Self-monitoring of blood glucose is essential for managing diabetes. Over 30 different blood glucose monitors cleared by the U.S. Food and Drug Administration (FDA) are available to consumers. The FDA relies solely on data provided by manufacturers for clearance. Few physicians or diabetic patients are aware that blood glucose monitor test strips that are exposed to humidity and/or excessive temperature give falsely elevated results. Manufacturers know this, and the information they provide to patients obfuscates the problem; furthermore, the FDA does not require the reporting of environmental effects on accuracy.
On more than one occasion, I administered excessive insulin based on results from the last few test strips from vials of 50 that had been opened for <1 month. I determined that the test strips were inaccurate by comparing them to new ones using glucose control solutions. I contacted the manufacturer, but representatives would provide no data on environmental effects. A Medline search showed no studies on this subject. Health Devices evaluates blood glucose monitors periodically but has never performed environmental testing.
The FDA requires manufacturers to report whenever they become aware of information that reasonably suggests that one of its devices 1) has or may have caused or contributed to a death or serious injury or 2) has malfunctioned and that the device or a similar device marketed by the manufacturer would be likely to cause or contribute to a death or serious injury if the malfunction were to recur. This information is available at the FDA website in the Manufacturer and User Facility Device Experience Database (MAUDE) (1). A search showed 691 reports for the test strips and 316 for the blood glucose monitor I used, many from health care providers. Many involve instances where the blood glucose monitor indicated high glucose levels, while symptoms were consistent with severe hypoglycemia. Hypoglycemia was confirmed by emergency medical services or in emergency departments. Manufacturers frequently respond to FDA inquiries with the statement “glucose controls were not used on the system.”
This is disingenuous, as the manufacturers are well aware that few diabetic patients use glucose control solutions. A personal survey of 24 pharmacies shows that 15 carry no solutions and 5 only one brand for one monitor because they don’t sell any. One recent study evaluating how well diabetic patients use blood glucose monitors reported that only 29 of 111 used glucose control solutions, often outdated and only rarely, and that most did not know what they were for (2).
The frequency and subject of MAUDE reports are not uniform across blood glucose monitors. Those using individually wrapped test strips, especially those containing a desiccant, have few reports, while those using test strips stored in vials account for most. (This is complicated because the two largest-selling devices both use vials with 50 test strips.) Studies addressing test strip reliability are lacking and urgently needed.
A human factors evaluation of one monitor marketed with a videotape beginning “It’s as easy as 1, 2, 3” found that it requires 52 substeps to use properly (3). The authors conclude, “It is not appropriate to blame the user for making an error when the root cause of the error may really be the design of the system itself.” Glucose monitors that are easier to use and that alert the user when the test strips should be discarded need to be designed. The medical community and regulatory agencies must insist on it.
