With the number of total joint replacement (TJRs) performed and associated costs growing annually, a new emphasis on improving the ‘‘value’’ of TJR has emerged. To achieve this goal, surgeons must continue to improve functional and pain-related outcomes after TJR, while reducing the numbers of postoperative complications and poor clinical results. Institutional and private practice registries have been developed to assist in this pursuit. These registries have collected a wide spectrum of data points from basic Level I to IV data.
It is important to consider whether an informed, competent, and unbiased team is analyzing the data. Previous reports have demonstrated difficulty with maintaining current clinical information when follow-up relied on return visits. There have been some improvements demonstrated with direct contact via phone or questionnaires sent to patients.
An additional challenge has been the lack of consensus on a minimal data set required for the establishment of a joint replacement registry. Traditional TJR registries were designed to collect data useful to monitoring implant survival and failure, defined by revision rates. National registries typically attempt to enroll 100% of patients from all hospitals to assure complete capture of data at the time of surgery and data from subsequent revision surgeries. The dates of both the initial surgery and revision are necessary to quantify the time to revision and failure rate. In this model, national registries incorporate large numbers of surgeries to identify relatively low annual failure rates. For example, New Zealand reports post-TKA implant failure rates of approximately 0.5%/year in the first 12 years.
Historically, follow-up with total joint patients has been 60-70% at one year and 30-40% at 2 years. Numerous privately maintained registries have demonstrated that for most follow-up intervals, the number of patients who are not current exceeds 50%.
An additional concern for the private practice physicians is the time and expense associated with routine, in-office follow-up of TJR patients. The longer a patient has not come in for evaluation, the more difficult it is to perform the evaluation. One example from a recent study of a 5-surgeon group reported that to make all patients with primary THA current at 10 years after surgery would involve attempting to contact 1173 patients. If all those patients could be contacted, it would take 30 clinical days at 40 patients/day to see every patient. That is a tremendous burden for any orthopaedic group to endure.
The issue of incomplete or inaccurate data is a tremendous challenge. When utilizing manually populated paper forms, any question can be skipped — with the form then rendered unreadable and invalid. Patients must complete a large number of forms, especially if more than one joint arthroplasty is being evaluated, which increases the magnitude of this problem. The TJR population also has many older patients who may not understand questions clearly, have difficulty filling out forms, or who are slow to make decisions in a rushed office environment. To compound the problem, incomplete forms can result from office staff error as well. Incomplete forms have been estimated to be a result of a patient error 85% of the time, and staff error 15% of the time.
A barrier to data collection can also be the number of surgeons reporting. This has been a major challenge for the Canadian total joint registry, with only 70% of Canadian orthopaedic surgeons participating in the Canadian joint registry, and <50% of total hip and total knee replacements being captured by the data. Another example of the challenge of gaining participation was highlighted in an article in The New York Times titled “A Call for a Warning System on Artificial Joints”. A pilot registry was attempted in Virginia by Dr. William Jiranek that resulted in only six of the estimated 200 doctors performing joint replacements in the state who actually participated.
PRO Application to Registry
The development and implementation software to assist in the collection of patient reported outcome (PRO) measures may be the only scalable way to gather this information in an inexpensive and less burdensome way. A recent meta-analysis found that the psychometric properties between electronic and paper-and-pencil methods for delivering PROs are equivalent. While there is only a small collection of literature on the implementation of electronic PROs within orthopedics, this remains extremely promising.
Registry data can be potentially flawed in numerous ways, but the primary concern remains the inconsistent follow-up. Utilization of PRO measures can provide a degree of consistency for joint replacement registry data. Electronic means of PRO data capture in orthopedics are an excellent fit because the patient can avoid a clinic visit, which enables patients to be followed across geographic areas. Enabling patients to enter outcome data remotely makes the collection of data more feasible, and reduces associated costs through reductions in human resource consumption. Electronic outcome reporting also provides the opportunity for patients to answer outcome questionnaires at various intervals, and provides high integrity data that was previously unattainable when patients moved or sought out care with another provider.
Should patient-reported outcomes be included in a US national joint registry?
Not a Practical Route
“I do not recommend patient-recorded outcomes for the U.S. national joint registry. From a cost and practicality standpoint, efforts should be made first to obtain level 1 data (patient age, diagnosis, laterality, surgical procedure, implants, surgeon case volume, hospital case volume, complications and re-operations). National registries from other countries have made significant scientific contributions and demonstrated improved patient outcomes and safety, using level 1 data. Level 2 data collection, such as patient body mass index, comorbidities, and detailed surgical and postoperative interventions, may one day be possible, with a network of electronic health records sharing aggregate data behind secure firewalls. The logistics and costs associated with collecting level 3 data, such as patient-reported outcome measures, would be prohibitive on a national scale.”
Robert S. Namba, MD
Orthopedics Today Editorial Board
Outcomes are Appropriate
“Patient-reported outcomes (PROs) convey knowledge of the patient’s health condition, without a third party being allowed to interpret or modify the data. PROs are already being monitored by the Centers for Medicare & Medicaid Services (CMS) and will contribute up to 30% of CMS’ evaluation of quality of care in determining reimbursements for inpatients. Given that our ultimate aim as physicians is to improve our patients’ health and satisfaction, it is entirely appropriate that we measure whether our interventions are successfully matching their health needs… It only makes sense that the joint replacement registry develops the capacity to include PROs in its database.”
Raj Rao, MD
Professor, Department of Orthopaedic Surgery
Medical College of Wisconsin
|Data Level||Data Collected|
|Level I||Patient-related data|
Incidence of deathProcedural data
Date of surgery
Type of procedure
Reoperating and/or revision
Height and weight
Surgical measures (technique, approach, fixation method, and timing)
|Level III||Patient-reported outcome|
|Level IV||Imaging diagnostics|