Predicting psychopathology symptom trajectories using machine learning: a 33‐year prospective study
Vorhersage von Psychopathologie-Symptomtrajektorien mit Machine Learning: eine prospektive 33-Jahres-Studie
Journal of Child Psychology and Psychiatry · 2026-11 · 67 (11) · 1855-1867
DOI: 10.1111/jcpp.70180 ↗Auf einen Blick
In einer Geburtskohorte wurden Externalisierungs- und Internalisierungssymptome über 25 Jahre analysiert. Drei Trajektorien ließen sich für beide Symptomgruppen identifizieren und mit genetischen sowie Umweltfaktoren vorhersagen.
Fragestellung
Lassen sich Externalisierungs- und Internalisierungssymptom-Trajektorien über 25 Jahre mit Machine Learning vorhersagen?
Methode
Longitudinale Geburtskohortenstudie; Psychopathologiesymptome von Alter 8 bis 33; Trajektorien mit Growth-Mixture-Modellen; Machine-Learning-Klassifikationsmodelle mit umfangreichen Risiko- und Schutzfaktoren.
Ergebnisse
N = 317, 171 Frauen. Drei Trajektorien für externe und interne Symptome (low, increasing, decreasing). Random forest: AUC 0,77 makro, 0,88 mikro; logistische Regression: AUC 0,75 makro, 0,83 mikro.
Schlussfolgerung der Autor:innen
Psychopathologie-Risikotrajektorien unterscheiden sich über die Entwicklung und werden von vielfältigen Risikofaktoren beeinflusst. Das kann frühe Identifikation und gezielte Prävention und Intervention unterstützen.
Kernaussagen
- Drei Trajektorien für externe und interne Symptome
- Genetische und Umweltfaktoren sagten die Trajektorien voraus
- Vorhersagegüte: AUC 0,77/0,88 bzw. 0,75/0,83
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
At a glance
Journal article in Journal of Child Psychology and Psychiatry, published online on 2026-06-15. Written by Seda Sacu and 5 co-authors. Institution: Department of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelbe…. Major MEDLINE topics: Mental Disorders, Machine Learning. Cited 1 times so far (Crossref); 49 references. The full text is freely available (open access, CC BY 4.0). Funded by: NHLBI NIH HHS, NIMH NIH HHS, Bundesministerium für Bildung und Forschung.
Abstract (original language)
Background
While previous literature has established a link between psychopathology symptom trajectories and psychosocial risk, most studies have not extended into adulthood or incorporated comprehensive risk measures. We aimed to identify symptom trajectories for externalizing and internalizing symptoms over a 25‐year span and predict them using machine learning.
Methods
We used data from a longitudinal birth cohort study ( N = 317, 171 females). Psychopathology symptoms were measured between age 8 and 33. Symptom trajectories were identified using growth mixture models. An extensive battery of risk and protective measures over the lifespan (e.g., polygenic scores, psychosocial risk factors, temperament) was used to train machine learning‐based classification models.
Results
We identified three trajectories for both externalizing and internalizing symptoms (entropy > 0.8, likelihood ratio test p < .05): Low, increasing (late‐adolescent emergence), and decreasing (childhood‐limited) symptoms classes. Both genetic (e.g., polygenic scores for cognition and mental disorders) and environmental factors (e.g., adverse family characteristics, poor social functioning) predicted symptom trajectories. For externalizing symptoms trajectories, random forest achieved fair discriminatory power with a multiclass macro‐average AUC of 0.77 and a multiclass micro‐average of 0.88. For internalizing symptoms trajectories, logistic regression achieved modest discriminatory power with a multiclass macro‐average AUC of 0.75 and a multiclass micro‐average of 0.83.
Conclusions
Psychopathology risk trajectories vary across development and are influenced by diverse risk factors. Understanding these risk factors can support early identification of at‐risk individuals and inform targeted prevention and intervention efforts.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Department of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany.
- German Center for Mental Health (DZPG), partner site Mannheim-Heidelberg-Ulm, Mannheim, Germany.
- Hector Institute for Artificial Intelligence in Psychiatry, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany.
- Department of Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany.
- Department of Genetic Epidemiology in Psychiatry, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany.
- Department of Psychiatry, NYU Grossman School of Medicine, New York, NY, USA.
Funding
- NHLBI NIH HHS
- NIMH NIH HHS
- Bundesministerium für Bildung und Forschung
Cite
APA
Sacu, S., Streit, F., Witt, S. H., Banaschewski, T., Holz, N. E., & Schultebraucks, K. (2026). Predicting psychopathology symptom trajectories using machine learning: a 33‐year prospective study. Journal of Child Psychology and Psychiatry, 67(11), 1855-1867. https://doi.org/10.1111/jcpp.70180
Vancouver
Sacu S, Streit F, Witt SH, Banaschewski T, Holz NE, Schultebraucks K. Predicting psychopathology symptom trajectories using machine learning: a 33‐year prospective study. Journal of Child Psychology and Psychiatry. 2026;67(11):1855-1867. doi:10.1111/jcpp.70180
BibTeX
@article{Sacu2026,
author = {Sacu, Seda and Streit, Fabian and Witt, Stephanie H and Banaschewski, Tobias and Holz, Nathalie E and Schultebraucks, Katharina},
title = {Predicting psychopathology symptom trajectories using machine learning: a 33‐year prospective study},
journal = {Journal of Child Psychology and Psychiatry},
year = {2026},
volume = {67},
number = {11},
pages = {1855--1867},
doi = {10.1111/jcpp.70180}
}
EUVIMED