EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access · CC BY-NC 4.0

Low-burden data-driven work addiction screening: Comparing demographic and psychosocial predictors using machine learning

Natalia A Woropay-Hordziejewicz iD, Edyta Charzyńska iD, Aleksandra Buźniak iD, Stanisław K Czerwiński iD, Zuzanna Schneider iD, Paweł A Atroszko iD

Journal of Behavioral Addictions · 15.07.2026 · 15 (2) · 900-914

DOI: 10.1556/2006.2026.00060 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift Journal of Behavioral Addictions, erschienen am 15.07.2026. Verfasst von Natalia A Woropay-Hordziejewicz und 5 weiteren Autor:innen. Beteiligte Einrichtung: 1Faculty of Psychology, University of Warsaw, Warsaw, Poland. Hauptthemen laut MEDLINE: Machine Learning, Behavior, Addictive, Occupational Stress. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Science Centre, Poland, Science Committee of the Republic of Armenia, NPO “Systemic Risk Institute”.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background and Aims Identifying employees at risk for work addiction is critical for early intervention, yet systematic screening approaches face practical challenges for organizations. This study compared two screening strategies: a demographic-only model that requires no additional data collection versus a psychosocial model that incorporates job satisfaction and job stress assessments.

Methods

We analyzed data from employees across 85 countries or territories. Dataset 1 (psychosocial model, n = 22,136) included demographic, occupational, and psychosocial variables. Dataset 2 (demographic model, n = 23,219) used only readily available information. Work addiction risk was assessed using the International Work Addiction Scale. Five machine learning algorithms (Logistic Regression, Random Forest, XGBoost, Neural Networks, Ensemble Voting) were evaluated on both full and Elastic Net-selected feature sets using cross-validation.

Results

The psychosocial model demonstrated superior performance. The best-balanced model (Logistic Regression with Elastic Net selection) achieved balanced accuracy of 0.72 ± 0.01, AUC-ROC of 0.80 ± 0.01, and Youden's J of 0.45 ± 0.02, correctly identifying 72.7% of at-risk individuals while maintaining 72.2% specificity. The demographic model demonstrated moderate performance (Ensemble Voting: balanced accuracy 0.61 ± 0.01, AUC-ROC 0.65 ± 0.01), yet retained practical utility, detecting 57.0% of at-risk cases. High-sensitivity configurations achieved detection rates of 71%, but with elevated false-positive rates, presenting important trade-offs for screening design. Discussion and Conclusions While psychosocial variables provide superior prediction, demographic-only models offer a practical screening alternative that can identify over half of at-risk individuals without an additional data-collection burden. This trade-off between predictive performance and implementation feasibility has important implications for organizational screening programs.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningBehavior, AddictiveOccupational StressHumansPredictive Learning ModelsBoosting Machine Learning AlgorithmsFemaleClassification AlgorithmsAdultMaleJob SatisfactionRandom ForestLogistic ModelsPrediction AlgorithmsMass Screening

Stichwörter der Autor:innen

job stressmachine learningoccupational healthscreeningwork addictionworkaholism

Autor:innen und Institutionen

  • 1Faculty of Psychology, University of Warsaw, Warsaw, Poland.
  • 2Institute of Psychology, Institute of Pedagogy, Faculty of Social Sciences, University of Silesia in Katowice, Katowice, Poland.
  • 3Institute of Psychology, Faculty of Social Sciences, University of Gdańsk, Gdańsk, Poland.
  • 4Institute of Psychology, Faculty of Social Sciences, University of Silesia in Katowice, Katowice, Poland.
  • 5Department of Psychological Sciences, FLAME University, Pune, India.

ORCID-Profile: Natalia A Woropay-Hordziejewicz, Edyta Charzyńska, Aleksandra Buźniak, Stanisław K Czerwiński, Zuzanna Schneider, Paweł A Atroszko

Förderung

  • National Science Centre, Poland
  • Science Committee of the Republic of Armenia
  • NPO “Systemic Risk Institute”
  • Estonian Research Council
  • Hungarian National Research, Development, and Innovation Office
  • János Bolyai Research Scholarship of the Hungarian Academy of Sciences

Zitieren

APA

Woropay-Hordziejewicz, N. A., Charzyńska, E., Buźniak, A., Czerwiński, S. K., Schneider, Z., & Atroszko, P. A. (2026). Low-burden data-driven work addiction screening: Comparing demographic and psychosocial predictors using machine learning. Journal of Behavioral Addictions, 15(2), 900-914. https://doi.org/10.1556/2006.2026.00060

Vancouver

Woropay-Hordziejewicz NA, Charzyńska E, Buźniak A, Czerwiński SK, Schneider Z, Atroszko PA. Low-burden data-driven work addiction screening: Comparing demographic and psychosocial predictors using machine learning. Journal of Behavioral Addictions. 2026;15(2):900-914. doi:10.1556/2006.2026.00060

BibTeX
@article{WoropayHordziejewicz2026,
  author = {Woropay-Hordziejewicz, Natalia A and Charzyńska, Edyta and Buźniak, Aleksandra and Czerwiński, Stanisław K and Schneider, Zuzanna and Atroszko, Paweł A},
  title = {Low-burden data-driven work addiction screening: Comparing demographic and psychosocial predictors using machine learning},
  journal = {Journal of Behavioral Addictions},
  year = {2026},
  volume = {15},
  number = {2},
  pages = {900--914},
  doi = {10.1556/2006.2026.00060}
}
Kostenlos starten →