A Novel Machine Learning Model Developed to Assist in Patient Selection for Outpatient Total Shoulder Arthroplasty

Abstract

Introduction: Patient selection for outpatient total shoulder arthroplasty (TSA) is important to optimizing patient outcomes. This study aims to develop a machine learning tool that may aid in patient selection for outpatient total should arthroplasty based on medical comorbidities and demographic factors. Methods: Patients undergoing elective TSA from 2011 to 2016 in the American College of Surgeons National Surgical Quality Improvement Program were queried. A random forest machine learning model was used to predict which patients had a length of stay of 1 day or less (short stay). A multivariable logistic regression was then used to identify which variables were significantly correlated with a short or long stay. Results: From 2011 to 2016, 4,500 patients were identified as having undergone elective TSA and having the necessary predictive features and outcomes recorded. The machine learning model was able to successfully identify short stay patients, producing an area under the receiver operator curve of 0.77. The multivariate logistic regression identified numerous variables associated with a short stay including age less than 70 years and male sex as well as variables associated with a longer stay including diabetes, chronic obstructive pulmonary disease, and American Society of Anesthesiologists class greater than 2. Conclusions: Machine learning may be used to predict which patients are suitable candidates for short stay or outpatient TSA based on their medical comorbidities and demographic profile.

Trends in total shoulder arthroplasty from 2005 to 2018: lower complications rates and shorter lengths of stay despite patients with more comorbidities

“…Notably, the shorter average hospital stay and decrease in non-home discharge observed in the 2015-2018 cohort of this current study occurred despite increased rates of comorbidities, which is typically associated with increased LOS and a higher proportion of non-home discharge. 3 , 5 , 6 , 17 , 28 This phenomenon may be the result of an improved understanding of which patients are good candidates for outpatient TSA, improved care coordination, patient education, and advances in both multimodal pain management and regional anesthesia. 4 , 10 , 11 Although not evaluated in this study, patients discharged home, especially on the day of surgery, are generally younger and have fewer comorbidities.…”

Outpatient Shoulder Arthroplasty—A Systematic Review

“…Following full text review, 26 articles were included in the review (Figure 1). Of the included studies, 18 articles discussed complications, readmissions, and safety, [16][17][18][19]21,22,[24][25][26][27][28][29][30][31][32][33]40,41 7 articles discussed patient selection, 21,[27][28][29][30][31]40 6 articles discussed pain management, 20,28,[34][35][36][37] 6 articles discussed the cost implications of outpatient surgery, 15,22,23,30,38,41 2 articles discussed patient satisfaction, 24,32 and 1 article discussed surgeon satisfaction. 15 The level of evidence of the included articles ranges from Level II to Level V.…”

“…The 18 studies examining complications and safety of outpatient SA fall into two categories: retrospective cohorts and series from single centers [16][17][18]22,24,28,29,32,33,40 and larger database or registry studies (Table 1). 19,21,[25][26][27]30,31,41 Seven studies found higher complications with inpatient SA, 18,21,27,28,30,31,33 one study found higher complications in outpatient SA, 25 four studies found no differences in complications or readmissions, 19,22,29,41 one study had mixed findings, 26 and five studies did not compare inpatient and outpatient SA. 16,17,24,32,40 Brolin et al found no statistical difference between inpatient (10%) and outpatient (13%) complication rates when comparing 30 inpatient and outpatient SA performed by a single surgeon, 29 and Arshi et al found higher rates of surgical site infections (SSI) requiring irrigation and debridement at 6 months (OR ¼ 1.49; 95% confidence interval (CI), 1.01-2.19; P ¼ .045) and 1 year (OR ¼ 1.65; 95% CI, 1.15-2.35; P < .001) in an outpatient SA cohort.…”

Machine learning models to predict length of stay and discharge destination in complex head and neck surgery

“…Discharge destination and prolonged hospitalization are important postoperative outcomes in complex surgeries including head and neck operations. Although ML models to predict these in neurological or orthopedic surgeries have been developed using the NSQIP database, 9,22‐25 there is a paucity of similar investigations in the otolaryngology literature. To our knowledge, this is the first manuscript that develops proof‐of‐concept ML algorithms to preoperatively predict DNHF and LOS following complex HN surgeries, and among the minority of studies to publish the developed ML models as a public interface for simulation and further examination by the readership.…”