A Novel Diathesis-Stress Model of Comorbid Early Onset Psychiatric Disorders

IMAGEN Consortium

doi:10.1101/2025.04.08.25325481

A Novel Diathesis-Stress Model of Comorbid Early Onset Psychiatric Disorders

IMAGEN Consortium

Research output: Contribution to journal › Article

185 Downloads (Pure)

Abstract

IMPORTANCE: Psychiatric comorbidity is the norm. Identifying transdiagnostic risk factors will inform our understanding of developmental pathways and early intervention targets.

OBJECTIVE: We recently reported that many psychiatric outcomes are predicted by a three-factor model composed of adolescent externalizing (EXT) behaviors, early life adversity, and dopamine autoreceptor availability. Here, we investigated whether this model could be reproduced in a large population-based sample using functional magnetic resonance imaging (fMRI) instead of positron emission tomography.

DESIGN: Data were collected by the IMAGEN consortium beginning in 2010 when cohort members were 14 years old, with follow-up testing at ages 16 and 19. These longitudinal data were used to predict psychiatric disorders by 19 years of age.

SETTING: Participants were recruited from secondary schools across Europe.

PARTICIPANTS: Adolescents (n = 1338) with fMRI, behavioural, diagnostic, and early life trauma data.

MAIN OUTCOMES AND MEASURES: Binary regression models tested whether a combination of EXT behaviors, childhood trauma, and mesocorticolimbic reward anticipation responses at age 14 or 19 predicted the presence of a disorder by age 19.

RESULTS: A total of 1338 participants had the required data (52.4% female). In all models, EXT and adversity scores were significant predictors ( p < 0.001). Reward anticipation responses in the ventral striatum, caudate, putamen, and anterior cingulate cortex (ACC) at age 14 ( p ≤ 0.05) and in the ventral striatum at age 19 ( p ≤ 0.029) were predictors in their respective models. The three- factor models overall were highly significant ( p < 1.0 x 10 -21 ), yielding greater predictive strength than each factor alone. They had an accuracy of nearly 75%, accounting for ≥ 11% (Nagelkerke R 2 ) of the variance in psychiatric disorders. The relationship between trauma and diagnoses was partially mediated by higher EXT (indirect path B = 0.0535, 95% CI = 0.0301- 0.0835), and moderated by fMRI responses in the ACC ( p = 0.0038) and putamen ( p = 0.0135) at age 14.

CONCLUSIONS AND RELEVANCE: The results extend our previous findings, increasing confidence in a novel diathesis-stress model of commonly comorbid early onset psychiatric disorders. The results have implications for diagnostic classification schemes and pleiotropic views of psychiatric disorder etiology.

KEY POINTS: Question: What factors contribute to a diathesis-stress model of commonly comorbid psychiatric disorders? Findings: A combination of childhood adversity, adolescent externalizing behavior, and lower mesocorticolimbic reward cue reactivity predicted psychopathology by age 19 in a large population-based cohort. Liability was modified by interactions between trauma and mesocorticolimbic responses, with trauma effects larger in those with smaller striatal and anterior cingulate cortex responses. Meaning: The study identifies a transdiagnostic diathesis-stress model of early onset psychiatric disorders.

Original language	English
Journal	MedRxiv
DOIs	https://doi.org/10.1101/2025.04.08.25325481
Publication status	Published - 10 Apr 2025

Access to Document

10.1101/2025.04.08.25325481

2025.04.08.25325481v1.full
medRxiv preprint doi: https://doi.org/10.1101/2025.04.08.25325481; this version posted April 10, 2025. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license .
Submitted manuscript, 416 KBLicence: CC BY-NC-ND

Cite this

@article{376de781b86443ecb8ac209fcfcee0d8,

title = "A Novel Diathesis-Stress Model of Comorbid Early Onset Psychiatric Disorders",

abstract = "IMPORTANCE: Psychiatric comorbidity is the norm. Identifying transdiagnostic risk factors will inform our understanding of developmental pathways and early intervention targets.OBJECTIVE: We recently reported that many psychiatric outcomes are predicted by a three-factor model composed of adolescent externalizing (EXT) behaviors, early life adversity, and dopamine autoreceptor availability. Here, we investigated whether this model could be reproduced in a large population-based sample using functional magnetic resonance imaging (fMRI) instead of positron emission tomography.DESIGN: Data were collected by the IMAGEN consortium beginning in 2010 when cohort members were 14 years old, with follow-up testing at ages 16 and 19. These longitudinal data were used to predict psychiatric disorders by 19 years of age.SETTING: Participants were recruited from secondary schools across Europe.PARTICIPANTS: Adolescents (n = 1338) with fMRI, behavioural, diagnostic, and early life trauma data.MAIN OUTCOMES AND MEASURES: Binary regression models tested whether a combination of EXT behaviors, childhood trauma, and mesocorticolimbic reward anticipation responses at age 14 or 19 predicted the presence of a disorder by age 19.RESULTS: A total of 1338 participants had the required data (52.4\% female). In all models, EXT and adversity scores were significant predictors ( p < 0.001). Reward anticipation responses in the ventral striatum, caudate, putamen, and anterior cingulate cortex (ACC) at age 14 ( p ≤ 0.05) and in the ventral striatum at age 19 ( p ≤ 0.029) were predictors in their respective models. The three- factor models overall were highly significant ( p < 1.0 x 10 -21 ), yielding greater predictive strength than each factor alone. They had an accuracy of nearly 75\%, accounting for ≥ 11\% (Nagelkerke R 2 ) of the variance in psychiatric disorders. The relationship between trauma and diagnoses was partially mediated by higher EXT (indirect path B = 0.0535, 95\% CI = 0.0301- 0.0835), and moderated by fMRI responses in the ACC ( p = 0.0038) and putamen ( p = 0.0135) at age 14. CONCLUSIONS AND RELEVANCE: The results extend our previous findings, increasing confidence in a novel diathesis-stress model of commonly comorbid early onset psychiatric disorders. The results have implications for diagnostic classification schemes and pleiotropic views of psychiatric disorder etiology.KEY POINTS: Question: What factors contribute to a diathesis-stress model of commonly comorbid psychiatric disorders? Findings: A combination of childhood adversity, adolescent externalizing behavior, and lower mesocorticolimbic reward cue reactivity predicted psychopathology by age 19 in a large population-based cohort. Liability was modified by interactions between trauma and mesocorticolimbic responses, with trauma effects larger in those with smaller striatal and anterior cingulate cortex responses. Meaning: The study identifies a transdiagnostic diathesis-stress model of early onset psychiatric disorders. ",

author = "Caswell, \{Charlotte B\} and Niki Hosseini-Kamkar and Cox, \{Sylvia M\} and Prada, \{Nicole Palacio\} and Maisha Iqbal and Maja Nikolic and Tobias Banaschewski and Barker, \{Gareth J\} and Bokde, \{Arun L W\} and R{\"u}diger Br{\"u}hl and Sylvane Desrivi{\`e}res and Herta Flor and Hugh Garavan and Penny Gowland and Antoine Grigis and Andreas Heinz and Jean-Luc Martinot and Martinot, \{Maire-Laure Paill{\`e}re\} and Eric Artiges and Frauke Nees and Orfanos, \{Dimitri Papadopoulos\} and Luise Poustka and Smolka, \{Michael N\} and Sarah Hohmann and Nilakshi Vaidya and Henrik Walter and Robert Whelan and Gunter Schumann and Tom{\'a}{\v s} Paus and Marco Leyton and \{IMAGEN Consortium\}",

year = "2025",

month = apr,

day = "10",

doi = "10.1101/2025.04.08.25325481",

language = "English",

journal = "MedRxiv",

publisher = "Cold Spring Harbor Laboratory Press",

}

TY - JOUR

T1 - A Novel Diathesis-Stress Model of Comorbid Early Onset Psychiatric Disorders

AU - Caswell, Charlotte B

AU - Hosseini-Kamkar, Niki

AU - Cox, Sylvia M

AU - Prada, Nicole Palacio

AU - Iqbal, Maisha

AU - Nikolic, Maja

AU - Banaschewski, Tobias

AU - Barker, Gareth J

AU - Bokde, Arun L W

AU - Brühl, Rüdiger

AU - Desrivières, Sylvane

AU - Flor, Herta

AU - Garavan, Hugh

AU - Gowland, Penny

AU - Grigis, Antoine

AU - Heinz, Andreas

AU - Martinot, Jean-Luc

AU - Martinot, Maire-Laure Paillère

AU - Artiges, Eric

AU - Nees, Frauke

AU - Orfanos, Dimitri Papadopoulos

AU - Poustka, Luise

AU - Smolka, Michael N

AU - Hohmann, Sarah

AU - Vaidya, Nilakshi

AU - Walter, Henrik

AU - Whelan, Robert

AU - Schumann, Gunter

AU - Paus, Tomáš

AU - Leyton, Marco

AU - IMAGEN Consortium

PY - 2025/4/10

Y1 - 2025/4/10

N2 - IMPORTANCE: Psychiatric comorbidity is the norm. Identifying transdiagnostic risk factors will inform our understanding of developmental pathways and early intervention targets.OBJECTIVE: We recently reported that many psychiatric outcomes are predicted by a three-factor model composed of adolescent externalizing (EXT) behaviors, early life adversity, and dopamine autoreceptor availability. Here, we investigated whether this model could be reproduced in a large population-based sample using functional magnetic resonance imaging (fMRI) instead of positron emission tomography.DESIGN: Data were collected by the IMAGEN consortium beginning in 2010 when cohort members were 14 years old, with follow-up testing at ages 16 and 19. These longitudinal data were used to predict psychiatric disorders by 19 years of age.SETTING: Participants were recruited from secondary schools across Europe.PARTICIPANTS: Adolescents (n = 1338) with fMRI, behavioural, diagnostic, and early life trauma data.MAIN OUTCOMES AND MEASURES: Binary regression models tested whether a combination of EXT behaviors, childhood trauma, and mesocorticolimbic reward anticipation responses at age 14 or 19 predicted the presence of a disorder by age 19.RESULTS: A total of 1338 participants had the required data (52.4% female). In all models, EXT and adversity scores were significant predictors ( p < 0.001). Reward anticipation responses in the ventral striatum, caudate, putamen, and anterior cingulate cortex (ACC) at age 14 ( p ≤ 0.05) and in the ventral striatum at age 19 ( p ≤ 0.029) were predictors in their respective models. The three- factor models overall were highly significant ( p < 1.0 x 10 -21 ), yielding greater predictive strength than each factor alone. They had an accuracy of nearly 75%, accounting for ≥ 11% (Nagelkerke R 2 ) of the variance in psychiatric disorders. The relationship between trauma and diagnoses was partially mediated by higher EXT (indirect path B = 0.0535, 95% CI = 0.0301- 0.0835), and moderated by fMRI responses in the ACC ( p = 0.0038) and putamen ( p = 0.0135) at age 14. CONCLUSIONS AND RELEVANCE: The results extend our previous findings, increasing confidence in a novel diathesis-stress model of commonly comorbid early onset psychiatric disorders. The results have implications for diagnostic classification schemes and pleiotropic views of psychiatric disorder etiology.KEY POINTS: Question: What factors contribute to a diathesis-stress model of commonly comorbid psychiatric disorders? Findings: A combination of childhood adversity, adolescent externalizing behavior, and lower mesocorticolimbic reward cue reactivity predicted psychopathology by age 19 in a large population-based cohort. Liability was modified by interactions between trauma and mesocorticolimbic responses, with trauma effects larger in those with smaller striatal and anterior cingulate cortex responses. Meaning: The study identifies a transdiagnostic diathesis-stress model of early onset psychiatric disorders.

AB - IMPORTANCE: Psychiatric comorbidity is the norm. Identifying transdiagnostic risk factors will inform our understanding of developmental pathways and early intervention targets.OBJECTIVE: We recently reported that many psychiatric outcomes are predicted by a three-factor model composed of adolescent externalizing (EXT) behaviors, early life adversity, and dopamine autoreceptor availability. Here, we investigated whether this model could be reproduced in a large population-based sample using functional magnetic resonance imaging (fMRI) instead of positron emission tomography.DESIGN: Data were collected by the IMAGEN consortium beginning in 2010 when cohort members were 14 years old, with follow-up testing at ages 16 and 19. These longitudinal data were used to predict psychiatric disorders by 19 years of age.SETTING: Participants were recruited from secondary schools across Europe.PARTICIPANTS: Adolescents (n = 1338) with fMRI, behavioural, diagnostic, and early life trauma data.MAIN OUTCOMES AND MEASURES: Binary regression models tested whether a combination of EXT behaviors, childhood trauma, and mesocorticolimbic reward anticipation responses at age 14 or 19 predicted the presence of a disorder by age 19.RESULTS: A total of 1338 participants had the required data (52.4% female). In all models, EXT and adversity scores were significant predictors ( p < 0.001). Reward anticipation responses in the ventral striatum, caudate, putamen, and anterior cingulate cortex (ACC) at age 14 ( p ≤ 0.05) and in the ventral striatum at age 19 ( p ≤ 0.029) were predictors in their respective models. The three- factor models overall were highly significant ( p < 1.0 x 10 -21 ), yielding greater predictive strength than each factor alone. They had an accuracy of nearly 75%, accounting for ≥ 11% (Nagelkerke R 2 ) of the variance in psychiatric disorders. The relationship between trauma and diagnoses was partially mediated by higher EXT (indirect path B = 0.0535, 95% CI = 0.0301- 0.0835), and moderated by fMRI responses in the ACC ( p = 0.0038) and putamen ( p = 0.0135) at age 14. CONCLUSIONS AND RELEVANCE: The results extend our previous findings, increasing confidence in a novel diathesis-stress model of commonly comorbid early onset psychiatric disorders. The results have implications for diagnostic classification schemes and pleiotropic views of psychiatric disorder etiology.KEY POINTS: Question: What factors contribute to a diathesis-stress model of commonly comorbid psychiatric disorders? Findings: A combination of childhood adversity, adolescent externalizing behavior, and lower mesocorticolimbic reward cue reactivity predicted psychopathology by age 19 in a large population-based cohort. Liability was modified by interactions between trauma and mesocorticolimbic responses, with trauma effects larger in those with smaller striatal and anterior cingulate cortex responses. Meaning: The study identifies a transdiagnostic diathesis-stress model of early onset psychiatric disorders.

U2 - 10.1101/2025.04.08.25325481

DO - 10.1101/2025.04.08.25325481

M3 - Article

C2 - 40297443

JO - MedRxiv

JF - MedRxiv

ER -

A Novel Diathesis-Stress Model of Comorbid Early Onset Psychiatric Disorders

Abstract

Access to Document

Fingerprint

Cite this