Quality Requirements and Standards
The Case for Tighter Glucose Quality Requirements
While the responses to FDA's Draft Guidelines for better Glucose Devices have been almost entirely critical, there is a case to be made for tighter glucose goals, particularly on the analytical side. Here's a summary of the studies that bolster the "pro" side for making changes to how glucose goals are set.
The Case for Tighter Glucose Guidelines
STEN WESTGARD, MS
August 2013
The FDA Draft Guidelines for Glucose Meters have stirred up quite a controversy. There has been an almost uniformly negative response to the combination of more extensive validation requirements, tighter analytical performance demands, and re-classification of method complexity from waived to moderate. It's hard to remember that only a few years ago, the outcry was in the other direction. In 2010, the FDA held a meeting that compiled a litany of problems with the current use of glucose meters.
Some key studies have been published between that 2010 meeting and today that make not only a patient safety, but an economic, argument for better analytical quality. Let's review them:
Tighter Glucose Quality means fewer insulin dosing errors
Glucose Meter Performance Criteria for Tight Glycemic Control Estimated by Simulation Modeling, Karon BS, Boyd JC, Klee GG, Clin Chem(2010) 56:7:1091-1097.
"Glucose meter technologies that operate within a 15% total allowable error are unlikely to produce large (>3-category) insulin dosing error during TQC. Increasing performance to 10% TEa should reduce the frequency of 2-category insulin dosing errors, although additional studies are necessary to determine the clinical impact of such errors during TQC. Current criteria that allow 20% total allowable error in glucose meters may not be optimal for patient management during TQC."
With a 20% TEa (the current glucose meter standard), the authors found that a 2-category error can be expected 6.1% of the time, and a 3-category error can be expected 0.3% of the time. Given the millions of meter readings made in the US, these error rates are not acceptable.
Empiric Validation of Simulation Models for Estimating Glucose Meter Performance Criteria for Moderate Levels of Glycemic Control, Karon BS, Boyd JC, Klee GG, Diabetes Technology and Therapeutics (2013), 15(12);996-1003
"Large (three or more category) dosing errors are predicted to occur only under the 20% TEa condition. Two category insulin dosing errors were common (6-20% of all insulin dosing decisions) when 20% TEa was assumed, but frequency decreased to only 0.2% of dosing decisions when 10% TEa was modeled. When insulin dosing error rates were measured empirically by comparing paired glucose meter and laboratory glucose values, insulin dosing rates were very similar to those predict for the 20% TEa condition."
The authors' observation of glucose meter and lab method discrepancy found that 39% of the pairs were off by 1 insulin dosing category, 4% were off by 2 dosing categories, and 0.3% were off by 3 or more insulin dosing categories. If a 10% goal was applied, the model indicated that 1-category errors could be reduced to 29%, 2-category errors could be reduced to 0.2%, and 3 or larger category errors would be reduced to less than 0.2%.
In other words, the tighter the quality, the fewer errors occur in the test results (and fewer insulin dosing errors).
Fewer errors means less expensive care
Clinical Implications and Economic Impact of Accuracy Differences among Commercially Available Blood Glucose Monitoring Systems , Budiman ES, Navendu S, Resch A, Journal of Diabetes Science and Technology (March 2013) 7;2:365-380.
Using a model of the German diabetic population (both Type 1 and Type 2), the authors estimated the effects of having patients switch from the least accurate to the most accurate blood glucose monitoring device:
"The results of our analysis show a difference in meter error-driven costs of hypoglycemia of approximately $339 million for T1DM [ed. Type 1 Diabetes Mellitus] and of approximately $121 million for T2DM [ed. Type 2 Diabetes Mellitus] patients. These differences were based on the finding that we expected almost 300,000 additional severe hypoglycemic episodes for T1DM patients and 105,000 severe hypoglycemic episodes for T2DM patients....These data support the view that use of more/most accurate meters can help patients avoid (severe) hypoglycemic episodes and ultimately have large budgetary implications, potentially saving the US health care system over $500 million a year."
Impact of a Reduced Error Range of SMBG in Insulin-treated Patients in Germany, Schnell O, Erbach M, Journal of Diabetes Science and Technology 2014, Vol 8(3):479-482.
The authors continued their approach, this time estimating the impact of reducing the allowable error from 20% to 15% and also from 20% to 10%. They calculated not only the reduction in hypoglycemic events, but also the reduction of myocardial infarctions (MI) that could occur if measurements were more accurate.
For a reduction of allowable error from 20% to 15%, the authors found "a HbA1c reduction of 0.14%. According to UKPDS Risk Engine, this translates into a 0.18% reduction in CHD. A 0.18% reduction in fatal and nonfatal MI translates into savings of... €676,904 per year [ed. on the German diabetic population of 2.3 million]....
Combining the savings from reduced severe hypolycemic episodes and reduced MIs, in the group of 390,000 type 1 diabetic patients in Germany, "this will add up to potential savings of €1.02 million. Savings in 2.3 million patients with insulin-treated diabetes will add up to €6.03 million."
For a reduction of allowable error from 20% to 10%, the authors found "a HbA1c reduction of 0.28%. According to UKPDS Risk Engine, this translates into a 0.36% reduction in CHD." This reduction translates into a savings of €1,353,808 per year for the German diabetic population of 2.3 million.
Combining the savings from reduced severe hypolycemic episodes and reduced MIs, in the group of 390,000 type 1 diabetic patients in Germany, "this will add to potential annual savings of €3.41 million. Analyzing the savings for 2.3 million patients with insulin-treated diabetes the sum will add up to €20.13 million." Again, if these rates are scaled up to the US diabetic population, the savings will increase even more.
Who saves? Who benefits?
The current outcry against the FDA Draft Guidelines manifests all the symptoms of a classic case of Loss Aversion. Right now the focus is on what labs and patients might lose, in convenience, capability, and cost (of either not being able to test or requiring more expensive staff to run the tests). There is also a stated fear that no one will build glucose meters in the future if it standards become tighter. Right now the laboratory marketplace has glucose devices for testing in many different settings, and it's very painful to envision how to cope with the loss of any of that.
However, much of the burden for these higher quality requirements will fall upon the manufacturers. They need to build better devices, which will entail more investment. But the market for these devices doesn't look to be going away, and for those manufacturers who complete the more rigorous standards, there will be fewer competitors. These new regulations (if they are implemented) will pose a significant barrier to entry for other manufacturers. There may be fewer players in the glucose device marketplace in the future, but while worrisome, it's nothing new to the diagnostic testing market. In other areas of diagnostics, consolidation has been occurring for years.
It is very true that much of the pain of transition to a higher standard for gluose devices will be borne by the laboratory. But the long term goal is to eliminate a larger problem, the pain that patients are suffering from inaccurate resuls and the extra costs being incurred to the entire healthcare system. Those burdens are spread out more broadly, and thus do not have as much of a voice in this current debate. Right now the patients are suffering and the healthcare sysem is spending money erroneously. Neither trend is sustainable.
If we asked labs whether they would like an expensive, error-prone system of glucose monitoring, most would probably ask for something better. But when that's what they have now, and it's all they know, it becomes difficult to convince them there is a better way. That's Loss Aversion.
Ultimately, the FDA will have to take into account all of this uproar and the inconvenience that more stringent guidelines will impose on the healthcare system. But we hope they don't abandon efforts to improve glucose device quality. The current performance is demonstrably unacceptable, and better devices would save lives and money. We can all agree on that goal.