Abstract
The risk of work disability can be predicted with machine learning and using various data sources. In this study, we compare two methods that use different data sources, occupational health care data and pension decision register data. The first method, medical report data classification for two classes (MHealth), reached an accuracy of 72% by using neural networks. This tool is used in occupational health care to help in the screening process of the patients. The second method, pension data prediction (MPension), achieved an accuracy of 69–78% depending on the algorithm used. Using pension decision register data, disability was best predicted by old age, high use of sickness benefits and rehabilitation allowance in previous years, and low or no earnings in previous years. Based on the experiments, machine learning techniques appear to be a potential tool to support expert work and decision-making.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
OECD. Sickness, Disability and Work: Breaking the Barriers. OECD Publishing, Paris, (2010)
Airaksinen, J., et al.: Development and validation of a risk prediction model for work disability: multicohort study. Sci. Rep. 7(13578), 1–12 (2017)
Salonen, L., Blomgren, J., Laaksonen, M.: From long-term sickness absence to disability retirement: diagnostic and occupational class differences within the working-age Finnish population. BMC Public Health 20(1), 1–11 (2020)
Salonen, L., Alexanderson, K., Rugulies, R., et al.: Combinations of Job demands and job control and future trajectories of sickness absence and disability pension an 11-year follow-up of two million employees in Sweden. J. Occup. Environ. Med. 62(10), 795–802 (2020)
Bethge, M., Spanier, K., Köhn, S., et al.: Self-reported work ability predicts health-related exit and absence from work, work participation, and death: longitudinal findings from a sample of German employees. Int. Arch. Occup. Environ. Health 94(4), 591–599 (2020). https://doi.org/10.1007/s00420-020-01608-4
Shiri, R., et al.: A screening tool for the risk of disability retirement due to musculoskeletal disoders. Scand. J. Work Environ. Health 44(1), 37–46 (2018)
van Amelsvoort, L.G.P.M., Kant, I.J., Beurskens, A.J.H.M., Schröer, C.A.P., Swaen, G.M.H.: Fatigue as a predictor of work disability. Occup. Environ. Med. 59(10), 712–713 (2002)
Reho, T.: Frequent attenders of occupational health primary care and work disability. Tampere University Dissertations 225 (2020)
Lindström, I., Pallasaho, P., Remes, J., et al.: Does lung function predict the risk of disability pension? An 11-year register-based follow-up study. BMC Public Health 20(165), 1–11 (2020)
Pham, A.D., et al.: Natural language processing of radiology reports for the detection of thromboembolic diseases and clinically relevant incidental findings. BMC Bioinform. 15(1), 266 (2014)
Huhta-Koivisto, T.: Work disability risk prediction with machine learning. Master thesis. Aalto University, School of Electrical Engineering, Espoo, 56 p. (2020)
Varis, J.: Eläketurvakeskuksen koneoppimiskokeilu – näin se tehtiin! Finnish Centre for Pensions, blog post. https://www.etk.fi/blogit/elaketurvakeskuksen-koneoppimiskokeilu-nain-se-tehtiin/. Accesed 17 Apr 2018
European Pensions: Finnish AI testing successfully identifies future retirees facing disability pension. https://www.europeanpensions.net/ep/Finnish-AI-successfully-identifies-future-retirees-facing-disability-pension.php. Accesed 17 Apr 2018
Howard, J., Ruder, S.: Universal language model fine-tuning for text classification, In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, vol. 1 (2018)
Devlin, J., Ming-Wei, C., Lee, K., Toutanova, K.: BERT: pre-training of deep bidirectional transformers for language understanding (2019). arXiv:1810.04805v2
Vaswani, A., et al.: Attention is all you need. In: Advances in Neural Information Processing Systems, pp. 5998–6008 (2017)
Bergström, G., Hagberg, J., Busch, H., Jensen, I., Björklund, C.: Prediction of sickness absenteeism, disability pension and sickness presenteeism among employees with back pain. J Occup. Rehabil. 24(2), 278–286 (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Saarela, K., Huhta-Koivisto, V., Nurminen, J.K. (2022). Work Disability Risk Prediction Using Machine Learning, Comparison of Two Methods. In: Daimi, K., Al Sadoon, A. (eds) Proceedings of the ICR’22 International Conference on Innovations in Computing Research. ICR 2022. Advances in Intelligent Systems and Computing, vol 1431. Springer, Cham. https://doi.org/10.1007/978-3-031-14054-9_2
Download citation
DOI: https://doi.org/10.1007/978-3-031-14054-9_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-14053-2
Online ISBN: 978-3-031-14054-9
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)