Background and Anticipated Impacts on VA Patient Care: Of the 5.6 million patients treated by the Veterans Health Administration (VHA) in FY11, 14.6% had a diagnosis of osteoarthritis, the fourth most common health condition treated by the VHA, and the leading cause of disability in the United States.(1,2) Advanced osteoarthritis of the hip and knee is treated with total joint arthroplasty (TJA), roughly 11,000 of which were performed in VHA in fiscal year 2011. The overall risks of complications and death in total joint arthroplasty are relatively low (5% and <1% respectively), but certain patient characteristics (e.g., diabetes, obesity, smoking, alcohol misuse) have been associated with dramatically higher risks. Currently, orthopedic surgeons in VHA have no systematic and validated way of using a patient's demographic and clinical profile to estimate their personalized risk of complications or death related to TJA. Patients also vary substantially in how much they benefit from this common and expensive procedure. As with the estimation of risks, orthopedic surgeons in VHA have no systematic and validated way to estimate patients' personalized likelihood of improvement. It is impossible to balance risks and benefits unless they can be accurately and reliably quantified, a capability currently unavailable to orthopedic surgeons in VHA. The development and validation of a risk, benefit, and satisfaction calculator, that draws from both electronic health record and manually entered data, would allow orthopedic surgeons to easily and accurately estimate both the expected risks and benefits of TJA for each candidate. These data could then be used within a shared decision making (SDM) framework for more accurate and reliable patient informed consent, including the identification of modifiable risk factors (e.g., weight, smoking) that should be addressed before these elective procedures, and more effective SDM that maximizes effectiveness, safety, patient preferences, satisfaction, and therefore value.
Aim 1 : Develop and validate a total joint arthroplasty risk calculator for the VHA population. Using the most recent three years of VHA Corporate Date Warehouse (CDW) and VA Surgical Quality Improvement Project (VASQIP) data including roughly 30,000 total joint arthroplasty cases, we will merge data on outcomes, especially death and major complications, and many known and candidate predictors of outcomes.
Aim 2 : Develop and validate a total joint arthroplasty benefit calculator for the VHA population. Patients (n = 1200) scheduled to undergo TJA at three high volume VA facilities (Palo Alto, San Francisco, Minneapolis) will be recruited to provide pre-operative symptom, functioning, expectation, and quality of life data and then followed up for reassessment one year later. These data will then be linked with VASQIP and CDW data in order to develop predictive benefit models, especially changes in functioning and pain, based on the pre-operative and other (demographic and clinical) data.
Aim 3 : Develop and validate a total joint arthroplasty patient satisfaction calculator. Because the predictors of risks and benefits may be different than the predictors of satisfaction, we will also develop a patient satisfaction model using data from the 1200 patients described in Aim 2.
Aim 4 : Develop and pilot test the calculator interface and related decision support functionality. Once the models are developed, we will develop and pilot test a CDW-integrated risk and benefit calculator and decision support system that might be used within a SDM framework. The goal will be to visually display where the patient exists in the risk/benefit space, identify possible opportunities to modify risk and benefit factors, and provide resources and information to facilitate possible prehabilitation.
Almost 11,000 total hip and knee replacement surgeries were conducted in VHA in 2011. Currently surgeons and patients have no systematic and validated method to estimate the patient-specific risks of the procedure (e.g., death, other complications including infection) or the estimated benefits for a specific patient. This project will take advantage of VA Surgical Quality Improvement and VA Corporate Warehouse data to develop and validate a surgical risk, benefit, and satisfaction calculator for total joint replacement candidates. Once developed, this decision-support tool can be used by surgeons and patients within a shared-decision making process to weigh the patient-tailored risks and probable benefits from total knee or hip replacement. The calculator will also help identify factors such as weight, smoking, and diabetic control that might be addressed prior to surgery in order to minimize risks and maximize the potential benefits of the procedure.
|Harris, Alex Hs (2017) Path From Predictive Analytics to Improved Patient Outcomes: A Framework to Guide Use, Implementation, and Evaluation of Accurate Surgical Predictive Models. Ann Surg 265:461-463|