glucose events.
Percentages are displayed for three different parameters:
•
Detection Rate – amount of time the System displays a Glucose Message correctly.
•
Missed Detection Rate – amount of time the System did not display a Glucose Message when it should
have.
•
False Notification Rate – amount of time the System displays a Glucose Message when it shouldn't
have.
Low and High Glucose Message percentages are based on comparison between Sensor glucose and true
blood glucose values within 15 minutes before or after the true blood glucose value.
Glucose Going Low and Glucose Going High Message percentages are based on comparison between
Sensor glucose and true blood glucose values up to 15 minutes before the true blood glucose value.
For example, the System was able to detect 77.6% of actual low glucose events (detection rate), but 54.7%
of the time a Low Glucose message was displayed in error (false notification rate) and 22.4% of the time a
Low Glucose message was not displayed when it should have been (missed detection rate).
Table 13: Detection of Hypoglycaemic and Hyperglycaemic Events
Precision
Precision of the System was evaluated by comparing the results from two separate Sensors worn on the
same subject at the same time. Table 14 provides data from two separate Sensors worn on 95 subjects at
the same time. Overall 15300 real-time pairs of GM measurements were obtained, with a mean CV of 5.7%.
Table 14: Overall between Sensor Precision
Sensor Wear Duration
The Sensor can be worn for up to 14 days. To estimate how long a Sensor will work over the wear
duration, 95 Sensors were evaluated to determine how many days of readings each Sensor provided. Of
the 95 Sensors, 82 (86.3%) lasted until the final day of use. 93 Sensors (97.9%) lasted more than 5 days.
ART41361-001_rev-A 2:18 pm 30 aug 2019