Background Predicting patient’s Length of Stay (LOS) before total joint replacement (TJR) surgery is vital for hospitals to optimally manage costs and resources. Many hospitals including in rural areas use publicly available models such as National Surgical Quality Improvement Program (NSQIP) calculator which, unfortunately, performs suboptimally when predicting LOS for TJR procedures. Objective The objective of this research was to develop a Machine Learning (ML) model to predict LOS for TJR procedures performed at a Perioperative Surgical Home implemented rural community hospital for better accuracy and interpretation than the NSQIP calculator. Methods A total of 158 TJR patients were collected and analyzed from a rural community hospital located in Montana. A random forest (RF) model was used to predict patient’s LOS. For interpretation, permuted feature importance and partial dependence plot methods were used to identify the important variables and their relationship with the LOS. Results The root mean square error for the RF model (0.7) was lower than the NSQIP calculator (1.21). The five most important variables for predicting LOS were BMI, Duke Activity Status-Index, diabetes, patient’s household income, and patient’s age. Conclusion This pilot study is the first of its kind to develop an ML model to predict LOS for TJR procedures that were performed at a small-scale rural community hospital. This pilot study contributes an approach for rural hospitals, making them more independent by developing their own predictions instead of relying on public models.

Background: The Perioperative Surgical Home (PSH) was designed by the American Society of Anesthesiologists (ASA) to improve perioperative care through more coordinated and communicative decision-making. PSH has proven success in large urban health centers by reducing surgery cancellation, operating room time, length of stay (LOS), and readmission rates. Yet, only limited studies have assessed the impact of PSH on surgical outcomes in rural areas. Thus, this study compared surgical outcomes – LOS, discharge disposition, and 90-day readmission – at a rural hospital from cohorts before and after the implementation of PSH. Method: An initial model of a PSH system was initiated in a local rural hospital. This newly created clinical team and outpatient clinic retrospectively assessed total joint surgeries occurring from November 2017 to April 2018 (before PSH, n = 324) as well as from November 2018 to April 2019 (after PSH, n = 326). General linear models were used to understand the impact of PSH on rural surgical outcomes. Results: Implementing PSH had a positive impact on LOS (P-value < 0.01), discharge disposition (P-value < 0.01), and readmission (P-value = 0.03).Conclusion: Implementation of the PSH system in a rural, community hospital reduced LOS, 90-day readmission, and increased direct to home discharge. Increased clinical management standardization across internal and external stakeholders was achieved because the PSH clinic enabled a centralized navigation point for clinicians, patients, and their families to better communicate and coordinate care.

BACKGROUND Perioperative surgical home (PSH) is a novel patient-centric surgical system developed by American Society of Anesthesiologist to improve outcomes and patient satisfaction. PSH has proven success in large urban health centers by reducing surgery cancellation, operating room time, length of stay (LOS), and readmission rates. Yet, only limited studies have assessed the impact of PSH on surgical outcomes in rural areas. AIM To evaluate the newly implemented PSH system at a community hospital by comparing the surgical outcomes using a longitudinal case-control study. METHODS The research study was conducted at an 83-bed, licensed level-III trauma rural community hospital. A total of 3096 TJR procedures were collected retrospectively between January 2016 and December 2021 and were categorized as PSH and non-PSH cohorts ( n = 2305). To evaluate the importance of PSH in the rural surgical system, a case-control study was performed to compare TJR surgical outcomes (LOS, discharge disposition, and 90-d readmission) of the PSH cohort against two control cohorts [Control-1 PSH (C1-PSH) ( n = 1413) and Control-2 PSH (C2-PSH) ( n = 892)]. Statistical tests including Chi-square test or Fischer’s exact test were performed for categorical variables and Mann-Whitney test or Student’s t -test were performed for continuous variables. The general linear models (Poisson regression and binomial logistic regression) were performed to fit adjusted models. RESULTS The LOS was significantly shorter in PSH cohort compared to two control cohorts (median PSH = 34 h, C1-PSH = 53 h, C2-PSH = 35 h) ( P value < 0.05). Similarly, the PSH cohort had lower percentages of discharges to other facilities (PSH = 3.5%, C1-PSH = 15.5%, C2-PSH = 6.7%) ( P value < 0.05). There was no statistical difference observed in 90-d readmission between control and PSH cohorts. However, the PSH implementation reduced the 90-d readmission percentage (PSH = 4.7%, C1-PSH = 6.1%, C2-PSH = 3.6%) lower than the national average 30-d readmission percentage which is 5.5%. The PSH system was effectively established at the rural community hospital with the help of team-based coordinated multi-disciplinary clinicians or physician co-management. The elements of PSH including preoperative assessment, patient education and optimization, and longitudinal digital engagement were vital for improving the TJR surgical outcomes at the community hospital. CONCLUSION Implementation of the PSH system in a rural community hospital reduced LOS, increased direct-to-home discharge, and reduced 90-d readmission percentages.