Getting More Information From Glucose Meter Evaluations

Abstract

Glucose meter evaluations are common in publications and inform whether the meter meets the ISO 15197 specification. The ISO 15197 specifications, which are universally cited, leave 1% of results unspecified, which can be thought of as typical performance of results (99%) versus rare performance (1%). Suggestions are provided to extract more information from these evaluations, including rare performance, since highly discrepant results or failure to obtain a result can be observed in a glucose meter that has met the ISO 15197 specification. It is also recommended that when manufacturers perform evaluations, they analyze adverse events contained in the FDA MAUDE database. Finally, we point out an important problem with the ISO 15197 specifications.

Analysis of a Point-of-Care HbA1c Assay: Is It Time for an HbA1c Error Grid?

“…3 As mentioned previously, most evaluations use relatively small sample sizes and rarely produce outliers. 4 This does not invalidate the study-rather one can consider the estimated performance to be for typical results. That is, this is the performance one can expect most of the time (eg 95%) with errors caused by imprecision and average bias.…”