Identifying most important predictors for suicidal thoughts and behaviours among healthcare workers active during the Spain COVID-19 pandemic : A machine-learning approach
Alayo, Itxaso (Universitat Pompeu Fabra)
Pujol, Oriol (Universitat de Barcelona)
Alonso, Jordi (Instituto de Salud Carlos III)
Ferrer Forés, Maria Montserrat (Instituto de Salud Carlos III)
Amigo, Franco (Instituto de Salud Carlos III)
Portillo-Van Diest, Ana (Instituto de Salud Carlos III)
Aragonès, E. (Institut Català de la Salut. Atenció Primària Camp de Tarragona)
Aragon Peña, A. (Fundación Investigación e Innovación Biosanitaria de Atención Primaria)
Asúnsolo Del Barco, Á. (The City University of New York)
Campos Martorell, Mireia (Catalunya. Generalitat)
Espuga Jordana, Meritxell (Hospital Universitari Vall d'Hebron)
González-Pinto, Ana (Centro de Investigación Biomédica en Red de Salud Mental)
Haro Abad, Josep Maria (Institut de Recerca Sant Joan de Déu)
López Fresneña, Nieves (Hospital General Universitario Gregorio Marañón)
Martínez De Salázar, A.D. (Hospital Universitario Torrecárdenas (Almeria, Andalusia))
Molina, Juan D. (Universidad Francisco de Vitoria)
Ortí-Lucas, Rafael M. (Hospital Clínic Universitari de Valencia)
Parellada, Mara (Hospital General Universitario Gregorio Marañón)
Pelayo-Terán, José Maria (Universidad de León)
Forjaz, M.J. (Instituto de Salud Carlos III)
Pérez-Zapata, A. (Hospital Universitario Príncipe de Asturias (Alcalá de Henares, Madrid))
Pijoan, J.I. (Instituto de Investigación Sanitaria Biobizkaia (Barakaldo, País Basc))
Plana, N. (Universidad de Alcalá)
Polentinos-Castro, E. (Universidad Rey Juan Carlos)
Puig, M.T. (Centro de Investigación Biomédica en Red en Enfermedades Cardiovasculares)
Rius i Gibert, Maria Cristina (Institut de Recerca Sant Pau)
Sanz, F. (Instituto Nacional de Bioinformatica - ELIXIR-ES)
Serra, Consol (Institut Hospital del Mar d'Investigacions Mèdiques)
Urreta-Barallobre, I. (Biodonostia Osasun Ikerketako Institutura (País Basc))
Bruffaerts, Ronny (Katholieke Universiteit te Leuven (1970-))
Vieta, Eduard (Institut d'Investigacions Biomèdiques August Pi i Sunyer)
Pérez-Solá, V. (Parc de Salut MAR de Barcelona)
Mortier, Philippe (Instituto de Salud Carlos III)
Vilagut, Gemma (Instituto de Salud Carlos III)
Universitat Autònoma de Barcelona. Departament de Medicina

Data:	2025
Resum:	Aims Studies conducted during the COVID-19 pandemic found high occurrence of suicidal thoughts and behaviours (STBs) among healthcare workers (HCWs). The current study aimed to (1) develop a machine learning-based prediction model for future STBs using data from a large prospective cohort of Spanish HCWs and (2) identify the most important variables in terms of contribution to the model's predictive accuracy. Methods This is a prospective, multicentre cohort study of Spanish HCWs active during the COVID-19 pandemic. A total of 8,996 HCWs participated in the web-based baseline survey (May-July 2020) and 4,809 in the 4-month follow-up survey. A total of 219 predictor variables were derived from the baseline survey. The outcome variable was any STB at the 4-month follow-up. Variable selection was done using an L1 regularized linear Support Vector Classifier (SVC). A random forest model with 5-fold cross-validation was developed, in which the Synthetic Minority Oversampling Technique (SMOTE) and undersampling of the majority class balancing techniques were tested. The model was evaluated by the area under the Receiver Operating Characteristic (AUROC) curve and the area under the precision-recall curve. Shapley's additive explanatory values (SHAP values) were used to evaluate the overall contribution of each variable to the prediction of future STBs. Results were obtained separately by gender. Results The prevalence of STBs in HCWs at the 4-month follow-up was 7. 9% (women = 7. 8%, men = 8. 2%). Thirty-four variables were selected by the L1 regularized linear SVC. The best results were obtained without data balancing techniques: AUROC = 0. 87 (0. 86 for women and 0. 87 for men) and area under the precision-recall curve = 0. 50 (0. 55 for women and 0. 45 for men). Based on SHAP values, the most important baseline predictors for any STB at the 4-month follow-up were the presence of passive suicidal ideation, the number of days in the past 30 days with passive or active suicidal ideation, the number of days in the past 30 days with binge eating episodes, the number of panic attacks (women only) and the frequency of intrusive thoughts (men only). Conclusions Machine learning-based prediction models for STBs in HCWs during the COVID-19 pandemic trained on web-based survey data present high discrimination and classification capacity. Future clinical implementations of this model could enable the early detection of HCWs at the highest risk for developing adverse mental health outcomes.
Ajuts:	Instituto de Salud Carlos III PI17/00521 Generalitat de Catalunya 2021/SGR-00624
Drets:	Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, la comunicació pública de l'obra i la creació d'obres derivades, fins i tot amb finalitats comercials, sempre i quan es reconegui l'autoria de l'obra original.
Llengua:	Anglès
Document:	Article ; recerca ; Versió publicada
Matèria:	Attempted suicide ; Interpretability ; Machine learning ; Mental health ; Suicidal ideation
Publicat a:	Epidemiology and Psychiatric Sciences, Vol. 34 (august 2025) , p. e28, ISSN 2045-7979

DOI: 10.1017/S2045796025000198
PMID: 40340775

13 p, 1.2 MB

El registre apareix a les col·leccions:
Documents de recerca > Documents dels grups de recerca de la UAB > Centres i grups de recerca (producció científica) > Ciències de la salut i biociències > Institut de Recerca Sant Pau
Articles > Articles de recerca
Articles > Articles publicats