2020 SCHOT Research Award: Development and Validation of a Multivariable Prediction Model of Hospital Stay in Elderly Chilean Patients Undergoing Elective Total Hip Arthroplasty Using Machine Learning

Abstract

Introduction The prediction of the length of hospital stay after elective total hip arthroplasty (THA) is crucial in the perioperative evaluation of the patients, and it plays a decisive role from the operational and economic point of view. Internationally, big data and artificial intelligence have been used to perform prognostic evaluations of this type. The present study aims to develop and validate, through the use of artificial intelligence (machine learning), a tool capable of predicting the hospital stay of patients over 65 years of age undergoing THA for osteoarthritis.

Material and Methods Using the electronic records of hospital discharges de-identified from the Department of Health Statistics and Information (Departamento de Estadísticas e Información de Salud, DEIS, in Spanish), the data of 8,970 hospital discharges of patients who had undergone THA for osteoarthritis between 2016 and 2018 were obtained. A total of 15 variables available in the DEIS registry, in addition to the percentage of poverty in the patient's borough of origin were included to predict the probability that a patient would have a shortened (< 3 days) or prolonged (> 3 days) stay after surgery. By using machine learning techniques, 8 prediction algorithms were trained with 80% of the sample. The remaining 20% was used to validate the predictive capabilities of the models created from the algorithms. The optimization metric was evaluated and ranked using the area under the receiver operating characteristic curve (AUC-ROC), which corresponds to how well a model can distinguish between two groups.

Results The XGBoost algorithm had the best performance, with an average AUC-ROC of 0.86 (standard deviation [SD]: 0.0087). Secondly, we observed that the linear support vector machine (SVM) algorithm obtained an AUC-ROC of 0.85 (SD: 0.0086). The relative importance of the explanatory variables showed that the region of residence, the administrative health service, the hospital where the patient was operated on, and the care modality are the variables that most determine the length of stay.

Discussion The present study developed machine learning algorithms based on free-access Chilean big data, which helped create and validate a tool that demonstrates an adequate discriminatory capacity to predict shortened versus prolonged hospital stay in elderly patients undergoing elective THA.

Conclusion The algorithms created through the use of machine learning allow to predict the hospital stay in Chilean patients undergoing elective total hip arthroplasty.

Keywords

hospital stay - machine learning - total hip arthroplasty

Publication History

Received: 18 March 2021

Accepted: 06 August 2021

Article published online:
22 December 2021

• Referencias

• 1 Problema de salud GES N° 12: Endoprótesis total de cadera en personas de 65 años y más con artrosis de cadera con limitación funcional severa. Orientación en Salud Superintendencia de Salud, Gobierno de Chile n.d. http://www.supersalud.gob.cl/difusion/665/w3-article-586.html (accessed December 18, 2020).
• 2 Grosso MJ, Neuwirth AL, Boddapati V, Shah RP, Cooper HJ, Geller JA. Decreasing Length of Hospital Stay and Postoperative Complications After Primary Total Hip Arthroplasty: A Decade Analysis From 2006 to 2016. J Arthroplasty 2019; 34 (03) 422-425 DOI: 10.1016/j.arth.2018.11.005.
• 3 Goyal N, Chen AF, Padgett SE. et al. Otto Aufranc Award: A Multicenter, Randomized Study of Outpatient versus Inpatient Total Hip Arthroplasty. Clin Orthop Relat Res 2017; 475 (02) 364-372 DOI: 10.1007/s11999-016-4915-z.
• 4 Paredes O, Ñuñez R, Klaber I. Successful initial experience with a novel outpatient total hip arthroplasty program in a public health system in Chile. Int Orthop 2018; 42 (08) 1783-1787 DOI: 10.1007/s00264-018-3870-6.
• 5 Greenky MR, Wang W, Ponzio DY, Courtney PM. Total Hip Arthroplasty and the Medicare Inpatient-Only List: An Analysis of Complications in Medicare-Aged Patients Undergoing Outpatient Surgery. J Arthroplasty 2019; 34 (06) 1250-1254 DOI: 10.1016/j.arth.2019.02.031.
• 6 Featherall J, Brigati DP, Faour M, Messner W, Higuera CA. Implementation of a Total Hip Arthroplasty Care Pathway at a High-Volume Health System: Effect on Length of Stay, Discharge Disposition, and 90-Day Complications. J Arthroplasty 2018; 33 (06) 1675-1680 DOI: 10.1016/j.arth.2018.01.038.
• 7 Ripollés-Melchor J, Abad-Motos A, Díez-Remesal Y. et al; Postoperative Outcomes Within Enhanced Recovery After Surgery Protocol in Elective Total Hip and Knee Arthroplasty (POWER2) Study Investigators Group for the Spanish Perioperative Audit and Research Network (REDGERM). Association Between Use of Enhanced Recovery After Surgery Protocol and Postoperative Complications in Total Hip and Knee Arthroplasty in the Postoperative Outcomes Within Enhanced Recovery After Surgery Protocol in Elective Total Hip and Knee Arthroplasty Study (POWER2). JAMA Surg 2020; 155 (04) e196024-e196024 DOI: 10.1001/jamasurg.2019.6024.
• 8 Manning DW, Edelstein AI, Alvi HM. Risk Prediction Tools for Hip and Knee Arthroplasty. J Am Acad Orthop Surg 2016; 24 (01) 19-27 DOI: 10.5435/JAAOS-D-15-00072.
• 9 Sconza C, Respizzi S, Grappiolo G, Monticone M. The Risk Assessment and Prediction Tool (RAPT) after Hip and Knee Replacement: A Systematic Review. Joints 2019; 7 (02) 41-45 DOI: 10.1055/s-0039-1693459.
• 10 Diaz Ledezma C, Radovic I. What's new in hip arthroplasty? South American perspective. Recent Advances in Orthopedics-2. Jaypee Brothers Medical Publishers (P) Ltd; 2018
• 11 Parvizi J, Gehrke T, Krueger CA. et al; International Consensus Group (ICM) and Research Committee of the American Association of Hip and Knee Surgeons (AAHKS). Resuming Elective Orthopaedic Surgery During the COVID-19 Pandemic: Guidelines Developed by the International Consensus Group (ICM). J Bone Joint Surg Am 2020; 102 (14) 1205-1212 DOI: 10.2106/JBJS.20.00844.
• 12 Donell ST, Thaler M, Budhiparama NC. et al. Preparation for the next COVID-19 wave: The European Hip Society and European Knee Associates recommendations. Knee Surg Sports Traumatol Arthrosc 2020; 28 (09) 2747-2755 DOI: 10.1007/s00167-020-06213-z.
• 13 Myers TG, Ramkumar PN, Ricciardi BF, Urish KL, Kipper J, Ketonis C. Artificial Intelligence and Orthopaedics: An Introduction for Clinicians. J Bone Joint Surg Am 2020; 102 (09) 830-840 DOI: 10.2106/JBJS.19.01128.
• 14 Cabitza F, Locoro A, Banfi G. Machine Learning in Orthopedics: A Literature Review. Front Bioeng Biotechnol 2018; 6: 75 DOI: 10.3389/fbioe.2018.00075.
• 15 Bini SA. Artificial Intelligence, Machine Learning, Deep Learning, and Cognitive Computing: What Do These Terms Mean and How Will They Impact Health Care?. J Arthroplasty 2018; 33 (08) 2358-2361 DOI: 10.1016/j.arth.2018.02.067.
• 16 Haeberle HS, Helm JM, Navarro SM. et al. Artificial Intelligence and Machine Learning in Lower Extremity Arthroplasty: A Review. J Arthroplasty 2019; 34 (10) 2201-2203 DOI: 10.1016/j.arth.2019.05.055.
• 17 Ramkumar PN, Haeberle HS, Bloomfield MR. et al. Artificial Intelligence and Arthroplasty at a Single Institution: Real-World Applications of Machine Learning to Big Data, Value-Based Care, Mobile Health, and Remote Patient Monitoring. J Arthroplasty 2019; 34 (10) 2204-2209 DOI: 10.1016/j.arth.2019.06.018.
• 18 Anderson AB, Grazal CF, Balazs GC, Potter BK, Dickens JF, Forsberg JA. Can Predictive Modeling Tools Identify Patients at High Risk of Prolonged Opioid Use After ACL Reconstruction?. Clin Orthop Relat Res 2020; 478 (07) 0-1618 DOI: 10.1097/CORR.0000000000001251.
• 19 Departamento de Estadisticas e Información de Salud. n.d. https://deis.minsal.cl/#publicaciones (accessed December 18, 2020).
• 20 Farley KX, Anastasio AT, Premkumar A, Boden SD, Gottschalk MB, Bradbury TL. The Influence of Modifiable, Postoperative Patient Variables on the Length of Stay After Total Hip Arthroplasty. J Arthroplasty 2019; 34 (05) 901-906 DOI: 10.1016/j.arth.2018.12.041.
• 21 Observatorio Social - Ministerio de Desarrollo Social y Familia. n.d. http://observatorio.ministeriodesarrollosocial.gob.cl/pobreza-comunal-2017#basedatos (accessed February 4, 2021).
• 22 Mackinnon A. The use and reporting of multiple imputation in medical research - a review. J Intern Med 2010; 268 (06) 586-593 DOI: 10.1111/j.1365-2796.2010.02274.x.
• 23 López V, Fernández A, García S, Palade V, Herrera F. An Insight into Classification with Imbalanced Data: Empirical Results and Current Trends on Using data Intrinsic Characteristics. Inf Sci 2013; 250: 113-141 DOI: 10.1016/j.ins.2013.07.007.
• 24 del Rio S, Benitez JM, Herrera F. Analysis of Data Preprocessing Increasing the Oversampling Ratio for Extremely Imbalanced Big Data Classification. 14th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom, vol. 2, 2015, p. 180–5. https://doi.org/10.1109/Trustcom.2015.579
• 25 Cerda J, Cifuentes L. Uso de curvas ROC en investigación clínica: Aspectos teórico-prácticos. Rev Chilena Infectol 2012; 29 (02) 138-141 DOI: 10.4067/S0716-10182012000200003.
• 26 Harris AHS, Kuo AC, Weng Y, Trickey AW, Bowe T, Giori NJ. Can Machine Learning Methods Produce Accurate and Easy-to-use Prediction Models of 30-day Complications and Mortality After Knee or Hip Arthroplasty?. Clin Orthop Relat Res 2019; 477 (02) 452-460 DOI: 10.1097/CORR.0000000000000601.
• 27 Collins GS, Moons KGM. Reporting of artificial intelligence prediction models. Lancet 2019; 393 (10181): 1577-1579 DOI: 10.1016/S0140-6736(19)30037-6.
• 28 Collins GS, Reitsma JB, Altman DG, Moons KGM. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD statement. BMJ 2015; 350: g7594 DOI: 10.1136/bmj.g7594.
• 29 Ramkumar PN, Navarro SM, Haeberle HS. et al. Development and Validation of a Machine Learning Algorithm After Primary Total Hip Arthroplasty: Applications to Length of Stay and Payment Models. J Arthroplasty 2019; 34 (04) 632-637 DOI: 10.1016/j.arth.2018.12.030.
• 30 Kunze KN, Polce EM, Sadauskas AJ, Levine BR. Development of Machine Learning Algorithms to Predict Patient Dissatisfaction After Primary Total Knee Arthroplasty. J Arthroplasty 2020; 35 (11) 3117-3122 DOI: 10.1016/j.arth.2020.05.061.
• 31 MeSH Browser. n.d. https://meshb.nlm.nih.gov/record/ui?ui=D000077558 (accessed December 19, 2020).
• 32 Grauer JN, Leopold SS. Editorial: large database studies–what they can do, what they cannot do, and which ones we will publish. Clin Orthop Relat Res 2015; 473 (05) 1537-1539 DOI: 10.1007/s11999-015-4223-z.
• 33 Kang HW, Bryce L, Cassidy R, Hill JC, Diamond O, Beverland D. Prolonged length of stay (PLOS) in a high-volume arthroplasty unit. Bone Jt Open 2020; 1 (08) 488-493 DOI: 10.1302/2633-1462.18.BJO-2020-0047.R1.
• 34 Burn E, Edwards CJ, Murray DW. et al. Trends and determinants of length of stay and hospital reimbursement following knee and hip replacement: evidence from linked primary care and NHS hospital records from 1997 to 2014. BMJ Open 2018; 8 (01) e019146 DOI: 10.1136/bmjopen-2017-019146.
• 35 Girbino KL, Klika AK, Barsoum WK. et al; Cleveland Clinic OME Arthroplasty Group. Understanding the Main Predictors of Length of Stay After Total Hip Arthroplasty: Patient-Related or Procedure-Related Risk Factors?. J Arthroplasty 2021; 36 (05) 1663-1670.e4 DOI: 10.1016/j.arth.2020.11.029.
• 36 Athey AG, Cao L, Okazaki K. et al. Survey of AAHKS International Members on the Impact of COVID-19 on Hip and Knee Arthroplasty Practices. J Arthroplasty 2020; 35 (7S): S89-S94 DOI: 10.1016/j.arth.2020.04.053.