Latest GAO Report Disputes Criticism of Controversial Test Strip Program

CMS claims that its competitive bidding program (CBP) is saving Medicare hundreds of millions of dollars per year on diabetes care by driving down the cost of self-monitoring supplies and reducing fraud.

In a new report, investigators from the General Accountability Office (GAO) tend to agree,¹ while scholars and patient advocates dispute these claims. CMS refutes criticism that its CBP has caused patients to stop monitoring themselves properly, that the program has led to more hospitalizations and deaths, and that extra hospital costs have exceeded savings on supplies.

¹

Just before Evidence-Based Diabetes Management™ went to press, CMS responded by e-mail to questions sent more than a month prior, which concerned the origins of the CBP and complaints from patient advocates that the program has not maintained acceptable safety and service levels for persons with diabetes. Information about the agency’s positions also comes from documents it has published and from a GAO report on the CBP issued in mid-September and publicly released on October 17, 2016.

Historically, Medicare used a fee schedule, which increased annually, to set reimbursement rates for physician-prescribed durable medical equipment, along with test strips and other products that insulin users need to monitor blood sugar. Congress mandated in 2003² that CMS switch to a competitive bidding system after reports from the GAO and other federal investigators found the old system to be plagued by above-market prices and outright fraud.^3,4 In its e-mail response, CMS said that earlier study results from the GAO and others showed that the old payment system caused Medicare to spend 3 to 4 times the amount paid by commercial insurers for some supplies.The CBP is still being rolled out in stages to various parts of the country. Under the program, qualified suppliers that wish to provide covered equipment to Medicare patients in particular areas are invited to give their best price on products and guarantee them for 3 years. Medicare will then offer contracts to as many bidding suppliers as needed to satisfy demand for a particular item, starting with the lowest bidder and working up through the list.

Medicare’s offer will equal the median of the bids from the winning suppliers, so companies that accept contract offers in any given market will get the same rate. (Rates can, and do, vary significantly from high-cost markets to low-cost markets.) Once the contracts go into effect in a particular market, local Medicare patients who have not received individual exemptions must buy from contracted suppliers, who, in turn, must sell to all Medicare patients at the contracted rate.⁵ In 2010, the initial phase of the program brought competitive bidding to 10 markets. In 2013, the second phase brought bidding to 100 markets, along with a national mail order program for diabetes testing supplies.6 If all goes according to schedule, suppliers will have to bid for contracts in all US markets by 2017.

CMS told the GAO that the CBP has produced dramatic savings. The agency reports that a combination of lower payment rates and less product overuse cut $3.6 billion from equipment spending in the 2-year period starting when phase 2 went into effect.¹ Much of the savings came from greatly reduced prices. Single payment amounts in competitive-bidding areas averaged 45% less than payments in fee-schedule areas. The numbers were even more eye-popping for the national mail order program for diabetes supplies: contract winners charged an average of 72% less to supply an item through the mail than stores charged for comparable products in areas still operating under the fee schedule. In its e-mail response, CMS said the agency has a “sophisticated active claims surveillance system” that keeps track of patient access to supplies and health outcomes. “To this point, the system has not detected any negative trends in access or health outcomes under the program,” the e-mail also stated. Results can be viewed on CMS’ website.

AADE Survey Uncovers Problems

Critics disagree that Medicare’s 20 contracted mail order suppliers are, in fact, supplying comparable products. Systems that use strips to test blood sugar vary enough that patients develop significant preferences for particular brands and some struggle to measure their blood sugar levels correctly when they are forced to switch from one brand to another. In theory, most patients with diabetes should be able to keep using their chosen brands under competitive bidding. Legislation requires that each company that wins a contract to supply a product under the CBP must offer a range of products that account for at least 50% of all sales for that product type. However, a 2014 survey by the American Association of Diabetes Educators (AADE) found that patients who want to get test strips through the mail are effectively forced to switch from popular and well-respected brands to cheaper brands with questionable accuracy.

“In this study, 7 diabetes educators contacted 23 suppliers and found that none of the suppliers offer products reflecting greater than 50% of the market, as intended by Congress, and that only 3

suppliers carry each brand of diabetes testing supplies (DTS) they reported as carrying to Medicare. These findings demonstrate that as a result of the CBP, Medicare beneficiaries have fewer choices and limited access to the DTS most commonly used. Beneficiaries participating in the CBP are effectively being made to either switch to different testing systems or purchase DTS through non—mail order settings. This study results also demonstrate that the information on Medicare’s website continues to be inaccurate and that the information from the suppliers is inconsistent,” read the AADE study.⁷

“When a beneficiary is forced to switch to a testing system that is unsuitable, unknown, confusing, or unreliable, testing compliance may diminish or even cease. Poor blood glucose management can increase the risk of complications.”

Beneficiaries Receive Fewer Supplies

Reductions in equipment prices were not the only source of the savings brought to light by phase 2 of the CBP. The rest of the savings came from reductions in equipment purchases: from 2012

to 2014, the number of beneficiaries receiving any equipment or supply covered by the CBP fell nationwide, but utilization fell far less in fee-schedule areas (6%) than it did in phase 2 bidding markets (17%). The introduction of competitive bidding had an even more dramatic effect on the percentage of patients with diabetes receiving supplies through the mail; there was a 39% decrease between 2012 and 2014, and that decline was only partially offset by a 13% increase in patients receiving such items at pharmacies and other retail outlets.

CMS believes that the CBP is also proving to be a very effective anti-fraud program. First, by eliminating the price premiums that Medicare had paid for many supplies, it has reduced the incentive

for fraud. Moreover, by implementing a thorough certification process for all bidders—and by dramatically reducing the total number of suppliers it does business with—the CBP has made it harder for the unscrupulous to get into the system and easier to find those who do. (CMS was already culling its supplier list: between 2011 and 2016, it revoked billing privileges for 543,100

suppliers.)¹

Using a health status tool that considers 7 measures—deaths, hospitalizations, emergency department visits, physician visits, admissions to skilled nursing facilities, average number of hospital days per month, and average number of nursing facility days per month—CMS has compared competitive-bidding areas and fee-schedule areas. It has also compared patient health status

in phase 2 markets immediately before program implementation and, again, 2 years later. In the GAO report, CMS officials said neither analysis found any significant differences in patient health.¹

However, over the past 2 years, others have disputed the validity of the methodology that CMS uses to look for evidence that the CBP disrupts vital product usage and worsens medical outcomes.

“CMS failed to establish (or report on) baseline values for [medical equipment] acquisition behaviors and health status, thus making it impossible to determine whether changes in either measure occurred. CMS also failed to construct a matched control group, which would have allowed the agency to determine whether CBP, as well as the significance of any changes seen compared with beneficiaries who were not affected by the CBP. Without appropriate baseline measures and a matched control group, CMS could not actually assess the impact of the CBP on changes in acquisition and health outcomes. Therefore, CMS’s claim of no disruption and no adverse outcome is unfounded,” wrote the authors of a report from the National Minority Quality Forum, a

group that advocates for the medical interests of disadvantaged minority groups.

published earlier this year in Diabetes Care11—that used a different methodology and found that the CBP had disastrous effects on patients with diabetes from phase 1 areas covered by Medicare who had been using testing strips correctly before the new program commenced.

behaviors a year after the CBP began showed a 23.0% increase in partial or no self-monitoring in the TEST group (P <.0001) and a 1.7% increase in partial or no self-monitoring among NONTEST patients. What’s more, the reduction in full self-monitoring associated with the introduction of the self-monitoring program led to significantly worse outcomes. “Propensity score—matched analysis showed that beneficiary migration from full to partial/no (self-monitoring blood glucose) acquisition in 2011 (1163 TEST vs 605 NONTEST) was associated with more deaths within the TEST cohort (102 vs 60), with higher inpatient hospital admissions and associated costs,” the study authors wrote. “Based on our findings, more effective monitoring protocols are needed to protect beneficiary safety.”

When asked to respond to the Diabetes Care article, CMS wrote in its e-mail, “The program has maintained beneficiary access to quality products from accredited suppliers in all (competitive bidding areas). Extensive real-time monitoring data have shown successful implementation with very few beneficiary complaints and no negative impact on beneficiary health status, based on such measures as hospitalizations, length of hospital stay, and number of emergency department visits compared with noncompetitive bidding areas.” CMS told the GAO that it takes other steps to monitor patient satisfaction with competitive bidding and claims that satisfaction levels appear to be high.

CMS tallies questions and complaints related to the CBP at its 1-800-MEDICARE number. It reported more than 100,000 queries or complaints in the first quarter of phase 2, but that number fell to 40,000 in the second quarter and continued falling slowly thereafter. Overall, CMS says, such calls constitute less than 1% of all calls received at 1-800-MEDICARE. Multiple complaints about a particular supplier not meeting its contractual obligations to beneficiaries leads CMS to contact the supplier “when appropriate” or sometimes to launch an investigation that sometimes entails dozens of secret shopper calls. Such investigations led to 43 contract termination notices between July 2013 and July 2015. Thirty-seven suppliers came into compliance, so only 6 contracts were actually terminated.

As for patient satisfaction levels with the Medicare equipment supply system, it ranged between 85% and 92% on a 6-metric survey just before competitive bidding and between 84% and 89% several years later.

Needless to say, critics of the CBP strongly dispute CMS’ claims. “It is absurd to claim that beneficiaries are happy with this system or that large numbers of them are struggling to procure supplies they need to test their blood sugar and measure insulin doses. It is absurd to claim that it was easy for beneficiaries to go online and get a new supplier after the program forced all but 20 suppliers to stop serving Medicare patients. The average age of a beneficiary is 78 years. If CMS wanted a real gauge of how easy most people found it to change suppliers, it could have just noted that participation in the program fell 39%—39%!” said Chris Parkin, MS, a co-author of both the Diabetes Care study and the paper from the National Minority Quality Forum. “There’s abundant evidence that competitive bidding is hurting people and wasting money. What there’s no evidence of is any inclination at CMS to fix the program.”

The GAO, however, found the assertions by CMS to be more creditable. The agency’s analysis took some issue with how CMS evaluated patient satisfaction, but issued a generally favorable review of its method for investigating health impact. “Based on our analysis, CMS’ methodologies and scoring algorithm used to evaluate health measure trends among CBP areas appear to be sound,” the report’s authors wrote. “However, we did not examine individual investigations that CMS conducted to assess aberrant changes in trends in particular competitive bidding areas and product categories, and whether these trends could be attributed to the CBP.”

whose initial test-strip acquisition patterns indicated full self-monitoring and those whose acquisition patterns indicated partial or no self-monitoring. Comparisons of same-patient

The investigators began with 43,939 beneficiaries in the 9 test markets (TEST) and 485,688 beneficiaries in the nontest markets (NONTEST). They then subdivided both groups into patients

Indeed, the authors of that report spend several dozen pages critiquing CMS’ study methodology in great detail.¹⁰ The main author of that piece co-authored another paper—which was

Of course, critics say the other possible explanation for the sharp decline in the usage of test strips and other medical equipment is that the competitive bidding system prevented many patients from receiving vital supplies and equipment. In the GAO report and elsewhere, CMS insists that its extensive program monitoring has yet to detect any significant declines in patient outcomes or patient satisfaction.

According to the GAO report, CMS believes the overwhelming majority of the decline stems from a sharp drop in careless overutilization and fraud.¹ The organization, which estimated that improper Medicare payments were almost $50 billion (of the program’s $586 billion expenditures) in fiscal year 2013,8 has undertaken a number of anti-fraud efforts in recent years. A number of those rules have been aimed at the durable medical equipment program. For example, CMS recently announced a national rollout for prior authorization rules that cut expenditures on certain fraud-prone items by as much as 80% in high-fraud markets where the rules were tested.⁹References

1. Medicare: CMS’s Round 2 durable medical equipment and national mail-order diabetes testing supplies competitive bidding program. Government Accountability Office website. http://www.gao.gov/assets/680/679771.pdf. Published September 15, 2016. Accessed October 17, 2016.

2. HR.1. Medicare Prescription Drug Improvement, Prescription, and Modernization Act of 2003.Congress.gov website. https://www.congress.gov/bill/108th-congress/house-bill/1. Accessed October 1, 2016.

3. Medicare: past experience can guide future competitive bidding for medical equipment and supplies. Government Accountability Office website. http://www.gao.gov/products/GAO-04-765. Published September 7, 2004. Accessed October 1, 2016.

4. Medicare: competitive bidding for medical equipment and supplies could reduce program payments, but adequate oversight is critical. Government Accountability Office website. http://www. gao.gov/products/GAO-08-767T. Published May 6, 2008. Accessed October 1, 2016.

5. What is the competitive bidding program? Medicare website. https://www.medicare.gov/what-medicare-covers/part-b/competitive-bidding-program.html. Accessed October 1, 2016.

6. Medicare expanding competitive bidding program to save billions [press release]. Baltimore, MD: CMS; November 30, 2011. https://www.cms.gov/Newsroom/MediaReleaseDatabase/Press-releases/2011-Press-releases-items/2011-11-30.html. October 1, 2016.

7. Competitive bidding program for mail-order diabetes testing supplies: Product Availability Survey. American Association of Diabetes Educators website. https://www.diabeteseducator.org/docs/ default-source/legacy-docs/_resources/advocacy/aade_study_on_suppliers_2014.pdf. Published January 2014. Accessed October 1, 2016.

8. King KM: Medicare fraud: progress made, but more action needed to address Medicare fraud, waste, and abuse. Government Accountability Office website. http://www.gao.gov/assets/ 670/662845.pdf. Published April 30, 2014.

9. CMS finalizes final rule creating prior authorization process for certain durable medical equipment, prosthetics, orthotics, and supplies items [press release]. Baltimore, MD: CMS; December 29, 2015. https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2015-Fact-sheetsitems/2015-12-29.html. October 1, 2016.

10. Centers for Medicare & Medicaid Services Competitive Bidding Program: assessment of impact on beneficiary acquisition of diabetes-testing supplies and durable medical equipment prosthetics, orthotics, and supplies—associated health outcomes. National Minority Quality Forum website. http://www.nmqf.org/wp-content/uploads/2015/11/National-Minority-Quality-Forum-Report-on-CMS-Competitive-Bidding-Program.pdf. Published November 13, 2015. Accessed October 1, 2016.

11. Puckrein GA, Nunlee-Bland G, Zangeneh F, et al. Impact of CMS competitive bidding program on Medicare beneficiary safety and access to diabetes testing supplies: a retrospective, longitudinal analysis. Diabetes Care. 2016; 39(4):563-571. doi: 10.2337/dc15-1264.

Download PDF: Latest GAO Report Disputes Criticism of Controversial Test Strip Program