Abstract
Purpose
Machine learning (ML) methods showed a higher accuracy in identifying individuals without cancer who were unable to return to work (RTW) compared to the classical methods (e.g. logistic regression models). We therefore aim to discuss the value of these methods in relation to RTW for cancer survivors.
Methods
Breast cancer (BC) survivors who were working at diagnosis within the CONSTANCES cohort were included in the study. RTW was assessed five years after the BC diagnosis (early retirement was considered as non-RTW). Age and occupation at diagnosis, and physical occupational job exposures assessed using the Job Exposure Matrix, JEM-CONSTANCES, were evaluated as predictors of RTW five years after BC diagnosis. The following four ML methods were used: (i) k-nearest neighbors; (ii) random forest; (iii) neural network; and (iv) elastic net.
Results
The training sample included 683 BC survivors (RTW: 85.7%), and the test sample 171 (RTW: 85.4%). The elastic net method had the best results despite low sensitivity (accuracy = 76.6%; sensitivity = 31.7%; specificity = 90.8%), and the random forest model was the most accurate (= 79.5%) but also the least sensitive (= 14.3%).
Conclusion
This study takes a first step towards opening up new possibilities for identifying the occupational determinants of cancer survivors’ RTW. Further work, including a larger sample size, and more predictor variables, is now needed.
Similar content being viewed by others
References
Gragnano A, Negrini A, Miglioretti M, Corbière M. Common psychosocial factors Predicting Return to Work after Common Mental Disorders, Cardiovascular Diseases, and cancers: a review of Reviews supporting a Cross-Disease Approach. J Occup Rehabil. 2018;28:215–31.
Tamminga SJ, de Wind A, Greidanus MA, Coenen P, Friberg E, Oldenburg H et al. Prognostic factors for return to work in breast cancer survivors. Cochrane Database of Systematic Reviews [Internet]. John Wiley & Sons, Ltd; 2022 [cited 2022 Jul 3]; Available from: https://www.cochranelibrary.com/cdsr/doi/https://doi.org/10.1002/14651858.CD015124/full
Wang L, Hong BY, Kennedy SA, Chang Y, Hong CJ, Craigie S, et al. Predictors of unemployment after breast Cancer surgery: a systematic review and Meta-analysis of Observational Studies. J Clin Oncol. 2018;36:1868–79.
Porro B, Michel A, Zinzindohoué C, Bertrand P, Monrigal E, Trentini F et al. Quality of life, fatigue and changes therein as predictors of return to work during breast cancer treatment. Scandinavian Journal of Caring Sciences [Internet]. 2019 [cited 2019 Mar 3]; Available from: https://doi.org/10.1111/scs.12646
Dumas A, Vaz Luis I, Bovagnet T, El Mouhebb M, Di Meglio A, Pinto S et al. Impact of Breast Cancer Treatment on Employment: Results of a Multicenter Prospective Cohort Study (CANTO).Journal of Clinical Oncology. 2020;JCO.19.01726.
Granell M, Senín A, Barata A, Cibeira M-T, Gironella M, López-Pardo J et al. Predictors of return to work after autologous stem cell transplantation in patients with multiple myeloma.Bone Marrow Transplant. 2021
Gross DP, Steenstra IA, Harrell FE, Bellinger C, Zaïane O. Machine learning for Work Disability Prevention: introduction to the Special Series. J Occup Rehabil. 2020;30:303–7.
Song X, Mitnitski A, Cox J, Rockwood K. Comparison of machine learning techniques with classical statistical models in Predicting Health Outcomes. MEDINFO 2004. IOS Press; 2004. pp. 736–40.
Houston A, Cosma G, Turner P, Bennett A. Predicting surgical outcomes for chronic exertional compartment syndrome using a machine learning framework with embedded trust by interrogation strategies. Sci Rep. 2021;11:24281.
Na K-S, Kim E. A machine learning-based predictive model of Return to Work after Sick leave. J Occup Environ Med. 2019;61:e191–9.
Lee J, Kim H-R. Prediction of return-to-original-work after an Industrial Accident using machine learning and comparison of techniques. J Korean Med Sci. 2018;33:e144.
Chen Y-C, Chen Y-L, Kuo D-P, Li Y-T, Chiang Y-H, Chang J-J, et al. Personalized prediction of postconcussive Working Memory decline: a feasibility study. J Pers Med. 2022;12:196.
Iosa M, Capodaglio E, Pelà S, Persechino B, Morone G, Antonucci G, et al. Artificial neural network analyzing Wearable device Gait Data for identifying patients with stroke unable to return to work. Front Neurol. 2021;12:650542.
Miotto R, Percha BL, Glicksberg BS, Lee H-C, Cruz L, Dudley JT, et al. Identifying Acute Low Back Pain Episodes in Primary Care Practice from Clinical Notes: Observational Study. JMIR Med Inform. 2020;8:e16878.
Deo RC. Machine learning in Medicine. Circulation. 2015;132:1920–30.
Shimizu H, Nakayama KI. Artificial intelligence in oncology. Cancer Sci. 2020;111:1452–60.
Cartus AR, Naimi AI, Himes KP, Jarlenski M, Parisi SM, Bodnar LM. Can Ensemble Machine Learning improve the accuracy of severe maternal morbidity screening in a Perinatal Database? Epidemiology. 2022;33:95–104.
Connor CW. Artificial Intelligence and Machine Learning in Anesthesiology. Anesthesiology. 2019;131:1346–59.
Christie SA, Conroy AS, Callcut RA, Hubbard AE, Cohen MJ. Dynamic multi-outcome prediction after injury: applying adaptive machine learning for precision medicine in trauma. PLoS ONE. 2019;14:e0213836.
Badreau M, Fadel M, Graszyck P, Descatha A. [Machine Learning: fundamentals for environmental and occupational health professionals]. Encycl Med Chir (Elservier Paris), Pathologie professionnelle,. 2023;acccepted.
Fauquemberg L, Le guen V, Badreau M, Gilbert F, Descatha A. Sectors-Job retention/disability matrix on Reunion Island: from description and prediction by Machine Learning to prevention of job loss. Arch MalProf Environ. 2023;accepted.
Porro B, Durand M-J, Petit A, Bertin M, Roquelaure Y. Return to work of breast cancer survivors: toward an integrative and transactional conceptual model. J Cancer Surviv. 2022;16:590–603.
Porro B, Campone M, Moreau P, Roquelaure Y. Supporting the Return to Work of Breast Cancer Survivors: From a Theoretical to a Clinical Perspective. International Journal of Environmental Research and Public Health. Multidisciplinary Digital Publishing Institute; 2022;19:5124.
Goldberg M, Carton M, Descatha A, Leclerc A, Roquelaure Y, Santin G, et al. CONSTANCES: a general prospective population-based cohort for occupational and environmental epidemiology: cohort profile. Occup Environ Med. 2017;74:66–71.
Zins M, Goldberg M. The french CONSTANCES population-based cohort: design, inclusion and follow-up. Eur J Epidemiol. 2015;30:1317–28.
Descatha A, Fadel M, Sembajwe G, Peters S, Evanoff BA, Job-Exposure Matrix. A Useful Tool for Incorporating Workplace Exposure Data Into Population Health Research and Practice. Frontiers in Epidemiology [Internet]. 2022 [cited 2022 Aug 10];2. Available from: https://www.frontiersin.org/articles/https://doi.org/10.3389/fepid.2022.857316
Evanoff BA, Yung M, Buckner-Petty S, Andersen JH, Roquelaure Y, Descatha A, et al. The CONSTANCES job exposure matrix based on self-reported exposure to physical risk factors: development and evaluation. Occup Environ Med. 2019;76:398–406.
Brünger M, Bernert S, Spyra K. Occupation as a Proxy for Job Exposures? Routine Data Analysis using the Example of Rehabilitation. Gesundheitswesen. 2020;82:41–51.
van der Laan MJ, Polley EC, Hubbard AE. Super Learner. Statistical Applications in Genetics and Molecular Biology [Internet]. De Gruyter; 2007 [cited 2022 Aug 13];6. Available from: https://www.degruyter.com/document/doi/https://doi.org/10.2202/1544-6115.1309/html
Hastie T, Tibshirani R, Friedman J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Second Edition. Springer Science & Business Media; 2009.
van Hoffen MFA, Norder G, Twisk JWR, Roelen CAM. Development of prediction models for sickness absence due to Mental Disorders in the General Working Population. J Occup Rehabil. 2020;30:308–17.
Gross DP, Zhang J, Steenstra I, Barnsley S, Haws C, Amell T, et al. Development of a computer-based clinical decision support tool for selecting appropriate rehabilitation interventions for injured workers. J Occup Rehabil. 2013;23:597–609.
Gross DP, Steenstra IA, Shaw W, Yousefi P, Bellinger C, Zaïane O. Validity of the Work Assessment Triage Tool for Selecting Rehabilitation Interventions for Workers’ Compensation Claimants with Musculoskeletal Conditions. J Occup Rehabil. 2020;30:318–30.
Six Dijkstra MWMC, Siebrand E, Dorrestijn S, Salomons EL, Reneman MF, Oosterveld FGJ, et al. Ethical considerations of using machine learning for decision support in Occupational Health: an Example Involving Periodic Workers’ Health assessments. J Occup Rehabil. 2020;30:343–53.
Funding
This manuscript was prepared as part of the SIRIC ILIAD program supported by the French National Cancer Institute (INCa), the French Ministry of Health and the Institute of Health and Medical Research (Inserm); INCa-DGOS-Inserm_12558 contract.
This manuscript was prepared as part of the TEC-TOP project (Pays de la Loire Region, Angers Loire Metropole, Univ Angers and CHU Angers).
Author information
Authors and Affiliations
Contributions
Marie Badreau (MSc): conception and design of the study, analysis, interpretation of data, drafting the article and the final approval of the version to be submitted; Marc Fadel (MSc, MD): conception and design of the study, analysis, interpretation of data, revising the article critically for important intellectual content and the final approval of the version to be submitted; Prof. Yves Roquelaure (PhD, MD): acquisition of data and the final approval of the version to be submitted; Mélanie Bertin (PhD): acquisition of data, interpretation of data, revising the article critically for important intellectual content and the final approval of the version to be submitted; Clémence Rapicault (MSc): revising the article critically for important intellectual content and the final approval of the version to be submitted; Fabien Gilbert (MSc): acquisition of data and the final approval of the version to be submitted; Bertrand Porro (PhD): interpretation of data, drafting the article, revising the article critically for important intellectual content and the final approval of the version to be submitted; Prof Alexis Descatha (PhD, MD): conception and design of the study, acquisition of data, revising the article critically for important intellectual content and the final approval of the version to be submitted.
Corresponding author
Ethics declarations
Conflict of Interest
Alexis Descatha is paid as editor in chief by Elsevier (Archives des maladies professionnelles et de l?environnement). Other authors have no conflict of interest to declare.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic Supplementary Material
Below is the link to the electronic supplementary material.
Table S1:Table S2:Table S3:
Some significant job codes description Physical exposures from JEM-CONSTANCES Descriptive analysis of the test sample
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Badreau, M., Fadel, M., Roquelaure, Y. et al. Comparison of Machine Learning Methods in the Study of Cancer Survivors’ Return to Work: An Example of Breast Cancer Survivors with Work-Related Factors in the CONSTANCES Cohort. J Occup Rehabil 33, 750–756 (2023). https://doi.org/10.1007/s10926-023-10112-8
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10926-023-10112-8