Questions about whether and. Medicare Competitive Bidding Program Realized Price Savings For Durable Medical Equipment Purchases

Transcription

1 By David Newman, Eric Barrette, and Katharine McGraves-Lloyd Medicare Competitive Bidding Program Realized Price Savings For Durable Medical Equipment Purchases doi: /hlthaff HEALTH AFFAIRS 36, NO. 8 (2017): Project HOPE The People-to-People Health Foundation, Inc. ABSTRACT From the inception of the Medicare program there have been questions regarding whether and how to pay for durable medical equipment, prosthetics, orthotics, and supplies. In 2011 the Centers for Medicare and Medicaid Services (CMS) implemented a competitive bidding program to reduce spending on durable medical equipment and similar items. Previously, CMS had used prices in an administrative fee schedule to reimburse for these items. We compared prices from Round 1 of the Medicare competitive bidding program, which were established for the periods and , to prices paid by national commercial insurers for the same types of items in Our results suggest that the initial years of the program produced prices comparable to those obtained, on average, by large commercial insurers sophisticated purchasers that presumably were able to negotiate prices with suppliers of durable medical equipment and similar items. David Newman is a senior advisor and formerly the executive director at the Health Care Cost Institute, in Washington, D.C. Eric Barrette (ebarrette@ healthcostinstitute.org) is the director of research at the Health Care Cost Institute. Katharine McGraves-Lloyd is a senior business information analyst at Anthem, in South Portland,Maine,andwasa researcheratthehealthcare Cost Institute while contributing to this research. Questions about whether and how to pay for durable medical equipment, prosthetics, orthotics, and supplies have existed since the inception of the Medicare program in Historically, the Centers for Medicare and Medicaid Services (CMS) paid for durable medical equipment and similar items using a statutory-based fee schedule. 1 Studies repeatedly suggested that Medicare could achieve significant savings by implementing an alternative payment scheme, such as a competitive bidding program. 2 Medicare s cost for durable medical equipment and similar items is not trivial: In 2010, the year before a competitive bidding program for these items was implemented, Medicare spent over $11.3 billion on the items. 3 The competitive bidding program was instituted in 2011 in nine Metropolitan Statistical Areas, and it is to become nationwide in CMS has reported that the program resulted in 35 percent savings relative to the use of the fee schedule in Round 1. Moreover, CMS anticipates that Medicare will save $25.7 billion in the period , while beneficiaries will collectively save an additional $17.1 billion. 4 While the savings are impressive, they reflect savings calculated from prices in the Medicare fee schedule, which were known to be high. It is also important to know how prices in the competitive bidding program compare to those paid by commercial insurers informed, sophisticated purchasers with the ability to negotiate prices with providers, presumably including suppliers of durable medical equipment and similar items. If competitive bidding program prices were systematically higher than the average prices that commercial insurers paid, this would suggest that program refinements could achieve additional savings or efficiencies in prices. We examined the prices paid by commercial insurers and Medicare in the initial round of the competitive bidding program, in In this article we first highlight the differences between the program prices and those in the previous year s administrative fee schedule, and we then August :8 Health Affairs 1367

2 compare commercial insurers prices and competitive bidding program prices. We discuss the results in the context of theoretical economic predictions about the program s results and conclude by discussing policy implications. Medicare Durable Medical Equipment Prices And Competitive Bidding Competitive bidding for durable medical equipment and similar items in Medicare has a long history. At first, rental of durable medical equipment and similar items was a covered benefit for Medicare enrollees, and by 1968 the law was amended to provide for rental or purchase. However, in 1972 the General Accounting Office recommended that Congress authorize the Department of Health, Education, and Welfare, through further amendments, to find more economical payment methods. 5 In the mid-1980s the Health Care Financing Administration (now CMS) began planning a demonstration project for competitive bidding for the supply of durable medical equipment and similar items on a rental or purchase basis or both. Given industry objections, however, Congress froze funding for the project. Competitive bidding for these items was resurrected in the Balanced Budget Act of CMS conducted a demonstration project in the Polk County, Florida, and San Antonio, Texas, Metropolitan Statistical Areas in the period It concluded that competitive bidding for the items resulted in substantial savings to the Medicare program without reducing beneficiaries access to or the quality of the items. 6 The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 directed the Department of Health and Human Services to establish and implement programs for competitive bidding in selected areas across the United States for certain durable medical equipment and similar items and authorized CMS to phase in competitive bidding in additional geographic areas over time. CMS launched the competitive bidding program in mid-2008 in ten Metropolitan Statistical Areas. However, the first competitive bidding program operated for only two weeks in July 2008, ending when Congress suspended it and directed CMS to have suppliers rebid prices after a bidding process redesign. 7 Rebidding, called Round 1 Rebid, was open to suppliers for 60 days starting in October 2009 in nine Metropolitan Statistical Areas, and prices for products in each of the areas were established for a three-year period, January 1, 2011 December 31, A sixty-day bidding period in the same areas, called Round 1 Recompete, began in October 2012, with the final prices in effect for the period January 1, 2014 December 31, A timeline of bidding and contract periods is included in online Appendix Exhibit A. 10 The Competitive Bidding Program The competitive bidding program established nine product categories of similar and related durable medical equipment and other products. Bidding was conducted separately for each Metropolitan Statistical Area. Suppliers who met CMS s eligibility requirements submitted bids, offering to supply products for the bid price in the given competitive bidding program area. Bona fide bids were required for each product in a category. 11 Eligible suppliers were also given information about the product markets and were required, as part of their bid, to estimate how much of the market demand they could satisfy. The bids for each product in a category were aggregated by CMS to generate a composite bid for the category. To establish the competitive bidding program price for each product in the category what CMS refers to as the single payment amount the composite bids for the categories were arrayed from low to high. Starting with the lowest bidder and then moving to each successively higher bidder, suppliers were offered contracts until their combined capacity was adequate to satisfy the market demand for the category. The price for each product was set at the median bid of the contracted suppliers. CMS implemented numerous protections for the bidding process, suppliers, and patients, such as intentionally including more suppliers than were needed to satisfy demand to prevent a reduction in access to durable medical equipment and similar items. The competitive bidding program s regulations have been described in more detail elsewhere. 1,12,13 Study Data And Methods Data The Health Care Cost Institute (HCCI) database was the source of our data for commercial prices of durable medical equipment and similar items. It consists of health insurance claims and enrollment data from three national commercial insurers, which collectively cover over one-quarter of people in the United States with employer-sponsored insurance. The HCCI claims data include procedure code, procedure code modifier, and unit (that is, quantity) information. These fields make it possible to identify claims for durable medical equipment and similar items, the type of claim (rental or purchase), and the amount of the product provided. The data also include a field for the allowed amount, 1368 Health Affairs August :8

3 Changes in program prices provide insight into whether suppliers bidding evolvedwithina competitive bidding area. or the actual amount paid for the claim. This amount comprises the insurer s payment and any out-of-pocket spending (copayments, coinsurance, or deductibles) by the patient. The enrollment file includes monthly enrollment records for each person, as well demographic information such as age, sex, and ZIP code of residence. The HCCI data were cleaned to account for payment adjustments, reversals, and apparently implausible claims or errors. The resulting data for our analysis included only claims with positive dollar amounts and positive units in other words, claims with negative or zero dollars or units were excluded from the analysis. The procedure code field in the data contains Healthcare Common Procedure Coding System (HCPCS) codes, which are used to identify durable medical equipment and similar items. These numeric and alphanumeric codes are used for billing for physician and nonphysician services, supplies, and equipment. Modifiers indicate whether the items are rented or purchased. Claims for the seven durable medical equipment and similar items included in our analysis were identified in the HCCI data by their combinations of codes and modifiers. Medicare fee schedule and competitive bidding program prices for the same items were obtained from reference files published by CMS. 13,14 CMS also publishes the ZIP codes in each competitive bidding program area. 13 These CMS files were matched to the HCCI enrollment file to identify eligible commercially insured members for our study. The study population was limited to members ages enrolled at any time in the period The seven items included in our analysis had sample sizes of at least thirty claims in every competitive bidding program area, which ensured robust estimates of average prices. Analysis In 2013 CMS compared competitive bidding program prices to prices in the fee schedule, but an arguably more rigorous test of how well the program performed is a comparison of program prices and those paid by commercial insurers. 4 Our analysis focused on prices for seven items (defined by unique code-modifier combinations) in the initial nine competitive bidding program areas. For each of the seven items, we computed the average Round 1 Rebid and Round 1 Recompete prices across the areas. A similar approach was used by the Government Accountability Office in comparing competitive bidding program prices to those in the fee schedule. 15 With the commercial price data, we first calculated an annual average price in each competitive bidding program area for each of the seven items. We then averaged these prices to provide a comparator for the average program prices. An alternative calculation, using averages of the median of commercial prices within program areas, produced substantially similar results. The average prices and results from the alternative calculation are in Appendix Exhibit B. 10 We analyzed seven prices: the rental prices of six distinct products as well as the new purchase price for one of the six products. The analysis included durable medical equipment and similar products from two categories. From the continuous positive airway pressure (CPAP) devices, respiratory assist devices, and related supplies and accessories category, we analyzed rental prices of three products and the rental and new purchase prices of one product. We also analyzed the rental prices of two products in the oxygen supplies and equipment category. Of the six durable medical equipment and similar products in our analysis, five were on CMS s list of the top 80 percent of cost and utilization for Round The sixth product, E0471 (a respiratory assist device with bilevel pressure capability and a backup rate feature), was included in our analysis because it provides an interesting comparator for E0470 (a respiratory assist device with bilevel pressure capability but with no backup rate feature). Although these two products are similar, their average Round 1 Rebid rental prices across the nine competitive bidding program areas differed substantially. The average competitive bidding program and commercial prices are provided in Appendix Exhibit B. 10 The level of utilization of the seven durable medical equipment and similar items we analyzed was lower in the population younger than age sixty-five with employer-sponsored insurance, compared to the level in the Medicare population. However, the seven items accounted for 28 percent both of the durable medical equipment and similar items claims in the 2013 HCCI August :8 Health Affairs 1369

4 data and of all Medicare claims for these items in Limitations This analysis is the first evaluation, to our knowledge, of the competitive bidding process using non-medicare prices, and it contributes to understanding the program s performance. However, there are notable limitations to the analysis. First, the analysis and findings are specific to the design of the program bidding process and the rules and regulations that governed the program during the study period. Specifically, the results from the analysis might not reflect the relationship between competitive bidding program prices and commercial prices for durable medical equipment and similar items not included in the analysis. Second, the findings are not generalizable to later years of Round 1 or any subsequent rounds because the design of the program has changed over time. In addition, the results are not necessarily generalizable to other competitive bidding environments because of differences in program design, market structure, and dynamics. Third, the analysis was restricted to seven prices for products from two categories (the category of respiratory assist devices and related supplies and accessories and the category of oxygen), for which there were sufficient claims in all competitive bidding program areas. Fourth, although 28 percent of the claims for durable medical equipment and similar items for both the commercially insured and Medicare populations were for the seven items in our study, the two populations differences in ages and comorbidities suggest that they probably also differed in what accounted for the other 72 percent of their claims. Focusing on the most common durable medical equipment and similar items in the commercial data is an appropriate comparison because these items are most likely to be subject to insurers attention and negotiations. Additionally, in a single study it is not feasible to analyze the prices for all durable medical equipment and similar items in all categories and all competitive bidding program areas. Fifth, the generalizability of the results is also limited by the data. The HCCI data are a convenience sample. Other payers prices for durable medical equipment and similar items, as well as their members use of the items, may differ from what we observed in the HCCI data. However, the HCCI database is one of the largest and most comprehensive sets of commercial claims data to include the actual amounts paid to providers. Approximately thirty million commercially insured people ages are represented in the HCCI data a population comparable in size to the approximately thirty-eight million fee-forservice Medicare enrollees in Additionally, Although some thirty years have passed, competitive bidding does not yet seem to be a settled matter. in six of the nine Round 1 competitive bidding program areas, one or both of the two largest commercial insurers in the market were HCCI data contributors. 18 Study Results We first compared the 2010 Medicare fee schedule prices to the competitive bidding program prices established for the period in Round 1 Rebid. The comparison is comparable to the calculation CMS used to project the competitive bidding program s savings, although it is limited to the seven prices included in our analysis.we also compared the fee schedule and program prices to the average prices that commercial insurers paid in 2010 as an alternative measure of program savings. Medicare Fee Schedule, Commercial, And Round 1 Rebid Prices On average, the Round 1 Rebid prices for the seven items were 34.7 percent lower than the prices in the 2010 fee schedule (Exhibit 1). This result is consistent with the 35 percent savings reported by CMS from the use of Round 1 Rebid prices, compared to the 2010 fee schedule prices. Our comparison of the Medicare fee schedule prices and the average prices that commercial insurers paid in 2010 also supports the conclusion that CMS overpaid for durable medical equipment and similar items. On average, commercial insurers paid 28.7 percent less than Medicare for the same set of durable medical equipment and similar items in The Round 1 Rebid prices for each item were slightly lower than the average 2010 commercial prices (Exhibit 1). This suggests that in the nine program Metropolitan Statistical Areas, the program resulted in prices that were generally comparable to but lower than the prices obtained by large commercial insurers. Average Commercial And Competitive Bidding Program Prices The difference between competitive bidding program prices and commercial prices could vary over time because commercial prices can change each year, while the 1370 Health Affairs August :8

5 Exhibit 1 Differences between average Medicare fee schedule, commercial, and Round 1 Rebid prices HCPCS product 2010 Percent difference between: Code and modifier E0470-rental E0471-rental E0562-new E0562-rental E0601-rental E0431-rental (oxygen) E1390-rental (oxygen) Description RAD, bilevel pressure capability, without backup rate feature, used with noninvasive interface RAD, bilevel pressure capability, with backup rate feature Humidifier, heated, used with positive airway device Humidifier, heated, used with positive airway device Avg. Medicare fee schedule price ($) Avg. commercial price ($) Round 1 Rebid price and 2010 fee schedule Two 2010 prices CPAP device Portable gaseous oxygen system, rental; includes portable container, regulator, flowmeter, humidifier, cannula or mask, and tubing Oxygen concentrator, single delivery port, capable of delivering 85 percent or greater oxygen concentration at the prescribed flow rate All combined a a a Round 1 Rebid price and 2010 commercial price SOURCE Authors analysis of data from the Health Care Cost Institute. NOTES Average commercial prices were calculated with the mean commercial price or the Medicare fee schedule price from each competitive bidding program (CBP) area. Negative values indicate a lower CBP price relative to the Medicare fee schedule price or commercial price (the average price paid by commercial insurers). HCPCS is Healthcare Common Procedure Coding System. CPAP is continuous positive airway pressure. RAD is respiratory assist device. a Not applicable. program prices are fixed for three years. We therefore compared the average commercial prices to the program prices for the period the years for which commercial data were available. In each year the majority of the seven items average competitive bidding program prices (five in 2011, four in 2012 and 2013, and six in 2014) were less than the average commercial insurer prices (Exhibit 2). The two respiratory assist products, E0470 and E0471, that have similar uses but differ substantially in program prices also differ in commercial prices (see Appendix Exhibit B). 10 For the less expensive item, E0470, the program price was lower than the average commercial price in three of the four years. However, for the more expensive item, E0471, the average commercial price was higher than the program price in all four years, and the difference between the prices grew each year. Because the program prices are set for three years, this implies that the average commercial price for E0471 increased over time. Average Prices In Round 1 Rebid And Round 1 Recompete Without access to the actual bids, it is impossible to know how suppliers bidding changed over time. However, it is possible to gain insight into the initial performance of the competitive bidding program by comparing Round 1 Rebid and Round 1 Recompete prices. Changes in suppliers costs or changes in the market conditions, such as supplier competition, could have affected suppliers bids as well as their contracted prices with commercial insurers. Changes in the program prices provide insight into whether suppliers bidding evolved within a competitive bidding area. Decreases in the program prices suggest that overall suppliers costs have decreased, so they are able to bid lower prices. However, commercial insurers are able to capitalize on lower supplier costs only if suppliers do not have sufficient market leverage to command higher prices. Therefore, we can assess the performance of the program at a high level by whether the decrease in program prices is greater than or equal to the decrease in commercial prices. Our results provide preliminary evidence that there may have been changes in the market for August :8 Health Affairs 1371

6 Exhibit 2 Differences between average commercial prices and average competitive bidding program prices, Difference between average commercial price and: Average Round 1 HCPCS product, Average commercial price Average Round 1 Rebid price Recompete price code and modifier E0470-rental $ $ $ $ % 0.5% 3.3% 2.9% E0471-rental E0562-new E0562-rental E0601-rental E0431-rental (oxygen) E1390-rental (oxygen) SOURCE Authors analysis of data from the Health Care Cost Institute. NOTES Product descriptions appear in Exhibit 1. Average commercial prices were calculated with the mean commercial price or the Medicare fee schedule price from each competitive bidding program (CBP) area. Negative values indicate a lower CBP price relative to the commercial price (the average price paid by commercial insurers). Round 1 Recompete prices were in effect in the period January 2014 December HCPCS is Healthcare Common Procedure Coding System. CPAP is continuous positive airway pressure. RAD is respiratory assist device. Exhibit 3 overall durable medical equipment and similar items as a result of the competitive bidding program. All program prices in our analysis decreased from the Round 1 Rebid to the Round 1 Recompete periods (Exhibit 3). Three of the average commercial prices increased, and four decreased. Two of those decreases were less than the decrease observed in program prices. This suggests, but does not prove, that the competitive bidding program led high-cost suppliers to Differences between average competitive bidding program prices and average commercial prices between Round 1 Rebid and Round 1 Recompete HCPCS product, code and modifier Difference between: Round 1 Rebid and Round 1 Recompete average prices E0470-rental 5.9% 6.4% E0471-rental E0562-new E0562-rental E0601-rental E0431-rental (oxygen) E1390-rental (oxygen) and 2011 average commercial prices SOURCE Authors analysis of data from the Health Care Cost Institute. NOTES Product descriptions appear in Exhibit 1. Average commercial prices were calculated with the mean commercial price or the Medicare fee schedule price from each competitive bidding program (CBP) area. Negative values indicate a lower price in the more recent period than the in the earlier period. HCPCS is Healthcare Common Procedure Coding System. CPAP is continuous positive airway pressure. RAD is respiratory assist device. exit the market, there were reductions in prices by all suppliers, or both. If suppliers exited the market, the remaining suppliers might have had more leverage to command higher prices from commercial insurers. Alternatively, if no suppliers exited the market, the program price might have revealed information to suppliers that they were able to use to their advantage in negotiating prices with insurers. Discussion In Medicare, the transition from using a fee schedule for durable medical equipment and similar items to setting prices through competitive bidding stretched over decades and has been controversial at every point. After the competitive bidding program was launched, Medicare claimed to have achieved significant savings compared to prices in the fee schedule. 4 We have offered an alternative evaluation of the program s initial performance by comparing Round 1 prices to commercial insurers average prices. Because national commercial insurers are large, informed, sophisticated purchasers, with the ability and leverage to negotiate with suppliers, the insurers prices can be used as a benchmark for market prices. Our results suggest that through a competitive bidding process, CMS was able to obtain prices comparable to those obtained by other sophisticated purchasers that were able to negotiate with suppliers. We found that the initial rounds of the competitive bidding program achieved prices for durable medical equipment and similar items that were similar to average prices paid by commercial insurers. Given the market presence of the insurers in 1372 Health Affairs August :8

7 The competitive bidding program has implications for the entire market for durable medical equipment. the HCCI data at the levels of the nation and the Metropolitan Statistical Area, we assumed that the HCCI data provided a reliable proxy for the average price in the commercial market. A conservative interpretation of our results is to assume that the average price in the HCCI data is higher than other payers prices. In many cases, however, the competitive bidding program price was substantially lower than the average price in the HCCI data (Exhibit 2). This implies that even if other commercial insurers paid 5 10 percent less than those in the HCCI data, many of the program prices were still comparable to commercial insurers prices. An analysis of supplier market power was beyond the scope of this study, but it should be considered in future work. For example, it is possible that a monopolist supplier could demand a higher price from insurers, compared to the price of a supplier in a market where many suppliers compete. Because there are thousands of suppliers and the national insurers contract with numerous national and local suppliers, it is improbable that there are monopolist prices for durable medical equipment and similar items. However, variation in supplier market power across competitive bidding program areas is likely, and the impact of supplier market power on commercial prices and program bidding warrants future research. There are multiple reasons to believe that CMS should pay an amount close to the price that large commercial insurers pay. First, CMS is likely to be the largest purchaser of durable medical equipment and similar items. Second, CMS has a reputation as a prompt payer (clean claims for durable medical equipment and similar items have a thirty-day payment ceiling before interest accrues). Third, CMS offeres successful bidders a three-year contract, which is likely longer than what other payers offered. Thus, contracting with CMS results in increased volume and reduced transaction costs through prompt payment and longer-term contracts. These factors should incentivize suppliers to bid accurately low enough to win a contract but high enough to cover their costs. It is also possible that the CMS competitive bidding program would have produced prices that differed from commercial insurers prices, because the auction process was untested. In the most extreme example, if suppliers in an area chose to bid prices at or near the published fee schedule, prices could have been higher than those paid by commercial insurers. In a more likely example, the competitive bidding program established a minimum number of suppliers and a target of 30 percent small-supplier participation in the program. To the extent that this encouraged smaller, higher-price suppliers to participate, program prices would be higher than in the absence of these program requirements. Although our study was not exhaustive, our results provide no evidence that competitive bidding program prices were systematically higher than those that commercial insurers paid, on average. This lack of evidence suggests that the program might not need to be refined for additional savings to be achieved. However, our results and additional empirical studies need to be aligned with the theoretical predictions of competitive bidding program performance to determine what, if any, modifications to the program will ensure its long-term success. Program modifications may also be needed to achieve other goals, such as improving access to durable medical equipment that patients need. The competitive bidding program includes numerous regulations, which researchers and policy makers suggest make it unsustainable. 19 Researchers have formally modeled and tested in controlled experiments the underlying economic theory of the program auction and suggested that the current design will not produce competitive prices or satisfy demand. 20,21 There have been recurring concerns that the gains from a competitive bidding program would disappear because unsuccessful bidders would leave the market and fewer suppliers would remain. 22 The predictions of unsustainably low Medicare payments imply that there will eventually be negative impacts on Medicare beneficiaries as a result of limited access. Policy Implications Although some thirty years have passed since competitive bidding for prices for durable medical equipment and similar items was first considered, and a program has finally been implemented, competitive bidding does not yet seem to be a settled matter. In 2015 Tom Price then a August :8 Health Affairs 1373

8 Republican congressman from Georgia and now health and human services secretary introduced H.R in the 114th Congress. 23 The bill, which had eighty-two cosponsors, was designed to amend the bidding process, address concerns about rural access and rural payment, and initiate yet another six-year demonstration project using a to-be-determined auction process based on market prices rather than the median price. As of March 2017, CMS had delayed future rounds of competitive bidding to allow the new administration to review the program. 24 However, it is important to recognize that the competitive bidding program resulted in prices similar to those paid by commercial insurers and reduced Medicare expenditures on durable medical equipment and similar items, at least in Round 1. If the concerns about the program s long-term sustainability can be resolved, competitive bidding for durable medical equipment and similar items may be an effective mechanism for achieving savings in Medicare, relative to historic fee schedule prices. Our results also suggest that the competitive bidding program has implications for the entire market for durable medical equipment and similar items. Whether the program is modified or replaced by an alternative payment scheme, any new policies should be evaluated in terms of the impact they will have on the Medicare program and suppliers, and subsequently on the overall market for durable medical equipment and similar items. Conclusion Our analysis compared prices for durable medical equipment and similar items set through CMS s competitive bidding program and those paid by large commercial insurers. The results suggest that the initial round of the program produced prices comparable to those paid by insurers. Given that the program has continued to evolve, additional evaluations of the program prices, relative to commercial prices, from later program rounds and for more durable medical equipment and similar items will likely be informative. The authors acknowledge the assistance of the Health Care Cost Institute and its data contributors, Aetna, Humana, and UnitedHealthcare, in providing the claims data analyzed in this study. NOTES 1 Congressional Research Service. Medicare durable medical equipment: the competitive bidding program [Internet]. Washington (DC): CRS; 2013 Jul 22 [cited 2017 Jun 21]. p. 2. Available from: 722_R43123_d7a4a02c3d1c1cab148 faad5bdb9fe81ee9a69e0.pdf 2 See, for example, Department of Health and Human Services, Office of Inspector General. A comparison of prices for power wheelchairs in the Medicare program [Internet]. Washington (DC): HHS; 2004 Apr [cited 2017 Jun 21]. Available from: oei pdf 3 Department of Health and Human Services CMS statistics [Internet]. Washington (DC): HHS; 2011 Jun [cited 2017 Jun 21]. (CMS Pub. No ). Available from: Data-and-Systems/Statistics-Trends- and-reports/cms-statistics- Reference-Booklet/Downloads/ CMS_Stats_2011.pdf 4 CMS.gov. Medicare s DMEPOS Competitive Bidding Program: frequently asked questions [Internet]. Baltimore (MD): Centers for Medicare and Medicaid Services; 2013 Apr [cited 2017 Jun 21]. Available from: outreach-and-education/outreach/ partnerships/downloads/dmepos partnerfaqsrevised pdf 5 Government Accountability Office. Need for legislation to authorize more economical ways of providing durable medical equipment under Medicare [Internet]. Washington (DC): GAO; 1972 May 12 [cited 2017 Jun 21. Available from: 6 Thomson TG. Final report to Congress: evaluation of Medicare s competitive bidding demonstration for durable medical equipment, prosthetics, orthotics, and supplies [Internet]. Washington (DC): Department of Health and Human Services; 2004 [cited 2017 Jun 21]. Available from: healthnews.files.wordpress.com/ 2009/11/cms_rtc.pdf 7 The bidding process was criticized by both industry and academics. See, for example, Crampton design beats current bidding system. HomeCare [serial on the Internet] Apr 11 [cited 2017 Jul 5]. Available from: topics/competitive-bidding/ cramton-design-beats-current The nine Metropolitan Statistical Areas were Charlotte Gastonia Concord, NC SC; Cincinnati Middletown, OH KY IN; Cleveland Elyria Mentor, OH; Dallas Fort Worth Arlington, TX; Kansas City, MO KS; Miami Fort Lauderdale Pompano Beach, FL; Orlando Kissimmee, FL; Pittsburgh, PA; and Riverside San Bernardino Ontario, CA. The ten Metropolitan Statistical Areas where CMS launched the competitive bidding program in mid were these nine and San Juan Caguas-Guaynabo, Puerto Rico. 9 Competitive bidding program prices for Metropolitan Statistical Areas included in Round 2 took effect July 1, These areas and prices were not included in this analysis. The Medicare Improvements for Patients and Providers Act of 2008 expanded the competitive bidding program to seventy more areas, and the Affordable Care Act expanded the program in 2010 to ninety-one areas in all. The three largest of these ninety-one Metropolitan Statistical Areas included multiple competitive bidding program areas, so that there are a hundred competitive bidding program areas in Round 2. Round Health Affairs August :8

9 Recompete established prices that were in effect for the period July 1, 2016 December 31, To access the Appendix, click on the Appendix link in the box to the right of the article online. 11 Palmetto GBA. Requirements to submit a bona fide bid [Internet]. Columbia (SC): Palmetto GBA; 2012 Aug [cited 2017 Jun 21]. Available from: bid.com/palmetto/cbic.nsf/files/ R1RC_Fact_Sheet_Bona_Fide_ Bid.pdf/$File/R1RC_Fact_Sheet_ Bona_Fide_Bid.pdf 12 Each round of bidding has an online home page that provides roundspecific details, including the competitive bidding program prices. For example, see the item in Note Centers for Medicare and Medicaid Services. Round 1 Rebid [Internet]. Washington (DC): CMS; [cited 2017 Jun 21]. Available from: dmecompetitivebid.com/palmetto/ cbicrd1rebid.nsf/docscat/home 14 Fee schedule information is available from CMS.gov. Durable medical equipment, prosthetics/orthotics, and supplies fee schedule [Internet]. Baltimore (MD): Centers for Medicare and Medicaid Services; [cited 2017 Jun 21]. Baltimore (MD): CMS; [cited 2017 Jun 21]. Available for download from: Service-Payment/DMEPOSFee Sched/DMEPOS-Fee-Schedule- Items/CMS html? DLPage=3&DLEntries=10 &DLSort=2&DLSortDir=descending 15 Government Accountability Office. Medicare: bidding results from CMS s durable medical equipment competitive bidding program [Internet]. Washington (DC): GAO; 2014 Nov [cited 2017 Jun 21]. (GAO Report No. GAO-15-63). Available from: 670/ pdf 16 Government Accountability Office. Medicare: review of the first year of CMS s durable medical equipment competitive bidding program s round 1 rebid [Internet].Washington (DC): GAO; 2012 May [cited 2017 Jun 21]. (GAO Report No. GAO ). Available from: 17 Calculated from CMS.gov. Medicare referring provider DMEPOS data CY2013 [Internet]. Baltimore (MD): Centers for Medicare and Medicaid Services; [last modified 2017 Jun 15; cited 2017 Jun 21]. Available from: Statistics-Data-and-Systems/ Statistics-Trends-and-Reports/ Medicare-Provider-Charge-Data/ DME2013.html 18 Based on an analysis of commercial insurers shares of employer-sponsored insurance markets in Metropolitan Statistical Areas, using data from American Medical Association. Competition in health insurance: a comprehensive study of U.S. markets, 2015 update. Chicago (IL): AMA; Moran Company. The state of expert judgement regarding Medicare s competitive bidding program for durable medical equipment [Internet]. Arlington (VA): Moran Company; 2016 Sep [cited 2017 Jun 21]. Available from: Paper_ pdf 20 Merlob B, Plott CR, Zhang Y. The CMS auction: experimental studies of a median-bid procurement auction with nonbinding bids. Q J Econ. 2012;127(2): Cramton P, Ellermeyer S, Katzman B. Designed to fail: the Medicare auction for durable medical equipment. Econ Inq. 2015;53(1): O Roark B, Foreman S. The impact of competitive bidding on the market for DME [Internet]. Mechanicsburg (PA): Pennsylvania Association of Medical Suppliers; 2008 Feb 18 [cited 2017 Jun 22]. Available from: documents/ dme EconomicImpactStudy.pdf 23 Congress.gov. H.R.4185 PACT Act of 2015 [Internet]. Washington (DC): US Congress; 2015 [cited 2017 Jul 10]. Available from: CMS.gov. DMEPOS Competitive bidding home [Internet]. Baltimore (MD): Centers for Medicare and Medicaid Services; 2017 Mar 10 [cited 2017 Jul 7]. Available for download from: Service-Payment/DMEPOS CompetitiveBid/index.html? redirect=/dmeposcompetitivebid/ August :8 Health Affairs 1375