Health policy study receives OREF Clinical Research Award
By Joen Kinnan
The 2006 OREF Clinical Research Award was presented to Kevin J. Bozic, MD, MBA; Patricia Katz, PhD; Jonathan Showstack, PhD; James Naessens, MPH; Harry E. Rubash, MD; Michael D. Ries, MD; and Daniel J. Berry, MD, for their paper, “Using Clinical and Economic Outcome Data to Influence Health Policy in the United States: The Case of Total Joint Replacement.”
Dr. Kevin J. Bozic
Total joint replacement (TJR)—one of the most commonly performed orthopaedic procedures—has high rates of clinical success in reducing pain and improving function and quality of life. However, both the number of TJR surgeries and the cost of implants have steadily increased over the past decade, leading to concerns that TJR procedures may be consuming a disproportionate share of scarce U.S. health care resources.
Dr. Bozic’s studies addressed two major health policy issues related to TJR. First, limitations in existing ICD-9 diagnosis and procedure codes related to TJR restrict the ability to use administrative claims data from large, public databases to evaluate patient outcomes and implant survivorship. Second, the lack of distinction in reimbursement between primary and revision TJR procedures has created strong financial disincentives for hospitals and surgeons to perform revision TJR procedures, potentially limiting access to care.
“Current ICD-9-CM diagnosis and procedure codes and Medicare diagnosis-related group (DRG) codes related to primary and revision TJR are too broad and inclusive to capture relevant differences in patient characteristics, procedure characteristics and resource utilization between primary and revision TJR procedures,” wrote Dr. Bozic.
“As a result, large public datasets—such as the Medicare Provider Analysis and Review database—that rely on administrative claims data are of limited value in evaluating TJR patient outcomes and detecting premature failures associated with specific TJR implants and techniques. Having more detailed, accurate and descriptive ICD-9 diagnosis and procedure codes would enhance public health efforts, such as the American Joint Replacement Registry Project (AJRR), that are intended to improve the overall quality of care and reduce revision rates in TJR.”
The AJRR is a collaborative undertaking among the AAOS, the Center for Medicare and Medicaid Services (CMS), and the Agency for Healthcare Research and Quality. The project’s goal is to capture relevant information on TJR procedures performed in the United States, such as factors related to the patient, the surgeon, the hospital, the implants used and the procedure. This information could be helpful in evaluating the quality, clinical outcomes and cost-effectiveness of TJR procedures and implants.
The study pointed out that the most commonly used ICD-9 diagnosis code associated with failed TJR is 996.4—mechanical complication of an internal orthopaedic device—regardless of the type or mechanism of failure. Furthermore, ICD-9 procedure codes categorize all revision TJR procedures simply as “revision of hip replacement” (81.53) or “revision of knee replacement” (81.55), regardless of the type or complexity of the procedure. These limitations make it difficult to use administrative claims data to evaluate factors related to implant longevity, specific implant failure mechanisms, and type of revision surgery.
Dr. Bozic’s study also addressed how discrepancies between resource utilization and reimbursement for primary and revision TJR can affect patient access to care.
“Given that hospital reimbursement for all primary and revision TJR procedures is the same under DRG 209, tertiary care referral hospitals that perform a high proportion of revision TJR procedures often incur substantial financial losses related to these procedures. The discrepancy between resource utilization and reimbursement for revision TJR procedures has created perverse financial disincentives that have deterred some hospitals from performing these procedures,” he wrote.
“If these trends continue, access to care and therefore quality of care and clinical outcomes for patients with failed TJRs could be jeopardized.”
Differences in primary and revision TJR
In two studies, Dr. Bozic and his team measured and compared differences in patient characteristics, procedure characteristics and hospital resource utilization between primary and various revision TJR procedures. They identified clinical, demographic and procedure characteristics predictive of higher resource use among patients undergoing primary and revision TJR.
They found significant differences in patient and procedure characteristics and resource utilization between different types of primary and revision TJR procedures. The studies revealed that hospital resource utilization is significantly higher for revision than for primary TJR and that patient age, gender, diagnosis, baseline medical health, magnitude of bone loss and type of revision procedure all significantly affect resource utilization.
“These findings could be used to help define additional, more descriptive ICD-9-CM diagnosis and procedure codes related to revision TJR procedures, thus creating more accurate and useful administrative claims data that could be used to evaluate factors related to implant longevity, specific failure mechanisms and patient outcomes in TJR,” said Dr. Bozic.
“Furthermore, our data could be used by health policy makers to revise the DRG codes related to TJR procedures to more closely match reimbursement to actual resource utilization.”
Health policy implications
In October 2004, Dr. Bozic and his colleagues presented the results of their work to the ICD-9-CM Care & Coordination Committee at the CMS headquarters in Baltimore. Based on their findings, the group proposed a series of additional, more descriptive ICD-9-CM diagnosis and procedure codes related to revision TJR. In April 2005, the committee officially accepted these recommendations, and the new codes went into effect in October 2005.
The research group was invited back to CMS in February 2005 to present their findings to the DRG Advisory Committee, and to make recommendations regarding hospital reimbursement for TJR procedures.
“We recommended that CMS create a separate DRG for revision TJR to recognize the higher resource intensity associated with these procedures,” wrote Dr. Bozic.
“Another issue that was addressed was the impact that an increase in hospital reimbursement for revision TJR would have on reimbursement for primary TJR. CMS clarified the budget-neutral formula that is used for modifying DRG weights and explained that any increase in reimbursement for revision TJR would be offset by a decrease in reimbursement for all 522 other DRGs. Therefore, the initial impact on hospital reimbursement for primary TJR would be negligible,” he continued.
Two months later, CMS announced that DRG 209 would be split into DRG 544 (primary hip and knee replacement) and DRG 545 (revision hip and knee replacement). In explaining its decision, CMS credited the importance of the data presented by Dr. Bozic’s research team.
“We agree with the commentators and the AAOS that the creation of a new DRG for revisions of hip and knee replacements should resolve payment issues for hospitals that perform the more difficult revisions of joint replacements,” said a CMS spokesman. CMS leadership also indicated a strong interest in maintaining an ongoing dialogue with the AAOS on orthopaedic surgery-related health policy issues.
Dr. Bozic, the principal investigator, is assistant professor in residence at the University of California, San Francisco. The studies were funded in part by the OREF Health Services Research Fellowship Grant and the AAOS/OREF Clinician-Scientist Traveling Fellowship Grant.