Predictive analytics: Is this the wave of the future for orthopaedic surgery?

Written by Richard A. Conn, MD, Medical Director, Stryker Performance Solutions & Mary Frances Delaune, MPH, Program Manager, Stryker Performance Solutions | July 19, 2016


While transparency in healthcare delivery and reporting was a mainstream topic in 2015, Predictive Analytics looks to play a major role in orthopaedic practice protocols for years to come.

With the Accountable Care Act and payment reform, hospitals and surgeons are finding themselves placed at risk for the outcomes of patient care. Whether in a bundled payment or Accountable Care Organization Payment Model, the risk of poor patient outcomes either clinically or financially is a real concern.

Enter the world of Predictive Analytics! Predictive Analytics has been defined as "the branch of the advanced analytics which is used to make predictions about unknown future events1." While historically used in many business segments, the use of Predictive Analytics has taken hold in orthopaedic surgery as a means to quantitatively predict various outcomes of joint replacement patient care.

A risk for hospitals and surgeons in a bundled payment model is the discharge status of a patient to either a skilled nursing facility or to home. Data has shown that the clinical outcome of patients discharged home is equivocal to those who are sent to skilled nursing facilities2. Both the Risk Assessment and Prediction Tool (RAPT)3 and the AM-PAC 64 clicks tool developed at the Cleveland Clinic have been developed to assist surgeons in predicting the discharge of joint replacement patients to either a skilled nursing facility or home. With only 6 questions, the RAPT is simple and predictable in terms of probable discharge disposition for total joint replacement (TJR) patients.

With a move towards performing more traditionally inpatient joint replacement procedures as a short-stay or outpatient procedure, the ability to accurately predict whether a TJR patient could be discharged from a facility within 24 hours becomes very helpful information. Dr. Michael Meneghini and Dr. Pete Caccavallo in Indianapolis, Indiana have developed a tool they refer to as Outpatient Arthroplasty Risk Assessment5 (OARA). The OARA tool is a 60 question document that allows the pre-assessment provider (hospitalist/anesthesia) to place a percentage risk of time of discharge on potential TJR patients.

Recently, IBM has announced that they will be utilizing Watson to scan data and create outcome predictive analytics in the field of orthopedic surgery. Many other forms of predictive tools have been developed including the AJRR Risk Stratification tool6.

As healthcare continues to change so will the need for more Predictive Analytic tools to assist facilities and providers in crafting the most effective care plan for TJR patients with a predictable outcome.