Psychiatry Research 312 (2022) 114526

Contents lists available at ScienceDirect

Psychiatry Research
journal homepage: www.elsevier.com/locate/psychres

Patient-reported exposures and outcomes link the gut-brain axis and

inflammatory pathways to specific symptoms of severe mental illness
Sarah J Fendrich a, b, #, Lauren R Koralnik c, #, Mharisi Bonner a, Deborah Goetz a, Peter Joe a,
Jakleen Lee a, Bridget Mueller a, Jessica Robinson-Papp a, Oded Gonen d, Jose C. Clemente e,
Dolores Malaspina a, *
a
Departments of Psychiatry, Neuroscience, and Genetics and Genomics, Icahn School of Medicine at Mount Sinai, New York, New York, USA
b
Department of Medical Ethics and Health Policy, Perelman School of Medicine, Philadelphia, Pennsylvania, USA
c
Department of Psychology, Barnard College of Columbia University, New York, New York, USA
d
Center for Advanced Imaging Innovation and Research (CAI2R), Department of Radiology, New York University Grossman School of Medicine, New York, NY 10016,
USA
e
Department of Genetics and Genomic Sciences, Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, New York, USA

A B S T R A C T

We developed a “gut-brain-axis questionnaire” (GBAQ) to obtain standardized person-specific “review of systems” data for microbiome-gut-brain-axis studies. In­
dividual items were compared to PANSS symptom measures using dimensional, transdiagnostic and traditional categorical approaches.
Method: Forty psychotic participants, independent of diagnoses, and 42 without psychosis (18 nonpsychotic affective disorders, 24 healthy controls) completed the
GBAQ and underwent research diagnostic and symptom assessments. The PANSS scales and its dysphoric mood, autistic preoccupation and activation factors were
computed.
Results: Transdiagnostic analyses robustly linked psychosis severity to constipation (p<.001), and Negative (p=.045) and General Psychopathology scores (p=.016)
with bowel hypomotility. Activation factor scores predicted numbers of psychiatric (p=.009) and medical conditions (p=.003), BMI (p=.003), skin (p<.001) and
other conditions. Categorical analyses comparing psychotic, nonpsychotic and control groups revealed behavioral differences: cigarette smoking (p=.013), alcohol
use (p=.007), diet (p’s <.05), exercise (p<.001). All subjects accurately self-reported their diagnosis.
Conclusions: The GBAQ is a promising tool. Transdiagnostic analyses associated psychotic symptoms to gut hypomotility, indicative of low gut vagal tone, consistent
with reduced cardiovagal activity in psychosis. Activation, similar to delirium symptoms, predicted medical comorbidity and systemic inflammatory conditions.
Group level comparisons only showed behavioral differences. Underpinnings of psychiatric disorders may include reduced gut vagal function, producing psychosis,
and systemic inflammation, impacting risks for psychotic and nonpsychotic conditions.

1. Introduction et al., 2017; Bahorik et al., 2017). Psychosis is unlikely to be the core
pathology, as socioemotional deficits, known as negative symptoms, are
1.1. xxxx better predictors of outcome (Rathnaiah et al., 2020; Strauss et al,
2016). Antipsychotic medications only target psychotic symptoms,
More than 13 million Americans had a serious psychiatric illness in including hallucinations and delusions, which are referred to as positive
2019, about 5.2% of US adults, with a higher prevalence for women symptoms, improving these in fewer than half of cases without reme­
(6.5%) than men (3.9%) (McCance-Katz 2019), with increasing rates in diating core negative and cognitive symptoms (Elkis, Buckley 2017;
2020 attributed to COVID (Substance Abuse and Mental Health Services Leucht 2017).
Administration). The greatest functional impairments are among those Knowledge of the factors mediating inflammation may lead to
with schizophrenia and related psychoses (Nordstroem et al., 2017; improved prevention and treatment approaches for psychosis (Mueller
Silberstein, Harvey 2019). The underpinnings of psychotic conditions 2018; Bishop et al., 2022). Inflammation is modulated to a great extent
remain enigmatic, although the association with medical comorbidity by the gut microbiome, and an aberrant microbiome (dysbiosis) is
and premature mortality suggest inflammatory underpinnings (Perry associated with food intolerance and chronic, low-grade inflammation

* Corresponding author at: Professor of Psychiatry, Neuroscience, Genetics & Genomics, Icahn School of Medicine at Mount Sinai, New York, New York, USA
E-mail address: dolores.malaspina@gmail.com (D. Malaspina).
#
Shared first authors

https://doi.org/10.1016/j.psychres.2022.114526
Received 26 January 2022; Received in revised form 16 March 2022; Accepted 20 March 2022
Available online 22 March 2022
0165-1781/© 2022 Elsevier B.V. All rights reserved.

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.
S.J. Fendrich et al. Psychiatry Research 312 (2022) 114526

in persons with psychiatric disorders (Caputi et al., 2021; Severance 2. Method

et al., 2015). The initial gut microbiota is optimally seeded at birth by
vaginal delivery, which plays a central role in developing the brain and 2.1. Participants and procedures
immune system and is relevant to schizophrenia (Hoffmann et al 2020).
Throughout life the microbiome impacts brain function and behavior Subjects: Participants were recruited from clinical treatment set­
through the microbiome-gut-brain axis (MGBA), which also includes the tings in the Mount Sinai Health System, or from postings in the hospital
vagal nerve, circulating cytokines and microbiome metabolites (Liu et or internet to participate in a study examining the MGBA and hippo­
al, 2022). While the microbiota composition is influenced by many life campal inflammation. Persons were included who had a psychotic dis­
course exposures, including diet, infections, antibiotics and other med­ order, independent of DSM-5 diagnosis, including any schizophrenia
ications and other factors (Rinninella et al., 2019; Clemente et al., 2018; related psychosis and psychotic bipolar disorder, or nonpsychotic af­
Carabotti et al., 2015), this information is not systematically assessed in fective disorders (psychiatric controls), or having no psychiatric disor­
most microbiome studies. der or personal or family history of psychosis (healthy controls). All
The need for a standardized self-report questionnaire to serve as a spoke English and were 18 – 55 years of age. Exclusion criteria included
review of systems for person-specific information on factors related to active major medical or neurological disorders, traumatic brain injury,
the microbiome-gut-brain-axis was evident as we commenced a study of significant risk of harm to self and others or metal in their body that
hippocampal inflammation and the microbiome in psychosis. We precluded imaging. The study was approved by the Icahn School of
developed a Gut-Brain-Axis Questionnaire (GBAQ) which is the focus of Medicine Institutional Review Board, all subjects signed informed con­
this report. The Research Domain Criteria (RDoC) framework for this sent and were paid by the hour for participation.
study is relevant for psychosis research (Cuthbert et al., 2021). The
National Institute of Mental Health (NIMH) launched the Research 2.2. Assessments
Domain Criteria RDoC in 2009 as an innovative way to define mental
symptoms across diagnostic groups that might better reveal their de­ Participants completed the self-report GBAQ (Supplementary Table
terminants than is accomplished using categorical approaches, such as 1), which included items on diet, lifestyle, and medical history, and stool
the DSM-5 (Insel et al 2010). The multilevel framework for our RDoC descriptions from the Bristol Stool Scale (Lewis, Heaton 1997). All had
approach includes observable symptoms and self-reports, herein comprehensive face-to-face clinical research assessments performed by
described, as well as microbiota, neurobiological and other measures. trained raters, having a Masters, MD or PhD degree or equivalent psy­
To our knowledge the GBAQ is the first comprehensive self-report chiatric assessment experience, with interrater reliability of κ = 0.95 for
questionnaire to probe a broad set of health measures associated with psychiatric diagnosis and κ = 0.80 for individual symptoms. Research
the microbiome or inflammation in psychiatric samples. The items span diagnostic interviews were conducted using the Diagnostic Interview for
four domains: conditions and allergies, health factors, diet, and lifestyle Genetic Studies (DIGS) (Nurnberger et al., 1994). Current psychiatric
(Supplementary Table 1). A number of the items are directly associated symptoms were assessed with Positive and Negative Syndrome Scale
with psychiatric comorbidity, including atopic and allergy conditions, as (Kay et al., 1987), which rates the severity of 30 different symptom
heightened sensitivity to benign antigens is associated with a range of items on a 7-point scale (1 = absent, 7 = extreme) to yield positive,
mental health outcomes (Kelly et al., 2019; Yaghmaie et al., 2013). Food negative, general psychopathology subscales and a total score. In addi­
sensitivities can be a consequence of dysbiosis and are associated with tion to the original subscales, the PANSS 5-factor solution (White et al.,
increased gut permeability, which is well- associated with systemic and 1997) was applied to the items to yield dysphoric mood, autistic pre­
central nervous system inflammation, especially gluten insensitivity occupation and activation factors for a finer resolution of symptom
(Daulatzai et al., 2015). Also queried are bowel health, perinatal expe­ clusters. (Supplementary Table 2)
riences, medications, and use of supplements. Constipation is common Data Analysis: This study examined responses to the GBAQ self-
in psychiatric cases, and is associated with imbalanced intestinal flora report items with respect to symptom severity, first across all partici­
and the use of anticholinergic medications (Shirazi et al, 2016; Janique pants in the transdiagnostic approach of the RDoC criteria. Next, with
et al 2016). Mode of birth delivery (C-section versus Vaginal delivery) participants classified into a psychotic group, independent of DSM-5
and infant nutrition seed the newborn’s gut microbiome impacting diagnoses; a nonpsychotic affective disorder group; or healthy com­
neurodevelopment and immune reactivity (Hoffman et al, 2020). Active parison group and compared across items. Data analysis provided
ingredients in some medications can reduce microbial diversity and descriptive statistics of demographic and self-reported information in
cause dysbiosis (Brusselaers et al., 2019). Diet is an important health the overall sample and by groups. A series of logistic regression analyses
factor, and chronically poor diets increase the risk for gut dysbiosis, examined the relationship between each self-report item from the GBAQ
through which the gut mucosal barrier can be perturbed to produce a and group classification. In two sets of models, the psychosis and
hyperpermeable “leaky gut” that permits interactions of gut microbiota nonpsychotic affective groups were first compared to controls by
and inflammatory molecules with the immune system and brain regressing on the binary outcome of a group (psychotic or nonpsychotic
(Kaplan et al., 2015). Other factors linked to the health of the micro­ affective) versus no diagnosis. In each of these analyses, the non-
biome include weight, exercise habits, oral health, and substance use relevant psychiatric group was excluded. To examine factors specif­
(John, Mullin, 2016; Donoso et al., 2022). ically related to psychosis, a series of logistic regression models was
constructed, comparing the psychosis group to nonpsychotic affective
1.2. Study aims group, excluding controls. Differences in symptom severity between the
psychosis and nonpsychotic affective groups were examined using un­
This report describes the associations between GBAQ items, to be paired, two-sided t-tests. To identify correlates of each domain of
used as person-specific data for microbiome-gut-brain axis (MGBA) symptomatology, a series of single-predictor linear regression models
studies, with PANSS derived symptoms in a cohort of subjects recruited were constructed based on demographic and self-reported information
for a MGBA study. In a transdiagnostic approach, psychosis-related from the GBAQ.
symptoms were dimensionally examined with the self-report items
across traditional categorical disease boundaries. Next, categorical an­ 3. Results
alyses compared items and symptoms across participants grouped as
psychotic cases, nonpsychotic affective disorders and healthy controls. Contingency tables tabulate the frequency of participants’ GBAQ
responses across all cases and stratified by the psychiatric group clas­
sifications in Supplementary Tables 3-8. Associations between self-

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.
S.J. Fendrich et al. Psychiatry Research 312 (2022) 114526

reported health factors and symptoms based on logistic regression are Table 2
included in Supplementary Tables 9-12. Self-reported mental health conditions by psychiatric diagnosis.
Overall Psychosis Nonpsychotic Controls
3.1. Sample description and self-reported psychiatric diagnosis Cases Cases
(N = 82) (N = 40) (N = 18) (N = 24)
Variables n (%) n (%) n (%) n (%)
The participant cohort (n = 82) had slightly more women (n = 46,
Condition
56%), with a mean age of 37.1 (SD 10.8) years, including 40 subjects
Schizophrenia 29 29 (72.5) 0 (0.0) 0 (0.0)
with psychotic conditions and 42 nonpsychotic participants (18 (35.4)
nonpsychotic affective cases, and 24 with no mental illness). The groups Bipolar Disorder 21 16 (40.0) 5 (27.8) 0 (0.0)
were similar in sex composition, age, and body mass index (Table 1). (25.6)
Participants self-reported psychiatric diagnoses from the GBAQ reliably Depression 19 5 (12.5) 13 (72.2) 1 (4.2)
(23.2)
accorded with research diagnoses; with the former group all self-
ADD 12 5 (12.5) 5 (27.8) 2 (8.4)
reporting schizophrenia, bipolar disorder or schizoaffective disorder (14.6)
and the latter group reporting diagnoses of depression or bipolar dis­ ASD 2 (2.4) 0 (0.0) 2 (11.1) 0 (0.0)
order, coinciding with the best-estimate diagnoses (Table 2). Many re­ PTSD 2 (2.4) 0 (0.0) 1 (5.6) 1 (4.2)
Substance Abuse 3 (3.7) 1 (2.5) 1 (5.6) 1 (4.2)
ported additional psychiatric comorbidities, including attention deficit
Alzheimer’s 1 (1.2) 1 (2.5) 0 (0.0) 0 (0.0)
disorder (ADD), autistic spectrum disorder, post-traumatic stress disor­ Disease
der (PTSD) or substance abuse. Three controls (12.5%) reported having Number of
a mental illness, one of whom reported three conditions, including Conditions
depression, ADD, PTSD, and substance abuse, which were subthreshold Zero 21 0 (0.0) 0 (0.0) 21
(25.6) (87.5)
with respect to the research diagnoses.
One 40 27 (60.0) 11 (61.1) 2 (8.4)
(48.8)
3.2. Symptom ratings comparing the psychotic and nonpsychotic affective Two 16 11 (27.5) 5 (27.8) 0 (0.0)
(19.5)
disorder groups
Three 5 (6.1) 2 (5.0) 2 (11.1) 1 (4.2)

Table 3 summarizes symptom severity ratings by psychiatric diag­ Abbreviations: ADD = Attention Deficit Disorder, ASD = Autism Spectrum
nosis. The psychotic and nonpsychotic affective cases had similar Disorder, PTSD = Post-Traumatic Stress Disorder.
negative symptom, activation and general psychopathology scores (p >
0.05),. The psychotic group displayed significantly higher positive and General Psychopathology scores (p=.016).
autistic preoccupation (p’s<0.001) and total symptoms scores Diet: Self-reported Vitamin D supplementation was associated with
(p=0.006). Conversely the nonpsychotic affective cases had higher higher total PANSS scores (p=.014) and higher general psychopathology
dysphoria scores (p=.023). scores (p=.003). Gluten allergy (including Celiac disease, gluten intol­
erance and wheat intolerance) predicted higher total PANSS (p=.04)
and general psychopathology scores (p=.023). Compared to no food
3.3. Dimensional assessments of self-report items with symptoms across
allergies, a single food allergy was associated with lower total PANSS
all cases
(p=.03) and Negative Symptom scores (p=.033), although those
reporting two food allergies tended to exhibit higher general psycho­
Quantitative symptom severity and self-report items across all cases
(Tables 4 and 5), pathology scores (p=.051). Total PANSS scores were 8 fold lower for
those eating fruit 5 to 7 days a week compared to no fruit p = 0.05).
Demographics: Women had less negative symptom severity than men
Total PANSS and general psychopathology scores were also lower for
(p=.022), and similarly lower scores on the autistic preoccupation factor
those consuming sugary sweets 3 to 7 days a week compared to never
(p=.002).
(-p= 0.043 and p= 0.023).
Health Factors: Psychosis severity was associated with ‘taking any
Comorbidities: The Activation factors, derived from the same PANSS
medication’ (p=.002) and more water consumption (p=.025), and, most
items, was significantly associated with the number of self-reported
robustly, with constipation (p<.001). Consistent with the constipation
medical conditions (p= 0.003) and number of psychiatric Disorders
finding, sel-reporting one or fewer bowel movements daily, compared to
(p= 0.009). Likewise, it was significantly linked to BMI (: p=0.003);
2 or 3, was associated with higher Negative Symptoms (p=.045) and
gastroesophageal reflux disorder or GERD (0.014); skin condition
(p=.000) and lactose Intolerance (p=.050). Of the aforementioned
Table 1
items, dysphoric mood was only associated with the number of medical
Demographic characteristics of study sample.
(p=0.050) and number of psychiatric conditions (p=0.025) with a trend
Overall Psychosis Nonpsychotic Controls association to skin conditions (p=0.054). The total number of psychi­
Cases Cases
(N = 82) (N = 40) (N = 18) (N = 24)
atric conditions was also related to higher general psychopathology
Variables n (%) n (%) n (%) n (%) p-value scores (p=.002).
a
Lifestyle: Self-reported exercise was associated with lower General
Sex 0.307 Psychopathology Scores, whether Indoors (p= 0.007); Outdoors (p=
Female 46 20 (50%) 13 (72%) 13 (54%) 0.048) or Both Indoors/Outdoors (; p= 0.031) compared versus no ex­
(56%) ercise. Lower autistic preoccupation severity scores were also linked to
Male 36 20 (50%) 5 (28%) 11 (46%) exercising; whether indoors (p=.038) or outdoors (p=.007). Reporting
(44%)
mean mean (SD) mean (SD) mean p-value
that exercise “depends” were as symptomatic as those who reported not
(SD) (SD) b exercising.
Age 37.1 38.1 (11.1) 39.7 (11.2) 33.4 0.120 Neonatal Factors: Self-reported vaginal birth was associated with
(10.8) (9.39) lower scores on general psychopathology than C-section born partici­
BMI 30.6 31.9 (7.34) 31.3 (6.86) 28 (6.98) 0.120
pants (p=0.01) and lower scores on dysphoric mood (p=0.39). Self-
(7.24)
reported formula feeding in infancy, versus breast feeding, was associ­
a
By chi-squared test ated with higher dysphoric mood factor scores (p=0.01).
b
By ANOVA analyses

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.
S.J. Fendrich et al. Psychiatry Research 312 (2022) 114526

Table 3
PANSS symptom severity by psychiatric diagnosis using linear regression.
All Cases Psychosis Cases Nonpsychotic Cases
(N = 59) (N = 40) (N = 18)
Variable mean (SD) mean (SD) mean (SD) t (df) p-value a
b
Positive Symptoms 13.5 (6.5) 16.05 (6.4) 7.94 (1.5) 8.0 (46) < 0.001**
Negative Symptoms b 10.8 (4.5) 11.2 (4.6) 10.1 (4.1) 0.90 (37) 0.400
b
General Psychopathology 23.9 (6.4) 24 (7.0) 23.8 (5.1) 0.10 (45) 0.900
Activation c 7.9 (2.3) 8.2 (2.4) 7.2 (1.7) 2 (46) 0.062
Autistic Preoccupation c 9.11 (2.26) 10.23 (2.42) 6.67 (1.69) 6 (54) < 0.001**
Dysphoric Mood c 9.7 (4.1) 8.9 (4.2) 11.4 (3.6) -2 (38) 0.023*
Total PANSS Score b 48.3 (13.9) 51.2 (14.8) 41.9 (9.2) 3.00 (50) 0.006**
a
By unpaired, two-sided t-test, comparing psychotic cases and case controls., Bolded p-values are < 0.07. ** P-Value < 0.01, * P-Value < 0.05
b
Original PANSS scales
c
PANSS-5 Factors

Table 4
Associations between health factors and symptom severity among all cases, derived from the original PANSS scoring by linear regression analyses.
Total PANSS Score Positive Symptom Severity Negative Symptom Severity General Psychopathology

Variables Estimate (95% C.I.) p-value Estimate (95% C.I.) p-value Estimate (95% C.I.) p-value Estimate (95% C.I.) p-value

Demographics
Gender (Ref = Male)
Female -3.20 (-10.70 - 4.25) 0.393 -1.83 (-5.32 - 1.65) 0.297 -2.70 (-4.99 - -0.41) 0.022* 1.33 (-2.12 - 4.78) 0.443
Conditions/Allergies
Number of food allergies (Ref = Zero)
One -8.80 (-16.69 - -0.90) 0.030* -2.92 (-6.80 - 0.96) 0.137 -2.81 (-5.39 - -0.23) 0.033* -3.07 (-6.69 - 0.56) 0.096
Two 6.62 (-4.38 - 17.63) 0.233 0.73 (-4.68 - 6.13) 0.788 0.85 (-2.74 - 4.45) 0.636 5.04 (-0.01 - 10.10) 0.051
Gluten Allergy (Ref = No)
Yes 20.40 (0.95 - 39.90) 0.040* 6.75 (-2.57 - 16.10) 0.153 3.27 (-3.15 - 9.70) 0.312 10.40 (1.49 - 19.30) 0.023*
Number of psychiatric 2.04 (-3.45 - 7.52) 0.460 -0.32 (-2.90 - 2.27) 0.806 -0.19 (-1.96 - 1.57) 0.826 2.55 (0.09 - 5.00) 0.042*
comorbidities
Neonatal Factors
Birth Delivery (Ref = C-Section)
Vaginal Birth -7.05 (-16.60 - 2.49) 0.144 -0.11 (-4.68 - 4.46) 0.962 -1.61 (-4.68 - 1.47) 0.300 -5.33 (-9.60 - -1.06) 0.016*
Unknown -2.26 (-15.70 - 11.21) 0.738 0.22 (-6.24 - 6.68) 0.946 0.00 (-4.35 - 4.35) 1.000 -2.48 (-8.51 - 3.55) 0.413
Health Factors
BMs Per Day (Ref = Normal; 2-3)
Low (<=1) 11.90 (-1.05 - 24.90) 0.071 0.46 (-5.79 - 6.71) 0.883 4.18 (0.10 - 8.27) 0.045* 7.26 (1.42 - 13.10) 0.016*
High (3+) -0.70 (-13.65 - 12.30) 0.914 1.26 (-4.99 - 7.51) 0.688 -1.62 (-5.70 - 2.47) 0.431 -0.34 (-6.18 - 5.50) 0.907
Nature of BMs (Ref = Normal)
Constipated 10.39 (1.58 - 19.20) 0.022* 6.66 (2.72 - 10.60) 0.001** 0.28 (-2.68 - 3.25) 0.849 3.45 (-0.71 - 7.60) 0.102
Diarrhea 4.35 (-5.37 - 14.10) 0.373 1.85 (-2.50 - 6.20) 0.397 -0.09 (-3.36 - 3.18) 0.954 2.59 (-1.99 - 7.18) 0.262
Any Medication (Ref = No)
Yes -7.38 (-14.80 - 0.08) 0.053 -5.30 (-8.63 - -1.98) 0.002** -0.58 (-3.05 - 1.89) 0.640 -1.50 (-5.05 - 2.05) 0.400
Vitamin D (Ref = No)
Yes 10.60 (2.19 - 19.00) 0.014* 3.15 (-0.93 - 7.23) 0.127 1.60 (-1.21 - 4.41) 0.259 5.85 (2.06 - 9.64) 0.003**
Diet
Fruits Daily (Ref = No)
Yes -8.83 (-17.70 - 0.00) 0.050* -3.81 (-7.97 - 0.35) 0.072 -2.52 (-5.37 - 0.32) 0.081 -2.49 (-6.66 - 1.68) 0.236
Sugary sweets (Ref = Never)
1-3 days/week -7.33 (-18.50 - 3.79) 0.192 -1.62 (-6.97 - 3.72) 0.545 -0.71 (-4.37 - 2.95) 0.699 -5.00 (-10.10 - 0.12) 0.055
3-7 days/week -11.50 (-22.60 - 0.043* -3.37 (-8.72 - 1.97) 0.211 -2.13 (-5.78 - 1.53) 0.249 -6.00 (-11.10 - -0.88) 0.023*
-0.38)
Water consumption (Ref = 3-7 days/week)
Never 10.65 (-9.25 – 30.50) 0.288 5.94 (-3.13 – 15.00) 0.195 -2.58 (-8.95 – 3.78) 0.419 7.29 (-1.97 – 16.55) 0.120
1-3 days/week 8.57 (-2.58 – 19.70) 0.129 5.87 (0.78 – 10.90) 0.025* 2.85 (-0.72 – 6.41) 0.116 -0.14 (-5.33 – 5.05) 0.958
Lifestyle
Exercise Location (Ref = Neither)
Indoors -7.11 (-17.00 - 2.78) 0.155 -0.37 (-5.05 - 4.32) 0.875 -0.59 (-3.82 - 2.63) 0.713 -6.14 (-10.49 - -1.79) 0.007*
Outdoors -2.92 (-14.00 - 8.12) 0.598 0.75 (-4.48 - 5.98) 0.775 1.23 (-2.38 - 4.83) 0.498 -4.89 (-9.75 - -0.04) 0.048*
Both Indoors/Outdoors -9.30 (-20.90 - 2.31) 0.114 -3.20 (-8.71 - 2.31) 0.249 -0.46 (-4.25 - 3.34) 0.810 -5.64 (-10.76 - -0.53) 0.031*
Depends -3.00 (-24.20 - 18.21) 0.778 0.50 (-9.55 - 10.55) 0.921 0.14 (-6.79 - 7.07) 0.967 -3.64 (-12.98 - 5.69) 0.437

Bolded p-values are < 0.07. ** P-Value < 0.01, * P-Value < 0.05.

3.4. Categorical assessments comparing psychotic, nonpsychotic affective, likely than nonpsychotic affective cases to use probiotics (p=0.038)
and control subject groups (Table 3).
Diet: Compared to healthy controls, both psychiatric groups
Categorical comparisons of self-report items across all participants consumed more filtered or bottled water than city or tap water (p’s<
(Table 6 and 7). 0.015) (Table 3). Nonpsychotic cases consumed more dairy substitutes
Health Factors: The psychosis cases were less likely to consume than healthy controls (p=0.048). Psychosis cases more frequently (3-7
supplements (other than vitamin B and D and multivitamins) than either days/week) consumed ready-made meals (e.g., boxed macaroni and
healthy controls (p=0.033) or nonpsychotic cases (p=0.011) and less cheese, ramen noodles) than controls (p=0.045).

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.
S.J. Fendrich et al. Psychiatry Research 312 (2022) 114526

Table 5
Associations between health factors and symptom severity using factors from the PANSS 5-Factor Model among cases by linear regression analysis.
Activation Autistic Preoccupation Dysphoric Mood

Variables Estimate (95% C.I.) p-value Estimate (95% C.I.) p-value Estimate (95% C.I.) p-value

Demographics
BMI 0.12 (0.04 - 0.20) 0.003** 0.05 (-0.07 - 0.16) 0.396 0.15 (-0.01 - 0.30) 0.061
Gender (Ref = Male)
Female 1.02 (-0.16 - 2.21) 0.089 -1.66 (-3.23 - -0.10) 0.037* 1.87 (-0.31 - 4.04) 0.091
Conditions
GERD (Ref = No)
Yes 1.91 (0.40 - 3.42) 0.014* 1.84 (-0.24 - 3.92) 0.082 1.18 (-1.72 - 4.07) 0.418
Skin Condition (Ref = No)
Yes 3.18 (1.66 - 4.69) 0.000** 0.17 (-2.15 - 2.49) 0.885 3.03 (-0.06 - 6.11) 0.054
Lactose Intolerance (Ref = No)
Yes 1.48 (0.00 - 2.96) 0.050* -0.92 (-2.95 - 1.11) 0.368 1.34 (-1.45 - 4.12) 0.341
Number of Medical Comorbidities 0.82 (0.30 - 1.34) 0.003** 0.14 (-0.62 - 0.90) 0.720 1.00 (0.00 - 2.00) 0.050*
Number of Psychiatric Comorbidities 1.14 (0.30 - 1.98) 0.009** -0.29 (-1.48 - 0.90) 0.624 1.80 (0.23 - 3.36) 0.025*
Health Factors
Bowel Health
BMs Per Day (Ref = Normal; 2-3)
Low (<= 1) -0.34 (-2.50 - 1.83) 0.757 0.72 (-2.17 - 3.60) 0.621 4.83 (1.08 - 8.58) 0.013*
High (3+) -0.14 (-2.30 - 2.03) 0.900 -0.49 (-3.37 - 2.40) 0.737 0.63 (-3.12 - 4.38) 0.738
Nature of Bowel Movements (ref = Normal)
Constipated 1.10 (-0.36 - 2.55) 0.137 2.35 (0.47 - 4.23) 0.015* 0.14 (-2.61 - 2.89) 0.919
Diarrhea 1.16 (-0.45 - 2.77) 0.154 1.74 (-0.34 - 3.81) 0.099 -0.51 (-3.54 - 2.53) 0.740
Perinatal
Birth Delivery (Ref = C-Section)
Vaginal Birth -0.72 (-2.26 - 0.82) 0.351 0.34 (-1.76 - 2.44) 0.746 -2.96 (-5.76 - -0.16) 0.039*
Unknown 1.01 (-1.16 - 3.18) 0.353 0.90 (-2.06 - 3.86) 0.546 -2.19 (-6.15 - 1.76) 0.271
Infant Nutrition (Ref = Primarily Breast Milk)
Primarily Formula 1.05 (-0.69 - 2.79) 0.232 -0.25 (-2.61 - 2.11) 0.833 3.40 (0.23 - 6.57) 0.036*
Mixture 0.57 (-1.16 - 2.30) 0.514 0.34 (-2.01 - 2.68) 0.774 2.02 (-1.12 - 5.17) 0.202
Unknown 2.07 (-0.26 - 4.39) 0.080 1.60 (-1.55 - 4.75) 0.313 2.00 (-2.22 - 6.22) 0.346
Supplements
Vitamin D (ref = No)
Yes 1.23 (-0.17 - 2.63) 0.084 1.86 (0.00 - 3.71) 0.050* 2.55 (0.00 - 5.10) 0.050
Diet
Fruits & Vegetables
Fruits daily (ref = No)
Yes -0.98 (-2.44 - 0.48) 0.185 -2.82 (-4.65 - -1.00) 0.003** 0.61 (-2.10 - 3.32) 0.654
Salty & Sweet
Sweet beverages (Ref = Never)
1-3 days/week 0.78 (-0.66 - 2.22) 0.282 0.97 (-1.03 - 2.97) 0.334 -1.40 (-4.11 - 1.31) 0.304
3-7 days/week 1.70 (0.30 - 3.10) 0.019* 0.93 (-1.02 - 2.87) 0.344 0.86 (-1.78 - 3.50) 0.516
Artificial sweeteners (Ref = Never)
1-3 days/week -0.07 (-1.38 - 1.23) 0.910 0.25 (-1.52 - 2.03) 0.776 -2.96 (-5.28 - -0.64) 0.013*
3-7 days/week 1.40 (-0.39 - 3.20) 0.123 0.85 (-1.59 - 3.30) 0.487 -1.24 (-4.42 - 1.95) 0.440
Lifestyle
Personal Health
Exercise Location (Ref = Neither)
Indoors -1.90 (-3.45 - -0.34) 0.018* -2.98 (-5.80 - -0.17) 0.038* -0.60 (-2.75 - 1.55) 0.577
Outdoors -1.55 (-3.28 - 0.19) 0.080 -4.39 (-7.53 - -1.25) 0.007** 0.46 (-1.94 - 2.87) 0.700
Both Indoors/Outdoors -1.81 (-3.64 - 0.01) 0.052 -2.74 (-6.05 - 0.56) 0.102 -0.89 (-3.42 - 1.64) 0.485
Depends -1.21 (-4.55 - 2.12) 0.468 -2.64 (-8.68 - 3.39) 0.384 2.21 (-2.40 - 6.83) 0.340

Lifestyle: Psychosis and nonpsychotic affective groups had similar report tool for microbiome and inflammatory disease research in psy­
weight gain over the past six month, both significantly more than chiatry. Support for its validity is bolstered by the accurate self-reporting
healthy controls (p=0.013; p=0.028). Psychosis cases were significantly of psychiatric research diagnoses by all of the 82 subjects. This finding
less likely to report having even occasional exercise (1-2 times/week) may overcome some concern that persons with psychiatric conditions
(p=0.026), and more likely to report that their exercise location ‘de­ could not accurately report their conditions and exposures.
pends” compared to healthy controls (p=0.000). Psychotic cases were Notably, a number of associations between BGAQ items and psy­
more likely to smoke cigarettes than healthy controls (p=0.003). Psy­ chiatric symptoms were robust against Bonferroni corrections for the
chosis and nonpsychotic affective groups were less likely to report oc­ multiple comparisons at p=.001. From the transdiagnostic analysis,
casional alcohol use (p=0.007; p=0.032) or low alcohol consumption these include the association of psychotic symptom severity with con­
(1-2 servings) compared to no alcohol consumption (p=0.004; p=0.050) stipation, and the association of activation factor severity with skin
than controls. Psychosis cases consuming less hard cider or beer than conditions, with similarly significant associations between activation
controls (p=0.007). Both psychiatric case groups included more par­ symptoms with the number of self-reported medical conditions and
ticipants who were abstinent or consuming alcohol in the highest cate­ number of psychiatric disorders and body mass index. This level of
gory (3+ weekly) relative to controls. significance applied to the categorical analyses highlights differences in
exercise, cigarette smoking and alcohol use patterns in the psychotic
4. Discussion group.
The association of psychosis severity with constipation, is consistent
The study showed promising results for using the GBAQ as a self- with reduced gut vagal (parasympathetic) function, echoed by

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.
S.J. Fendrich et al. Psychiatry Research 312 (2022) 114526

Table 6 Table 7
Associations between health and diet factors and categorical psychiatric Association between lifestyle factors and categorical psychiatric conditions.
conditions. Psychosis Cases vs Psychosis Cases vs Nonpsychotic Cases
Psychosis Cases vs Psychosis Cases vs Nonpsychotic Nonpsychotic Controls vs Controls
Nonpsychotic Controls Cases vs Controls Cases
Cases
Variables OR p-val. OR p-val. OR p-val.
Variables OR p-val. OR p-val. OR p-val. (95% C. (95% C. (95% C.
(95% (95% (95% I.) I.) I.)
C.I.) C.I.) C.I.)
Personal
Health Factors Health
Supplements Weight
Probiotic (Ref Change
= No) (Ref =
Yes 0.22 0.038* 0.32 0.103 1.42 0.613 Stable)
(0.05 - (0.07 - (0.36 - Increased 1.17 0.809 14.78 0.013* 12.67 0.028*
0.91) 1.24) 5.61) (0.34 - (2.56 - (1.80 -
Other 4.22) 281.72) 258.78)
Supplement Decreased 1.33 0.716 2.11 0.282 1.58 0.595
(Ref = No) (0.30 - (0.56 - (0.27 -
Yes 0.11 0.011* 0.16 0.033* 1.50 0.555 7.21) 9.06) 8.75)
(0.01 - (0.02 - (0.38 - Exercise (Ref
0.52) 0.76) 5.91) = Rarely or
Diet Never)
Dairy & Fermented Food Yes (at least 1- 0.78 0.664 0.24 0.026* 0.31 0.113
Dairy 2 times/ (0.24 - (0.06 - (0.07 -
substitutes week) 2.39) 0.79) 1.27)
(Ref = No) Exercise
Yes 0.53 0.273 2.00 0.225 3.75 0.048* Location
(0.17 - (0.67 - (1.04 - (Ref =
1.63) 6.51) 14.73) Neither)
Water Indoors 5.92 0.120 1.35 0.265 0.23 0.223
Drinking (0.86 - (0.35 - (0.01 -
source (Ref 119.81) 5.63) 1.90)
= City) Outdoors 0.54 0.432 0.44 0.381 0.82 0.795
Filtered/ 0.80 0.714 4.40 0.010* 5.50 0.014* (0.11 - (0.10 - (0.17 -
Bottled (0.23 - (1.48 - (1.48 - 2.55) 1.84) 3.79)
2.59) 14.59) 23.01) Both Indoors/ NA 0.993 0.41 0.135 0.00 0.994
Unsure/Other 1.20 0.884 NA 0.990 NA 0.994 Outdoors (0.00 - (0.05 - (NA -
(0.12 - (0.00 - (0.00 - NA) 2.99) NA)
26.97) NA) NA) Depends 0.97 0.966 5.54 0.000** 5.71 0.150
Meal Preparation/Source (0.23 - (0.80 - (0.69 -
Ready-made 4.20) 112.16) 124.00)
meals (Ref = Cigarettes
3-7 days/ (Ref = No)
week) Yes 1.69 0.358 6.76 0.003** 4.00 0.056
Never 0.38 0.270 0.17 0.113 0.44 0.538 (0.55 - (2.11 - (1.01 -
(0.05 - (0.01 - (0.02 - 5.32) 26.62) 18.19)
1.89) 1.10) 5.53) Alcohol Use
1-3 days/week 0.38 0.270 0.11 0.045* 0.29 0.336 Alcohol (Ref
(0.05 - (0.01 - (0.01 - = Never)
1.89) 0.67) 3.44) Occasionally 0.79 0.709 0.11 0.007** 0.14 0.032*
(1-2 times/ (0.22 - (0.02 - (0.02 -
Bolded p-values are < 0.07. ** P-Value < 0.01, * P-Value < 0.05. week) 2.68) 0.47) 0.75)
Regularly (3+ 0.63 0.590 0.63 0.723 1.00 1.000
times/ (0.11 - (0.05 - (0.06 -
associations between having one or fewer bowel movements daily,
week) 3.80) 15.11) 26.86)
compared to 2-3, with more severe dysphoria, negative symptoms, and Drinks
general psychopathology scores. While antipsychotics and other medi­ Consumed
cations have anticholinergic effects that can impact vagal function, (Ref =
vagal dysfunction in psychiatric conditions, usually indexed by beat-to None)
Low (1-2 0.58 0.428 0.09 0.004** 0.16 0.050
-beat heart rate variability (HRV), is independent of any medication
servings) (0.15 - (0.01 - (0.02 -
effects (Bengsten et al, 2021; Umetani et al 1998; Quintana et al, 2016). 2.18) 0.40) 0.91)
We previously associated reduced HRV to features of schizophrenia, High (3+ 0.67 0.590 0.67 0.707 1.00 1.000
including information processing deficits and hallucinations (Malaspina servings) (0.15 - (0.07 - (0.09 -
2.98) 6.31) 11.32)
et al., 1997; Malaspina et al., 2002; Mujica-Parodi et al., 2005; Kimhy
Cider/Beer
et al., 2017) as have other investigators (Jung et al., 2019; Clamor et al., (Ref = No)
2016; Zhang et al., 2020). This data replicates and extends the reduced Yes 0.47 0.207 0.23 0.007** 0.48 0.247
vagal function in psychosis to include the gut, based on constipation and (0.15 - (0.07 - (0.13 -
hypomotility, and links bowel function to the broad syndrome of 1.53) 0.65) 1.65)
schizophrenia, including positive and negative symptoms and general Bolded p-values are < 0.07. ** P-Value < 0.01, * P-Value < 0.05.
psychopathology.
The nature or frequency of bowel movements was not related to the
activation symptoms, so activation severity might tap a separate
vulnerability for psychiatric illness. If so it the factor has impacts beyond
the brain and behavior, as activation was significantly associated with

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.
S.J. Fendrich et al. Psychiatry Research 312 (2022) 114526

the number of medical comorbidities in addition to the number of with neonatal experiences in these adults. Cases reporting vaginal birth
different psychiatric condition. Activation associated conditions include showed less dysphoria than those reporting Cesarean Section. Similarly,
body mass index, now recognized to be an inflammatory condition, those reporting bottle feeding had more dysphoria than breast fed sub­
which elevates interleukin 6 and participates in risk pathways to car­ jects. The optional seeding of a newborn’s gut, which occurs during
diometabolic diseases, and skin diseases such as psoriasis (Ellulu et al., vaginal delivery and breast feeding, controls the early development of
2017). Activation items include uncooperativeness, poor rapport, ten­ the immune system and the brain, and could well be pertinent to
sion, excitement, and hostility, overlapping with features of delirium, a continuing or treatment refractory symptoms. The gut of Caesarian
condition of confused thinking and reduced awareness of the environ­ Section babies are first colonized by microbiota from the skin or mouth
ment, which can additionally include disturbed perception and emotion of parents or other persons (Hoffman et al 2020).
(Setters B, Solber, 2017). While delirium typically accompanies severe The associations of the GBAQ items with symptom severity sets the
acute conditions, including metabolic imbalance, infection, surgery, or stage for future studies. Findings from the RDoC transdiagnostic
drug intoxication or withdrawal, it can present in a chronic form in approach suggest separate factors may be underpinnings of psychiatric
persons with mild chronic systemic inflammation (Cunningham et al., symptoms. The first is reduced gut vagal activity, illustrated by the very
2017). Delirium, like schizophrenia, is associated with elevations of the strong association of constipation with psychosis, with infrequent bowel
proinflammatory IL-2 and IL-6 cytokine (Tucker et al, 2017) supporting movements furthermore associated with negative and general psycho­
an overlapping mechanism between schizophrenia and delirium. Skin pathology symptoms. A second factor is inflammation, represented by
conditions associated with activation severity may likewise be consis­ the association of activation symptoms severity with the numbers of
tent with systemic inflammation. It is recently proposed that the psychiatric and medical conditions, and other inflammatory disorders. If
Gut-Brain Axis and the Gut-Skin Axis be merged into a Gut-Brain-Skin so, the former may be more specific to psychosis whereas inflammation
Axis, which emerges from imbalance of the immune system with the may be common to psychotic and nonpsychotic disorders. Group level
gut microbiota (Vojvodic et al 2019). One of the few studies assessing categorical analyses, on the other hand demonstrated lifestyle and
activation factor severity found that it independently predicted a lower health conditions that may be consequences of having a psychiatric
likelihood of hospital discharge. (White et al., 2004). Future studies can disorder. All of the subjects were in active treatment so the symptoms
examine if treatment-refractory psychotic conditions with have in­ being measured are residual and thus of keen importance to the field.
flammatory underpinnings, are associated with activation symptoms, The inflammatory pathophysiology is presumed to be ongoing in such
and, if so, whether anti-inflammatory treatments are beneficial. persons. Moreover, this is a real world naturalistic study whose findings
Our analysis included self-reported celiac disease, gluten intolerance will be directly relevant to the symptomatic urban population.
and wheat sensitivity into a single gluten allergy factor, as all are In conclusion, these findings suggest the GBAQ can be a valuable tool
associated with psychopathology (Singh et al., 1976; Wijarnpreecha for obtaining a person-specific “review of systems” for items associated
et al., 2018), showing increased PANSS and general psychopathology with the gut microbiome. The validity of the self-reported diagnoses
ratings for self-reported gluten allergy. These conditions, and other food against the research diagnoses offer evidence that psychiatric cases
allergies, are related to gut dysbiosis, which can increase gut perme­ accurately complete the questionnaires. This first study of the GBAQ is
ability such that antibodies and toxins can enter the circulation to pro­ not without limitations. We identified sex differences in some psychosis-
duce systemic inflammation (Di Tommaso et al., 2021), as well as skin related symptoms and outcomes, but are underpowered for applying
conditions (Lee et al 2021; De Pessemier et al., 2021). sex-stratified analyses for some survey items that were rarely endorsed
The analyses comparing subgroups of our participants showed psy­ at this stage of our study. Additionally, some results may be confounded
chotic and nonpsychotic affective disorder groups were more than 12- by medication use; for instance, the greater changes in weight in the past
fold likely to experience weight gain, common with medication use, six months for cases than healthy controls and greater constipation in
and 4-fold more likely to smoke cigarettes than controls, in line with psychosis cases could be a consequence of antipsychotic treatment.
other reports (Hartz et al., 2014). The psychosis cases had less exercise, Future research should include non-medicated cases to exclude this
so introducing exercise as an complementary intervention in persons potential confound, specifically minimizing exposure to anticholinergic
with psychosis may be promising (García-Garcés et al., 2021; Kimhy medications. The current results await confirmation from other cohorts
et al., 2021). Alcohol consumption profiles differed for psychiatric and analyses with the microbiota, cytokine, ANS and imaging measures
groups as both were less likely to have just occasional alcohol use. Both in the parent grant.
groups also consumed more filtered than tap water compared to the
controls. Filtered water can lack essential mineral elements typically Funding
found in tap water (e.g., Ca, Mg, Se, F, Zn). Mineral deficiencies can
impair brain function (Shayganfard et al., 2021), with a meta-analyses Support NIH 1R01MH110418 (DM)
showing benefits of supplemental minerals for persons with psychosis
or depression (Cherak et al., 2021). Supplementary materials
Other evidence of less healthy behaviors for the psychosis group was
their frequent use of processed, ready-made meals and reduced pro­ Supplementary material associated with this article can be found, in
biotics and supplement use than non-psychotic affective cases. Likewise, the online version, at doi:10.1016/j.psychres.2022.114526.
reduced fruit and vegetable intake predicted autistic preoccupation
severity. Notably cause-and-effect could not be established for the items References
distinguishing subgroups of participants. Some counterintuitive findings
may reflect clinical advice on diet and supplements. For example Bahorik, A.L., Satre, D.D., Kline-Simon, A.H., Weisner, C.M., Campbell, C.I., 2017.
Serious mental illness and medical comorbidities: Findings from an integrated health
vitamin D supplementation predicted more severe total symptoms, care system. Journal of Psychosomatic Research 100, 35–45. https://doi.org/
general psychopathology, autistic preoccupation, and dysphoria. Simi­ 10.1016/j.jpsychores.2017.07.004.
larly, both psychiatric groups reported higher frequency of consuming Bengtsson, J., Olsson, E., Igelström, H., Persson, J., Bodén, R., 2021. Ambulatory Heart
Rate Variability in Schizophrenia or Depression: Impact of Anticholinergic Burden
seafood than controls, consistent with evidence that omega-3 fatty acids and Other Factors. J Clin Psychopharmacol 41, 121–128. https://doi.org/10.1097/
from dietary fish can benefit psychiatric symptoms (Cadenhead et al., JCP.0000000000001356.
2019; Pawelczyk et al., 2018; Tang et al., 2020). Omega-3 availability Bishop, J.R., Zhang, L., Lizano, P., 2022. Inflammation Subtypes and Translating
Inflammation-Related Genetic Findings in Schizophrenia and Related Psychoses: A
also regulates gut microbiota composition and inflammation (Costantini Perspective on Pathways for Treatment Stratification and Novel Therapies. Harv Rev
et al., 2017). Psychiatry 30, 59–70. https://doi.org/10.1097/HRP.0000000000000321.
It was impressive that current symptoms showed any associations

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.
S.J. Fendrich et al. Psychiatry Research 312 (2022) 114526

Brusselaers, N., 2019. Prescribed Drugs and the Microbiome. Gastroenterology Clinics of Lewis, S.J., Heaton, K.W., 1997. Stool form scale as a useful guide to intestinal transit
North America 48, 331–342. https://doi.org/10.1016/j.gtc.2019.04.002. time. Scand J Gastroenterol 32, 920–924. https://doi.org/10.3109/
Cadenhead, K.S., Minichino, A., Kelsven, S., Addington, J., Bearden, C., Cannon, T.D., 00365529709011203.
Cornblatt, B.A., Mathalon, D., McGlashan, T.H., Perkins, D.O., Seidman, L.J., Liu, L., Huh, J.R., Shah, K., 2022. Microbiota and the gut-brain-axis: Implications for new
Tsuang, M., Walker, E.F., Woods, S.W., Yao, J., 2019. Metabolic abnormalities and therapeutic design in the CNS. eBioMedicine 77, 103908. https://doi.org/10.1016/j.
low dietary Omega 3 are associated with symptom severity and worse functioning ebiom.2022.103908.
prior to the onset of psychosis: Findings from the North American Prodrome Malaspina, D., Bruder, G., Dalack, G.W., Storer, S., Van Kammen, M., Amador, X.,
Longitudinal Studies Consortium. Schizophr Res 204, 96–103. https://doi.org/ Glassman, A., Gorman, J., 1997. Diminished cardiac vagal tone in schizophrenia:
10.1016/j.schres.2018.09.022. Associations to brain laterality and age of onset. Biological Psychiatry 41, 612–617.
Caputi, V., Popov, J., Giron, M.C., O Apos Mahony, S., 2021. Gut Microbiota as a https://doi.org/10.1016/S0006-3223(96)00161-8.
Mediator of Host Neuro-Immune Interactions: Implications in Neuroinflammatory Malaspina, D., Dalack, G., Leitman, D., Corcoran, C., Amador, X.F., Yale, S.,
Disorders. Mod Trends Psychiatry 32, 40–57. https://doi.org/10.1159/000510416. Glassman, A., Gorman, J.M., 2002. Low Heart Rate Variability Is Not Caused by
Carabotti, M., Scirocco, A., Maselli, M.A., Severi, C., 2015. The gut-brain axis: Typical Neuroleptics in Schizophrenia Patients. CNS spectr 7, 53–57. https://doi.
interactions between enteric microbiota, central and enteric nervous systems. Ann org/10.1017/S1092852900022264.
Gastroenterol 28, 203–209. McCance-Katz, E.F., n.d. The National Survey on Drug Use and Health: 2019 63.
Clamor, A., Lincoln, T.M., Thayer, J.F., Koenig, J., 2016. Resting vagal activity in Mujica-Parodi, L.R., Yeragani, V., Malaspina, D., 2005. Nonlinear Complexity and
schizophrenia: meta-analysis of heart rate variability as a potential endophenotype. Spectral Analyses of Heart Rate Variability in Medicated and Unmedicated Patients
Br J Psychiatry 208, 9–16. https://doi.org/10.1192/bjp.bp.114.160762. with Schizophrenia1. Neuropsychobiology 51, 10–15. https://doi.org/10.1159/
Clemente, J.C., Manasson, J., Scher, J.U., 2018. The role of the gut microbiome in 000082850.
systemic inflammatory disease. BMJ 360, j5145. https://doi.org/10.1136/bmj. Müller, N., 2018. Inflammation in Schizophrenia: Pathogenetic Aspects and Therapeutic
j5145. Considerations. Schizophrenia Bulletin 44, 973–982. https://doi.org/10.1093/
Costantini, L., Molinari, R., Farinon, B., Merendino, N., 2017. Impact of Omega-3 Fatty schbul/sby024.
Acids on the Gut Microbiota. IJMS 18, 2645. https://doi.org/10.3390/ Nordstroem, A.-L., Talbot, D., Bernasconi, C., Berardo, C.G., Lalonde, J., 2017. Burden of
ijms18122645. illness of people with persistent symptoms of schizophrenia: A multinational cross-
Cuthbert, B.N., Morris, S.E., 2021. Evolving Concepts of the Schizophrenia Spectrum: A sectional study. Int J Soc Psychiatry 63, 139–150. https://doi.org/10.1177/
Research Domain Criteria Perspective. Front Psychiatry 12, 641319. https://doi.org/ 0020764016688040.
10.3389/fpsyt.2021.641319. Nurnberger, J.I., Blehar, M.C., Kaufmann, C.A., York-Cooler, C., Simpson, S.G., Harkavy-
Daulatzai, M., 2015. Non-Celiac Gluten Sensitivity Triggers Gut Dysbiosis, Friedman, J., Severe, J.B., Malaspina, D., Reich, T., 1994. Diagnostic interview for
Neuroinflammation, Gut-Brain Axis Dysfunction, and Vulnerability for Dementia. genetic studies. Rationale, unique features, and training. NIMH Genetics Initiative.
CNSNDDT 14, 110–131. https://doi.org/10.2174/1871527314666150202152436. Arch Gen Psychiatry 51, 849–859. https://doi.org/10.1001/
De Pessemier, B., Grine, L., Debaere, M., Maes, A., Paetzold, B., Callewaert, C., 2021. archpsyc.1994.03950110009002 discussion 863-864.
Gut-Skin Axis: Current Knowledge of the Interrelationship between Microbial Pawełczyk, T., Piątkowska-Janko, E., Bogorodzki, P., Gębski, P., Grancow-Grabka, M.,
Dysbiosis and Skin Conditions. Microorganisms 9, 353. https://doi.org/10.3390/ Trafalska, E., Żurner, N., Pawełczyk, A., 2018. Omega-3 fatty acid supplementation
microorganisms9020353. may prevent loss of gray matter thickness in the left parieto-occipital cortex in first
Di Tommaso, N., Gasbarrini, A., Ponziani, F.R., 2021. Intestinal barrier in human health episode schizophrenia: A secondary outcome analysis of the OFFER randomized
and disease. Int J Environ Res Public Health 18, 12836. https://doi.org/10.3390/ controlled study. Schizophr Res 195, 168–175. https://doi.org/10.1016/j.
ijerph182312836. schres.2017.10.013.
Donoso, F., Cryan, J.F., Olavarría-Ramírez, L., Nolan, Y.M., Clarke, G., 2022. Quintana, D.S., Alvares, G.A., Heathers, J.A.J., 2016. Guidelines for Reporting Articles
Inflammation, lifestyle factors, and the microbiome-gut-brain axis: relevance to on Psychiatry and Heart rate variability (GRAPH): recommendations to advance
depression and antidepressant action. Clin Pharmacol Ther. https://doi.org/ research communication. Transl Psychiatry 6. https://doi.org/10.1038/tp.2016.73
10.1002/cpt.2581. e803–e803.
Ellulu, M.S., Patimah, I., Khaza’ai, H., Rahmat, A., Abed, Y., 2017. Obesity and Rathnaiah, M., Liddle, E.B., Gascoyne, L., Kumar, J., Zia Ul Haq Katshu, M., Faruqi, C.,
inflammation: the linking mechanism and the complications. aoms 4, 851–863. Kelly, C., Gill, M., Robson, S., Brookes, M., Palaniyappan, L., Morris, P., Liddle, P.F.,
https://doi.org/10.5114/aoms.2016.58928. 2020. Quantifying the Core Deficit in Classical Schizophrenia. Schizophrenia
García-Garcés, L., Sánchez-López, M.I., Cano, S.L., Meliá, Y.C., Marqués-Azcona, D., Bulletin Open 1, sgaa031. https://doi.org/10.1093/schizbullopen/sgaa031.
Biviá-Roig, G., Lisón, J.F., Peyró-Gregori, L., 2021. The short and long-term effects of Rinninella, E., Raoul, P., Cintoni, M., Franceschi, F., Miggiano, G.A.D., Gasbarrini, A.,
aerobic, strength, or mixed exercise programs on schizophrenia symptomatology. Sci Mele, M.C., 2019. What is the Healthy Gut Microbiota Composition? A Changing
Rep 11, 24300. https://doi.org/10.1038/s41598-021-03761-3. Ecosystem across Age. Environment, Diet, and Diseases. Microorganisms 7, E14.
Hartz, S.M., Pato, C.N., Medeiros, H., Cavazos-Rehg, P., Sobell, J.L., Knowles, J.A., https://doi.org/10.3390/microorganisms7010014.
Bierut, L.J., Pato, M.T., 2014. Comorbidity of severe psychotic disorders with Setters, B., Solberg, L.M., 2017. Delirium. Primary Care. Clinics in Office Practice 44,
measures of substance use. JAMA Psychiatry 71, 248–254. https://doi.org/10.1001/ 541–559. https://doi.org/10.1016/j.pop.2017.04.010.
jamapsychiatry.2013.3726. Severance, E.G., Prandovszky, E., Castiglione, J., Yolken, R.H., 2015. Gastroenterology
Hoffman, K.W., Lee, J.J., Corcoran, C.M., Kimhy, D., Kranz, T.M., Malaspina, D., 2020. Issues in Schizophrenia: Why the Gut Matters. Curr Psychiatry Rep 17, 27. https://
Considering the Microbiome in Stress-Related and Neurodevelopmental Trajectories doi.org/10.1007/s11920-015-0574-0.
to Schizophrenia. Front Psychiatry 11, 629. https://doi.org/10.3389/ Shayganfard, M., 2021. Are Essential Trace Elements Effective in Modulation of Mental
fpsyt.2020.00629. Disorders? Update and Perspectives. Biol Trace Elem Res. https://doi.org/10.1007/
Insel, T., Cuthbert, B., Garvey, M., Heinssen, R., Pine, D.S., Quinn, K., Sanislow, C., s12011-021-02733-y.
Wang, P., 2010. Research Domain Criteria (RDoC): Toward a New Classification Shirazi, A., Stubbs, B., Gomez, L., Moore, S., Gaughran, F., Flanagan, R.J., MacCabe, J.H.,
Framework for Research on Mental Disorders. AJP 167, 748–751. https://doi.org/ Lally, J., 2016. Prevalence and Predictors of Clozapine-Associated Constipation: A
10.1176/appi.ajp.2010.09091379. Systematic Review and Meta-Analysis. Int J Mol Sci 17, E863. https://doi.org/
John, G.K., Mullin, G.E., 2016. The Gut Microbiome and Obesity. Curr Oncol Rep 18, 45. 10.3390/ijms17060863.
https://doi.org/10.1007/s11912-016-0528-7. Silberstein, J., Harvey, P.D., 2019. Impaired introspective accuracy in schizophrenia: an
Jung, W., Jang, K.-I., Lee, S.-H., 2019. Heart and Brain Interaction of Psychiatric Illness: independent predictor of functional outcomes. Cognitive Neuropsychiatry 24,
A Review Focused on Heart Rate Variability, Cognitive Function, and Quantitative 28–39. https://doi.org/10.1080/13546805.2018.1549985.
Electroencephalography. Clin Psychopharmacol Neurosci 17, 459–474. https://doi. Singh, M.M., Kay, S.R., 1976. Wheat gluten as a pathogenic factor in schizophrenia.
org/10.9758/cpn.2019.17.4.459. Science 191, 401–402. https://doi.org/10.1126/science.1246624.
Kaplan, B.J., Rucklidge, J.J., Romijn, A., McLeod, K., 2015. The Emerging Field of Strauss, G.P., Vertinski, M., Vogel, S.J., Ringdahl, E.N., Allen, D.N., 2016. Negative
Nutritional Mental Health: Inflammation, the Microbiome, Oxidative Stress, and symptoms in bipolar disorder and schizophrenia: A psychometric evaluation of the
Mitochondrial Function. Clinical Psychological Science 3, 964–980. https://doi.org/ brief negative symptom scale across diagnostic categories. Schizophrenia Research
10.1177/2167702614555413. 170, 285–289. https://doi.org/10.1016/j.schres.2015.12.014.
Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale Tang, W., Wang, Y., Xu, F., Fan, W., Zhang, Yaoyao, Fan, K., Wang, W., Zhang, Yi,
(PANSS) for schizophrenia. Schizophr Bull 13, 261–276. https://doi.org/10.1093/ Zhang, C., 2020. Omega-3 fatty acids ameliorate cognitive dysfunction in
schbul/13.2.261. schizophrenia patients with metabolic syndrome. Brain Behav Immun 88, 529–534.
Kelly, K., Ratliff, S., Mezuk, B., 2019. Allergies, asthma, and psychopathology in a https://doi.org/10.1016/j.bbi.2020.04.034.
nationally-representative US sample. J Affect Disord 251, 130–135. https://doi.org/ Tucker, P., Pfefferbaum, B., Nitiéma, P., Khan, Q., Aggarwal, R., Walling, E.E., 2017.
10.1016/j.jad.2019.03.026. Possible link of Interleukin-6 and Interleukin-2 with psychiatric diagnosis, ethnicity,
Kimhy, D., Tay, C., Vakhrusheva, J., Beck-Felts, K., Ospina, L.H., Ifrah, C., Parvaz, M., disaster or BMI. Cytokine 96, 247–252. https://doi.org/10.1016/j.
Gross, J.J., Bartels, M.N., 2021. Enhancement of aerobic fitness improves social cyto.2017.04.025.
functioning in individuals with schizophrenia. Eur Arch Psychiatry Clin Neurosci Umetani, K., Singer, D.H., McCraty, R., Atkinson, M., 1998. Twenty-Four Hour Time
271, 367–376. https://doi.org/10.1007/s00406-020-01220-0. Domain Heart Rate Variability and Heart Rate: Relations to Age and Gender Over
Leucht, S., Leucht, C., Huhn, M., Chaimani, A., Mavridis, D., Helfer, B., Samara, M., Nine Decades. Journal of the American College of Cardiology 31, 593–601. https://
Rabaioli, M., Bächer, S., Cipriani, A., Geddes, J.R., Salanti, G., Davis, J.M., 2017. doi.org/10.1016/S0735-1097(97)00554-8.
Sixty Years of Placebo-Controlled Antipsychotic Drug Trials in Acute Schizophrenia: Vojvodic, A., Peric-Hajzler, Z., Matovic, D., Vojvodic, P., Vlaskovic-Jovicevic, T.,
Systematic Review, Bayesian Meta-Analysis, and Meta-Regression of Efficacy Sijan, G., Dimitrijevic, S., Stepic, N., Wollina, U., Badr, B.A.E., Badawi, A.,
Predictors. AJP 174, 927–942. https://doi.org/10.1176/appi.ajp.2017.16121358. Goldust, M., Tirant, M., Nguyen Van, T., Fioranelli, M., Lotti, T., 2019. Gut
Microbiota and the Alteration of Immune Balance in Skin Diseases: From

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.
S.J. Fendrich et al. Psychiatry Research 312 (2022) 114526

Nutraceuticals to Fecal Transplantation. Open Access Maced J Med Sci 7, Wijarnpreecha, K., Jaruvongvanich, V., Cheungpasitporn, W., Ungprasert, P., 2018.
3034–3038. https://doi.org/10.3889/oamjms.2019.827. Association between celiac disease and schizophrenia: a meta-analysis. European
White, L., Harvey, P.D., Opler, L., Lindenmayer, J.P., 1997. Empirical assessment of the Journal of Gastroenterology & Hepatology 30, 442–446. https://doi.org/10.1097/
factorial structure of clinical symptoms in schizophrenia. A multisite, multimodel MEG.0000000000001048.
evaluation of the factorial structure of the Positive and Negative Syndrome Scale. Yaghmaie, P., Koudelka, C.W., Simpson, E.L., 2013. Mental health comorbidity in
The PANSS Study Group. Psychopathology 30, 263–274. https://doi.org/10.1159/ patients with atopic dermatitis. J Allergy Clin Immunol 131, 428–433. https://doi.
000285058. org/10.1016/j.jaci.2012.10.041.
White, L., Opler, L.A., Harvey, P.D., Parrella, M., Friedman, J.I., 2004. Activation Zhang, Y., Zhou, B., Qiu, J., Zhang, L., Zou, Z., 2020. Heart rate variability changes in
symptoms and discharge in elderly chronic schizophrenic inpatients. J Nerv Ment Dis patients with panic disorder. Journal of Affective Disorders 267, 297–306. https://
192, 880–883. https://doi.org/10.1097/01.nmd.0000146883.74872.07. doi.org/10.1016/j.jad.2020.01.132.

Downloaded for Anonymous User (n/a) at Icahn School of Medicine at Mount Sinai from ClinicalKey.com by Elsevier on
September 29, 2023. For personal use only. No other uses without permission. Copyright ©2023. Elsevier Inc. All rights reserved.