Poor performance of blood glucose monitors prompts call to action

The battle for better accuracy for blood glucose monitoring systems began years ago with a demand for stricter premarket standards. Emerging research and new regulatory and reimbursement issues, however, have cast the problem in a new light. Now, growing concern has brought all of the stakeholders, including health care professionals, FDA officials, patient advocacy groups and industry representatives, to the table to consider their next move.

“In the last few years, scientists have discovered that some of the blood glucose monitors that are cleared by the FDA do not perform up to the standards for which they were cleared,” David C. Klonoff, MD, clinical professor of medicine at the University of California, San Francisco and president of the Diabetes Technology Society, said in an interview.

On May 25, the Diabetes Technology Society hosted a meeting in which Ronald L. Brazg, MD, FACE, of the Rainier Clinical Research Center in Washington, presented data on seven commonly used, FDA-cleared blood glucose monitoring systems. The results, which showed that about 25% of the evaluated monitors failed to perform at the standards currently required for FDA clearance, added to an already mounting pile of anecdotal evidence.

Unfortunately, external factors, such as an influx of cheaper, lower-quality systems on the market, the introduction of the CMS competitive bidding program, and a lack of consistency in postmarket adverse event reporting by manufacturers, are compounding the problem, according to experts who spoke with Endocrine Today.

 

There is reason to be hopeful, though, they said. During a follow-up meeting Sept. 9 in Bethesda, Md., Klonoff proposed the implementation of a postmarket surveillance program for blood glucose monitoring systems. Although he and other speakers said the plan needs fine-tuning, many attendees were optimistic about the program’s ability to ensure that systems continue to meet FDA standards after clearance and are also encouraged by the collaboration among health care societies, researchers and government officials.

“This was a wonderful opportunity to bring the FDA, the industry and the consumers together to objectively talk about what the concerns and the problems are. It’s the only way we’re going to reach a solution,” Robert Ratner, MD, chief scientific and medical officer of the American Diabetes Association, said in an interview.

Setting the standard

Where to set the bar in terms of accuracy for blood glucose monitoring systems has long been a point of discussion, according to Barry H. Ginsberg, MD, PhD, president of Diabetes Technology Consultants.

For in vitro glucose monitoring systems, the current International Organization for Standardization (ISO) 15197 criteria published in 2003 (ISO 15197:2003) require that a system provide results within 20% of a laboratory standard at least 95% of the time for glucose values of at least 75 mg/dL and within 15 mg/dL for values lower than 75 mg/dL.

Earlier this year, however, new, stricter ISO 15197 criteria (ISO 15197:2013) were proposed. These criteria tighten the standards, requiring 99% of a glucose monitoring system’s results to fall within 15% of a laboratory standard for values of at least 100 mg/dL. The FDA will likely move toward adopting more stringent criteria for clearance, but it remains to be seen whether the agency will opt to follow the proposed ISO 15197:2013 standards, Klonoff said.

Although many clinicians would like to see stricter standards in place, raising the bar across the board tends to make products more expensive, Ginsberg told Endocrine Today. Instead, because not all patients with diabetes require the same level of accuracy from their glucose monitoring systems, he suggests maintaining the ISO 15197:2003 standards for FDA clearance but also adding accuracy values to each product’s label.

 “What we need is to keep the 20% standard, but instead label the product and educate patients and doctors so they can select the system that best meets their needs,” he said, including systems with higher accuracy values for patients who require them. “Blood glucose monitoring is a consumer product, and people choose when they have a choice. When you buy a car, you know what the horsepower is, so when you buy a blood glucose monitoring system, why shouldn’t you know what the accuracy is?”  

Quality concerns

Concerns about cost are not limited to Ginsberg, especially with the appearance of cheap, low-quality meters on the market and changes to Medicare policies, according to George Grunberger, MD, FACP, FACE, chairman of the Grunberger Diabetes Institute in Bloomfield Hills, Mich.

In a study published in the Journal of Diabetes Science and Technology, and presented by Brazg at the May 25 meeting, researchers evaluated capillary blood samples from 100 patients using three test strip lots from seven currently available blood glucose monitoring systems: Accu-Chek Aviva Plus (Roche), Advocate Redi-Code, Element (Infopia USA), Embrace (Omnis Health), Prodigy Voice, TRUEBalance (Nipro Diagnostics) and WaveSense Presto (AgaMatrix). Only three systems — Accu-Chek Aviva Plus, Element and WaveSense Pro — met the ISO 15197:2003 accuracy criteria. After imposing the newly proposed ISO 15197:2013 standards, only one meter — the Accu-Chek Aviva Plus — met the criteria. Further, the other six systems failed to meet the accuracy criteria in at least two of the three lots tested, indicating lot-to-lot variability, according to the abstract.

These data, although disappointing, are symptomatic of even larger problems, according to some experts. In an editorial comment on the study, Andreas Pfützner, MD, PhD, of the Institute for Clinical Research and Development in Mainz, Germany, said stricter postmarket requirements are needed for blood glucose meters, but he also admonished cheaper, nonbranded meters.

“Low-cost meters use older technologies and their manufacturers do not usually seriously invest in new technology or constant quality assurance efforts. This may explain the study results, which show superior performance of the branded meter,” Pfützner wrote. “Finally, the article pinpoints the ‘quality vs. price’ dilemma faced by the prescribing physician and their patients in daily routine, which may additionally be aggravated by budget constraints and prescription rules in reimbursed markets.”

Most prominent companies have to do business with many regulators and meet internal and external quality standards, Grunberger told Endocrine Today.

“When these sorts of imposters flood the market with inferior products, I see the outrage from patients, my colleagues and the manufacturers,” he said.  

Cost of competitive bidding

The availability of low-cost, low-quality blood glucose monitoring systems is worrisome on its own, but the CMS Durable Medical Equipment, Prosthetics, Orthotics, and Supplies (DMEPOS) Competitive Bidding Program may be exacerbating the situation, Grunberger said.

According to information from the CMS website, the program allows suppliers to compete to become Medicare contract suppliers by submitting bids to furnish certain items and the CMS in turn awards contracts to enough suppliers to meet beneficiary demand.

As some manufacturers win Medicare contracts, particularly those that may have sacrificed quality to achieve low costs, other companies may feel cornered, according to Robert A. Vigersky, MD, past-president of The Endocrine Society and director of the Diabetes Institute at the Walter Reed National Military Medical Center in Bethesda, Md.

“This competitive bidding has the potential to either drive out the major meter manufacturers if they decide that they can’t manufacture strips at that price with a sufficient profit margin and/or they’ll lower their quality so that, in the end, patients will be left with lower-quality manufacturers or major manufacturers who have opted to operate at a lower quality,” he said in an interview.

During a panel discussion at the Sept. 9 meeting, Elizabeth A. Koller, MD, of the Coverage and Analysis Group for the CMS, offered insight into the agency’s point of view. She said Medicare is primarily concerned about the outcomes of the diagnostic and therapeutic interventions for its unique patient population. For instance, most Medicare patients with diabetes have type 2 diabetes and are more likely to die of cardiovascular disease, which is not mitigated by glucose control. However, Koller added, this naturally may not be the case for all patients with diabetes.

“Basically, the goals and what is needed by our patient population might not represent what might be appropriate for other patient populations,” she said.

Even so, patients get caught in the confusion, according to Grunberger. Often, he said, patients on Medicare experience inaccurate glucose readings while using a new monitoring system that he, the prescribing physician, did not order and with which he was not familiar. Without this knowledge, a patient may experience negative outcomes, and Grunberger would be unable to identify why.

“In older patients, symptoms of hypoglycemia, for example, are not typical, so they are critically dependent on detecting true hypoglycemia using a finger stick measurement,” Grunberger said. “When caring for Medicare patients, you’re dealing with the most vulnerable patient population. This can literally be a life or death situation.”  

Regulatory obstacles  

According to Courtney H. Lias, PhD, director of the division of chemistry and toxicology devices in the Office of In Vitro Diagnostics and Radiological Health at the FDA’s Center for Devices and Radiological Health, manufacturers are feeling the pressure.

“They have expressed concern that, if the quality of the products marketed by different companies may not be equivalent and if companies are cutting corners [to keep prices low], how can we, the FDA, prevent that or how can we make that matter in terms of products that are selected and bought by customers or reimbursed by payers?” she told Endocrine Today.

Although the FDA has a few tools in its toolbox to address a product’s postmarket performance — including random inspections, adverse event reporting and product recalls — certain challenges hamper their effectiveness.

The agency, for instance, can inspect foreign companies, which are some of the largest manufacturers of low-cost, nonbranded blood glucose monitoring systems, but it cannot go in unannounced, Lias said. Moreover, the FDA relies on companies, both inside and outside of the United States, for adverse event reporting. Some manufacturers are conservative and report all adverse events, whereas others will only report serious, high-risk events or opt not to report events at all. If the FDA is not aware that a problem exists, the agency cannot investigate it.

“Blood glucose monitors account for more than 30,000 medical device reports (MDRs), more than any other medical device. That’s because an MDR is supposed to be anything that could potentially be dangerous,” Ginsberg said. “And one problem is that some foreign companies take the rules, for the most part, very loosely. They see them more as guidelines than regulations.”

The FDA is exploring several avenues to remove these obstacles, according to Lias.

“Internally, we’re working on ways to enhance our surveillance, like with our adverse event reporting, and ways to improve accountability for manufacturers,” she said. “Mainly, we’re looking into improving guidance for manufacturers for MDRs to clarify our expectation of what should be reported so there is more consistent decision making across companies.”  

Creative solutions for care  

Although one strategy will likely not address every aspect of a multifaceted issue such as postmarket performance of blood glucose monitoring systems, everyone is eager to find an answer, according to Klonoff.

At the Diabetes Technology Society follow-up meeting Sept. 9, Klonoff proposed one potential solution: implementation of a postmarket surveillance program for blood glucose monitoring systems. He said the program will test the performance of FDA-cleared systems and inform patients, health care providers and payers about each system’s real-world performance.

Goals of the program include creating a laboratory network using validated methods traceable to a higher order; coordinating testing by sending blood glucose monitoring systems to network labs for standardized measurements; maintaining or publishing performance data, as well as indicating systems that are compliant with FDA standards; and monitoring performance of the network labs, Klonoff said.

The program would have a steering committee composed of national and international experts in blood glucose monitoring from various fields, as well as industry and the FDA. They would be charged with directing and implementing the surveillance program and reviewing policies, protocols and reports from the network labs. There would also be an administrative center that would obtain blood glucose monitoring systems from commercial sources, send them to the network labs and monitor results. They would also analyze data, coordinate policies with the FDA, issue statements on performance and more. Klonoff said network labs, based inside and outside of the United States, will test the analytical performance of the monitoring systems and maintain certification for any of their glucose reference methods.

In terms of the testing process, the labs will analyze a system’s performance with a reference method calibrated by materials traceable to a higher order and method, Klonoff said, and bins will be created via glycolysis or spiking for testing across the physiological range.

The steering committee and the FDA would collaborate to decide the source of blood specimens used for testing, how many specimens and the number of lots per blood glucose monitor, and the number of glycemic bins and cutoff levels. They would also decide the reference method, the format for reporting results and at which intervals to retest after poor performance. A statement on performance results would be issued for each system that is tested and be made available through the program’s website, according to Klonoff.

“Patients, health care professionals and payers will benefit if the FDA improves postmarket surveillance of blood glucose monitors,” he said. “And the best time to create the blood glucose monitor surveillance program is now.”  

‘The devil is in the details’ 

Reaction to Klonoff’s proposal was largely positive, with FDA officials, industry representatives and health care professionals welcoming a solution to the problem. Even so, many challenges lie ahead.

From the FDA’s perspective, a postmarket surveillance program provides another checkpoint to ensure that the agency can identify failures in the system at the manufacturing level or in the supply chain, so they can be corrected. Further, the program may also serve as a deterrent for companies looking to cut corners, Lias told Endocrine Today.

However, she said, the program is expensive. Additionally, at the meeting, speakers debated the pros and cons of a program that is voluntary vs. mandatory for all manufacturers. Klonoff said he only had interest in instituting a mandatory program, but to do so, the FDA would require new regulations, and the rule-making process could take 5 to 6 years, according to Lias.

Other concerns expressed by researchers, industry representatives and members of various medical societies and organizations mostly pertained to specifics. Finding the best way to address the challenges of testing glucose in the lab, for example, or details on reporting the data were among the questions raised after Klonoff outlined the basic goals and structure of the postmarket surveillance program. During the industry discussion panel, manufacturers expressed concern about the ramifications in their arena, with several wondering how the data gleaned from the program could affect reimbursement or marketing of their products.

Enforcement is also key, according to Ratner.

“My concerns, as were spoken during the meeting, include how the program is going to be financed, how data are going to be expressed and how it will be enforced,” he said. “My hope is that all of this will be done by the FDA because they have the legislative and regulatory tools. As Dr. Klonoff said, the ‘devil is in the details.’”

Nevertheless, the program is in the spirit of many principles by which endocrinologists practice medicine, according to Vigersky.

“This is a preventive medicine equivalent,” he said. “As it exists now, the FDA doesn’t know a problem exists until after the fact. … They are inundated with adverse event reports and they have a lot of difficulty in figuring out exactly what is going on and where the error is. If you can identify that a certain manufacturer’s strips, as purchased off the shelf, are having problems, then the FDA can suspend distribution of these strips until the manufacturer corrects the problem or just drops out of the market.”

All told, Vigersky said he is hopeful as Klonoff takes the next steps toward forming a steering committee and seeking funding for the program.

“There’s reason to be optimistic that an FDA- and industry-supported surveillance program will be put in place over the next couple of years, and this will ensure higher quality products, better patient outcomes and fewer adverse events,” he said.

Moreover, Ratner said, placing the struggles with blood glucose monitoring systems in context is encouraging.

“We are moving in the right direction, even if we’re never moving as fast as people would like,” he said. “But we’ve come a long way in 30 years from the days of monitoring glucose through urine testing. We still have a long way to go, but we are now to the point where we can provide a lot more guidance to people with diabetes.” – by Melissa Foster

For more information: 

• Brazg RL. J Diabetes Sci Technol. 2013;7:144-152.
• Pfützner A. J Diabetes Sci Technol. 2013;7:153-155.
• Barry H. Ginsberg, MD, PhD, can be reached at Diabetes Technology Consultants, 501 Lydia Lane, Wykoff, NJ, 07481; email: diabetes_consultants@yahoo.com.
• George Grunberger, MD, can be reached at The Grunberger Diabetes Institute, 43494 Woodward Avenue, Suite 208, Bloomfield Hills, Michigan 48302; email: grunberger@comcast.net.
• David C. Klonoff, MD, can be reached at the Diabetes Technology Society, 1157 Chess Drive, Suite 100, Foster City, CA 94404; email: dklonoff@diabetestechnology.org.
• Courtney H. Lias, PhD, can be reached at the Center for Devices and Radiological and Health, FDA, 10903 New Hampshire Ave., Silver Spring, MD, 20993; email: Jennifer.Haliski@fda.hhs.gov.
• Robert Ratner, MD, can be reached at American Diabetes Association, 1701 North Beauregard St., Alexandria, VA 2231; email: rratner@diabetes.org.
• Robert A. Vigersky, MD, can be reached at theWalter Reed National Military Medical Center, Endocrinology, Diabetes and Metabolism Service, 8901 Wisconsin Avenue | Bethesda, MD 20889; email: Robert.A.Vigersky.mil@health.mil.
• Disclosure: Ginsberg reports being a consultant for Agamatrix. Grunberger reports no relevant financial disclosures. Klonoff reports being a consultant to Sanofi. Lias, Ratner and Vigersky report no relevant financial disclosures.

 

 

Do all patients with diabetes require the same level of accuracy from their glucose meters?  

 

Yes  

Accuracy is very important with blood glucose monitoring systems, regardless of whether you have type 1 or type 2 diabetes.

 

Type 2 diabetes is a different disease than type 1 diabetes. However, patients with type 2 diabetes also require reliable, highly accurate meters. First of all, many patients with type 2 diabetes end up with pancreatic exhaustion and therefore progress to taking insulin because their pancreases aren’t working. Additionally, sulfonylureas are used very prominently in type 2 diabetes, which have the potential to produce unnecessarily low blood sugars. For older people, who are especially prone to hypoglycemia, an inaccurate reading from a glucose meter could be harmful and cause a lot of mayhem.

Blood glucose monitoring systems are not perfect. Even when blood is drawn and measured in the laboratory, a certain amount of variance exists and we have to accept some variance with every test. However, you don’t have to accept undue variance. You want a meter that is accurate, reliable, trustworthy and has been properly evaluated by the FDA.

The bottom line is that no matter what the patient’s age is or whether it’s type 1 or type 2 diabetes, accuracy is essential in our blood glucose meters.

Philip Levy, MD, MACE, is a clinical professor of medicine at the University of Arizona College of Medicine and a practicing clinician at the Phoenix Endocrinology Clinic. He is also an Endocrine Today Editorial Board member. Disclosure: Levy reports financial ties with AstraZeneca/Bristol-Myers Squibb, Daiichi Sankyo, Eli Lilly/Boehringer Ingelheim, Janssen, Novo Nordisk and Sanofi.

 

Not necessarily  

Accuracy of blood glucose monitoring is very important for all patients. It provides them with feedback about the interaction between food, exercise and blood glucose-lowering medications.

However, it may be variably critical in different patients. For example, accuracy is very important in a patient with either type 1 or type 2 diabetes who is dosing insulin based on pre-meal blood glucoses or in a pregnant patient with diabetes in whom blood glucoses need to be tightly controlled to prevent an increased risk for maternal and fetal complications. On the other hand, this degree of accuracy may not be needed in a patient with type 2 diabetes who is treated with diet and exercise or with minimal oral hypoglycemic therapy.

 

Accuracy is also important when insulin dosing needs to be performed in the setting of kidney disease, which can reduce insulin clearance and when alterations in carbohydrate intake and/or exercise may be producing acute changes in blood glucose levels.

Several currently used glucometers are mandated to reach standards of accuracy, which has certainly increased their reliability.

Blood glucose monitoring helps promote patient engagement and their sense of control over their diabetes, while simultaneously increasing safety of diabetes therapy, and is therefore an important component of treatment.

Shubhada Jagasia, MD, MMHC, is an associate professor of medicine and associate director of clinical affairs in the division of endocrinology, diabetes and metabolism at Vanderbilt University Medical Center. Disclosure: Jagasia reports no relevant financial disclosures.