Abstract
Purpose Post-injury health service utilization (HSU) contributes to injury outcomes, but limited studies investigated their relationship. This study aims to group injured patients in transport accidents based on minimal historical information of their HSU so that the groups are meaningfully associated with the outcome of interest. Methods The data include 20,692 injured patients who had compensation claims over 3 years. We propose a hybrid approach, combining unsupervised and supervised machine learning methods. Based on the first week post-injury data, we identify a proper clustering of patients best associated with total cost to recovery, as well as the discovery of HSU patterns. This allows developing models to accurately predict the outcome of interest using the discovered patterns. Furthermore, we propose to use decision tree classifiers to accurately classify future patients into the discovered clusters using their first week post-injury information. Results Our hybrid approach has identified eight patient groups. The compactness of the resulted clusters, assessed by Average Silhouette Width metric, is 0.71 indicating well-defined clusters. The resulted patient groups are highly predictive of injury outcomes. They improve the cost predictability more than twice in comparison with predictors such as gender, age and injury type. These groups also have substantial association with patients’ recovery. The transparency and interpretability of decision trees allow integrating the resulting classification rules conveniently in operational processes. Conclusions This study provides a framework to discover knowledge and useful insights for health service providers and policy makers to control injury outcomes, and consequently to reduce the severity of transport accidents.
Similar content being viewed by others
References
Eyler L, Hubbard A, Juillard C. Assessment of economic status in trauma registries: a new algorithm for generating population-specific clustering-based models of economic status for time-constrained low-resource settings. Int J Med Inform. 2016;94:49–58.
United-Nations. Open Working Group proposal for Sustainable Development Goals. 2014. https://sustainabledevelopment.un.org/focussdgs.html. Accessed 15 Nov 2017.
Khorshidi HA, Gunawan I, Nikfalazar S. Application of fuzzy risk analysis for selecting critical processes in implementation of SPC with a case study. Group Decis Negot. 2015;25(1):203–220.
Baker SP, et al. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma Acute Care Surg. 1974;14(3):187–196.
van Dijck JTJM, et al. In-hospital costs after severe traumatic brain injury: a systematic review and quality assessment. PLoS ONE. 2019;14(5):e0216743.
Scheetz LJ, Zhang J, Kolassa J. Classification tree modeling to identify severe and moderate vehicular injuries in young and middle-aged adults. Artif Intell Med. 2009;45(1):1–10.
Mitchell RJ, Cameron CM, McClure R. Patterns of health care use of injured adults: a population-based matched cohort study. Injury. 2017;48(7):1393–1399.
Phillips-Wren G, Sharkey P, Dy SM. Mining lung cancer patient data to assess healthcare resource utilization. Expert Syst Appl. 2008;35(4):1611–1619.
Yoo I, et al. Data mining in healthcare and biomedicine: a survey of the literature. J Med Syst. 2012;36(4):2431–2448.
Esfandiari N, et al. Knowledge discovery in medicine: current issue and future trend. Expert Syst Appl. 2014;41(9):4434–4463.
Collie A, Prang KH. Patterns of healthcare service utilisation following severe traumatic brain injury: an idiographic analysis of injury compensation claims data. Injury. 2013;44(11):1514–1520.
Chang HT, et al. Utilization and patterns of community healthcare services for senior residents in long-term care facilities in Taiwan: a nationwide study. J Chin Med Assoc. 2013;76(1):42–47.
Lyon AR, et al. Patterns and predictors of mental healthcare utilization in schools and other service sectors among adolescents at risk for depression. School Mental Health. 2013;5(3):155–165.
Weymann D, et al. High-cost users of prescription drugs: a population-based analysis from British Columbia, Canada. Health Serv Res. 2017;52(2):697–719.
Forestier G, et al. Finding discriminative and interpretable patterns in sequences of surgical activities. Artif Intell Med. 2017;82:11–19.
Brewster AL, et al. Patterns of collaboration among health care and social services providers in communities with lower health care utilization and costs. Health Serv Res. 2018;53:2892–2909.
Berecki-Gisolf J, Collie A, McClure R. Reduction in health service use for whiplash injury after motor vehicle accidents in 2000-2009: results from a defined population. J Rehabil Med. 2014;45(10):1034–1041.
Prang KH, Ruseckaite R, Collie A. Healthcare and disability service utilization in the 5-year period following transport-related traumatic brain injury. Brain Inj. 2012;26(13–14):1611–1620.
Khorshidi HA, Marembo M, Aickelin U. Predictors of return to work for occupational rehabilitation users in work-related injury insurance claims: insights from mental health. Journal of Occupational Rehabilitation. 2019. https://doi.org/10.1007/s10926-019-09835-4.
Prang KH, Hassani-Mahmooei B, Collie A. Compensation Research Database: population-based injury data for surveillance, linkage and mining. BMC Res Notes. 2016;9(1):1–11.
Department of Health, MBS online. 2017. http://www9.health.gov.au/mbs/search.cfm?adv=1.
James G, et al. An introduction to statistical learning: with applications in R. New York: Springer-Verlag; 2013.
Hastings C, et al. Low moments for small samples: a comparative study of order statistics. Ann Math Stat. 1947;18(3):413–426.
Olson CH, et al. Clustering of elderly patient subgroups to identify medication-related readmission risks. Int J Med Informatics. 2016;85(1):43–52.
Nikfalazar, S., et al. A new iterative fuzzy clustering algorithm for multiple imputation of missing data. In IEEE International Conference on Fuzzy Systems (FUZZ-IEEE). 2017. Naples, Italy.
Bezdek JC. Pattern recognition with fuzzy objective function algorithms. New York: Plentum Press; 1981.
Maechler, M., et al. cluster: Cluster Analysis Basics and Extensions. 2018. https://cran.r-project.org/web/packages/cluster/cluster.pdf. Accessed 19 Mar 2019.
Sun H, Wang S, Jiang Q. FCM-based model selection algorithms for determining the number of clusters. Pattern Recogn. 2004;37(10):2027–2037.
Delen D, et al. Analysis of healthcare coverage: a data mining approach. Expert Syst Applications. 2009;36:995–1003.
Ripley, B. tree: Classification and Regression Trees. 2018. https://CRAN.R-project.org/package=tree. Accessed 10 Jan 2019.
Hennig C, Liao TF. How to find an appropriate clustering for mixed-type variables with application to socio-economic stratification. J R Stat Soc. 2013;62(3):309–369.
Mihaljević B, et al. Classifying GABAergic interneurons with semi-supervised projected model-based clustering. Artif Intell Med. 2015;65(1):49–59.
Kaufman L, Rousseeuw PJ. Finding groups in data: an introduction to cluster analysis. New Jersey: Wiley; 2005.
Lecun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521(7553):436–444.
Raskutti B, Kowalczyk A. Extreme re-balancing for SVMs: a case study. SIGKDD Explor Newsl. 2004;6(1):60–69.
Funding
This study is funded by Transport Accident Commission (TAC) through the Institute of Safety, Compensation and Recovery Research (ISCRR).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
Hadi Akbarzadeh Khorshidi, Behrooz Hassani-Mahmooei and Gholamreza Haffari declare that they have no conflict of interest.
Ethical Approval
This article does not contain any studies with human participants performed by any of the authors.
Informed Consent
Statement not required. This study was performed using a de-identified administrative dataset, with ethics approval granted by Monash University Human Research Ethics Committee (CF09/3150—2009001727).
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Akbarzadeh Khorshidi, H., Hassani-Mahmooei, B. & Haffari, G. An Interpretable Algorithm on Post-injury Health Service Utilization Patterns to Predict Injury Outcomes. J Occup Rehabil 30, 331–342 (2020). https://doi.org/10.1007/s10926-019-09863-0
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10926-019-09863-0