in biopsychological
stress research. The original TSST consists of two parts: a public speak-
ing task (a mock job interview) and a mental arithmetic task in front
of an evaluative panel with a microphone and a camera (Kirschbaum
et al., 1993). This test can lead to a 2- to 4-fold increase above baseline
in salivary cortisol (e.g. Kirschbaum et al., 1993), and the TSST shows a
response rate of above 70% for healthy participants (Kirschbaum et al.,
10 C. Lange et al. / Schizophrenia Research 182 (2017) 4–12
1993; Schommer et al., 2003). A meta-analysis concluded that the TSST
provokes the most robust cortisol responses to psychosocial stress
(Dickerson and Kemeny, 2004). Overall, the TSST causes substantial
changes in cortisol levels, mood as well as increases in HR. It is a reliable,
effective and standardized tool, which enables comparisons between
studies. Furthermore, the TSST allows variations with the purpose of
adapting to the specific needs of the participants: variations include,
e.g., a control condition (“placebo TSST”: a speaking/cognitive task com-
bination without uncontrollability and without being socially evaluative
(Het et al., 2009)), a version for children (public speaking/cognitive chal-
lenge combination where children have to finish telling a story as excit-
ing as possible followed by a subtracting task with the panel being
instructed to provide adequate positive feedback (Buske.Kirschbaum
et al., 1997)), groups (includes all elements of the original TSST in groups
of 6 participants (Von Dawans et al., 2011)), and patients with schizo-
phrenia spectrum disorder (public speaking/cognitive challenge combi-
nation where the participants are instructed to talk about their physical
appearance without the usage of recording equipment (Brenner et al.,
2009)).
In addition, several other paradigms can be used to experimentally
induce social stress. The following stress tests are used in original re-
search included in the current review not employing the original
TSST: 1) a modified TSST – a combination of a public speaking task
and a mental arithmetic task without any recording equipment. The
topic of the public speaking task is to talk about the optical appearance
(instead of a mock job interview) (Brenner et al., 2009; Brenner et al.,
2011); 2) an unsolvable anagram task – unknown to the participants,
16 out of the 20 presented anagrams are unsolvable, yet, the partici-
pants were told that they would receive money if they solve 5 or
more under time pressure (Dinzeo et al., 2008); 3) a role play task – es-
pecially designed for psychiatric patients to evaluate their capabilities to
solve interpersonal problems through dialogue (Horan and Blanchard,
2003); 4) a public speaking task – speaking in public while being re-
corded (Van Venrooij et al., 2012; Jansen et al., 2000; Jansen et al.,
1998); 5) a noise and social stress task – speaking while being recorded
under loud noise (Lincoln et al., 2015); 6) a psychological distress chal-
lenge – computerized cognitive tasks (Paced Auditory Serial Addition
Task (PASAT) or Mirror-Tracing Persistence Task (MTPT)) with a combi-
nation of a loud noise when an error was made (Nugent et al., 2014); 7)
a socially evaluated cold-pressor test – placement of a hand and wrist in
ice cold water while being recorded for facial expression (Rubio et al.,
2015); 8) a speeded mental challenge task while being watched by
the experimenter (Steen et al., 2011).
Psychosocial stress tasks trigger a cortisol response; yet, effects differ
based on the task itself. A meta-analysis of 208 laboratory stress studies
found that emotion induction and noise exposure did not provoke a sig-
nificant cortisol response, although this could have been a result of hav-
ing included only a small amount of such studies in the analysis.
Cognitive tasks, public speaking/verbal interaction tasks and public
speaking/cognitive task combination reveal significant cortisol re-
sponses. Furthermore, the public speaking task combined with solving
a cognitive challenge reached the highest effect size. Additionally, so-
cial-evaluative threat predicted these outcomes: when participants
fear to be negatively judged or when the performance is uncontrollable
(e.g. through noise or incorrect feedback), a stronger activation of the
HPA axis was provoked. Participants' cortisol level increased with the
number of social evaluative threats. Physical presence of the experi-
menter, rather than recording the task, also increases the cortisol re-
lease (Dickerson and Kemeny, 2004). In summary, not all psychosocial
stressor tasks lead to identical induction of HPA axis, subjective and
vegetative stress response, and the strongest HPA axis activation
can be achieved when using a public speaking/cognitive challenge
combination.
The repetition of psychosocial stress tests in the same participants
may lead to habituation effects, as has been shown for the TSST
(Frisch et al., 2015; Allen et al., 2014; Kudielka et al., 2009). Additionally,
further factors have a moderating effect on HPA axis response such as
age, gender (Allen et al., 2014), time of day, and time of assessment
(Dickerson and Kemeny, 2004). For example, older men may have a
higher cortisol release in response to a psychosocial stress test, men in
general release up to twice as much cortisol compared to women,
women in the follicular phase of their menstrual cycle and with intake
of oral contraceptives release significantly less cortisol, regular nicotine
consumption leads to a reduced HPA axis activity in response to stress,
and early morning testing could interfere with the cortisol awakening
response or with higher absolute cortisol levels in the morning (for re-
view see Kudielka et al., 2009).
Investigating stress related HPA axis activity in schizophrenia is a
challenge for research, as psychotic symptoms can involve the loss of
contact with reality, e.g. through false beliefs (delusions), perceptual ex-
periences not shared by others (hallucinations), and bizarre behavior
(Mueser and McGurk, 2004). Those symptoms may cause problems in
the execution of the TSST, as stress and fear might intensify the psychot-
ic symptoms, and perception problems and a focus on internal experi-
ences may decrease subjective stress. Cognitive processes and coping
strategies might also play a significant role in the stress induction in
patients with schizophrenia. Research did show that patients with
schizophrenia tend to use more passive and avoidant strategies in psy-
chosocial stress situations, which might also help to explain the stress
vulnerability of those patients (Jansen et al., 2000). Cognition, on the
other hand, also plays a crucial role, as a situation can only be considered
stressful if the individual interprets the situation accordingly, and as im-
paired cognitive functioning may increase the subjective stress experi-
enced during demanding cognitive tasks.
In order to overcome such issues, some study protocols omitted re-
cording the psychosocial stress tasks, as this may trigger paranoid
symptoms (Brenner et al., 2009). Furthermore, some research groups
have decided to exchange the job interview and rather let the partici-
pants talk about their optical appearance, as patients may not be able
to relate to a job interview situation (e.g. Brenner et al., 2009; Brenner
et al., 2011). It could therefore be speculated, that the modified TSST
may be less stressful for the participants. Yet, the studies included in
the current review employing modified TSSTs tended to exert a stronger
stress response, indicating that the changes may have no relevant effect
on the efficacy of the stress task.
5.3. Further methodological considerations
Several methodological aspects limit the results reported in the in-
cluded studies and complicate their interpretation. These issues mainly
refer to the methods of obtaining and reporting the psychobiological re-
sults and to inter-individual differences in response to stress:
Most of the studies included in this review conducted a psychosocial
stress task in medicated patients. Only one study reported results from
medication naïve patients which show that more research in this area is
needed. The duration and type of antipsychotic medication differed be-
tween the studies, and it is therefore difficult to draw conclusions on the
effect of the medication on subjective responses to stress, cortisol levels,
and heart rate. Research has shown that antipsychotic medication may
have a normalizing effect on the HPA axis activity in patients with
schizophrenia. Studies investigating basal cortisol and cortisol awaken-
ing response found a cortisol level reduction following antipsychotic
treatment (Ryan et al., 2004; Zhang et al., 2005; Popovic et al., 2007;
Mondelli et al., 2010; Venkatasubramanian et al., 2010). Differences in
antipsychotic medication might therefore account for some of the
non-significant group differences in HPA axis functioning of chronically
ill SSD patients, and for the different findings between (medication-
free) first-episode and (medicated) chronically ill SSD patients. Heart
rate and HR variability are also sensitive to antipsychotic medication.
For example, it has been found that olanzapine increased HR while thi-
oridazine decreased HR (Silke et al., 2002). Thus, differences regarding
antipsychotic medication may also have influenced the HR results
 C. Lange et al. / Schizophrenia Research 182 (2017) 4–12
reported above, e.g., the increased baseline HR and the increased HR as a
response to psychosocial stress.
As we see stress related HPA axis differences in UHR, first-episode
and chronic patients, one may also assume differences within the
chronically ill patient group based on clinical characteristics. Out of
the studies investigating psychosocial stress response, only one study
performed further analyses regarding positive and negative symptoms.
Jansen et al. (2000) used the PANSS (Positive and Negative Symptom
Scale), however, did not find an association between symptoms and
stress related cortisol release. The patients in the study were on medica-
tion, which may be reason for not finding a relationship between corti-
sol response to a stressor and symptomatology of the patients (Jansen et
al., 2000). In addition, the influence of positive and negative symptoms
on basal cortisol is also not well investigated. While some studies have
found a correlation between negative symptoms and basal cortisol
(Shirayama et al., 2002a; Zhang et al., 2005; Iancu et al., 2007), most
studies did not examine this association.
Furthermore, the discrepancy in the described results could be due
to dissimilar patient characteristics. Allowing comorbid diagnoses
(Steen et al., 2011) or including patients in varying phases of their ill-
ness (acutely ill vs. remitted patients, in- vs. outpatients) (Jansen et
al., 1998; Lincoln et al., 2015) may contribute to different findings re-
garding subjective responses to stress levels, HPA axis response and
HR. In addition, SSD and even its individual diagnostic groups (e.g.,
schizophrenia) constitute a spectrum of different disorders with differ-
ent etiology, course of illness and outcome. Differences in composition
between study samples with respect to these disorders can be assumed
to have an influence on biological correlates of psychosocial stress.
Moreover, the discrepancy in the described results could be due to
heterogeneous cortisol measures. The sampling procedures were not
standardized as one study relies on only one baseline and one post-
stress cortisol sample. This, however, does not fairly represent HPA
axis activity caused by psychosocial stress and the peak of the cortisol
response might have been missed with such a long time delay (Steen
et al., 2011).
Lastly, a publication bias concerning the studies used in this review
cannot be ruled out, for example concerning the findings for a blunted
cortisol response in chronically ill SSD patients.
5.4. Implications for future research and clinical practice
Recent consensus papers have dealt with the question on what bio-
markers could be useful as diagnostic indicators for schizophrenia spec-
trum disorders and to identify endophenotypes. Whereas biological
markers from electroencephalography (EEG), event-related potentials
(ERP), sleep monitoring, neuroimaging, and neurogenetics were incor-
porated, psychosocial stress response and HPA axis functioning were
not taken into consideration (Thibaut et al., 2015; Schmitt et al.,
2016). However, these parameters could constitute promising targets
for new biomarkers and endophenotypes, as we already see HPA axis
dysregulation in UHR (Pruessner et al., 2011). Furthermore, testing is
non-invasive (e.g., when investigating saliva cortisol), easy-to-perform,
standardized and low-priced. Yet, higher sample size and better clinical
characterization are needed in order to enable differentiation between
subgroups in further studies and to explore the potential of stress re-
sponse and HPA axis dysfunction as diagnostic markers.
The identification of endophenotypes, potentially together with
other neurophysiological, neuropsychological, neuroimaging and ge-
netic markers could help to identify more homogeneous patient sub-
groups with different treatment needs, course and outcome and
therefore also improve prognosis. Possible treatment approaches re-
garding patients with specific reactions to psychosocial stress and HPA
axis response could for example include trauma specific interventions
and improving resources to counteract social stress (e.g., through social
competence training). Paradigms experimentally inducing psychosocial
stress could be used to choose suitable patient subgroup for these
11
interventions, and to enable prediction and monitoring of treatment
response.
5.5. Conclusion and suggestion for future research
Taken together, these results indicate that first-onset patients felt
higher subjective response to stress during and after a stress task and re-
leased less cortisol in response to a psychosocial stress task compared to
the healthy controls. Yet, only one study has investigated these issues in
first-onset medication free patients. Chronically ill patients, on the other
hand, showed no differences in the subjective and cortisol response
compared to healthy controls. First-episode and chronically ill patients
showed a similar heart rate response than healthy controls. A possible
pathophysiological overdrive of the HPA axis in prodromal, UHR and
first-episode schizophrenia spectrum patients and a subsequent dysreg-
ulation during the course of the illness should be further investigated.
Future studies should include and control for factors that have been
associated with HPA axis abnormalities, such as comorbid disorders and
phase of illness. Studies regarding the influence of antipsychotic medi-
cation of the psychosocial stress response are needed, especially studies
in medication-free and medicated first-episode patients. This would
allow a more detailed investigation of the effects of medication on the
HPA axis reactivity.
Considering the small number of studies examining the effects of ex-
perimentally induced psychosocial stress in schizophrenia spectrum
disorder and the pathophysiological relevance of psychosocial stress
for the course of illness, future research in this area is encouraged.
Role of funding
All authors received regular salaries through their primary employ-
er. None of the authors have received special funding for activities relat-
ed to the preparation and publication of this review.
Contributions
CL conducted the literature research and wrote the first draft of the
manuscript. CL, LD, MW, and CGH participated in the discussion and in-
terpretation of the findings. CGH revised the manuscript providing im-
portant intellectual content. LD, MW, SB, and UEL contributed during
literature research and during the writing process. All authors contrib-
uted to and have approved the final manuscript.
Conflict of interest
All authors declare that they have no conflict of interest related to this work.
Acknowledgement
N/A.
References
Allen, A.P., Kennedy, P.J., Cryan, J.F., Dinan, T.G., Clarke, G., 2014. Biological and psycholog-
ical markers of stress in humans: focus on the trier social stress test. Neurosci.
Biobehav. Rev. 38, 94–124.
Altamura, A.C., Boin, F., Maes, M., 1999. HPA axis and cytokines dysregulation in schizo-
phrenia: potential implications for the antipsychotic treatment. Eur.
Neuropsychopharmacol. 10 (1), 1–4.
Beards, S., Gayer-Anderson, C., Borges, S., Dewey, M.E., Isher, H.L., Morgan, C., 2013. Life
events and psychosis: a review and meta-analysis. Envior Schizo 39 (4), 740–747.
Borges, S., Gayer-Anderson, C., Mondelli, V., 2013. A systematic review of the activity
of the hypothalamic-pituitary-adrenal axis in first episode psychosis.
Psychoneuroendocrinology 38 (5), 603–611.
Brenner, K., Liu, A., Laplante, D.P., et al., 2009. Cortisol response to a psychosocial stressor
in schizophrenia: blunted, delayed, or normal? Psychoneuroendocrinology 34,
859–868.
Brenner, K., St-Hilaire, A., Liu, A., Laplante, D.P., King, S., 2011. Cortisol response and cop-
ing style predict quality of life in schizophrenia. Schizophr. Res. 128, 23–29.
Buske.Kirschbaum, A., Jobst, S., Wustmans, A., Kirschbaum, C., Rauh, W., Hellhammer, D.,
1997. Attenuated free cortisol response to psychsocial stress in children with atopic
dermatitis. Psychosom. Med. 59, 419–426.
12 C. Lange et al. / Schizophrenia Research 182 (2017) 4–12
Ciufolini, S., Dazzan, P., Kempton, M.J., Pariante, C., Mondelli, V., 2014. HPA axis response
to social stress is attenuated in schizophrenia but normal in depression: evidence
from a meta-analysis of existing studies. Neurosci. Biobehav. Rev. 47, 359–368.
Claes, S., 2009. Glucocorticoid receptor polymorphisms in major depression. Ann. N. Y.
Acad. Sci. 1179 (1), 216–228.
Dickerson, S.S., Kemeny, M.E., 2004. Acute stressors and cortisol responses: a theoretical
integration and synthesis of laboratory research. Psychol. Bull. 130, 355–391.
Dinzeo, T.J., Cohen, A.S., Nienow, T.M., Docherty, N.M., 2004. Stress and arousability in
schizophrenia. Schizophr. Res. 71 (1), 127–135.
Dinzeo, T.J., Cohen, A.S., Nienow, T.M., Docherty, N.M., 2008. Arousability in schizophre-
nia: relationship to emotional and physiological reactivity and symptom severity.
Acta Psychiatr. Scand. 117 (6), 432–439.
Frisch, J.U., Häusser, J.A., Mojzisch, A., 2015. The trier social stress test as a pardigm to
study how people respond to thereat in social interactions. Front. Psychol. 6 (14),
1–15.
Guest, P.C., Schwarz, E., Krishnamurthy, D., et al., 2011. Altered levels of circulating insulin
and other neuroendocrine hormones associated with the onset of schizophrenia.
Psychoneuroendocrinology 36 (7), 1092–1096.
Het, S., Rohleder, N., Schoofs, D., Kirschbaum, C., Wolf, O.T., 2009. Neuroendocrine and
psychometric evaluation of a placebo version of the “Trier Social Stress Test. Psychi-
atry 34, 1075–1086.
Holsboer, F., Spengler, D., Heuser, I., 1992. The role of corticotropin-releasing hormone in
the pathogenesis of Cushing's disease, anorexia nervosa, alcoholism, affective disor-
ders and dementia. Prog. Brain Res. 93, 385–417.
Horan, W.P., Blanchard, J.J., 2003. Emotional responses to psychosocial stress in schizo-
phrenia: the role of individual differences in affective traits and coping. Schizophr.
Res. 60, 271–281.
Horan, W.P., Ventura, J., Nuechterlein, K.H., Subotnik, K.L., Hwang, S.S., Mintz, J., 2005.
Stressful life events in recent-onset schizophrenia: reduced frequencies and altered
subjective appraisals. Schizophr. Res. 75, 363–374.
Horstmann, S., Binder, E.B., 2011. Glucocorticoids as predictors of treatment response in
depression. Harv Rev Psychiatry 19 (3), 125–143.
Huttunen, M.O., Niskanen, P., 1978. Prenatal loss of father and psychiatric disorders. Arch.
Gen. Psychiatry 35, 429–431.
Iancu, I., Tchernihovsky, E., Maayan, R., Poreh, A., Dannon, P., Kotler, M., Weizman, A.,
Strous, R.D., 2007. Circulatory neurosteroid levels in smoking and non-smoking
chronic schizophrenia patients. Eur. Neuropsychopharmacol. 17, 541–545.
Jansen, L.M., Gispen-de Wied, C.C., Gademan, P.J., De Jonge, R.C., van der Linden, J.A., Kahn,
R.S., 1998. Blunted cortisol response to a psychosocial stressor in schizophrenia.
Schizophr. Res. 33 (1–2), 87–94.
Jansen, L.M., Gispen-de Wied, C.C., Kahn, R.S., 2000. Selective impairments in the stress re-
sponse in schizophrenic patients. Psychopharmacology (Berlin) 149 (3), 319–325.
Kirschbaum, C., Hellhammer, D.H., 1999. Noise and stress - salivary cortisol as a non-inva-
sive measure of allostatic load. Noise Health 1 (4), 57–66.
Kirschbaum, C., Strasburger, C.J., Langkrär, J., 1993. Attenuated cortisol response to psy-
chological stress but not to CRH or ergometry in young habitual smokers. Pharmacol.
Biochem. Behav. 44, 527–531.
Kraan, T., Velthorst, E., Smith, F., de Haan, L., van der Gaag, M., 2015. Trauma and recent
life events in individuals at ultra high risk for psychosis: review and meta-analysis.
Schizophr. Res. 161 (2-3), 143–149.
Kudielka, B.M., Hellhammer, D.H., Wüst, S., 2009. Why do we spond so diffenerently?
Reviewing determinants of human salivary cortisol resonses to challenge.
Psychoneuro 34, 2–18.
Lincoln, T.M., Köther, U., Hartmann, M., Kempkensteffen, J., Moritz, S., 2015. Responses to
stress in patients with psychotic disorders compared to persons with varying levels
of vulnerability to psychosis, persons with depression and healthy controls. J Behav
Ther & Exp Psychiat 47, 92–101.
Lukoff, D., Snyder, K., Ventura, J., Nuechterlein, K.H., 1984. Life events, familial stress, and
coping in the developmental course of schizophrenia. Schizophr. Bull. 10 (2),
258–292.
Maslina, D., Corcoran, C., Kleinhaus, K.R., Perrin, M.C., Fennig, S., Nahon, D., Friedlander, Y.,
Harlap, S., 2008. Acute maternal stress in pregnancy and schizopghrenia in offspring:
a cohort prospective study. BMC Psychiatry 2008, 1–9.
Meltzer, H.Y., Lee, M.A., Jayathilake, K., 2001. The blunted plasma cortisol response to
Apomorphine and its relationship to treatment response in patients with schizophre-
nia. Neuropsychopharmacology 24 (3), 278–290.
Mizrahi, R., 2015. Social stress and psychosis risk: common neurochemical substrates?
[review]. Neuropsychopharmacology.
Moher, D., Liberati, A., Tetzlaff, J., Altman, D.G., The PRISMA Group, 2009. Preferred
reporting items for systematic reviews and meta-analyses: the PRISMA statement.
PLoS Med. 6 (6), e1000097. http://dx.doi.org/10.1371/journal.pmed1000097.
Mondelli, V., Dazzan, P., Hepgul, N., et al., 2010. Abnormal cortisol levels during the day
and cortisol awakening response in first-episode psychosis: the role of stress and of
antipsychotic treatment. Schizophr. Res. 116, 234–242.
Mueser, K.T., McGurk, S.R., 2004. Schizophrenia. Lancet 363, 2063–2072.
Norman, R.M., Malla, A.K., 1993. Stressful life events and schizophrenia. I: a review of the
research. Br. J. Psychiatry 162, 161–166.
Nugent, K.L., Chiappelli, J., Rowland, L.M., Daughters, S.B., Hong, L.E., 2014. Distress intol-
erance and clinical functioning in persons with schizophrenia. Psychiatry Res. 220,
31–33.
Petrowski, K., Wintermann, G.B., Kirschbaum, C., Bornstein, S.R., 2012. Dissociation be-
tween ACTH and cortisol response in DEX–CRH test in patients with panic disorder.
Psychoneuroendocrinology 37 (8), 1199–1208.
Popovic, V., Doknic, M., Maric, N., et al., 2007. Changes in neuroendocrine and metabolic
hormones induced by atypical antipsychotics in normal-weight patients with schizo-
phrenia. Neuroendocrinology 85, 249–256.
PROSPERO, n.d. International Prospective Register of Systematic Reviewshttp://www.crd.
york.ac.uk/prospero/display_record.asp?ID=CRD42015026525 (status: 11/05/2015).
Pruessner, M., Iyer, S.N., Faridi, K., Joober, R., Malla, A.K., 2011. Stress and protective factors
in individuals at ultra-high risk for psychosis, first episode psychosis and healthy con-
trols. Schizophr. Res. 129, 29–35.
Pruessner, M., Vracotas, N., Joober, R., Pruessner, J.C., Malla, A.K., 2013. Blunted cortisol
awakening response in men with first episode psychosis: relationship to parental
bonding. Psychoneuroendocrinology 38, 229–240.
Rubio, G., López-Muñoz, F., Jurado-Barba, R., et al., 2015. Stress induced by the socially
evaluated cold-pressor test cause equivalent deficiencies of sensory gating in male
subjects with schizophrenia and healthy controls. Psychiatry Res. 228 (3), 281–288.
Rusch, N., Heekeren, K., Theodoridou, A., et al., 2015. Stigma as a stressor and transition to
schizophrenia after one year among young people at risk of psychosis. Schizophr. Res.
166 (1–3), 43–48.
Ryan, M.C., Sharifi, N., Condren, R., Thakore, J.H., 2004. Evidence of basal pituitary-adrenal
overactivity in first episode, drug naïve patients with schizophrenia.
Psychoneuroendocrinology 29 (8), 1065–1070.
Schmitt, A., Rujescu, D., Gawlik, M., Hasan, A., Hashimoto, K., Iceta, S., Jarema, M.,
Kambeitz, J., Kasper, S., Keeser, D., Kornhuber, J., Koutsouleris, N., Lanzenberger, R.,
Malchow, B., Saoud, M., Spies, M., Stöber, G., Thibaut, F., Riederer, P., Falkai, P.,
WFSBP Task Force on Biological Markers, 2016. Concensus paper of the WFSBP task
force on biological markers: criteria for biomarkers and endophenotypes of schizo-
phrenia part II: cognition, neuroimaging and genetics. World J. Biol. Psychiatry 17,
406–428.
Schommer, N.C., Hellhammer, D.H., Kirschbaum, C., 2003. Dissociation between reactivity
of the hypothalamus-pituitary-adrenal axis and the sympathetic-adrenal-medullary
system to repeated psychosocial stress. Psychosom. Med. 65 (3), 450–460.
Schreiber, W., Lauer, C.J., Krumrey, K., Holsboer, F., Krieg, J.C., 1996. Dysregulation
of the hypothalamic-pituitary-adrenocortical system in panic disorder.
Neuropsychopharmacology 15 (1), 7–15.
Shirayama, Y., Hashimoto, K., Suzuki, Y., Higuchi, T., 2002. Correlation of plasma
neurosteroid levels to the severity of negative symptoms in male patients with
schizophrenia. Schizophr. Res. 58 (1), 69–74.
Silke, B., Campbell, C., King, D.J., 2002. The potential cardiotoxicity of antipsychotic drugs
as assessed by heart rate variability. J. Psychopharmacol. 16 (4), 355–360.
Smith, N.T., Lenzenweger, M.F., 2012. Increased stress responsivity in Schizotypy leads to
diminished spatial working memory performance. PD: TRT 4 (4), 324–331.
Steen, N.E., Lorentzen, S., Barrett, E.A., et al., 2011. Sex-specific cortisol levels in bipolar
disorder and schizophrenia during mental challenge- relationship to clinical charac-
teristics and medication. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 35 (4),
1100–1107.
Thibaut, F., Boutros, N.N., Jarema, M., Oranje, B., Hasan, A., Daskalakis, Z.F., Wichniak, A.,
Schmitt, A., Riederer, P., Falkai, P., WFSBP Task Force on Biological Markers, 2015.
Concensus paper of the WFSBP task force on biological markers: criteria for bio-
markers and endophenotypes of schizophrenia part I: neurophysiology. World
J. Biol. Psychiatry 16, 280–290.
van Venrooij, J.A.E.M., Fluitman, S.B.A.H.A., Lijmer, J.G., et al., 2012. Impaired neuroendo-
crine and immune response to acute stress in medication-naive patients with a first
episode of psychosis. Schizophr. Bull. 38 (2), 272–279.
van Winkel, R., Stefanis, N.C., Myin-Germeys, I., 2008. Psychosocial stress and psychosis. A
review of the neurobiological mechanisms and the evidence for gene-stress interac-
tion. Schizophr. Bull. 34 (6), 1095–1105.
Venkatasubramanian, G., Chittiprol, S., Neelakantachar, N., Shetty, T., Gangadhar, B.N.,
2010. Effect of antipsychotic treatment on insulin-like growth factor-1 and cortisol
in schizophrenia: a longitudinal study. Schizophr. Res. 119 (1–3), 131–137.
Von Dawans, B., Kirschbaum, C., Heinrichs, M., 2011. The trier social stress test for groups
(TSST-G): a new research tool for controlled simultaneous social stress exposure in a
group format. Psychoneuro 36, 514–522.
Walker, E.F., Diforio, D., 1997. Schizophrenia: a neural diathesis-stress model. Psychol.
Rev. 104 (4), 667.
Walker, E., Mittal, V., Tessner, K., 2008. Stress and the hypothalamic pituitary adrenal axis
in the developmental course of schizophrenia. Annu. Rev. Clin. Psychol. 4, 189–216.
Walter, M., Wiesbeck, G.A., Bloch, N., et al., 2008. Psychobiological responses to drug cues
before and after methadone intake in heroin-dependent patients: a pilot study. Eur.
Neuropsychopharmacol. 18 (5), 390–393.
Walter, M., Wiesbeck, G.A., Degen, B., et al., 2011. Heroin reduces startle and cortisol re-
sponse in opioid-maintained heroin-dependent patients. Addict. Biol. 16 (1),
145–151.
Wolf, O.T., 2003. HPA axis and memory. Best Pract. Res. Clin. Endocrinol. Metab. 17 (2),
287–299.
Zhang, X.Y., Zhou, D.F., Cao, L.Y., Wu, G.Y., Shen, Y.C., 2005. Cortisol and cytokines in
chronic and treatment-resistant patients with schizophrenia: association with psy-
chopathology and response to antipsychotics. Neuropsychopharmacology 30,
1532–1538.