The Journal of Arthroplasty xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty Mario Farías-Kovac, MD, Caleb R. Szubski, BA, Mark Hebeish, DDS, Alison K. Klika, MS, Kirtishri Mishra, BS, Wael K. Barsoum, MD Department of Orthopaedic Surgery, Orthopaedic and Rheumatologic Institute, Cleveland Clinic, Cleveland, Ohio
a r t i c l e
i n f o
Article history: Received 2 January 2014 Accepted 14 February 2014 Available online xxxx Keywords: total hip arthroplasty total knee arthroplasty price capitation implant selection premium implants
a b s t r a c t While price capitation strategies may help to control total hip (THA) and knee arthroplasty (TKA) implant costs, its effect on premium implant selection is unclear. Primary THA and TKA cases 6 months before and after capitated pricing implementation were retrospectively identified. After exclusions, 716 THA and 981 TKA from a large academic hospital and 2 midsize private practice community hospitals were reviewed. Academic hospital surgeons increased premium THA implant usage (66.5% to 70.6%; P = 0.28), while community surgeons selected fewer premium implants (36.4%) compared to academic surgeons, with no practice change (P = 0.95). Conversely, premium TKA implant usage significantly increased (73.4% to 89.4%; P b 0.001) for academic surgeons. Community surgeons used premium TKA implants at greater rates in both periods, with all cases having ≥1 premium criterion. © 2014 Elsevier Inc. All rights reserved.
Over the last 2 decades, nontransformative, interval innovation of orthopedic implants has been the rule, with a focus on improving fixation, motion, kinematics, stability, and durability [1]. Implant manufacturers have spent billions of dollars on the development, marketing, and selling of their newest technologies as the demand for joint arthroplasties have exponentially risen. In 2010, total hip (THA) and knee arthroplasty (TKA) accounted for 1,053,000 procedures, with an annual average increase of 4.3% in the number of cases from 1993 to 2011 [2,3]. Driven by a rapidly aging population of baby boomers and a rise in the number of young patients (b 65 years) receiving joint arthroplasty, primary THA and TKA have a projected increase of more than 174% and 673%, respectively, by 2030 [4,5]. Rising implant costs and diminishing reimbursement are a serious challenge to hospitals [6]. As orthopedic implants have evolved, there has been an associated increase in cost. However, these changes are significantly outpacing Medicare's adjustments in reimbursement. From 1991 to 2005, there was a 156% rise in hip and knee implant list
Conflicts of interest and source of funding: No financial support was received for this study. Dr. Barsoum would like to acknowledge the following disclosures: consultant to Stryker Orthopaedics; research support from Stryker Orthopaedics, Zimmer, Cool Systems, Orthovita, DJO, Active Implants, The Medicines Company, and the State of Ohio; royalties from Stryker Orthopaedics, Zimmer, Exactech, and Shukla Medical; stock options in OtisMed Corporation, Custom Orthopaedic Solutions, and iVHR; board member at KEF Healthcare. No other authors have relevant financial relationships to disclose. The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2014.02.020. Reprint requests: Alison K. Klika, MS, Cleveland Clinic, 9500 Euclid Avenue, A41, Cleveland, OH 44195.
prices while the average Medicare payment to hospitals only increased by 19% ($8489 to $10,109) [7]. In 2010, the cost of a primary hip and knee implant was $6398 and $5324, respectively, accounting for approximately half of the total Medicare compensation for an uncomplicated primary joint arthroplasty at the time ($11,653) [8,9]. In 1997 orthopedic programs generated 25% of hospitals' profits, but by 2001 their financial impact had decreased to only 2% [10]. Hospitals have developed and tested several strategies (i.e. implant selection standardization, group purchasing consignment, valuebased purchasing, volume-incentive vendor contracts, single price/ case price purchasing, physician gain sharing, establishment of standardized clinical and surgical pathways) to decrease implant and service costs [11–18]. Capitated pricing, in which a flat purchase price is negotiated for implant line items regardless of technology or manufacturer, has emerged as a successful option for decreasing implant costs in some health care institutions [13]. This model also maintains physician autonomy by offering a full list of implant options. Based on positive outcomes reported on experiences utilizing capitated pricing strategies, our health system (which includes a large academic tertiary referral center and a number of community-based satellite hospitals) replaced a discount-based vendor-buyer system with a capitated pricing model for primary total joint implants in 2011. This capitated pricing system assigned a single price cap for implant line items, regardless of technology. The purpose of this study was to evaluate whether the implementation of this type of capitated pricing system affected usage rates of primary THA and TKA premium implants (i.e. more expensive to produce, newer technology) by salaried surgeons at a large academic hospital. As physicians are no longer burdened by minimizing costs
0883-5403/0000-0000$36.00/0 – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.arth.2014.02.020
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
2
M. Farías-Kovac et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
for the patient or maximizing revenue for the hospitals under capitated pricing, we hypothesized that physicians would be more likely to select premium implants for their patients. A secondary objective was to compare the implant selection trends at the academic hospital to those at two midsize private practice–based community hospitals from the same health system. The surgeons at these community hospitals receive the same implant prices as the large academic hospital, yet their reimbursement systems are independent of the health system. Our secondary hypothesis was that private practice surgeons at these community hospitals are less likely to be affected by this same implant purchasing model. Methods After approval by the institutional review board, all consecutive primary THA (CPT code 27130) and TKA (CPT code 27447) surgeries 6 months before and after implementation of capitated pricing (7/1/ 2011) were retrospectively queried. Study sites included an academic, tertiary referral hospital and two community hospitals within the same health system. Surgeon education regarding the new pricing policy was conducted for a 1-month period following implementation, and these data were excluded from the analysis. This process established preprice (1/1/2011–6/30/2011) and postprice capitation (8/1/2011–1/31/2012) periods. An initial cohort of 1881 cases (THA, n = 794; TKA, n = 1087) were identified that met the inclusion criteria. Simultaneous and staged bilateral procedures were counted as two separate cases, one for each limb. All staged bilateral procedures were retained in the study cohorts because there were no instances that spanned both preprice and postprice capitation periods. Exclusion criteria (Fig. 1)
Inclusion Criteria All consecutive primary total hip arthroplasty (THA) (CPT code 27130) and total knee arthroplasty (TKA) (CPT code 27447) cases performed at a tertiary academic hospital and two community hospitals in following time periods before and after price capitation was implemented (7/1/2011): Pre-capitated (1/1/2011-6/30/2011) Post-capitated (8/1/2011-1/31/2012) *OMITTED 1 month surgeon education period (7/1/2011-7/31/2011)* THA (n=794); TKA (n=1087)
Excluded THA patients (n=78) - Procedure surgeon not represented in both pre- and post-capitated periods (n=29) - Bipolar prosthesis (n=14) - Complex pathologies (i.e. dwarfism, bone tumor lesion, presence of previous hardware) (n=11) - Metal-on-metal prosthesis (n=10) - Nickel allergy (n=8) - Unipolar prosthesis (n=3) - One-of-a-kind prosthesis (n=2) - Resurfacing procedure (n= 1)
Final THA Cohort (n=716) n=463 Academic Hospital n=253 Community Hospital Analyzed variables: - Patient demographics - Hospital location - Implant fixation - Implant characteristics - Bearing surface
Excluded TKA patients (n=106) - Procedure surgeon not represented in both pre- and post-capitated periods (n=29) - Total stabilizing system (n=20) - Nickel allergy (n=20) - Unicompartmental TKA (n=16) - Complex pathologies (i.e. dwarfism, bone tumor lesion, presence of previous hardware, previous fracture) (n=15) - Revision system (n=4) - Cobalt allergy (n=1) - Titanium allergy (n=1)
Final TKA Cohort (n=981) n=666 Academic Hospital n=315 Community Hospital Analyzed variables: - Patient demographics - Hospital location - Implant fixation - Implant characteristics - Bearing surface
Fig. 1. Inclusion/Exclusion flow diagram.
were established to remove cases that did not use primary total joint arthroplasty implants (e.g., unipolar, bipolar, unicompartmental) and to minimize situations in which the surgeon may not have had complete freedom of implant selection (e.g., complex pathology, metal allergies, use of specific implant type, metal-on-metal designs). Cases performed by a surgeon who was not represented in both preprice and postprice capitation periods were excluded due to the potential bias. After exclusions, a total of 463 THA and 666 TKA from the large academic hospital, and 253 THA and 315 TKA from the 2 midsize community hospitals comprised the final study cohort. A review of patient demographics and implant characteristics for each case was performed using the electronic medical record. Demographics variables included age at surgery, gender, body mass index (BMI), and laterality. Using implant catalog numbers and operative notes, the manufacturer, model, and material of each implant subcomponent were noted. Additionally, bearing surface, design characteristics, and component fit (i.e. cemented or press-fit) were collected. Classification of each THA and TKA case's implants as premium or non-premium was based on a slightly expanded version of that used by Gioe et al [1]. Premium THA implants were defined by the existence of one of the following bearing surfaces: second (2G) or third generation (3G) highly cross-linked polyethylene liner with a ceramic or oxidized-zirconium femoral head, ceramic liner with a ceramic femoral head, or mobile-bearing system. Large femoral head size and press-fit femoral stems were not regarded as premium factors. Second-generation highly cross-linked polyethylene liners included: AltrX Altralinked (DePuy, Warsaw, IN), Marathon (DePuy, Warsaw, IN), Longevity (Zimmer, Warsaw, IN), and R3 XLPE (Smith & Nephew, Memphis, TN). Third-generation highly cross-linked polyethylene liners included only X3 (Stryker, Mahwah, NJ). Premium TKA implants were defined by the existence of at least one of the following criteria: mobile-bearing design, high-flexion design, oxidized-zirconium femoral component, and/or highly cross-linked polyethylene bearing surface. Study data were collected and managed using REDCap (Research Electronic Data Capture), which is a secure, Web-based application designed to support data capture for research studies [19]. The data were analyzed using R software (Version 3.0.2, Vienna, Austria). Continuous variables were described using means, standard deviations, and 95% confidence intervals for means. Categorical variables were described using counts and percentages. Continuous demographic variables were compared using Welch's two-sample t-test. Categorical demographic variables were compared using Pearson's chi-squared test with Yates' continuity correction. Hip and knee premium implant usage rates were compared between precapitation and postcapitation periods within the subsets defined by academic and community hospitals using logistic regression fit using generalized estimating equations (GEE) to account for staged bilateral cases, on which observations cannot be assumed to be independent. Age, gender, and BMI were included in these models and so the resulting Pvalues testing for differences between precapitation and postcapitation are adjusted P-values. Rates of specific premium and nonpremium implant subgroups were compared using Fisher's exact test for count data or Pearson's chi-squared test, as appropriate, ignoring repeated measures and the effects of age, gender, and BMI. Premium and non-premium THA implant characteristics are mutually exclusive. Conversely, premium TKA characteristics are not mutually exclusive (i.e. a patient might receive an implant that possesses more than one premium characteristic). Rates of usage of particular TKA implant types were therefore adjusted for multiplicity. Results A final cohort of 716 THA and 981 TKA procedures performed by 22 surgeons was analyzed (Fig. 1). The mean years of clinical practice
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
M. Farías-Kovac et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
3
Table 1 Demographic Data for the THA and TKA Study Cohorts. Academic Hospital
Variable Total hip arthroplasty Number of cases; surgeons Age at surgery (yrs) a Female c Body mass indexa Left laterality c Total knee arthroplasty Number of cases Age at surgery (yrs) Female c Body mass indexa Left laterality a b c d
c
a
Pre-Capitated
Post-Capitated
215; 10 61.9 (13.3) [60.0–63.6] 124 (57.7%) 30.2 (6.5) [29.3–31.1] 96 (44.6%)
248; 10 61.8 (13.2) [60.2–63.5] 139 (56.0%) 29.9 (6.3) [29.2–30.7] 128 (51.6%)
364; 12 66.0 (10.2) [64.9–67.0] 224 (61.5%) 31.8 (6.8) [31.1–32.5] 177 (48.6%)
302; 12 65.6 (9.9) [64.5–66.7] 169 (56.0%) 32.4 (6.9) [31.7–33.2] 156 (51.7%)
Community Hospital
P Value
0.97b 0.72d 0.67b 0.14d
0.64b 0.17d 0.26b 0.48d
Overall
Pre-Capitated
Post-Capitated
463; 10 61.8 (13.2) [60.6–63.0] 263 (56.8%) 30.1 (6.4) [29.5–30.6] 224 (48.4%)
121; 7 67.3 (10.5) [65.4–69.2] 72 (59.5%) 30.7 (8.3) [29.2–32.2] 66 (54.5%)
132; 7 67.9 (10.2) [66.1–69.7] 74 (56.0%) 30.8 (7.5) [29.5–32.1] 61 (46.2%)
666; 12 65.8 (10.1) [65.1–66.6] 393 (59.0%) 32.1 (6.8) [31.6–32.6] 333 (50.0%)
155; 9 69.7 (10.4) [68.0–71.4] 107 (69.0%) 32.7 (7.1) [31.6–33.8] 82 (52.9%)
160; 9 66.7 (10.3) [65.1–68.3) 98 (61.2%) 33.6 (6.8) [32.5–34.7] 83 (51.9%)
P Value
0.64b 0.58d 0.92b 0.19d
0.01b 0.18d 0.26b 0.94d
Overall 253; 7 67.6 (10.3) [66.3–68.9] 146 (57.7%) 30.8 (7.9) [29.8–31.7] 127 (50.2%)
315; 9 68.2 (10.4) [67.0–69.3] 205 (65.1%) 33.2 (7.0) [32.4–33.9] 165 (52.4%)
P Value (Acad vs. Comm)
b0.001b 0.82d 0.20b 0.64d
b0.001b 0.08d 0.03b 0.53d
Result values are expressed as mean (standard deviation) [95% confidence interval]. Welch's two-sample t-test. Result values are expressed as number of cases (percentage). Pearson's chi-squared test with Yates' continuity correction.
among academic and community hospital surgeons were 19.8 ± 7.3 years (95% confidence interval, 15.2–24.5 years) and 23.3 ± 7.9 years (95% confidence interval, 17.6–29.0 years), respectively (P = 0.30). The percentage of adult reconstruction fellowship-trained surgeons was 66.7% (n = 8/12) at the academic setting and 40.0% (n = 4/10) at the community hospitals (P = 0.21). The demographic makeup of the THA and TKA study cohorts was similar across hospital types, with only a slightly younger population being operated on at the academic hospital for both procedures (P b 0.001) (Table 1). There were no significant differences in demographic data between precapitation and postcapitation periods for each hospital type and procedure (Table 1). Implementation of a capitated pricing model at our institution had no significant effect on the type of THA implant selected by either academic center or community hospital surgeons (Table 2). At the academic hospital, rates of premium THA implant usage slightly increased from 66.5% (n = 143) in the precapitation period to 70.6% (n = 175) in the postcapitation period, although this increase was not statistically significant (P = 0.28). Community hospital surgeons selected considerably fewer premium implants (36.4%) in both periods compared to their academic hospital peers and did not change their practice after the new pricing model was instituted (P = 0.95). Furthermore, there were no cases of polyethylene 2G liner
paired with oxidized-zirconium head, ceramic liner with ceramic head, or mobile-bearing systems (three of the five premium criteria) in either period at the community hospitals. Premium TKA implant usage following implementation of the new pricing model significantly increased at the academic hospital from 73.4% (n = 267) to 89.4% (n = 270) (P b 0.001) (Table 3). Mobilebearing and high-flexion designs, which were not utilized in the precapitation period, were used more frequently in the postcapitation period (1. 0%, P = 0.09; 5.3%, P b 0.001; respectively). Highly crosslinked polyethylene use at the academic hospital increased from 73.4% (n = 267) to 83.1% (n = 251) (P = 0.005), and comprised the majority of this hospital type's premium implant usage. No TKA cases with two or more premium criteria were used in either period by academic center surgeons. The community hospital surgeons selected premium TKA implants at much greater rates than their academic peers (Table 3). In fact, all TKA cases at the community hospitals had at least one premium criterion, with no statistical significant changes in utilization of any subgroup of implant. High-flexion knee system usage increased from 10.3% (n = 16) to 18.8% (n = 30) (P = 0.11) following implementation of price capitation. Mobile-bearing designs comprised a large portion of the community hospitals' premium implant usage before (46.5%) and after (50.0%) capitated pricing (P = 0.57). There was a
Table 2 Distribution of THA Implants by Premium Criteria Before and After Implementation of Capitated Pricing, Classified by Hospital Type. Academic Hospital Variable Premium Polyethylene 2G liner—ceramic head Polyethylene 3G liner—ceramic head Polyethylene 2G liner—oxidized-zirconium head Ceramic liner—ceramic head Mobile-bearing system Non-premium Polyethylene 2G liner—metal head Polyethylene 3G liner—metal head
Community Hospital
Pre-Capitated (n = 215)
Post-Capitated (n = 248)
P Value
Pre-Capitated (n = 121)
Post-Capitated (n = 132)
143 (66.5%) 19 (8.8%) 101 (47.0%) 4 (1.9%) 17 (7.9%) 2 (0.9%)
175 (70.6%) 17 (6.9%) 132 (53.2%) 11 (4.4%) 15 (6.0%) 0
0.28a N0.99b N0.99b N0.99b N0.99b N0.99c
44 (36.4%) 2 (1.7%) 42 (34.7%) 0 0 0
48 (36.4%) 4 (3.0%) 44 (33.3%) 0 0 0
0.95a N0.99c N0.99b
28 (13.0%) 44 (20.5%)
37 (14.9%) 36 (14.5%)
N0.99b 0.82b
33 (27.3%) 44 (36.4%)
37 (28.0%) 47 (35.6%)
N0.99b N0.99b
P Value
Result values are expressed as number of cases (raw percentage). a Logistic regression (GEE); adjusted for age, gender, and BMI. b Pearson's chi-squared test with Yates' continuity correction; corrected for multiple testing. c Fisher's exact test for count data; corrected for multiple testing.
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
4
M. Farías-Kovac et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
Table 3 Distribution of TKA Implants by Premium Criteria Before and After Implementation of Capitated Pricing, Classified by Hospital Type. Academic Hospital
Community Hospital
Variable
Pre-Capitated (n = 364)
Post-Capitated (n = 302)
P Value
Pre-Capitated (n = 155)
Post-Capitated (n = 160)
Premium (≥1 below criteria) Mobile-bearing design High-flexion design Oxidized-zirconium femoral component Highly cross-linked polyethylene
267 (73.4%) 0 0 0 267 (73.4%)
270 (89.4%) 3 (1.0%) 16 (5.3%) 0 251 (83.1%)
b0.001a 0.09b b0.001b
155 72 16 0 70
160 80 30 0 53
0.005b
(100.0%) (46.5%) (10.3%) (45.2%)
P Value
(100.0%) (50.0%) (18.8%)
0.57b 0.11b
(33.1%)
0.11b
Result values are expressed as number of cases (raw percentage). a Logistic regression (GEE); adjusted for age, gender, and BMI. b Fisher's exact test for count data; multiplicity corrected.
decrease in the use of highly cross-linked polyethylene from 45.2% (n = 70) to 33.1% (n = 53) (P = 0.11). No oxidized-zirconium femoral components were used in either time period or at either hospital type. The majority of the community hospital TKA cases had only one premium criterion, with three cases before and after capitated pricing having two premium criteria. Discussion As of 2010, Medicare, Medicaid, and private insurers were the major sources of reimbursement for THA and TKA in the United States, representing 53%, 14%, and 31% of these cases, respectively [3]. Accordingly, 67% of THA and TKA reimbursement nationally are subjected to diagnosis-related group (DRG) weights and payments by the Centers for Medicare & Medicaid Services. A recent study examining amounts paid to manufacturers by 61 institutions across 8 different states in 2008 showed an average total surgical cost for THA and TKA of $12,548 and $11,666, respectively, with implant cost accounting for 50.2% and 43.5% of the surgical cost, respectively. A multivariate regression analysis of these data for both procedures found higher implant costs were associated with age, payer (Medicare), complications, discharge disposition (discharge to acute or postacute care facility), and a diagnosis of fracture (for TKA only) [20]. Prior to establishing a capitated pricing model for primary total joint implants, our institution utilized a two-vendor buying system based on volume discounts. This previous approach required an integrated physician organization and limited the number of implant vendors, theoretically reducing the competition to stimulate increased willingness to discount prices [21]. Under this model the physician's choice of orthopedic implants for THA and TKA was largely restricted to those offered by two medical device companies in order to assure high demand by the hospitals in our health system. This high volume was matched with discounted prices by the vendors. Our institution ultimately migrated away from this pricing model in search of higher surgeon satisfaction and patient outcomes through a wider range of implant options while maintaining acceptable costs. Implementation of a capitated pricing model helped remove our institution's reliance on discount- and volume-based purchasing as well as cut total implant costs by approximately 10% overall for THA and TKA, while increasing the range of manufacturers and on-theshelf implant options available to surgeons. It is also important to note that the term premium implants may be misleading. We use this term to define implants that were marketed as premium implants and had list pricing commensurate with their perceived premium status. To our knowledge, there is only one other detailed report in the literature of a capitated pricing system used for orthopedic implant purchasing. Taylor et al [13] presented similar positive financial results at their institution after implementation of a four-tier price capitation system for purchasing THA and TKA implants, reducing costs per implant by 26.1%. This tiered categorization matrix placed implants with similar levels of technology or innovation in categories (I-IV) based on implant fit and bearing surface, and negotiated a price for each category. The classification implemented in our study was a
modified version of that used by Gioe et al [1], focusing on bearing surface technology and classifying implants as premium or nonpremium (e.g. 2 categories). Taylor et al [13] found no statistically significant change in overall implant selection at their institution. However, usage of their most premium and expensive THA and TKA implants (category IV) increased from 55.8% to 63.4% and 26.8% to 29.0%, respectively, with implementation of the new system. A theoretical disadvantage of tiered capitated pricing systems is that cost is not removed entirely from the implant selection equation as a price difference between premium and non-premium implants is retained. No studies to date have reported the effect of a single price cap system, like the one used at our institution, on implant selection. There are noted obstacles that can be encountered when trying to establish a price capitation system. In particular, capitated pricing models typically require large volumes to negotiate beneficial pricing and depending on buying power, they might not cover the purchasing of certain goods (e.g., newest technology), which would require separate pricing contracts or purchasing at the manufacturer's list price. Also, the specifics of a capitated pricing model can vary greatly depending on the necessities or goals of the buyer and the demands of the seller. There are substantial differences in practice patterns between academic and community hospitals for both THA and TKA. The data from our institution demonstrate that the influence of capitated pricing is very different for hip and knee implants, with a trend toward increased premium implant usage for primary TKA in the academic setting only. No statistically significant changes in premium implant selection were found for primary THA in either location. Additionally, private practice community hospital surgeons used premium TKA implants at very high rates (100.0%) and premium THA implants at very low rates (36.4%). We believe that these differences in implant selection between procedure type may be due to the higher levels of satisfaction among patients after THA compared to TKA [22–24]. Surgeons may pursue novel and premium technologies at greater rates for TKA in search of better outcomes for these patients. As hypothesized, the private practice community hospitals appear to be less affected than the academic center in premium THA and TKA implant usage following implementation of price capitation. In fact, while some individual premium criterion rates increased or decreased between time periods, neither THA nor TKA overall premium implant usage statistically changed at the community hospitals. However the ceiling effect observed in TKA at the community hospitals with 100% overall premium criteria implant use for both periods makes comparison between practices difficult. Nonetheless, TKA premium criterion subgroups were not affected by this ceiling effect and still had no statically significant changes. Observed variation in implant selection between hospital settings may highlight the dichotomy in reimbursement systems. The present study had several notable limitations. First, the retrospective design did not allow for control of potential confounding variables between precapitation and postcapitation periods, such as differences in patient populations and surgeons. However, the basic demographics were similar across time periods and exclusion criteria were established to minimize variables that could influence implant
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
M. Farías-Kovac et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
selection. Second, the new capitated pricing policy was accompanied by the addition of five new manufacturers to the previous list of preferred vendors, expanding to a 7-vendor system. We believe that this concurrent change had little effect on implant selection. Of the 22 surgeons who performed surgeries in both periods, only 1 surgeon switched to a different implant manufacturer in the postcapitation period and all manufacturers offered at least one premium system. Third, although no implants were retired from the market during the studied periods, two implants were voluntarily recalled by the manufacturers within months of the end of the postcapitation period. The knowledge of the issues that prompted these recalls could have affected its use by surgeons in both academic and community hospitals prior to its recall. Fourth, the length of the study was limited to 6 months before and after implementation of price capitation. This represents a snapshot in time, and implant selection practice patterns could change. The results of this study represent a first effort at examining the impact of price capitation on premium implant selection at academic and community hospital settings. Further reporting of the financial implications and implant usage trends of different capitated pricing systems is necessary to clarify if price capitation can help sustain implant cost reductions and what institutions might benefit most from this type of pricing model. References 1. Gioe TJ, Sharma A, Tatman P, et al. Do “premium” joint implants add value? Analysis of high cost joint implants in a community registry. Clin Orthop Relat Res 2011;469(1):48. 2. Centers for Disease Control and Prevention-National Center for Health Statistics. National Hospital Discharge Survey: number of all-listed procedures for discharges from short-stay hospitals, by procedure category and age: United States. http:// www.cdc.gov/nchs/data/nhds/4procedures/2010pro4_numberprocedureage.pdf; 2010. Accessed July 2013. 3. Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project. http://hcupnet.ahrq.gov/HCUPnet.jsp . Accessed July 2013. 4. Kurtz S, Ong K, Lau E, et al. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007; 89(4):780.
5
5. Kurtz SM, Lau E, Ong K, et al. Future young patient demand for primary and revision joint replacement: National Projections from 2010 to 2030. Clin Orthop Relat Res 2009;467(10):2606. 6. Healy WL, Rana AJ, Iorio R. Hospital economics of primary total knee arthroplasty at a teaching hospital. Clin Orthop Relat Res 2011;469(1):87. 7. Mendenhall S. Hip and knee implant prices rise 8.9%. Orthop Netw News 2005;16(1):2. 8. Mendenhall S. Hospital resources and implant cost management—a 2010 update. Orthop Netw News 2011;22(3):9. 9. Mendenhall S. 2011 CMS payment update. Orthop Netw News 2010;21(4):22. 10. Physician Executive Council. Orthopaedics practicum: reconcile hospital and physician agendas to improve service line performance. Washington: Advisory Board Company; 2004. 11. Iorio R, Healy WL, Kirven FM, et al. Knee implant standardization: an implant selection and cost reduction program. Am J Knee Surg 1998;11(2):73. 12. Healy WL, Iorio R. Implant selection and cost for total joint arthroplasty: conflict between surgeons and hospitals. Clin Orthop Relat Res 2007;457:57. 13. Taylor B, Fankhauser RA, Fowler T. Financial impact of a capitation matrix system on total knee and total hip arthroplasty. J Arthroplasty 2009;24(5):783. 14. Healy WL, Ayers ME, Iorio R, et al. Impact of a clinical pathway and implant standardization on total hip arthroplasty: a clinical and economic study of shortterm patient outcome. J Arthroplasty 1998;13(3):266. 15. Ho DM, Huo MH. Are critical pathways and implant standardization programs effective in reducing costs in total knee replacement operations? J Am Coll Surg 2007;205(1):97. 16. Krummenauer F, Guenther K-P, Kirschner S. Cost effectiveness of total knee arthroplasty from a health care providers' perspective before and after introduction of an interdisciplinary clinical pathway—is investment always improvement? BMC Health Serv Res 2011;11:338. 17. Obremskey WT, Dail T, Jahangir AA. Value-based purchasing of medical devices. Clin Orthop Relat Res 2012;470(4):1054. 18. Healy WL, Iorio R, Lemos MJ, et al. Single price/case price purchasing in orthopaedic surgery: experience at the Lahey Clinic. J Bone Joint Surg Am 2000;82(5):607. 19. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42(2):377. 20. Robinson JC, Pozen A, Tseng S, et al. Variability in costs associated with total hip and knee replacement implants. J Bone Joint Surg Am 2012;94A(18):1693. 21. Robinson JC. Value-based purchasing for medical devices. Heal Aff Proj Hope 2008;27(6):1523. 22. Issa K, Naziri Q, Johnson AJ, et al. Evaluation of patient satisfaction with physical therapy following primary THA. Orthopedics 2013;36(5):e538. 23. Hamilton D, Henderson GR, Gaston P, et al. Comparative outcomes of total hip and knee arthroplasty: a prospective cohort study. Postgrad Med J 2012; 88(1045):627. 24. Scott CEH, Bugler KE, Clement ND, et al. Patient expectations of arthroplasty of the hip and knee. J Bone Joint Surg Br 2012;94(7):974.
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty Mario Farías-Kovac, MD, Caleb R. Szubski, BA, Mark Hebeish, DDS, Alison K. Klika, MS, Kirtishri Mishra, BS, Wael K. Barsoum, MD Department of Orthopaedic Surgery, Orthopaedic and Rheumatologic Institute, Cleveland Clinic, Cleveland, Ohio
a r t i c l e
i n f o
Article history: Received 2 January 2014 Accepted 14 February 2014 Available online xxxx Keywords: total hip arthroplasty total knee arthroplasty price capitation implant selection premium implants
a b s t r a c t While price capitation strategies may help to control total hip (THA) and knee arthroplasty (TKA) implant costs, its effect on premium implant selection is unclear. Primary THA and TKA cases 6 months before and after capitated pricing implementation were retrospectively identified. After exclusions, 716 THA and 981 TKA from a large academic hospital and 2 midsize private practice community hospitals were reviewed. Academic hospital surgeons increased premium THA implant usage (66.5% to 70.6%; P = 0.28), while community surgeons selected fewer premium implants (36.4%) compared to academic surgeons, with no practice change (P = 0.95). Conversely, premium TKA implant usage significantly increased (73.4% to 89.4%; P b 0.001) for academic surgeons. Community surgeons used premium TKA implants at greater rates in both periods, with all cases having ≥1 premium criterion. © 2014 Elsevier Inc. All rights reserved.
Over the last 2 decades, nontransformative, interval innovation of orthopedic implants has been the rule, with a focus on improving fixation, motion, kinematics, stability, and durability [1]. Implant manufacturers have spent billions of dollars on the development, marketing, and selling of their newest technologies as the demand for joint arthroplasties have exponentially risen. In 2010, total hip (THA) and knee arthroplasty (TKA) accounted for 1,053,000 procedures, with an annual average increase of 4.3% in the number of cases from 1993 to 2011 [2,3]. Driven by a rapidly aging population of baby boomers and a rise in the number of young patients (b 65 years) receiving joint arthroplasty, primary THA and TKA have a projected increase of more than 174% and 673%, respectively, by 2030 [4,5]. Rising implant costs and diminishing reimbursement are a serious challenge to hospitals [6]. As orthopedic implants have evolved, there has been an associated increase in cost. However, these changes are significantly outpacing Medicare's adjustments in reimbursement. From 1991 to 2005, there was a 156% rise in hip and knee implant list
Conflicts of interest and source of funding: No financial support was received for this study. Dr. Barsoum would like to acknowledge the following disclosures: consultant to Stryker Orthopaedics; research support from Stryker Orthopaedics, Zimmer, Cool Systems, Orthovita, DJO, Active Implants, The Medicines Company, and the State of Ohio; royalties from Stryker Orthopaedics, Zimmer, Exactech, and Shukla Medical; stock options in OtisMed Corporation, Custom Orthopaedic Solutions, and iVHR; board member at KEF Healthcare. No other authors have relevant financial relationships to disclose. The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2014.02.020. Reprint requests: Alison K. Klika, MS, Cleveland Clinic, 9500 Euclid Avenue, A41, Cleveland, OH 44195.
prices while the average Medicare payment to hospitals only increased by 19% ($8489 to $10,109) [7]. In 2010, the cost of a primary hip and knee implant was $6398 and $5324, respectively, accounting for approximately half of the total Medicare compensation for an uncomplicated primary joint arthroplasty at the time ($11,653) [8,9]. In 1997 orthopedic programs generated 25% of hospitals' profits, but by 2001 their financial impact had decreased to only 2% [10]. Hospitals have developed and tested several strategies (i.e. implant selection standardization, group purchasing consignment, valuebased purchasing, volume-incentive vendor contracts, single price/ case price purchasing, physician gain sharing, establishment of standardized clinical and surgical pathways) to decrease implant and service costs [11–18]. Capitated pricing, in which a flat purchase price is negotiated for implant line items regardless of technology or manufacturer, has emerged as a successful option for decreasing implant costs in some health care institutions [13]. This model also maintains physician autonomy by offering a full list of implant options. Based on positive outcomes reported on experiences utilizing capitated pricing strategies, our health system (which includes a large academic tertiary referral center and a number of community-based satellite hospitals) replaced a discount-based vendor-buyer system with a capitated pricing model for primary total joint implants in 2011. This capitated pricing system assigned a single price cap for implant line items, regardless of technology. The purpose of this study was to evaluate whether the implementation of this type of capitated pricing system affected usage rates of primary THA and TKA premium implants (i.e. more expensive to produce, newer technology) by salaried surgeons at a large academic hospital. As physicians are no longer burdened by minimizing costs
0883-5403/0000-0000$36.00/0 – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.arth.2014.02.020
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
2
M. Farías-Kovac et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
for the patient or maximizing revenue for the hospitals under capitated pricing, we hypothesized that physicians would be more likely to select premium implants for their patients. A secondary objective was to compare the implant selection trends at the academic hospital to those at two midsize private practice–based community hospitals from the same health system. The surgeons at these community hospitals receive the same implant prices as the large academic hospital, yet their reimbursement systems are independent of the health system. Our secondary hypothesis was that private practice surgeons at these community hospitals are less likely to be affected by this same implant purchasing model. Methods After approval by the institutional review board, all consecutive primary THA (CPT code 27130) and TKA (CPT code 27447) surgeries 6 months before and after implementation of capitated pricing (7/1/ 2011) were retrospectively queried. Study sites included an academic, tertiary referral hospital and two community hospitals within the same health system. Surgeon education regarding the new pricing policy was conducted for a 1-month period following implementation, and these data were excluded from the analysis. This process established preprice (1/1/2011–6/30/2011) and postprice capitation (8/1/2011–1/31/2012) periods. An initial cohort of 1881 cases (THA, n = 794; TKA, n = 1087) were identified that met the inclusion criteria. Simultaneous and staged bilateral procedures were counted as two separate cases, one for each limb. All staged bilateral procedures were retained in the study cohorts because there were no instances that spanned both preprice and postprice capitation periods. Exclusion criteria (Fig. 1)
Inclusion Criteria All consecutive primary total hip arthroplasty (THA) (CPT code 27130) and total knee arthroplasty (TKA) (CPT code 27447) cases performed at a tertiary academic hospital and two community hospitals in following time periods before and after price capitation was implemented (7/1/2011): Pre-capitated (1/1/2011-6/30/2011) Post-capitated (8/1/2011-1/31/2012) *OMITTED 1 month surgeon education period (7/1/2011-7/31/2011)* THA (n=794); TKA (n=1087)
Excluded THA patients (n=78) - Procedure surgeon not represented in both pre- and post-capitated periods (n=29) - Bipolar prosthesis (n=14) - Complex pathologies (i.e. dwarfism, bone tumor lesion, presence of previous hardware) (n=11) - Metal-on-metal prosthesis (n=10) - Nickel allergy (n=8) - Unipolar prosthesis (n=3) - One-of-a-kind prosthesis (n=2) - Resurfacing procedure (n= 1)
Final THA Cohort (n=716) n=463 Academic Hospital n=253 Community Hospital Analyzed variables: - Patient demographics - Hospital location - Implant fixation - Implant characteristics - Bearing surface
Excluded TKA patients (n=106) - Procedure surgeon not represented in both pre- and post-capitated periods (n=29) - Total stabilizing system (n=20) - Nickel allergy (n=20) - Unicompartmental TKA (n=16) - Complex pathologies (i.e. dwarfism, bone tumor lesion, presence of previous hardware, previous fracture) (n=15) - Revision system (n=4) - Cobalt allergy (n=1) - Titanium allergy (n=1)
Final TKA Cohort (n=981) n=666 Academic Hospital n=315 Community Hospital Analyzed variables: - Patient demographics - Hospital location - Implant fixation - Implant characteristics - Bearing surface
Fig. 1. Inclusion/Exclusion flow diagram.
were established to remove cases that did not use primary total joint arthroplasty implants (e.g., unipolar, bipolar, unicompartmental) and to minimize situations in which the surgeon may not have had complete freedom of implant selection (e.g., complex pathology, metal allergies, use of specific implant type, metal-on-metal designs). Cases performed by a surgeon who was not represented in both preprice and postprice capitation periods were excluded due to the potential bias. After exclusions, a total of 463 THA and 666 TKA from the large academic hospital, and 253 THA and 315 TKA from the 2 midsize community hospitals comprised the final study cohort. A review of patient demographics and implant characteristics for each case was performed using the electronic medical record. Demographics variables included age at surgery, gender, body mass index (BMI), and laterality. Using implant catalog numbers and operative notes, the manufacturer, model, and material of each implant subcomponent were noted. Additionally, bearing surface, design characteristics, and component fit (i.e. cemented or press-fit) were collected. Classification of each THA and TKA case's implants as premium or non-premium was based on a slightly expanded version of that used by Gioe et al [1]. Premium THA implants were defined by the existence of one of the following bearing surfaces: second (2G) or third generation (3G) highly cross-linked polyethylene liner with a ceramic or oxidized-zirconium femoral head, ceramic liner with a ceramic femoral head, or mobile-bearing system. Large femoral head size and press-fit femoral stems were not regarded as premium factors. Second-generation highly cross-linked polyethylene liners included: AltrX Altralinked (DePuy, Warsaw, IN), Marathon (DePuy, Warsaw, IN), Longevity (Zimmer, Warsaw, IN), and R3 XLPE (Smith & Nephew, Memphis, TN). Third-generation highly cross-linked polyethylene liners included only X3 (Stryker, Mahwah, NJ). Premium TKA implants were defined by the existence of at least one of the following criteria: mobile-bearing design, high-flexion design, oxidized-zirconium femoral component, and/or highly cross-linked polyethylene bearing surface. Study data were collected and managed using REDCap (Research Electronic Data Capture), which is a secure, Web-based application designed to support data capture for research studies [19]. The data were analyzed using R software (Version 3.0.2, Vienna, Austria). Continuous variables were described using means, standard deviations, and 95% confidence intervals for means. Categorical variables were described using counts and percentages. Continuous demographic variables were compared using Welch's two-sample t-test. Categorical demographic variables were compared using Pearson's chi-squared test with Yates' continuity correction. Hip and knee premium implant usage rates were compared between precapitation and postcapitation periods within the subsets defined by academic and community hospitals using logistic regression fit using generalized estimating equations (GEE) to account for staged bilateral cases, on which observations cannot be assumed to be independent. Age, gender, and BMI were included in these models and so the resulting Pvalues testing for differences between precapitation and postcapitation are adjusted P-values. Rates of specific premium and nonpremium implant subgroups were compared using Fisher's exact test for count data or Pearson's chi-squared test, as appropriate, ignoring repeated measures and the effects of age, gender, and BMI. Premium and non-premium THA implant characteristics are mutually exclusive. Conversely, premium TKA characteristics are not mutually exclusive (i.e. a patient might receive an implant that possesses more than one premium characteristic). Rates of usage of particular TKA implant types were therefore adjusted for multiplicity. Results A final cohort of 716 THA and 981 TKA procedures performed by 22 surgeons was analyzed (Fig. 1). The mean years of clinical practice
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
M. Farías-Kovac et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
3
Table 1 Demographic Data for the THA and TKA Study Cohorts. Academic Hospital
Variable Total hip arthroplasty Number of cases; surgeons Age at surgery (yrs) a Female c Body mass indexa Left laterality c Total knee arthroplasty Number of cases Age at surgery (yrs) Female c Body mass indexa Left laterality a b c d
c
a
Pre-Capitated
Post-Capitated
215; 10 61.9 (13.3) [60.0–63.6] 124 (57.7%) 30.2 (6.5) [29.3–31.1] 96 (44.6%)
248; 10 61.8 (13.2) [60.2–63.5] 139 (56.0%) 29.9 (6.3) [29.2–30.7] 128 (51.6%)
364; 12 66.0 (10.2) [64.9–67.0] 224 (61.5%) 31.8 (6.8) [31.1–32.5] 177 (48.6%)
302; 12 65.6 (9.9) [64.5–66.7] 169 (56.0%) 32.4 (6.9) [31.7–33.2] 156 (51.7%)
Community Hospital
P Value
0.97b 0.72d 0.67b 0.14d
0.64b 0.17d 0.26b 0.48d
Overall
Pre-Capitated
Post-Capitated
463; 10 61.8 (13.2) [60.6–63.0] 263 (56.8%) 30.1 (6.4) [29.5–30.6] 224 (48.4%)
121; 7 67.3 (10.5) [65.4–69.2] 72 (59.5%) 30.7 (8.3) [29.2–32.2] 66 (54.5%)
132; 7 67.9 (10.2) [66.1–69.7] 74 (56.0%) 30.8 (7.5) [29.5–32.1] 61 (46.2%)
666; 12 65.8 (10.1) [65.1–66.6] 393 (59.0%) 32.1 (6.8) [31.6–32.6] 333 (50.0%)
155; 9 69.7 (10.4) [68.0–71.4] 107 (69.0%) 32.7 (7.1) [31.6–33.8] 82 (52.9%)
160; 9 66.7 (10.3) [65.1–68.3) 98 (61.2%) 33.6 (6.8) [32.5–34.7] 83 (51.9%)
P Value
0.64b 0.58d 0.92b 0.19d
0.01b 0.18d 0.26b 0.94d
Overall 253; 7 67.6 (10.3) [66.3–68.9] 146 (57.7%) 30.8 (7.9) [29.8–31.7] 127 (50.2%)
315; 9 68.2 (10.4) [67.0–69.3] 205 (65.1%) 33.2 (7.0) [32.4–33.9] 165 (52.4%)
P Value (Acad vs. Comm)
b0.001b 0.82d 0.20b 0.64d
b0.001b 0.08d 0.03b 0.53d
Result values are expressed as mean (standard deviation) [95% confidence interval]. Welch's two-sample t-test. Result values are expressed as number of cases (percentage). Pearson's chi-squared test with Yates' continuity correction.
among academic and community hospital surgeons were 19.8 ± 7.3 years (95% confidence interval, 15.2–24.5 years) and 23.3 ± 7.9 years (95% confidence interval, 17.6–29.0 years), respectively (P = 0.30). The percentage of adult reconstruction fellowship-trained surgeons was 66.7% (n = 8/12) at the academic setting and 40.0% (n = 4/10) at the community hospitals (P = 0.21). The demographic makeup of the THA and TKA study cohorts was similar across hospital types, with only a slightly younger population being operated on at the academic hospital for both procedures (P b 0.001) (Table 1). There were no significant differences in demographic data between precapitation and postcapitation periods for each hospital type and procedure (Table 1). Implementation of a capitated pricing model at our institution had no significant effect on the type of THA implant selected by either academic center or community hospital surgeons (Table 2). At the academic hospital, rates of premium THA implant usage slightly increased from 66.5% (n = 143) in the precapitation period to 70.6% (n = 175) in the postcapitation period, although this increase was not statistically significant (P = 0.28). Community hospital surgeons selected considerably fewer premium implants (36.4%) in both periods compared to their academic hospital peers and did not change their practice after the new pricing model was instituted (P = 0.95). Furthermore, there were no cases of polyethylene 2G liner
paired with oxidized-zirconium head, ceramic liner with ceramic head, or mobile-bearing systems (three of the five premium criteria) in either period at the community hospitals. Premium TKA implant usage following implementation of the new pricing model significantly increased at the academic hospital from 73.4% (n = 267) to 89.4% (n = 270) (P b 0.001) (Table 3). Mobilebearing and high-flexion designs, which were not utilized in the precapitation period, were used more frequently in the postcapitation period (1. 0%, P = 0.09; 5.3%, P b 0.001; respectively). Highly crosslinked polyethylene use at the academic hospital increased from 73.4% (n = 267) to 83.1% (n = 251) (P = 0.005), and comprised the majority of this hospital type's premium implant usage. No TKA cases with two or more premium criteria were used in either period by academic center surgeons. The community hospital surgeons selected premium TKA implants at much greater rates than their academic peers (Table 3). In fact, all TKA cases at the community hospitals had at least one premium criterion, with no statistical significant changes in utilization of any subgroup of implant. High-flexion knee system usage increased from 10.3% (n = 16) to 18.8% (n = 30) (P = 0.11) following implementation of price capitation. Mobile-bearing designs comprised a large portion of the community hospitals' premium implant usage before (46.5%) and after (50.0%) capitated pricing (P = 0.57). There was a
Table 2 Distribution of THA Implants by Premium Criteria Before and After Implementation of Capitated Pricing, Classified by Hospital Type. Academic Hospital Variable Premium Polyethylene 2G liner—ceramic head Polyethylene 3G liner—ceramic head Polyethylene 2G liner—oxidized-zirconium head Ceramic liner—ceramic head Mobile-bearing system Non-premium Polyethylene 2G liner—metal head Polyethylene 3G liner—metal head
Community Hospital
Pre-Capitated (n = 215)
Post-Capitated (n = 248)
P Value
Pre-Capitated (n = 121)
Post-Capitated (n = 132)
143 (66.5%) 19 (8.8%) 101 (47.0%) 4 (1.9%) 17 (7.9%) 2 (0.9%)
175 (70.6%) 17 (6.9%) 132 (53.2%) 11 (4.4%) 15 (6.0%) 0
0.28a N0.99b N0.99b N0.99b N0.99b N0.99c
44 (36.4%) 2 (1.7%) 42 (34.7%) 0 0 0
48 (36.4%) 4 (3.0%) 44 (33.3%) 0 0 0
0.95a N0.99c N0.99b
28 (13.0%) 44 (20.5%)
37 (14.9%) 36 (14.5%)
N0.99b 0.82b
33 (27.3%) 44 (36.4%)
37 (28.0%) 47 (35.6%)
N0.99b N0.99b
P Value
Result values are expressed as number of cases (raw percentage). a Logistic regression (GEE); adjusted for age, gender, and BMI. b Pearson's chi-squared test with Yates' continuity correction; corrected for multiple testing. c Fisher's exact test for count data; corrected for multiple testing.
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
4
M. Farías-Kovac et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
Table 3 Distribution of TKA Implants by Premium Criteria Before and After Implementation of Capitated Pricing, Classified by Hospital Type. Academic Hospital
Community Hospital
Variable
Pre-Capitated (n = 364)
Post-Capitated (n = 302)
P Value
Pre-Capitated (n = 155)
Post-Capitated (n = 160)
Premium (≥1 below criteria) Mobile-bearing design High-flexion design Oxidized-zirconium femoral component Highly cross-linked polyethylene
267 (73.4%) 0 0 0 267 (73.4%)
270 (89.4%) 3 (1.0%) 16 (5.3%) 0 251 (83.1%)
b0.001a 0.09b b0.001b
155 72 16 0 70
160 80 30 0 53
0.005b
(100.0%) (46.5%) (10.3%) (45.2%)
P Value
(100.0%) (50.0%) (18.8%)
0.57b 0.11b
(33.1%)
0.11b
Result values are expressed as number of cases (raw percentage). a Logistic regression (GEE); adjusted for age, gender, and BMI. b Fisher's exact test for count data; multiplicity corrected.
decrease in the use of highly cross-linked polyethylene from 45.2% (n = 70) to 33.1% (n = 53) (P = 0.11). No oxidized-zirconium femoral components were used in either time period or at either hospital type. The majority of the community hospital TKA cases had only one premium criterion, with three cases before and after capitated pricing having two premium criteria. Discussion As of 2010, Medicare, Medicaid, and private insurers were the major sources of reimbursement for THA and TKA in the United States, representing 53%, 14%, and 31% of these cases, respectively [3]. Accordingly, 67% of THA and TKA reimbursement nationally are subjected to diagnosis-related group (DRG) weights and payments by the Centers for Medicare & Medicaid Services. A recent study examining amounts paid to manufacturers by 61 institutions across 8 different states in 2008 showed an average total surgical cost for THA and TKA of $12,548 and $11,666, respectively, with implant cost accounting for 50.2% and 43.5% of the surgical cost, respectively. A multivariate regression analysis of these data for both procedures found higher implant costs were associated with age, payer (Medicare), complications, discharge disposition (discharge to acute or postacute care facility), and a diagnosis of fracture (for TKA only) [20]. Prior to establishing a capitated pricing model for primary total joint implants, our institution utilized a two-vendor buying system based on volume discounts. This previous approach required an integrated physician organization and limited the number of implant vendors, theoretically reducing the competition to stimulate increased willingness to discount prices [21]. Under this model the physician's choice of orthopedic implants for THA and TKA was largely restricted to those offered by two medical device companies in order to assure high demand by the hospitals in our health system. This high volume was matched with discounted prices by the vendors. Our institution ultimately migrated away from this pricing model in search of higher surgeon satisfaction and patient outcomes through a wider range of implant options while maintaining acceptable costs. Implementation of a capitated pricing model helped remove our institution's reliance on discount- and volume-based purchasing as well as cut total implant costs by approximately 10% overall for THA and TKA, while increasing the range of manufacturers and on-theshelf implant options available to surgeons. It is also important to note that the term premium implants may be misleading. We use this term to define implants that were marketed as premium implants and had list pricing commensurate with their perceived premium status. To our knowledge, there is only one other detailed report in the literature of a capitated pricing system used for orthopedic implant purchasing. Taylor et al [13] presented similar positive financial results at their institution after implementation of a four-tier price capitation system for purchasing THA and TKA implants, reducing costs per implant by 26.1%. This tiered categorization matrix placed implants with similar levels of technology or innovation in categories (I-IV) based on implant fit and bearing surface, and negotiated a price for each category. The classification implemented in our study was a
modified version of that used by Gioe et al [1], focusing on bearing surface technology and classifying implants as premium or nonpremium (e.g. 2 categories). Taylor et al [13] found no statistically significant change in overall implant selection at their institution. However, usage of their most premium and expensive THA and TKA implants (category IV) increased from 55.8% to 63.4% and 26.8% to 29.0%, respectively, with implementation of the new system. A theoretical disadvantage of tiered capitated pricing systems is that cost is not removed entirely from the implant selection equation as a price difference between premium and non-premium implants is retained. No studies to date have reported the effect of a single price cap system, like the one used at our institution, on implant selection. There are noted obstacles that can be encountered when trying to establish a price capitation system. In particular, capitated pricing models typically require large volumes to negotiate beneficial pricing and depending on buying power, they might not cover the purchasing of certain goods (e.g., newest technology), which would require separate pricing contracts or purchasing at the manufacturer's list price. Also, the specifics of a capitated pricing model can vary greatly depending on the necessities or goals of the buyer and the demands of the seller. There are substantial differences in practice patterns between academic and community hospitals for both THA and TKA. The data from our institution demonstrate that the influence of capitated pricing is very different for hip and knee implants, with a trend toward increased premium implant usage for primary TKA in the academic setting only. No statistically significant changes in premium implant selection were found for primary THA in either location. Additionally, private practice community hospital surgeons used premium TKA implants at very high rates (100.0%) and premium THA implants at very low rates (36.4%). We believe that these differences in implant selection between procedure type may be due to the higher levels of satisfaction among patients after THA compared to TKA [22–24]. Surgeons may pursue novel and premium technologies at greater rates for TKA in search of better outcomes for these patients. As hypothesized, the private practice community hospitals appear to be less affected than the academic center in premium THA and TKA implant usage following implementation of price capitation. In fact, while some individual premium criterion rates increased or decreased between time periods, neither THA nor TKA overall premium implant usage statistically changed at the community hospitals. However the ceiling effect observed in TKA at the community hospitals with 100% overall premium criteria implant use for both periods makes comparison between practices difficult. Nonetheless, TKA premium criterion subgroups were not affected by this ceiling effect and still had no statically significant changes. Observed variation in implant selection between hospital settings may highlight the dichotomy in reimbursement systems. The present study had several notable limitations. First, the retrospective design did not allow for control of potential confounding variables between precapitation and postcapitation periods, such as differences in patient populations and surgeons. However, the basic demographics were similar across time periods and exclusion criteria were established to minimize variables that could influence implant
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
M. Farías-Kovac et al. / The Journal of Arthroplasty xxx (2014) xxx–xxx
selection. Second, the new capitated pricing policy was accompanied by the addition of five new manufacturers to the previous list of preferred vendors, expanding to a 7-vendor system. We believe that this concurrent change had little effect on implant selection. Of the 22 surgeons who performed surgeries in both periods, only 1 surgeon switched to a different implant manufacturer in the postcapitation period and all manufacturers offered at least one premium system. Third, although no implants were retired from the market during the studied periods, two implants were voluntarily recalled by the manufacturers within months of the end of the postcapitation period. The knowledge of the issues that prompted these recalls could have affected its use by surgeons in both academic and community hospitals prior to its recall. Fourth, the length of the study was limited to 6 months before and after implementation of price capitation. This represents a snapshot in time, and implant selection practice patterns could change. The results of this study represent a first effort at examining the impact of price capitation on premium implant selection at academic and community hospital settings. Further reporting of the financial implications and implant usage trends of different capitated pricing systems is necessary to clarify if price capitation can help sustain implant cost reductions and what institutions might benefit most from this type of pricing model. References 1. Gioe TJ, Sharma A, Tatman P, et al. Do “premium” joint implants add value? Analysis of high cost joint implants in a community registry. Clin Orthop Relat Res 2011;469(1):48. 2. Centers for Disease Control and Prevention-National Center for Health Statistics. National Hospital Discharge Survey: number of all-listed procedures for discharges from short-stay hospitals, by procedure category and age: United States. http:// www.cdc.gov/nchs/data/nhds/4procedures/2010pro4_numberprocedureage.pdf; 2010. Accessed July 2013. 3. Agency for Healthcare Research and Quality. Healthcare Cost and Utilization Project. http://hcupnet.ahrq.gov/HCUPnet.jsp . Accessed July 2013. 4. Kurtz S, Ong K, Lau E, et al. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007; 89(4):780.
5
5. Kurtz SM, Lau E, Ong K, et al. Future young patient demand for primary and revision joint replacement: National Projections from 2010 to 2030. Clin Orthop Relat Res 2009;467(10):2606. 6. Healy WL, Rana AJ, Iorio R. Hospital economics of primary total knee arthroplasty at a teaching hospital. Clin Orthop Relat Res 2011;469(1):87. 7. Mendenhall S. Hip and knee implant prices rise 8.9%. Orthop Netw News 2005;16(1):2. 8. Mendenhall S. Hospital resources and implant cost management—a 2010 update. Orthop Netw News 2011;22(3):9. 9. Mendenhall S. 2011 CMS payment update. Orthop Netw News 2010;21(4):22. 10. Physician Executive Council. Orthopaedics practicum: reconcile hospital and physician agendas to improve service line performance. Washington: Advisory Board Company; 2004. 11. Iorio R, Healy WL, Kirven FM, et al. Knee implant standardization: an implant selection and cost reduction program. Am J Knee Surg 1998;11(2):73. 12. Healy WL, Iorio R. Implant selection and cost for total joint arthroplasty: conflict between surgeons and hospitals. Clin Orthop Relat Res 2007;457:57. 13. Taylor B, Fankhauser RA, Fowler T. Financial impact of a capitation matrix system on total knee and total hip arthroplasty. J Arthroplasty 2009;24(5):783. 14. Healy WL, Ayers ME, Iorio R, et al. Impact of a clinical pathway and implant standardization on total hip arthroplasty: a clinical and economic study of shortterm patient outcome. J Arthroplasty 1998;13(3):266. 15. Ho DM, Huo MH. Are critical pathways and implant standardization programs effective in reducing costs in total knee replacement operations? J Am Coll Surg 2007;205(1):97. 16. Krummenauer F, Guenther K-P, Kirschner S. Cost effectiveness of total knee arthroplasty from a health care providers' perspective before and after introduction of an interdisciplinary clinical pathway—is investment always improvement? BMC Health Serv Res 2011;11:338. 17. Obremskey WT, Dail T, Jahangir AA. Value-based purchasing of medical devices. Clin Orthop Relat Res 2012;470(4):1054. 18. Healy WL, Iorio R, Lemos MJ, et al. Single price/case price purchasing in orthopaedic surgery: experience at the Lahey Clinic. J Bone Joint Surg Am 2000;82(5):607. 19. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42(2):377. 20. Robinson JC, Pozen A, Tseng S, et al. Variability in costs associated with total hip and knee replacement implants. J Bone Joint Surg Am 2012;94A(18):1693. 21. Robinson JC. Value-based purchasing for medical devices. Heal Aff Proj Hope 2008;27(6):1523. 22. Issa K, Naziri Q, Johnson AJ, et al. Evaluation of patient satisfaction with physical therapy following primary THA. Orthopedics 2013;36(5):e538. 23. Hamilton D, Henderson GR, Gaston P, et al. Comparative outcomes of total hip and knee arthroplasty: a prospective cohort study. Postgrad Med J 2012; 88(1045):627. 24. Scott CEH, Bugler KE, Clement ND, et al. Patient expectations of arthroplasty of the hip and knee. J Bone Joint Surg Br 2012;94(7):974.
Please cite this article as: Farías-Kovac M, et al, Effect of Price Capitation on Implant Selection for Primary Total Hip and Knee Arthroplasty, J Arthroplasty (2014), http://dx.doi.org/10.1016/j.arth.2014.02.020
