Keywords
Epidemiological surveillance
Impairment Severity
Neural Networks
Predictive Modelling
Abstract
Introduction. Occupational diseases remain an underreported public health problem. In Italy, national insurance claims provide a useful view of recognized occupational diseases, although they do not measure the full burden of disease in the workforce. Few studies have combined descriptive analysis of insurance claims with predictive modelling to assess factors associated with recognized impairment.
Methods. We conducted a retrospective observational analysis of 310,189 occupational disease claims from the Italian National Institute for Insurance against Accidents at Work (INAIL) open data, covering the period 2020 to 2024. Multivariable linear regression was used to estimate associations between demographic, clinical, and occupational factors and the degree of impairment. Neural-network models with embedding layers were developed to classify claims with recognized impairment (>0%) and to predict impairment severity.
Results. Musculoskeletal disorders represented the largest share of claims, whereas neoplasms and asbestos-related diseases were associated with the highest impairment. Diagnostic category and causal agent were the strongest predictors of severity among claims. The classification model achieved 98.15% accuracy and 99.99% sensitivity for recognized impairment. The regression model attained a mean absolute error of 2.31% and R2=0.71, outperforming the linear regression baseline R2=0.55).
Conclusion. Combining descriptive epidemiology with machine learning yields actionable insights for prioritizing high-risk claims and supporting administrative decision-making in occupational health surveillance. These findings represent a proof of concept given inherent administrative data limitations, including underreporting, lack of workforce denominators, and potential data bias. Future developments should incorporate explainable artificial intelligence techniques to bridge the gap between predictive performance and clinical interpretability.
References
Takala J, Hämäläinen P, Sauni R, Nygård CH, Gagliardi D, Neupane S. Global-, regional- and country-level estimates of the work-related burden of diseases and accidents in 2019. Scand J Work Env Hea 2024;50:73-82. https://doi.org/10.5271/sjweh.4132.
Hämäläinen P, Takala J, Kiat TB. Global estimates of occupational accidents and work-related illnesses 2017. Technical report, Workplace Safety and Health Institute, Singapore, Available at: https://www.icohweb.org/site/images/news/pdf/Report%20Global%20Estimates%20of%20Occupational%20Accidents%20and%20Work-related%20Illnesses%202017%20rev1.PDF. Accessed on: 27/12/2025.
World Health Organization and International Labour Organization. Who/ilo joint estimates of the work-related burden of disease and injury, 2000–2016: Global monitoring report. Technical report, World Health Organization and International Labour Organization, Geneva, 2021. Available at: https://www.who.int/publications/i/item/9789240034945. Accessed on: 09/01/2026.
Azaroff LS, Levenstein C, Wegman DH. Occupational injury and illness surveillance: Conceptual filters explain underreporting. Am J Public Health 2002;92:1421-9. https://doi.org/10.2105/AJPH.92.9.1421
Khoe LC, Fitriana Saldi SR, Isbayuputra M, Mansyur M, Wiseman V, Asante A. Global reporting and underreporting of occupational diseases: A systematic review. PloS one 2026;21:e0345318. https://doi.org/10.1371/journal.pone.0345318.
Nagy K, Kudász F. Introduction to occupational diseases. OSHwiki, European Agency for Safety and Health at Work, 2022. Published 08 April 2014, last updated 09 September 2022. Available at: https://oshwiki.osha.europa.eu/en/themes/introduction-occupational-diseases. Accessed on: 15/01/2026.
European Agency for Safety and Health at Work. Work-related diseases and occupational diseases: The issue. Technical report, Publications Office of the European Union, Luxembourg, 2017. Available at: https://osha.europa.eu/en/publications/work-related-diseases-and-occupational-diseases-issue. Accessed on: 22/01/2026.
Eurostat. European occupational diseases statistics (EODS). Experimental Statistics, European Commission, 2025. Available at: https://ec.europa.eu/eurostat/web/experimental-statistics/european-occupational-diseases-statistics. Accessed on: 12/01/2026.
Larese Filon F, Spadola O, Colosio C, Van Der Molen H. Trends in Occupational Diseases in Italian Industry and Services From 2006 To 2019. Med Lav 2023;114:e2023035. https://doi.org/10.23749/mdl.v114i4.14637.
European Agency for Safety and Health at Work. Methodologies to identify work-related diseases – Review of sentinel and alert approaches, Publications Office, 2017. Available at: https://data.europa.eu/doi/10.2802/053155. Accessed on: 15/01/2026.
Lecca LI, Pili S, Lai M, Murru A, Campo G, Pizzuti A, Mattioli S, Campagna M. Diagnosing and Reporting of Occupational Diseases: An Assessment Study of Reports from an Italian Workplace Safety Prevention Program Service. Med Lav 2025;116:16609. https://doi.org/10.23749/mdl.v116i1.1660.
Decreto legislativo. Disposizioni in materia di assicurazione contro gli infortuni sul lavoro e le malattie professionali (d.lgs. 23 febbraio 2000, n. 38), 2000. Gazzetta Ufficiale n. 60, 13 marzo 2000. Available at: https://www.normattiva.it/uri-res/N2Ls?urn:nir:stato:legge:2000;38~art12. Accessed on: 21/12/2025.
Ferrante P. Respiratory Diseases with High Occupational Fraction in Italy: Results from the Italian Hospital Discharge Registry (2010-2021). Healthcare (Basel) 2024;12:2565. https://doi.org/10.3390/healthcare12242565.
van der Molen HF, Marsili C, Vitali A, Colosio C. Trends in occupational diseases in the Italian agricultural sector, 2004-2017. Occup Environ Med 2020;77:340-3. https://doi.org/10.1136/oemed-2019-106168.
Vivian GA, Bauder RA, Khoshgoftaar TM. A comprehensive survey on machine learning for workplace injury analysis: risk prediction, return to work strategies, and demographic insights. J Big Data 2025;12:167. https://doi.org/10.1186/s40537-025-01229-z.
Armenteros-Cosme P, Arias-González M, Alonso-Rollán S, Márquez-Sánchez S, Carrera A. Advancements in Artificial Intelligence and Machine Learning for Occupational Risk Prevention: A Systematic Review on Predictive Risk Modeling and Prevention Strategies. Sensors 2025;25:5419. https://doi.org/10.3390/s25175419.
Luo N, Xu X, Jiang B, Zhang Z, Huang J, Zhang X, Tan Q, Wang X, Bai S, Liu S, Pan Y, Tang, C, Zhu P. Explainable machine learning framework to predict the risk of work-related neck and shoulder musculoskeletal disorders among healthcare professionals. Front Public Health 2024;12:1414209. https://doi.org/10.3389/fpubh.2024.1414209.
Montanari P. Occupational Diseases Risk Prediction by Neural Networks. Health 2025;17:579-93. https://doi.org/10.4236/health.2025.175037.
Decreto interministeriale. Nuove tabelle delle malattie professionali nell’industria e nell’agricoltura (9 aprile 2008), 2008. Gazzetta Ufficiale n. 169, 21 luglio 2008. Available at: https://www.gazzettaufficiale.it/atto/serie_generale/caricaDettaglioAtto/originario?atto.dataPubblicazioneGazzetta=2008-07-21&atto.codiceRedazionale=08A05160&elenco30giorni=false. Accessed on: 12/12/2025.
Decreto ministeriale. Aggiornamento delle tabelle delle malattie professionali (10 ottobre 2023), 2023. Gazzetta Ufficiale n. 269, 18 novembre 2023. Available at: https://www.gazzettaufficiale.it/eli/id/2023/11/18/23A06309/sg. Accessed on: 21/01/2026.
Decreto legislativo. Attuazione dell’articolo 1 della legge 3 agosto 2007, n. 123, in materia di tutela della salute e della sicurezza nei luoghi di lavoro (d.lgs. 9 aprile 2008, n. 81), 2008. Gazzetta Ufficiale n. 101, 30 aprile 2008. Available at: https://www.normattiva.it/uri-res/N2Ls?urn:nir:stato:decreto.legislativo:2008-04-09;81!vig=. Accessed on: 24/03/2026.