Accuracy of Perceiving Blood Glucose in IDDM

Cox, Daniel J.; Clarke, William L.; Gönder-Frederick, Linda; Pohl, Stephen L.; Hoover, Claudia W.; Snyder, Andrea; Zimbelman, Linda K.; Carter, William R.; Bobbitt, Sharon A.; Pennebaker, James W.

doi:10.2337/diacare.8.6.529

Cited by 165 publications

(99 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Table 1 gives the percentage in each zone of the Clarke error grid (16,17). Figure 2 shows the Clarke error grid plotted with color-coded points giving the density of measurements in each 1 mg/dl square of the grid.…”

Section: Clinical Accuracy Overallmentioning

confidence: 99%

Accuracy of the 5-Day FreeStyle Navigator Continuous Glucose Monitoring System

et al. 2007

View full text Add to dashboard Cite

OBJECTIVE -The purpose of this study was to compare the accuracy of measurements of glucose in interstitial fluid made with the FreeStyle Navigator Continuous Glucose Monitoring System with Yellow Springs Instrument laboratory reference measurements of venous blood glucose.RESEARCH DESIGN AND METHODS -Fifty-eight subjects with type 1 diabetes, aged 18 -64 years, were enrolled in a multicenter, prospective, single-arm study. Each subject wore two sensors simultaneously, which were calibrated with capillary fingerstick measurements at 10, 12, 24, and 72 h after insertion. Measurements from the FreeStyle Navigator system were collected at 1-min intervals and compared with venous measurements taken once every 15 min for 50 h over the 5-day period of sensor wear in an in-patient clinical research center. Periods of high rates of change of glucose were induced by insulin and glucose challenges. RESULTS -Comparison of the FreeStyle Navigator measurements with the laboratory ref-erence method (n ϭ 20,362) gave mean and median absolute relative differences (ARDs) of 12.8 and 9.3%, respectively. The percentage in the clinically accurate Clarke error grid A zone was 81.7% and that in the in the benign error B zone was 16.7%. During low rates of change (ϽϮ1 mg ⅐ dl Ϫ1 ⅐ min Ϫ1 ), the percentage in the A zone was higher (84.9%) and the mean and median ARDs were lower (11.7 and 8.5%, respectively).CONCLUSIONS -Measurements with the FreeStyle Navigator system were found to be consistent and accurate compared with venous measurements made using a laboratory reference method over 5 days of sensor wear (82.5% in the A zone on day 1 and 80.9% on day 5).

show abstract

Section: Clinical Accuracy Overallmentioning

confidence: 99%

Accuracy of the 5-Day FreeStyle Navigator Continuous Glucose Monitoring System

et al. 2007

View full text Add to dashboard Cite

show abstract

“…Evaluating the accuracy of CGSs, however, is not straightforward, especially if taken in the context of established accuracy measures such as correlation or regression, consensus error grid analysis (EGA) (3), or the clinically based EGA introduced by our group 18 years ago (4,5). The problem is that these measures judge the quality of the approximation of reference BG (RBG) from readings taken at isolated static points in time, regardless of the temporal structure of the data.…”

mentioning

confidence: 99%

Evaluating the Accuracy of Continuous Glucose-Monitoring Sensors

et al. 2004

View full text Add to dashboard Cite

OBJECTIVE -The objective of this study was to introduce continuous glucose-error grid analysis (CG-EGA) as a method of evaluating the accuracy of continuous glucose-monitoring sensors in terms of both accurate blood glucose (BG) values and accurate direction and rate of BG fluctuations and to illustrate the application of CG-EGA with data from the TheraSense Freestyle Navigator.RESEARCH DESIGN AND METHODS -We approach the design of CG-EGA from the understanding that continuous glucose sensors (CGSs) allow the observation of BG fluctuations as a process in time. We account for specifics of process characterization (location, speed, and direction) and for biological limitations of the observed processes (time lags associated with interstitial sensors). CG-EGA includes two interacting components: 1) point-error grid analysis (P-EGA) evaluates the sensor's accuracy in terms of correct presentation of BG values and 2) rateerror grid analysis (R-EGA) assesses the sensor's ability to capture the direction and rate of BG fluctuations.RESULTS -CG-EGA revealed that the accuracy of the Navigator, measured as a percentage of accurate readings plus benign errors, was significantly different at hypoglycemia (73.5%), euglycemia (99%), and hyperglycemia (95.4%). Failure to detect hypoglycemia was the most common error. The point accuracy of the Navigator was relatively stable over a wide range of BG rates of change, and its rate accuracy decreased significantly at high BG levels.CONCLUSIONS -Traditional self-monitoring of BG device evaluation methods fail to capture the important temporal characteristics of the continuous glucose-monitoring process. CG-EGA addresses this problem, thus providing a comprehensive assessment of sensor accuracy that appears to be a useful adjunct to other CGS performance measures.

show abstract

“…The actual clinical relevance is therefore limited." In fact, EGA was initially employed in a clinical setting, evaluating the accuracy of patient BG measurements, relative to reference laboratory readings (12), and the accuracy of patient BG estimates relative to their meter readings (14,16,35). EGA was specifically developed to identify the types and frequency of clinically significant errors (i.e., does meter/patient overestimate hypoglycemia [upper zones D and E], underestimate hyperglycemia [lower zones D and E], or misinterpret euglycemia for hyper-or hypoglycemia [C zones]).…”

mentioning

confidence: 99%