Although patient factors may substantially determine the need for revision surgery, at least some revisions are related to device problems. Many joint replacement surgeons will recall the considerable failures of threaded acetabular shells, metal-backed patellar components, first-generation ceramic heads, and titanium bearing surfaces. The clinical impact of failures such as these could be minimized with prompt reporting and ongoing data collection. In addition, patient complications that may arise from surgical techniques, perioperative care patterns, or other factors not related to the device could be brought to light more quickly and lead to rapid response changes. Past failures illustrate the pressing need for ongoing data collection to be a standard companion to treatment so that orthopaedic surgeons can be provided promptly with the information that they need to avoid harming patients.
The professional responsibility of the orthopaedic surgeon to guide his or her practice with evidence entails a commitment to "using modern quality assurance strategies" in furthering the clinical community's knowledge5. Two points in the "Code of Ethics and Professionalism for Orthopaedic Surgeons," articulated by the AAOS, are particularly germane to this issue6. In the Academy's public declaration of its ethical responsibilities it asserts that:IV. A. The orthopaedic surgeon continually should strive to maintain and improve medical knowledge and skill, and should make available to patients and colleagues the benefits of his or her professional attainments… .IX. A. The honored ideals of the medical profession imply that the responsibility of the orthopaedic surgeon extends not only to the individual but also to society as a whole. Activities that have the purpose of improving the health and well-being of the patient and/or the community in a cost-effective way deserve the interest, support, and participation of the orthopaedic surgeon.
IV. A. The orthopaedic surgeon continually should strive to maintain and improve medical knowledge and skill, and should make available to patients and colleagues the benefits of his or her professional attainments… .
IX. A. The honored ideals of the medical profession imply that the responsibility of the orthopaedic surgeon extends not only to the individual but also to society as a whole. Activities that have the purpose of improving the health and well-being of the patient and/or the community in a cost-effective way deserve the interest, support, and participation of the orthopaedic surgeon.
When read in the light of the harm caused by a lack of data, these statements clearly suggest, as Gruen et al. concluded, that orthopaedic surgeons "have a professional responsibility to ensure that our interventions are more effective than alternatives and, for this reason, outcomes-based research must be supported."5 To fulfill these responsibilities, the professional must undertake a program of ongoing data collection and long-term follow-up to track surgical outcomes as a regular feature of orthopaedic practice. In other words, the orthopaedic surgeon's acknowledged professional responsibility to "improve medical knowledge" and to improve the "well-being of the patient" supports the creation of national registries for follow-up of new treatments and devices.
Collecting the data to improve patient care is certainly a noble goal. But we know that anything that can be used for good purposes can also be used in ways that may cause harm. Whenever data are collected, it is important to consider whether the project is asking useful and appropriate study questions, whether the information being gathered will be useful in answering the study questions, and whether more information, less information, or other information should be gathered. It is also important to consider whether the data are being analyzed by an informed, competent, and unbiased team.
In addition, the collected data may be seriously impacted by the number of surgeons reporting. A major concern with the Canadian total joint registry has been the considerable number of surgeons failing to report their data. Likewise, those who do report are not reporting on all of their cases. As Bourne noted, only 70% of Canadian orthopaedic surgeons participate in the Canadian joint registry and <50% of total hip and total knee replacements are captured by the data7.
"A Call for a Warning System on Artificial Joints," a recent article in The New York Times examining the need for a national joint registry in the United States, reported on a pilot registry study that Dr. William Jiranek attempted in Virginia8. Only six of the estimated 200 doctors performing joint replacements in the state actually participated. Mandatory participation is necessary for the establishment of an effective registry that will efficiently and accurately report results. When participation is optional, people are disinclined to participate and prefer to leave the work to others. Selective reporting or failure to report can lead to skewed data and misinterpretation.
For any participating joint replacement surgeon, collecting and reporting data will require some investment of time and effort. In addition, expenses are likely to be incurred either directly, in the form of staff salary, or indirectly, as extra working time for the physician. These will, in effect, further erode the reimbursement fees for joint arthroplasties and serve as a disincentive for doctors to participate voluntarily. Because many doctors would be reluctant to participate in a registry that will cost time, effort, and money, voluntary participation would likely bias the data and slow the process of collecting the critical information. Furthermore, surgeons may decline participation because they are fearful of revealing their outcomes when the outcomes may appear worse than those of other surgeons. Therefore, unless reporting is mandated and unless the requirement is enforced, even individuals who appreciate the need for registries will be disinclined to report data when it is easy for them to imagine that others are not doing their fair share because doing so would expose them to the risk of looking bad relative to their peers.
Furthermore, surgeons may have concerns about who will have access to the data and how it will be used. In the past, reports on physician and institution performance with certain procedures, namely, cardiac catheterizations and bypass surgeries, have been used to demonstrate that some physicians and institutions perform better or worse than others do. Yet, data that are not risk-adjusted are likely to be misleading. Surgeons and institutions are concerned that data collected for a registry with the purpose of learning about the outcomes related to a particular device or technique may be used to imply that the clinical outcomes are related to institutional or physician performance. Personal and institutional reputations may be tarnished by unfortunate results. There is also concern that disclosed registry data may be used as a basis of liability or malpractice litigation.
Recent court decisions may also hamper efforts by professional organizations to mandate participation in the collection of necessary registry information because courts may consider the reporting requirements to be a restraint of trade, usually defined as any action or condition that tends to prevent free competition in business9. Courts may not find the value of providing registry information important enough to justify interference with an individual practitioner's liberty to conceal information about his or her private practice and outcomes.
In establishing a joint registry, these concerns will require careful attention. A thoughtfully constructed joint registry that includes appropriate safeguards could actually be a useful resource in protecting doctors from charges of negligence or lax vigilance. The registry is also likely to be helpful in efficiently identifying problems, thereby saving physicians substantial time in diagnosing and rectifying complications associated with faulty devices and ineffectual procedures.
Another ethical dilemma is presented by federal regulations that govern human subject research. The HIPAA (Health Insurance Portability and Accountability Act) regulations are designed to protect patient privacy. These privacy protections could limit collection of identifiable data, tracking of individual patient outcomes, and recontacting patients to confirm outcomes.
The regulations of the Office for Human Research Protections (OHRP) and the enforcement of Common Rule regulations could entail additional obstacles10. These regulations could be interpreted as requiring approval of the research protocol by the institutional review board at each involved institution and eliciting the informed consent of each participant. Whereas a registry should include every patient who receives the studied device, conformance with these research requirements would make the process of collecting and tracking data far more cumbersome. It would also make the process far less inclusive because, if given the choice, individual patients may refuse to do their part in advancing knowledge by refusing to participate in the registry and leaving the job to kindhearted others. An example of how these rules inhibit similar research will illustrate the problems that existing regulations could generate.
In 2001, Dr. Peter Pronovost, a critical-care specialist from Johns Hopkins, decided to try using a checklist to address infections in hospital intensive care units. His published results showed that implementing the checklist for doctors and nurses improved the consistency of their practice and saved >1500 lives in the first eighteen months11. When his work was brought to the attention of OHRP, they halted the study because the researchers at Johns Hopkins and some of the other participating institutions had not treated this study as "human subjects research" and had not secured institutional review board approval or individual informed consent from the subjects in the 100 or more participating hospitals12.
The Common Rule draws an artificial distinction between research and treatment. Research requires investigators to abide by all Common Rule regulations and to obtain informed consent from each participant. Treatment does not. Quality improvement studies are an extremely important area for data collection that aids hospitals in determining how best to improve service in order to serve patients. Although quality improvement and public health studies are undoubtedly a form of research, by bureaucratic decree they are stipulated to be exempt from institutional review board oversight and OHRP regulations and, therefore, do not require either institutional review board approval or the informed consent of each patient or subject. Pronovost had considered his work to be quality improvement rather than a study of a new and unproven intervention. In light of his experience and the definitions that have been incorporated into regulations, it is not at all clear how a registry would be seen. Would a registry be counted as research or would registry entries be treated as a public health or quality improvement endeavor?
If registries can be treated as "public health" activities or "quality improvement" rather than as "research," then it would not be necessary to obtain informed consent from individual patients or institutional review board approval from each institution. The issues regarding who would have access to patient information, demographics, and medical records may remain problematic and must be carefully addressed when establishing a registry.
In conclusion, registries are among the most cost-effective interventions in medicine. In our case scenario, R.P. may have benefited if a joint registry had been in place to detect a trend of implant failures prior to his initial procedure. At the very least, his implant failure may have served to more rapidly signal a faulty device, thereby preventing additional revisions. Early detection of problems spares patient suffering and saves health-care dollars. It has been estimated that even a 2% reduction in joint revisions would save >$650 million per year3.
The importance of consistently participating in a registry system should be recognized as an important responsibility of the profession. Many of the foreseeable ethical dilemmas associated with the establishment of a registry can be addressed by thoughtful attention during the planning process. Others require the cooperation and collaboration of governmental agencies. Although the profession needs society's support in establishing and maintaining a joint arthroplasty registry, we applaud the stand of the AAOS in cooperation with the specialty societies. Our profession acknowledges the important place of registries as part of the profession's ethical duty to serve patients and the community and to live up to the standards established in our "Code of Ethics and Professionalism."
Note: The authors thank Ilene Tsui for research assistance in the preparation of this manuscript.