Use of reverse total shoulder arthroplasty in the Medicare population

J Shoulder Elbow Surg (2015) -, 1-7

www.elsevier.com/locate/ymse

Use of reverse total shoulder arthroplasty in the Medicare population Judd S. Day, PhDa, E. Scott Paxton, MDb,c, Edmund Lau, MSa, Victoria A. Gordon, BAb, Joseph A. Abboud, MDb,*, Gerald R. Williams, MDb a

Exponent, Inc., Philadelphia, PA, USA Rothman Institute, Thomas Jefferson University, Philadelphia, PA, USA c Warren Alpert Medical School of Brown University, Providence, RI, USA b

Background: Reverse shoulder arthroplasty (RSA) has been Food and Drug Administration approved in the United States since 2004 but did not obtain a unique code until 2010. Therefore, the use of this popular procedure has yet to be reported. The purpose of this study was to examine the use and reimbursement of RSA compared with total shoulder arthroplasty (TSA) and shoulder hemiarthroplasty (SHA). Methods: We analyzed the 100% sample of the 2011 Medicare Part A claims data for patients aged 65 years or older. Patient demographic characteristics, diagnoses, provider information, reimbursements, and lengths of stay were extracted from the claims data. Results: In 2011, a total of 31,002 shoulder arthroplasty procedures were performed; 37% were RSAs, 42% were TSAs, and 21% were SHAs. Osteoarthritis was the primary diagnosis code in 91% of TSAs, 37% of SHAs, and 45% of RSAs. A primary diagnosis of osteoarthritis with no secondary code for rotator cuff tear was found in 22% of patients undergoing RSA. The mean length of stay for RSA (2.6 days; SD, 2.1 days) was longer than that for TSA (2.1 days; SD, 1.5 days) and shorter than that for SHA (3.5 days; SD, 3.6 days) (P < .001). Lower-volume surgeons (<10 arthroplasties per year) performed most shoulder arthroplasties: 57% of RSAs, 65% of TSAs, and 97% of SHAs. Seventy percent of RSAs were implanted by surgeons who performed more RSAs than TSAs and SHAs combined. Conclusions: RSA is performed with similar frequency to TSA and almost twice as much as SHA in the Medicare population. Lower-volume surgeons perform most RSAs, and a majority of surgeons perform more RSAs than all anatomic shoulder arthroplasties combined. Level of evidence: Epidemiology Study, Database Analysis. Ó 2015 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Shoulder arthroplasty; reverse shoulder arthroplasty; Medicare claims data

The patient data are deidentified and publicly available; therefore, the study was exempt from requiring institutional review board approval. *Reprint requests: Joseph A. Abboud, MD, Rothman Institute, 925 Chestnut St, Philadelphia, PA 19107, USA. E-mail address: [email protected] (J.A. Abboud).

Reverse total shoulder arthroplasty has been Food and Drug Administration approved in the United States since 2004. Early reports from European surgeons showed complication rates as high as 50% and recommended that reverse total shoulder arthroplasty be reserved for patients older than 70 years with low functional demands and ‘‘severe shoulder dysfunction caused by an irreparable

1058-2746/$ - see front matter Ó 2015 Journal of Shoulder and Elbow Surgery Board of Trustees. http://dx.doi.org/10.1016/j.jse.2014.12.023

2 rotator cuff tear associated with other glenohumeral lesions.’’6,24 More recently, there has been increasing interest in expanded indications including acute proximal humeral fractures, late fracture sequelae, osteoarthritis (OA) with severe glenoid deformity and an intact cuff, post-traumatic arthritis, and revision shoulder arthroplasty.20,23 Despite these expanded indications and the widespread use of reverse shoulder arthroplasty (RSA), its use has yet to be reported in the United States. Until 2010, the International Classification of Diseases code for RSA was the same as that for an anatomic total shoulder arthroplasty (TSA). As a result, little has been known about the overall use and cost of this new technology. With the implementation of a new International Classification of Diseases code for RSA, its use can now be examined. Multiple studies have evaluated trends in the use of all shoulder arthroplasties but none specific to RSA.3,7,12,14 One of these studies, by Kim et al,14 reported a sharp increase in the incidence of TSA in 2004 with an increased linear slope in the number of shoulder arthroplasties performed after 2004 compared with before. They postulated that this was related to the Food and Drug Administration approval of RSA in the United States and pointed out that further studies were needed to identify the use of RSA along with its safety and efficacy. The US Medicare population includes more than 40 million persons aged 65 years or older and is the single largest insurer for the elderly. Medicare claims data have been used previously to examine treatment patterns, outcomes, and trends in hip, knee, and shoulder arthroplasty.2,3,10-12,15-18,21 The 2011 Medicare Part A claims data allow for the examination of the use of TSA, shoulder hemiarthroplasty (SHA), and RSA in the United States. The purposes of this study were to examine and compare the current use of RSA with that of TSA and SHA in the Medicare population in the United States and to provide data on the potential impact of cost. Our hypotheses are that RSA accounts for a substantial portion of the volume and costs associated with shoulder arthroplasty in the Medicare population and that RSAs are implanted for a variety of diagnoses, in addition to cuff tear arthropathy.

Methods In this retrospective study, we analyzed claims submitted by hospitals and hospital outpatient clinics (Medicare Part A) from January 1 through December 31, 2011. Individual claims submitted to Medicare for payment were deidentified and encrypted by the Centers for Medicare and Medicaid Services (CMS) and were then made available for qualified health care researchers. These claims records reflect the medical service for elderly Medicare beneficiaries who receive care through the conventional fee-for-service program, not including beneficiaries enrolled in the Medicare Advantage plans (ie, Medicare Part C) or beneficiaries who were Medicare eligible because of disability or end-stage renal disease. The CMS made both 100% and 5% samples of these

J.S. Day et al. claims records available for investigators, and the 100% data from 2011 were used in this study. Claims associated with TSA, SHA, and RSA were identified and extracted using the appropriate procedure codes in accordance with the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Upper extremity revision procedures, identified by ICD-9-CM code 81.97, were excluded from this analysis. Patient demographic characteristics, diagnoses, provider information, reimbursements, and length-of-stay data were also extracted from the claims data using the ICD-9-CM where appropriate. Diagnoses were categorized into 1 of 9 categories based on a predetermined algorithm (Table I). These diagnoses were evaluated for the primary diagnosis associated with the RSA, as well as a separate evaluation for any secondary diagnoses. Provider data were categorized by procedure volume for the operating surgeon. Providers were categorized by the number of arthroplasties performed per year in Medicare patients as very low volume (1-5), low volume (6-10), moderate volume (11-20), or high volume (>20) using the identifier for the operating surgeon. This was based on clinically meaningful cutoff values based on those used by Jain et al.10,13 Length-of-stay data and hospital payments were analyzed for each procedure type. Data were stratified for patients with a principal diagnosis of fracture/ dislocation with comparison to other diagnoses. To estimate the proportion of the national procedure volume that was captured by our Medicare Part A analysis, we compared the Medicare Part A procedure counts with those reported by the Healthcare Cost and Utilization Project (HCUP) Nationwide Inpatient Sample (NIS) (http://hcupnet.ahrq.gov/).8 This provided an estimate of the total nationwide volume for all ages and payers in the United States. Standard errors for procedure counts from the HCUP NIS data have been presented as reported by the online database. Standard errors were not required for the Medicare procedure counts because the 100% sample was used. Mean values for age, length of stay, and claim payments were compared for the Medicare population between procedure types using a 1-way analysis of variance with post hoc Bonferroni testing (IBM SPSS Statistics, version 20; IBM, Armonk, NY, USA).

Results There were a total of 31,002 shoulder arthroplasty procedures that met our inclusion criteria in the Medicare Part A population. Of these, 37% were RSAs, 42% were TSAs, and 21% were SHAs (Table II). Two thirds (66%) of all RSA recipients were female patients. This proportion of female recipients was greater than that for TSA but smaller than the proportion of female recipients of SHAs (Table II). Within the studied population, recipients of RSAs were approximately 2 years older than recipients of TSAs (P < .001) but approximately the same age as recipients of SHAs (P ¼ .614) (Table II). The proportions of admissions that were classified as emergent or urgent were 10%, 5%, and 30% for RSA, TSA, and SHA, respectively (Table II). The mean length of stay for RSA was 2.6 days, varying from 2.3 days for high-volume surgeons to 2.8 days for very low–volume surgeons (h2 ¼ 0.01). This was

Reverse shoulder in the Medicare population Table I

3

ICD-9-CM diagnosis and treatment codes used to identify index procedure and diagnoses ICD-9-CM code

Index procedure Total shoulder arthroplasty Shoulder hemiarthroplasty Reverse total shoulder arthroplasty Diagnosis Osteoarthritis Rheumatoid and other inflammatory arthritis Arthropathy/traumatic arthropathydnoninflammatory Arthropathy with rotator cuff tear Fracture/dislocation

Late fracture sequelae Aseptic necrosis Mechanical complication Other

81.80 81.81 81.88 715.00, 715.09, 715.10, 715.11, 715.12, 715.20, 715.21, 715.22, 715.30, 715.31, 715.32, 715.80, 715.89, 715.90, 715.91, 715.92 274.00, 274.01, 274.02, 274.03, 274.10, 696.0, 710.0, 710.1, 711.01, 711.02, 711.09, 714.0, 714.1, 714.2, 714.30, 714.31, 714.32, 714.33, 714.4, 714.89, 714.9, 719.21, 719.22, 719.29, 719.31, 719.32, 719.39, 725, 729.0 716.10, 716.11, 716.12, 716.19, 716.41, 716.42, 716.49, 716.80, 716.81, 716.82, 716.89, 716.91, 716.92, 716.99 726.10, 726.13, 726.19, 727.60, 727.61, 728.83, 840.3, 840.4, 840.5, 905.8 718.21, 718.22, 718.29, 718.31, 718.32, 718.39, 810.00, 810.03, 810.10, 810.13, 811.00, 811.01, 811.02, 811.03, 811.09, 811.10, 811.11, 811.12, 811.13, 811.19, 812.00, 812.01, 812.02, 812.03, 812.09, 812.10, 812.11, 812.12, 812.13, 812.19, 812.20, 812.21, 812.30, 812.31, 812.40, 812.41, 812.42, 812.43, 812.44, 812.49, 812.50, 812.51, 812.52, 812.53, 812.54, 812.59, 813.00, 818.0, 818.1, 819.0, 819.1, 831.00, 831.01, 831.02, 831.03, 831.04, 831.09, 831.10 733.81, 733.82, 905.2, 905.6, V54.09, V54.10, V54.11, V54.20, V54.21 733.40, 733.41, 733.49 996.40, 996.41, 996.42, 996.43, 996.44, 996.45, 996.46, 996.47, 996.49

ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification.

Table II

Medicare population demographic and admission-related data

n (%) Admission type Emergency Urgent Elective Other Gender (female/male) Overall Fracture/dislocation Other diagnosis Age, y, mean (SD) Overall Fracture/dislocation Other diagnosis Length of stay, d, mean (SD) Overall Fracture/dislocation Other diagnosis Payment, $, mean (SD) Overall Fracture/dislocation Other diagnosis

RSA

TSA

SHA

11,480 (37%)

13,152 (42%)

4% 6% 90% 1%

1% 4% 94% 1%

20% 10% 69% 1%

66%/34% 84%/16% 64%/36%

58%/42% 78%/22% 58%/42%

73%/27% 85%/15% 63%/37%

6,370 (21%)

75.7 (6.3) 78.1 (6.9) 75.4 (6.1)

73.4 (5.8) 76.9 (7.3) 73.3 (5.8)

75.8 (7.1) 77.4 (7.3) 74.5 (6.6)

2.6 (2.1) 4.1 (3.3) 2.4 (1.7)

2.1 (1.5) 4.6 (4.0) 2.1 (1.4)

3.5 (3.6) 4.5 (3.6) 2.7 (3.5)

12,625 (4,619) 13,576 (5,718) 12,502 (4,444)

11,989 (4,416) 13,710 (10,314) 11,965 (4,270)

12,958 (5,906) 13,605 (5,316) 12,439 (6,293)

RSA, reverse shoulder arthroplasty; SHA, shoulder hemiarthroplasty; TSA, total shoulder arthroplasty.

4

J.S. Day et al. Table III

Comparative demographic data from National Inpatient Survey

Procedure Total Age group 18-44 y 45-64 y 65-84 y 85 y Sex Male Female Payer Medicare Medicaid Private insurance Uninsured Other

RSA

TSA

SHA

Discharges

SE

Discharges

SE

Discharges

SE

21,916 (100.00%)

1,589

29,414 (100.00%)

1,665

15,860 (100.00%)

701

83 3,948 16,141 1,713

(0.38%) (18.02%) (73.65%) (7.81%)

21 430 1,132 157

575 9,923 18,024 849

(1.95%) (33.74%) (61.28%) (2.89%)

80 647 1,007 82

7,868 (35.90%) 14,002 (63.89%)

641 984

14,483 (49.24%) 14,855 (50.50%)

903 813

17,215 314 3,291 128 878

(78.55%) (1.43%) (15.02%) (0.58%) (4.00%)

1,209 61 335 33 108

18,468 522 9,152 100 1,028

(62.79%) (1.77%) (31.11%) (0.34%) (3.49%)

1,031 78 657 25 100

902 5,507 8,126 1,271

(5.69%) (34.72%) (51.23%) (8.02%)

116 296 401 106

5,890 (37.14%) 9,902 (62.44%)

328 434

9,570 696 4,483 268 787

461 95 260 39 80

(60.34%) (4.39%) (28.27%) (1.69%) (4.96%)

RSA, reverse shoulder arthroplasty; SHA, shoulder hemiarthroplasty; TSA, total shoulder arthroplasty.

approximately one half of a day longer than the mean length of stay for a TSA (P < .001) and almost one day shorter than the mean length of stay for an SHA (P < .001) (Table II). Hospital claim payments were $12,625, $11,989, and $12,958 for RSA, TSA, and SHA, respectively (P < .001 between groups). Patients with a diagnosis of fracture/dislocation were more likely to be female patients, were older, had a longer stay, and had increased claim payments, regardless of treatment type (Table II). Comparative data from the 2011 HCUP NIS database showed that, overall, approximately 67,190 shoulder arthroplasties were performed in 2011 in the United States. The percentage of all arthroplasties in this database that were RSAs was 33%; TSAs, 44%; and SHAs, 24%. This database showed that 74% of all RSAs, 60% of TSAs, and 55% of SHAs were performed in patients aged older than 65 years (http://hcupnet.ahrq.gov/).8 In total, approximately 52% of all RSAs, 45% of TSAs, and 40% of SHAs were captured by our analysis of Medicare Part A claims compared with the HCUP NIS data (Table III). However, the NIS database does not stratify patients who are both older than 65 years and using Medicare Part A as opposed to Part C. In 2011, approximately 25% of all Medicare recipients were enrolled in Part C.9 This is likely the cause of the differences between the Medicare Part A–reported volumes compared with the volumes listed in the NIS database for Medicare patients. The HCUP NIS volumes for Medicare patients include age younger than 65 years and Part C participants. These patients were excluded from the Medicare Part A analysis. Regardless, the breakdown of each type of arthroplasty remained similar within these groups. OA was the predominant primary diagnosis code in the Medicare database, with 91% of all TSAs, 37% of all

SHAs, and 45% of all RSAs coded for this diagnosis primarily. Rotator cuff tear was a primary diagnosis for 21% of all RSAs. Fracture/dislocation was a primary diagnosis for 45% of all SHAs (Table IV). When we considered all diagnosis codes (including secondary diagnosis codes), OA was included for 93% of all TSAs, 46% of all SHAs, and 59% of all RSAs. For all RSAs, a diagnosis code for a rotator cuff tear was included in 53%, a code for a fracture in 13%, and a code for late fracture sequelae in 4% (Table IV). A primary diagnosis of OA with no secondary code for rotator cuff tear was found in 22% of patients undergoing RSA. The majority of all shoulder arthroplasties (97% of SHAs, 65% of TSAs, and 57% of RSAs) were performed by very low–volume or low-volume operating surgeons who performed fewer than 10 shoulder arthroplasties on Medicare Part A patients in 2011. No surgeon performed more than 20 SHAs, but 80 high-volume surgeons performed 22% of RSAs and 68 high-volume surgeons performed 16% of TSAs (Fig. 1). Twenty-five percent of providers performed more RSAs than TSAs plus SHAs, and these providers performed 70% of RSAs. Approximately 30% of RSAs were performed by providers who performed at least 3 times as many RSAs as TSAs plus SHAs (Fig. 1).

Discussion As hypothesized, this evaluation of the 2011 Medicare database has shown that RSA is performed at a nearly equal rate to TSA (37% vs 42%) and about twice as often as SHA (37% vs 21%) in Medicare Part A patients. The most common diagnosis for RSA is OA, followed by rotator cuff tear. Surprisingly, about one quarter of RSAs were coded

Reverse shoulder in the Medicare population Table IV

5

Analysis of diagnosis codes RSA

Osteoarthritis Rheumatoid and inflammatory arthritis Arthropathy/traumatic arthropathydnon-RA RTC tear Fracture/dislocation Fracture sequelae Aseptic necrosis Mechanical failure Other

TSA

SHA

Primary diagnosis

Any diagnosis

Primary diagnosis

Any diagnosis

Primary diagnosis

Any diagnosis

45.3% 0.7%

58.8% 7.2%

91.2% 0.7%

93.1% 5.3%

37.2% 0.6%

46.0% 5.7%

14.7%

20.2%

3.5%

4.1%

3.2%

4.5%

20.7% 11.4% 2.3% 0.6% 1.8% 2.5%

52.8% 12.9% 3.9% 1.8% 2.2%

1.0% 1.4% 0.3% 1.0% 0.2% 0.8%

11.4% 2.1% 0.7% 1.9% 0.3%

2.9% 44.5% 2.9% 2.1% 1.3% 5.5%

15.6% 48.0% 4.1% 3.6% 1.9%

RA, rheumatoid arthritis; RSA, reverse shoulder arthroplasty; RTC, rotator cuff tear; SHA, shoulder hemiarthroplasty; TSA, total shoulder arthroplasty. Discharge data were analyzed to determine the proportion of total cases for each procedure type that contained codes in each category as either the primary diagnosis code or in any diagnosis field.

for a diagnosis of OA without a diagnosis of rotator cuff tear. Our findings were consistent with those reported in the Australian National Joint Replacement Registry, which found primary diagnoses of OA, rotator cuff arthropathy, and fracture/dislocation in 51%, 31%, and 14% of reverse total shoulder procedures, respectively.4 We have previously reported a steady increase in both procedure volumes and rates of shoulder arthroplasty.3 Our current study would suggest that the introduction of the RSA in the United States is at least partially responsible for this increase, as was previously postulated by Kim et al.14 The NIS database reported that approximately 67,000 shoulder arthroplasties were performed in 2011; this is a 43% increase from just 3 years prior, with approximately 47,000 reported using the NIS database for 2008. This is an average increase of about 13% per year, which is at the upper limit of previous projections for shoulder arthroplasty growth.3 If this rate of increase were to continue, the number of shoulder arthroplasties would double from the 2011 numbers by 2017 and could climb to over 200,000 by 2020. Interestingly, the number of SHAs performed in 2011 was only 15,860. This was a 21% reduction compared with the 20,178 SHAs performed in 2008.14 It is quite likely that many patients who would have previously undergone an SHA are now undergoing an RSA. We found that 22% of RSAs during the period studied were implanted with a diagnosis code for OA but no code for rotator cuff tear. This is a concerning finding and could represent overuse of RSA in patients with OA without glenoid deformity because the use of RSA for OA without bony deformity has not been reported. Mizuno et al20 recently published promising outcomes with the use of RSA for OA without any rotator cuff deficiency with biconcave, posteriorly worn glenoids in patients aged between 68 and 82 years; this was presumably not a common indication during the study period because this study was published after 2011. In addition, 51% of all shoulder

Fig 1 Surgical volumes for individual surgeons. RSA, reverse shoulder arthroplasty; SHA, shoulder hemiarthroplasty; TSA, total shoulder arthroplasty.

arthroplasties are performed by lower-volume surgeons (<10 per year), who presumably would see fewer patients with severe bone deformity. This trend deserves further scrutiny. In this study, we report that use of RSA has grown to nearly the same procedure volume as TSA in the Medicare Part A population aged older than 65 years. The claims payments differed by approximately $600 between RSA and TSA ($12,625 for RSA and $11,989 for TSA). This difference may be explained by the greater number of nonelective procedures and the longer length of stay for RSAs. Although the volume of shoulder arthroplasties has increased, there has been little effect on the per-procedure cost to the Medicare system in 2011. However, the CMS recently reported in their fiscal year 2014 report that the average cost of RSA is about $2,000 more than the average for other upper extremity replacements in the same Medicare Severity-Diagnosis Related Group (MS-DRG). MSDRGs are used to determine hospital reimbursements and group patients and procedures that are thought to use

6 similar hospital resources. Regardless of this increased average cost, the CMS rejected requests to reclassify the MS-DRG for RSA to accommodate the increased costs of the procedure.19 Considering that the reported average sales price for reverse total shoulder implants is approximately $10,000 and the price for anatomic shoulder implants ranges from $4,000 to $9,000,22 this lack of differentiation in claims payments may result in financial strain for treating institutions in the future. This is something that needs to be considered going forward because RSA represents a substantial percentage of the shoulder arthroplasties being performed. The relationship between surgeon volume and relative use of SHA, TSA, and RSA shows a trend that warrants further discussion. First, the percentage of arthroplasties performed by very low–volume or low-volume surgeons is 57% for RSA, 65% for TSA, and 97% for SHA. In addition, 70% of RSAs are performed by surgeons who perform more RSAs than all anatomic implants combined, and 30% of RSAs are performed by providers who perform over 3 times as many RSAs as all anatomic implants combined. The explanation for this trend is unknown. The impact that this trend has on use, cost, and revision burden is likely to be substantial to hospitals and the health care system for years to come. With enrollment of a large proportion of adults aged 65 years or older, the Medicare Part A claims database provides a rich source of data. The use of medical claims data presents a cost-effective means to analyze health care data for large populations, which would be impossible to achieve by chart abstraction. However, there are limitations to the use of claims data. Whereas Medicare Part A data capture a large proportion of the total procedures performed, our estimates of provider volume are underestimated because they do not account for procedures performed in younger patients or through health management organizations as a part of Medicare Advantage (ie, Medicare Part C). Our analysis required the assumption that providers’ overall volume and treatment patterns were correlated to the volume and treatment patterns observed in the Medicare population. Payment data reflect the Medicare payments to the institution but do not include physician reimbursement or payments from private sources or other insurers and may therefore underestimate the total cost per procedure. Furthermore, the diagnosis data included on claims forms are primarily for the purpose of gaining reimbursement and were not designed for the purpose of research.5 It is certainly possible that coding errors exists that could affect the data, although we do not have any evidence that this is the case. Despite these limitations, our use of Medicare Part A data has allowed us to analyze more than 30,000 shoulder arthroplasty procedures, or approximately half of all procedures performed in the United States during the period studied. At a minimum, our data indicate that Kim et al14 were likely correct in postulating that the vast

J.S. Day et al. increase in use of shoulder arthroplasty was the result of the introduction of RSA. This is also consistent with the findings of the Australian National Joint Replacement Registry, which reported that 34% of shoulder replacements during the same period were reverse shoulder procedures.4 Although ideal percentages of the differing types of shoulder arthroplasty are unknown, the large percentage of this relatively new technology should be noted4 and is important information. Boguski et al1 showed wide variation in the percentage of RSAs across hospitals and suggested that this variation signals uncertainty about the best application of each device. Without strict indications for RSA and accurate epidemiologic statistics on specific shoulder pathology, it is impossible to truly know whether RSA is being used accurately, but this study is the first to report utilization patterns for this new technology to begin the discussion.

Conclusion We investigated the use of shoulder arthroplasty as it relates to RSA in the Medicare population. Surprisingly, we found that RSA is performed with similar frequency to anatomic TSA and almost twice as much as SHA in the Medicare Part A population. Almost a quarter of RSAs are implanted for OA without a diagnosis of rotator cuff tear, which may represent overuse of this technology in this patient population. Lower-volume surgeons perform the majority of reverse total shoulder procedures. In addition, a substantial number of surgeons perform more reverse procedures than anatomic shoulder arthroplasties and hemiarthroplasties combined. As more data become available, future studies will allow us to further address changing arthroplasty use patterns, indications, outcomes, and hospital payments and to gain a clearer understanding of the effect of RSA in the United States.

Disclaimer The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.

References 1. Boguski RM, Miller BS, Carpenter JE, Mendenhall S, Hughes RE. Variation in use of reverse total shoulder arthroplasty across hospitals.

Reverse shoulder in the Medicare population

2.

3.

4.

5. 6.

7.

8. 9.

10.

11.

12.

13.

J Shoulder Elbow Surg 2013;22:1633-8. http://dx.doi.org/10.1016/j. jse.2013.09.002 Bozic KJ, Kurtz SM, Lau E, Ong K, Vail TP, Berry DJ. The epidemiology of revision total hip arthroplasty in the United States. J Bone Joint Surg Am 2009;91:128-33. http://dx.doi.org/10.2106/JBJS.H.00155 Day JS, Lau E, Ong KL, Williams GR, Ramsey ML, Kurtz SM. Prevalence and projections of total shoulder and elbow arthroplasty in the United States to 2015. J Shoulder Elbow Surg 2010;19:1115-20. http://dx.doi.org/10.1016/j.jse.2010.02.009 National Joint Replacement Registry. Demographics and outcomes of shoulder arthroplasty. In: National Joint Replacement Registry. Australian Orthopedic Association; 2013. p. 37. Ferver K, Burton B, Jesilow P. The use of claims data in healthcare research. Open Public Health J 2009;2:11-24. Guery J, Favard L, Sirveaux F, Oudet D, Mole D, Walch G. Reverse total shoulder arthroplasty. Survivorship analysis of eighty replacements followed for five to ten years. J Bone Joint Surg Am 2006; 88:1742-7. http://dx.doi.org/10.2106/JBJS.E.00851 Hasan SS, Leith JM, Smith KL, Matsen FA III. The distribution of shoulder replacement among surgeons and hospitals is significantly different than that of hip or knee replacement. J Shoulder Elbow Surg 2003;12:164-9. http://dx.doi.org/10.1067/mse.2003.23 HCUPnet Web site. Available from: http://hcupnet.ahrq.gov/. Accessed July 18, 2013. Hasan SS, Leith JM, Smith KL, Matsen FA III. http://www.cms.gov/ Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Rep ortsTrustFunds/downloads/tr2012.pdf. Jain N, Pietrobon R, Hocker S, Guller U, Shankar A, Higgins LD. The relationship between surgeon and hospital volume and outcomes for shoulder arthroplasty. J Bone Joint Surg Am 2004;86:496-505. Jain NB, Guller U, Pietrobon R, Bond TK, Higgins LD. Comorbidities increase complication rates in patients having arthroplasty. Clin Orthop Relat Res 2005:232-8. Jain NB, Higgins LD, Guller U, Pietrobon R, Katz JN. Trends in the epidemiology of total shoulder arthroplasty in the United States from 1990-2000. Arthritis Rheum 2006;55:591-7. http://dx.doi.org/10.1002/ art.22102 Jain NB, Kuye I, Higgins LD, Wanner JP. Surgeon volume is associated with cost and variation in surgical treatment of proximal humeral fractures. Clin Orthop Relat Res 2013;471:655-64. http://dx.doi.org/ 10.1007/s11999-012-2481-6

7 14. Kim SH, Wise BL, Zhang Y, Szabo RM. Increasing incidence of shoulder arthroplasty in the United States. J Bone Joint Surg Am 2011; 93:2249-54. http://dx.doi.org/10.2106/JBJS.J.01994 15. Kurtz S, Mowat F, Ong K, Chan N, Lau E, Halpern M. Prevalence of primary and revision total hip and knee arthroplasty in the United States from 1990 through 2002. J Bone Joint Surg Am 2005;87:148797. http://dx.doi.org/10.2106/JBJS.D.02441 16. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007;89:780-5. http://dx.doi.org/10. 2106/JBJS.F.00222 17. Kurtz SM, Lau E, Schmier J, Ong KL, Zhao K, Parvizi J. Infection burden for hip and knee arthroplasty in the United States. J Arthroplasty 2008;23:984-91. http://dx.doi.org/10.1016/j.arth.2007.10.017 18. Kurtz SM, Ong KL, Schmier J, Mowat F, Saleh K, Dybvik E, et al. Future clinical and economic impact of revision total hip and knee arthroplasty. J Bone Joint Surg Am 2007;89(Suppl 3):144-51. http:// dx.doi.org/10.2106/JBJS.G.00587 19. Medicare Program; Hospital Inpatient Prospective Payment Systems for Acute Care Hospitals and the LongTerm Care Hospital Prospective Payment System and Fiscal Year 2014 Rates; Quality Reporting Requirements for Specific Providers; Hospital Conditions of Participation; Payment Policies Related to Patient Status. Washington, DC: Department of Health and Human Services; 2013. p. 50534-6. 20. Mizuno N, Denard PJ, Raiss P, Walch G. Reverse total shoulder arthroplasty for primary glenohumeral osteoarthritis in patients with a biconcave glenoid. J Bone Joint Surg Am 2013;95:1297-304. http://dx. doi.org/10.2106/jbjs.l.00820 21. Ong KL, Manley MT, Kurtz SM. Have contemporary hip resurfacing designs reached maturity? A review. J Bone Joint Surg Am 2008; 90(Suppl 3):81-8. http://dx.doi.org/10.2106/JBJS.H.00574 22. A 2011 extremity update. Orthopedic Network News 2011;22:9-12. 23. Smith CD, Guyver P, Bunker TD. Indications for reverse shoulder replacement: a systematic review. J Bone Joint Surg Br 2012;94:57783. http://dx.doi.org/10.1302/0301-620X.94B5.27596 24. Werner CM, Steinmann PA, Gilbart M, Gerber C. Treatment of painful pseudoparesis due to irreparable rotator cuff dysfunction with the Delta III reverse-ball-and-socket total shoulder prosthesis. J Bone Joint Surg Am 2005;87:1476-86. http://dx.doi.org/10.2106/JBJS.D. 02342