Table 1

Efficacy and safety outcomes*

Participants 11–14 years of age (n = 45)Participants 4–10 years of age (n = 36)
Control nights
(n = 941)Intervention nights
(n = 955)P valueControl nights
(n = 755)Intervention nights
(n = 769)P value
Bedtime glucose, baseline
 Blood glucose (mg/dL)168 ± 20166 ± 18162 ±18165 ± 15
Primary analysis
 Time spent <70 mg/dL (%)10.1 (5.9, 13.8)4.6 (2.9, 7.3)<0.0016.2 (3.0, 7.6)3.1 (1.6, 5.0)<0.001
Secondary analysis
 Hypoglycemia outcomes
  Time spent <60 mg/dL (%)5.8 (2.9, 8.1)1.9 (1.4, 3.9)<0.0012.3 (1.1, 4.6)1.2 (0.5, 2.8)<0.001
  Time spent <50 mg/dL (%)2.8 (0.9, 3.9)0.8 (0.2, 1.9)<0.0010.5 (0.1, 2.5)0.6 (<0.1, 1.2)0.03
 Low blood glucose index2.33 (1.62, 3.03)1.08 (0.80, 1.84)<0.0011.45 (0.74, 2.00)0.76 (0.47, 1.31)<0.001
 Percent of nights with ≥1 value
  <70 mg/dL39340.0133280.04
  <60 mg/dL2921<0.00124190.01
  <50 mg/dL1712<0.00114100.005
 Percent of nights with events of glucose <60 mg/dL continuously§ for
  >10 min2719<0.0012215<0.001
  >25 min2114<0.0011710<0.001
  >60 min137<0.001750.03
  >120 min83<0.00151<0.001
Overall outcomes
 Overnight mean glucose (mg/dL)144 ± 18152 ± 19<0.001153 ± 14160 ± 160.004
 Time in range 70–180 mg/dL (%)64 ± 1066 ± 100.1063 ± 1063 ± 110.50
Hyperglycemia outcomes
 Time spent >180 mg/dL (%)25 (18, 32)30 (23, 33)0.0431 (23, 37)32 (24, 43)0.23
 Time spent >250 mg/dL (%)4 (3, 8)6 (4, 9)0.187 (4, 10)6 (5, 12)0.77
Morning glucose outcomes
 Blood glucose (mg/dL)159 ± 29176 ± 28<0.001154 ± 25158 ± 220.11
 Percent of mornings with blood glucose
  <50 mg/dL<10NA<1<1NA
  <60 mg/dL110.01110.15
  <70 mg/dL41<0.00151<0.001
  70–180 mg/dL6457<0.00166700.11
  >180 mg/dL3242<0.00129290.97
  >250 mg/dL10110.57970.18
  With blood ketones ≥1.0 mmol/L0.41.00.162.92.30.39
  With blood ketones ≥0.6 mmol/L1.72.20.398.09.90.10
  With urine ketones ≥ “small”1.41.70.532.63.50.28
  • Data are mean ± SD, median (IQR), or %. NA, not applicable.

  • *Glucose results from CGM unless specified as blood glucose.

  • †One participant with 5 intervention nights and 5 control nights was excluded for <80 h of CGM glucose data.

  • §Hypoglycemia event assessment based on CGM values below threshold was added to statistical analysis plan post hoc.

  • ‡Too few events for formal statistical comparison.