Table 2.
Integration window length (samples) | 5 | 7 | 13 | ||||||
Integral threshold (mg·min/dl) | 5 | 10 | 20 | 5 | 10 | 20 | 5 | 10 | 20 |
Hypo detection time cohort (min) | -65 [-110, -35] | -95 [-120, -55] | -145 [-180, -120] | -55 [-95, -30] | -85 [-120, -50] | -130 [-180, -115] | -35 [-73, -15] | -55 [-95, -25] | -120 [-160, -85] |
Hypo detection time patient median (min) | -65 [-94, -33] | -90 [-118, -56] | -170 [-180, -124] | -60 [-83, -31] | -80 [-110, -47] | -150 [-180, -123] | -35 [-66, -16] | -55 [-88, -26] | -120 [-166, -101] |
BG level at alarm (mg/dl) | 58 [55, 60] | 64 [62, 66] | 78 [74, 79] | 55 [53, 57] | 62 [60, 64] | 75 [71, 77] | 52 [48, 54] | 55 [51, 60] | 72 [61, 75] |
false alarms | 1 [0, 2] | 1 [0, 2] | 2 [1, 3] | 1 [0, 1] | 1 [0, 2] | 2 [1, 3] | 0 [0, 1] | 1 [0, 1] | 1 [1, 2] |
Data are median (IQR) for all 700 Monte Carlo runs (100 per patient).