An Analysis of Medicare Payment Policy for Total Joint Arthroplasty

Size: px

Start display at page:

Download "An Analysis of Medicare Payment Policy for Total Joint Arthroplasty"

Osborne Peters
5 years ago
Views:

1 The Journal of Arthroplasty Vol. 23 No. 6 Suppl An Analysis of Medicare Payment Policy for Total Joint Arthroplasty Kevin J. Bozic, MD, MBA,*y Harry E. Rubash, MD,z Thomas P. Sculco, MD, and Daniel J. Berry, MD Abstract: Medicare facility payment policy for lower extremity total joint arthroplasty (TJA) has undergone extensive changes since The purpose of this study was to compare patient and procedure characteristics and resource use among TJA procedures and to identify predictors of resource use in TJA. Clinical, demographic, and economic data were analyzed from 6483 primary or revision TJA patients from 4 high-volume centers between October 2005 and June Descriptive analyses were conducted to evaluate differences between procedure types, and multivariable linear regression analyses were undertaken to identify predictors of resource use. Both patient severity of illness and surgical complexity influenced resource use associated with TJA procedures. As the primary goal of Medicare payment policy is to set payment rates proportional to relative resource use, both severity of illness and surgical complexity should be incorporated for payment equity and to minimize incentives for selection bias among hospitals that perform TJA procedures. Key words: Costs, revision hip, revision knee, outcomes Published by Elsevier Inc. Medicare is the primary payer for more than 60% of the roughly hip and knee arthroplasty procedures that are performed annually in the United States, and major arthroplasty procedures of the lower extremity represent the single largest procedural cost in the Medicare budget [1]. The Centers for Medicare & From the *Institute for Health Policy Studies, University of California, San Francisco, San Francisco, California; ydepartment of Orthopaedic Surgery, University of California, San Francisco, San Francisco, California; zdepartment of Orthopaedic Surgery, Massachusetts General Hospital, Boston, Massachusetts; Department of Orthopaedic Surgery, Hospital for Special Surgery, New York, New York; and Department of Orthopaedic Surgery, Mayo Clinic, Rochester, Minnesota. Submitted January 4, 2008; accepted April 6, Benefits or funds were received from Orthopaedic Research and Education Foundation. Reprint requests: Kevin J. Bozic, MD, MBA, UCSF Department of Orthopaedic Surgery and Institute for Health Policy Studies, 500 Parnassus, MU 320W, San Francisco, CA Published by Elsevier Inc /08/ $34.00/0 doi: /j.arth Medicaid Services' (CMS) decision to create a separate diagnosis-related group (DRG 545) for revision total joint arthroplasty (TJA) procedures in October 2005 resulted in more equitable reimbursement for hospitals that perform a disproportionate share of complex revision TJA procedures, recognizing the higher resource use associated with these procedures [2]. This important payment policy change will hopefully provide increased access to care for patients with failed total joint arthroplasties and help referral and tertiary care TJA centers continue to provide a high standard of care for these challenging patients. The addition of new, more descriptive International Classification of Diseases, Ninth, Clinical Modification (ICD-9-CM) diagnosis and procedure codes in October 2005 has provided the opportunity to further analyze differences in clinical characteristics and resource intensity among TJA procedures and patients. In 2006, policy makers recommended using All-Patient Refined Diagnosis-Related Group (APR- DRG) severity of illness (), a measure of medical comorbidities, as the primary predictor of resource use, without accounting for procedure characteristics, such 133

2 134 The Journal of Arthroplasty Vol. 23 No. 6 Suppl. 1 September 2008 Table 1a. ICD-9-CM Procedure Codes for Primary and THA Procedures ICD-9 Principal Procedure Code Primary THA (81.51) Both (0.70) Acetabular (00.71) Femoral (00.72) Head/Liner Exchange (00.73) Total n % n % n % n % n % n % Site UCSF % % 6 4.0% 6 6.7% 4 3.6% % Mayo % % % % % % MGH % % 3 2.0% 1 1.1% 4 3.6% % HSS % % % % % % Total % % % % % % MGH indicates Massachusetts General Hospital; HSS, Hospital for Special Surgery. as surgical complexity. The purpose of this study was to compare patient characteristics, procedure characteristics, and resource use among TJA procedures, and to identify useful predictors of resource use in TJA. Materials and Methods A retrospective analysis of clinical, demographic, treatment, and economic data from 6483 primary and revision hip and knee arthroplasty procedures from 4 high-volume TJA centers was conducted (Tables 1a and 1b). Patients were included in the study if they received either a primary or revision TJA from October 2005 through June Clinical factors included in the analysis were principal diagnosis on admission, APR-DRG classification, presence of extensive bone loss, and surgical complexity. Routine surgical complexity was defined as all primary total hip arthroplasty (THA) and total knee arthroplasty (TKA) procedures, and revision procedures that involved revision of modular parts only (isolated femoral head and liner THA revisions and isolated tibial insert TKA revisions). Complex surgical complexity was defined as all other types of revision THA and TKA procedures. Demographic factors included age, sex, and insurance type. Treatment factors included type of primary or revision TJA procedure and mean operative time. Economic data included mean total costs per procedure (using hospital-based decision support systems from each center), length of stay, and discharge disposition. Costs were normalized to 1 center to account for differences in procedure cost across centers. Statistical Analyses Descriptive analyses were conducted, as well as comparisons by ICD-9-CM procedure code (using analysis of variance and post hoc Dunnett's t test methodology). A linear regression was performed to predict total standardized costs by APR-DRG and surgical complexity. Surgical complexity was defined by both patient factors (eg, major bone loss) and procedure characteristics (eg, primary vs revision, type of revision). Results Patient Demographic and Clinical Characteristics The majority of primary and revision THA patients were women, and there were no differences in sex among procedure types, other than a significantly higher percentage of women (63%) who underwent isolated acetabular liner and femoral head exchange procedures (Tables 2a and 2b). Similarly, the majority of primary and Table 1b. ICD-9-CM Procedure Codes by Site for Primary and TKA Procedures ICD-9 Principal Procedure Code Primary TKA (81.54) All (00.80) Tibial (00.81) Femoral (00.82) Patellar (00.83) Tibial Insert (00.84) Total n % n % n % n % n % n % n % Site UCSF % % 3 2.6% 5 9.1% % 1 2.2% % Mayo % % % % % % % MGH % 6 4.0% 6 5.1% 2 3.6% 0 0.0% 4 8.7% % HSS % % % % % % % Total % % % % % % %

3 Medicare Payment Policy Bozic et al 135 Table 2a. Patient Demographics for Primary and THA Procedures (All Sites) Primary THA (81.51) (n = 2717) Both (00.70) (n = 187) Acetabular (00.71) Femoral (00.72) (n = 90) Head/Liner Exchange (00.73) (n = 111) Sex, male 1312 (48) 84 (45) 55 (37) 46 (51) 57 (51) Age, y 63.6 ± ± ± ± ± 13.1 Medicare, yes 1289 (47) 90 (48) 83 (55) 53 (59) 53 (47) Major osseous defect, yes 59 (2) 114 (61) 93 (62) 40 (44) 28 (25) (43) 43 (23) 18 (12) 5 (6) 15 (14) (49) 106 (57) 115 (77) 72 (80) 82 (74) (8) 35 (19) 15 (10) 12 (13) 13 (12) 4 14 (1) 3 (2) 2 (1) 1 (1) 1 (1) Values are shown as n (%) or mean ± SD. revision TKA patients were women, with the exception of femoral component only revisions (55% men) and tibial liner exchanges (52% men) (P b.01 for both comparisons). The average age for primary THA patients was 63.8 years, and there were no significant differences in age between patients who underwent primary and revision procedures. The average age for primary TKA patients was 68.5 years, and patients who underwent all component revisions (64.2), tibial component revisions (64.7), and tibial liner exchanges (64.5) were significantly younger than primary TKA patients (P b.01 for all comparisons). Forty-seven percent of primary THA and 61% of primary TKA patients had Medicare insurance, and there were no significant differences in payer type across procedures. Only 2% of primary THA patients had major bone loss, and all types of revision THA procedures had a significantly higher incidence of major bone loss, ranging from 25% for isolated head and liner exchanges to 62% for acetabular component revisions (P b.01 for all comparisons). Only 1% of primary TKA patients had major bone loss, and all types of revision TKA procedures (with the exception of isolated patellar component revisions) had a significantly higher incidence of major bone loss, ranging from 4% for isolated tibial liner exchanges to 30% for all component revision TKA procedures (P b.01 for all comparisons). All types of revision THA patients had significantly higher APR-DRG scores (indicative of more comorbid medical conditions) than primary THA patients (P b.01 for all comparisons). Similarly, all types of revision TKA patients, with the exception of isolated patellar component revisions, had significantly higher APR-DRG scores than primary TKA patients (P b.01 for all patients). Treatment Factors The average operative time (skin-to-skin) for primary THA was 92.6 minutes. All types of revision THA procedures had significantly longer average operative times (P b.01), ranging from minutes for isolated head and liner exchange procedures to minutes for both component revisions (Tables 3a and 3b). The most common revision THA procedure was both component revisions. The average operative time for primary TKA was 97.7 minutes. All types of revision TKA procedures had significantly longer average operative times (P b.01), with the exception of isolated patellar component revisions (84.6 minutes) and isolated tibial insert exchanges (88.9 minutes). The most common revision TKA procedure was both component revisions. Table 2b. Patient Demographics for Primary and TKA Procedures (All Sites) Primary TKA (81.54) (n = 2809) All (00.80) Tibial (00.81) (n = 117) Femoral (00.82) (n = 55) Patellar (00.83) (n = 14) Tibial Insert (00.84) (n = 46) Sex. male 1053 (38) 70 (47) 50 (43) 30 (55) 6 (43) 24 (52) Age, years 68.5 ± ± ± ± ± ± 11.1 Medicare, yes 1713 (61) 66 (44) 61 (52) 30 (55) 6 (43) 22 (48) Major osseous defect, 23 (1) 45 (30) 17 (15) 11 (20) 1 (7) 2 (4) yes (43) 51 (34) 41 (35) 20 (36) 6 (43) 14 (30) (48) 78 (52) 65 (56) 28 (51) 8 (57) 23 (50) (9) 20 (13) 11 (9) 7 (13) 0 (0) 8 (17) 4 19 (1) 1 (1) 0 (0) 0 (0) 0 (0) 1 (2) Values are shown as n (%) or mean ± SD.

136 The Journal of Arthroplasty Vol. 23 No. 6 Suppl. 1 September 2008 Table 3a. Operative Times and Length of Hospital Stay for Primary and THA Procedures (All Sites) Primary THA (81.

4 136 The Journal of Arthroplasty Vol. 23 No. 6 Suppl. 1 September 2008 Table 3a. Operative Times and Length of Hospital Stay for Primary and THA Procedures (All Sites) Primary THA (81.51) (n = 2717) Both (00.70) (n = 187) Acetabular (00.71) Femoral (00.72) (n = 90) Head/Liner Exchange (00.73) (n = 111) OR Time (skin-to-skin, min) 92.6 ± ± ± ± ± 57.3 Length of stay (d) 4.4 ± ± ± ± ± 3.6 Values are shown as mean ± SD. OR indicates operating room. Table 3b. Operative Times and Length of Hospital Stay for Primary and THA Procedures (All Sites) Primary TKA (81.54) (n = 2809) All (00.80) Tibial (00.81) (n = 117) Femoral (00.82) (n = 55) Patellar (00.83) (n = 14) Tibial Insert Exchange (00.84) (n = 46) OR Time (skinto-skin, 97.7 ± ± ± ± ± ± 52.0 min) Length of stay (d) 4.7 ± ± ± ± ± ± 5.8 Values are shown as mean ± SD. Fig. 1. Mean total costs by type of revision THA procedure. Economic Factors The average costs for primary THA, normalized to 1 institution (UCSF), were $ (Tables 3a and 3b). The average costs for all types of revision THA procedures were significantly higher (P b.01), with the exception of isolated head and liner exchanges, ranging from a low of $ for isolated acetabular component procedures to a high of $ for both component revisions (Fig. 1). The average costs for primary TKA, normalized to 1 institution (UCSF), were $ The average costs for all types of revision TKA procedures were significantly higher (P b.01), with the exception of isolated tibial insert exchanges, ranging from a low of $ for isolated femoral component revisions to a high of $ for both component revisions (Fig. 2). Isolated patellar component revisions were significantly less costly than primary TKA procedures ($20 310, P b.01). Multivariable Regression Analysis Both and surgical complexity were significantly correlated with resource use associated with TJA procedures, with costs increasing with higher values and more complex surgical procedures (Tables 4a and 4b). Even after controlling for, a higher surgical complexity procedure (defined as any revision THA or TKA procedure other than a revision of modular parts) still resulted in higher costs for each category for both THA (between 25% and 34% higher) and TKA (between 22% and 29% higher) procedures. Discussion Medicare payment policy for lower extremity TJA procedures has evolved considerably over the past

5 Medicare Payment Policy Bozic et al 137 Fig. 2. Mean total costs by type of revision TKA procedure. Table 4a. Independent Contributors of Severity of Illness and Surgical Complexity for Primary and THA Procedures When Complexity Code = 1 Complexity Code n Mean Total Cost ($) 95% Confidence Limits Increase ($) Increase (%) % % % Complexity code 1 is defined as all revision THA procedures except isolated head and liner exchange procedures (ICD-9 procedure code is 00.73). Table 4b. Independent Contributors of Severity Illness and Surgical Complexity for Primary and TKA Procedures When Complexity Code = 1 Complexity Code n Mean Total Cost ($) 95% Confidence Limits Increase ($) Increase (%) % % % Complexity code 1 is defined as all revision TKA procedures except isolated patellar component procedures (ICD-9 procedure code is 00.83) and isolated tibial insert exchange procedures (ICD-9 procedure code 00.84). 3 years. Before 2005, all lower extremity arthroplasty procedures, regardless of patient characteristics (eg, comorbidities) or type of procedure (eg, primary vs revision), were lumped together under a single DRG (DRG 209). Previous investigators had reported significant differences in resource use between primary and revision TJA procedures [3-11]. In October 2005, in response to a request from the American Association of Hip and Knee Surgeons and the American Academy of Orthopaedic Surgeons, CMS created separate DRGs for primary (DRG 544) and revision (DRG 545) TJA procedures [2]. These changes were intended to more accurately reflect differences in clinical characteristics and resource use between primary and revision TJA procedures. However, in 2006, policy makers recommended further refinements to the Medicare Inpatient Prospective Payment System which would have once again eliminated separate DRGs for primary and revision

6 138 The Journal of Arthroplasty Vol. 23 No. 6 Suppl. 1 September 2008 Table 5. CMS Inpatient Prospective Payment System Changes from FY07 to FY08 Related to TJA Procedures FY07 FY08 Description of hip or knee arthroplasty with MCC 467 of hip or knee arthroplasty with CC 468 of hip or knee arthroplasty without CC/MCC Major joint arthroplasty or reattachment of lower extremity with MCC 470 Major joint arthroplasty or reattachment of lower extremity without MCC MCC indicates major complications and comorbid conditions. TJA procedures. The rationale for this change was that differences in resource use could be accounted for by differences in patient alone, regardless of the complexity of the surgical procedure (eg, primary vs revision). Although our current study confirms that is a useful predictor of resource use associated with TJA procedures, our data also indicate that both and surgical complexity are important independent predictors of resource use in TJA. Because the intent of Medicare DRGs is to set payment rates proportional to relative resource use, both and surgical complexity factors that significantly affect resource use should be incorporated for payment equity and to avoid incentives which may create selection bias among orthopedic providers and institutions. Policy Implications The results of this study were shared with CMS in December Based in part on our findings, the new Medicare Severity DRGs related to TJA procedures that were implemented in October 2007 were revised to recognize both patient and procedure characteristics as important contributors to resource use in TJA (Table 5) [12]. By more appropriately matching hospital reimbursement to resource use, these payment policy changes will help hospitals that perform complex TJA procedures in medically ill patients be appropriately compensated for these resource-intensive procedures. Conclusions By working with CMS to continue to evaluate policyrelevant differences in clinical characteristics and resource use among hip and knee arthroplasty patients and procedures, we recommended further refinements to the orthopedic DRGs based on differences in and surgical complexity. These changes were adopted by CMS in the FY2008 Final Rule for the Inpatient Prospective Payment System. We believe these changes more closely accomplish CMS's goal of matching hospital reimbursement to resource use for primary and revision TJA procedures, and minimize incentives for hospitals and surgeons to selectively treat patients with less complex orthopedic conditions whose procedures are less resource intensive. These types of collaborative efforts have the potential to greatly improve the accuracy of the Medicare payment system as it relates to TJA procedures by more appropriately matching hospital reimbursement to actual resource use associated with these procedures, with the ultimate goal of improving patient access to care and clinical outcomes for TJA patients. References 1. Mendenhall S Hip and knee implant review. Orthop Netw News 2006;17:1 [July, 2006]. 2. Bozic K. CMS Changes ICD-9 and DRG codes for revision TJA. AAOS Bull 2005;53: Barber T, Healy W. The hospital cost of total hip arthroplasty. J Bone Joint Surg Am 1993;75-A: Barrack R. Economics of revision total hip arthroplasty. Clin Orthop 1995;319: Barrack R. Cost analysis of revision total hip arthroplasty. A 5-year follow-up study. Clin Orthop 1999; 369: Barrack R. The evolving spectrum of revision hip arthroplasty. Orthopedics 1999;22: Bozic K, Katz P, Showstack J, et al. The 2006 OREF Clinical Research Award Paper: using clinical and economic data to influence health policy in the United States. Paper presented at: AAOS Annual Meeting Chicago, IL. 8. Bozic KJ, Durbhakula S, Berry DJ, et al. Differences in patient and procedure characteristics and hospital resource use in primary and revision total joint arthroplasty: a multicenter study. J Arthroplasty 2005;20(7 Suppl 3): Bozic KJ, Katz P, Cisternas M, et al. Hospital resource utilization for primary and revision total hip arthroplasty. J Bone Joint Surg Am 2005;87-A: Iorio R, Healy W, Richards J. Comparison of the hospital cost of primary and revision total hip arthroplasty after cost containment. Orthopedics1999;22: Lavernia C, Drakeford M, Tsao A, et al. and primary hip and knee arthroplasty A cost analysis. Clin Orthop 1995;311: Mendenhall S Hip and knee implant review. Orthop Netw News 2007;18:1 July, 2007.

ARTICLE IN PRESS. All-Patient Refined Diagnosis- Related Groups in Primary Arthroplasty

ARTICLE IN PRESS. All-Patient Refined Diagnosis- Related Groups in Primary Arthroplasty The Journal of Arthroplasty Vol. 00 No. 0 2009 All-Patient Refined Diagnosis- Related Groups in Primary Arthroplasty Carlos J. Lavernia, MD,*y Artit Laoruengthana, MD,y Juan S. Contreras, MD,y and Mark