Identifying most important predictors for suicidal thoughts and behaviours among healthcare workers active during the Spain COVID-19 pandemic: a machine-learning approach


Por: Alayo I, Pujol O, Alonso J, Ferrer M, Amigo F, Portillo-Van Diest A, Aragonès E, Aragon Peña A, Asúnsolo Del Barco Á, Campos M, Espuga M, González-Pinto A, Haro JM, López-Fresneña N, Martínez de Salázar AD, Molina JD, Ortí-Lucas RM, Parellada M, Pelayo-Terán JM, Forjaz MJ, Pérez-Zapata A, Pijoan JI, Plana N, Polentinos-Castro E, Puig MT, Rius C, Sanz F, Serra C, Urreta-Barallobre I, Bruffaerts R, Vieta E, Pérez-Solá V, Mortier P and Gemma Vilagut Saiz

Publicada: 8 may 2025 Ahead of Print: 8 may 2025
Resumen:
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.Study registration NCT04556565

Filiaciones:
Alayo I:
 Hospital del Mar Research Institute, Barcelona, Spain

 Biosistemak Institute for Health Systems Research, Bilbao, Bizkaia, Spain

 Red de Investigación en Cronicidad, Atención Primaria y Promoción de la Salud RICAPPS-(RICORS), Instituto de Salud Carlos III (ISCIII), Madrid, Spain

 Department of Medicine and Life Sciences (MELIS), Universitat Pompeu Fabra, Barcelona, Spain

Pujol O:
 Departament de Matemàtiques i Informàtica, Universitat de Barcelona, Barcelona, Spain

Alonso J:
 Hospital del Mar Research Institute, Barcelona, Spain

 Department of Medicine and Life Sciences (MELIS), Universitat Pompeu Fabra, Barcelona, Spain

 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

Ferrer M:
 Hospital del Mar Research Institute, Barcelona, Spain

 Department of Medicine and Life Sciences (MELIS), Universitat Pompeu Fabra, Barcelona, Spain

 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

Amigo F:
 Hospital del Mar Research Institute, Barcelona, Spain

 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

Portillo-Van Diest A:
 Hospital del Mar Research Institute, Barcelona, Spain

 Department of Medicine and Life Sciences (MELIS), Universitat Pompeu Fabra, Barcelona, Spain

 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

Aragonès E:
 Institut d'Investigació en Atenció Primària IDIAP Jordi Gol, Barcelona, Spain

 Atenció Primària Camp de Tarragona, Institut Català de la Salut, Tarragona, Spain

Aragon Peña A:
 Epidemiology Unit, Regional Ministry of Health, Community of Madrid, Madrid, Spain

 Fundación Investigación e Innovación Biosanitaria de AP, Comunidad de Madrid, Madrid, Spain

Asúnsolo Del Barco Á:
 Department of Surgery, Medical and Social Sciences, Faculty of Medicine and Health Sciences, University of Alcala, Alcalá de Henares, Spain

 Ramón y Cajal Institute of Sanitary Research (IRYCIS), Madrid, Spain

 Department of Epidemiology and Biostatistics, Graduate School of Public Health and Health Policy, The City University of New York, New York, NY, USA

Campos M:
 Service of Prevention of Labor Risks, Medical Emergencies System, Generalitat de Catalunya, Barcelona, Spain

Espuga M:
 Occupational Health Service, Hospital Universitari Vall d'Hebron, Barcelona, Spain

González-Pinto A:
 BIOARABA, Hospital Universitario Araba-Santiago, UPV/EHU, Vitoria-Gasteiz, Spain

 CIBER Salud Mental (CIBERSAM), Madrid, Spain

Haro JM:
 CIBER Salud Mental (CIBERSAM), Madrid, Spain

 Parc Sanitari Sant Joan de Déu, Institut de Recerca Sant Joan de Deu (IRSJD), Sant Boi de Llobregat, Barcelona, Spain

López-Fresneña N:
 Hospital General Universitario Gregorio Marañón, Madrid, Spain

Martínez de Salázar AD:
 UGC Salud Mental, Hospital Universitario Torrecárdenas, Almería, Spain

Molina JD:
 CIBER Salud Mental (CIBERSAM), Madrid, Spain

 Villaverde Mental Health Center, Clinical Management Area of Psychiatry and Mental Health, Psychiatric Service, Hospital Universitario 12 de Octubre, Madrid, Spain

 Research Institute Hospital 12 de Octubre (i+12), Madrid, Spain

 Facultad de Medicina, Universidad Francisco de Vitoria, Madrid, Spain

Ortí-Lucas RM:
 Servicio de Medicina Preventiva y Calidad Asistencial, Hospital Clínic Universitari de Valencia, Valencia, Spain

Parellada M:
 CIBER Salud Mental (CIBERSAM), Madrid, Spain

 Hospital General Universitario Gregorio Marañón, Madrid, Spain

Pelayo-Terán JM:
 CIBER Salud Mental (CIBERSAM), Madrid, Spain

 Servicio de Psiquiatría y Salud Mental, Hospital el Bierzo, Gerencia de Asistencia Sanitaria del Bierzo (GASBI). Gerencia Regional de Salud de Castilla y Leon (SACYL), Ponferrada, León, Spain

 Area de Medicina Preventiva y Salud Pública, Departamento de Ciencias Biomédicas, Universidad de León, León, Spain

Forjaz MJ:
 Red de Investigación en Cronicidad, Atención Primaria y Promoción de la Salud RICAPPS-(RICORS), Instituto de Salud Carlos III (ISCIII), Madrid, Spain

 National Center of Epidemiology, Instituto de Salud Carlos III (ISCIII), Madrid, Spain

Pérez-Zapata A:
 Hospital Universitario Príncipe de Asturias, Servicio de Prevención de Riesgos Laborales, Spain

Pijoan JI:
 Red de Investigación en Cronicidad, Atención Primaria y Promoción de la Salud RICAPPS-(RICORS), Instituto de Salud Carlos III (ISCIII), Madrid, Spain

 Clinical Epidemiology Unit-Hospital Universitario Cruces/ OSI EEC, Biobizkaia Health Research Institute, Barakaldo, Spain

Plana N:
 Red de Investigación en Cronicidad, Atención Primaria y Promoción de la Salud RICAPPS-(RICORS), Instituto de Salud Carlos III (ISCIII), Madrid, Spain

 Ramón y Cajal University Hospital, IRYCIS, Department of Surgery, Medical and Social Sciences, Faculty of Medicine and Health Sciences, University of Alcalá, Alcala de Henares, MAD, Spain

Polentinos-Castro E:
 Service of Prevention of Labor Risks, Medical Emergencies System, Generalitat de Catalunya, Barcelona, Spain

 Research Unit, Primary Care Management, Madrid Health Service, Madrid, Spain

 Department of Medical Specialities and Public Health, King Juan Carlos University, Madrid, Spain

Puig MT:
 Universitat Autònoma de Barcelona (UAB), Barcelona, Spain

 Department of Epidemiology and Public Health, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain

 Biomedical Research Institute Sant Pau (IIB Sant Pau), Barcelona, Spain

 CIBER Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain

Rius C:
 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

 Agència de Salut Pública de Barcelona, Barcelona, Spain

Sanz F:
 Department of Medicine and Life Sciences (MELIS), Universitat Pompeu Fabra, Barcelona, Spain

 Research Progamme on Biomedical Informatics (GRIB), Hospital del Mar Research Institute, Barcelona, Spain

 Instituto Nacional de Bioinformatica - ELIXIR-ES, Barcelona, Spain

Serra C:
 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

 CiSAL-Centro de Investigación en Salud Laboral, Hospital del Mar Research Institute/University Pompeu Fabra, Barcelona, Spain

 Occupational Health Service, Hospital del Mar, Barcelona, Spain

Urreta-Barallobre I:
 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

 Osakidetza Basque Health Service, Donostialdea Integrated Health Organisation, Donostia University Hospital, Clinical Epidemiology Unit, San Sebastián, Spain

 Biodonostia Health Research Institute, Clinical Epidemiology, San Sebastián, Spain

Bruffaerts R:
 Center for Public Health Psychiatry, Universitair Psychiatrisch Centrum, KU Leuven, Leuven, Belgium

Vieta E:
 CIBER Salud Mental (CIBERSAM), Madrid, Spain

 Institute of Neuroscience, Hospital Clinic, University of Barcelona, IDIBAPS, Barcelona, Spain

Pérez-Solá V:
 CIBER Salud Mental (CIBERSAM), Madrid, Spain

 Universitat Autònoma de Barcelona (UAB), Barcelona, Spain

 Institute of Neuropsychiatry and Addiction (INAD), Parc de Salut Mar, Barcelona, Spain

Mortier P:
 Hospital del Mar Research Institute, Barcelona, Spain

 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain

Gemma Vilagut Saiz:
 Hospital del Mar Research Institute, Barcelona, Spain

 CIBER de Epidemiología y Salud Pública (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain
ISSN: 20457960





Epidemiology and Psychiatric Sciences
Editorial
CAMBRIDGE UNIV PRESS, EDINBURGH BLDG, SHAFTESBURY RD, CB2 8RU CAMBRIDGE, ENGLAND, Reino Unido
Tipo de documento: Article
Volumen: 34 Número:
Páginas:
WOS Id: 001484077700001
ID de PubMed: 40340775
imagen Green Submitted, gold

MÉTRICAS