TABLE 1.

Baseline characteristics of ACT1ON participants included and excluded from the primary analysis (n = 68)

	Included (n = 40)	Excluded (n = 28)	P
Age (y)1	25.5 ± 3.2	25.4 ± 3.0	0.77
Female gender2	28 (70.0)	21 (75.0)	0.79
Non-Hispanic White race and ethnicity2,3	30 (75.0)	15 (53.6)	0.08
UNC site2	25 (37.5)	14 (50.0)	0.33
BMI (kg/m2)1	29.8 (28.1, 33.9)	30.8 (27.3, 33.8)	0.80
Hemoglobin A1c (%)1	7.8 ± 1.3	8.0 ± 1.5	0.31
Diabetes duration (y)1	15.6 ± 6.2	11.8 ± 6.0	0.84
Insulin pump use2	23 (57.5)	17 (60.7)	0.81
Diet4	Included (n = 36)	Excluded (n = 22)
Total fiber intake (g/d)1	12.6 (9.5, 21.3)	14.3 (10.5, 19.3)	0.68

ACT1ON, Advancing Care for Type 1 Diabetes and Obesity Network; UNC, University of North Carolina. ¹Continuous variables are presented as mean ± SD if normally distributed or as median (IQR) if nonnormally distributed. Group differences were tested with the independent t test and Mann-Whitney U test for normally and nonnormally distributed continuous variables, respectively. ²Categorical variables are presented as n (%). The χ² test of equal proportions was used to test for group differences in categorical variables. ³Race and ethnicity were collapsed into other or non-Hispanic White race and ethnicity. To avoid participant identification, we expressed frequencies with fewer than 3 individuals as “n < 3.” Included participants: African American, n = 3 (7.9%); Asian, n < 3; >1 race, n < 3; and White, n = 32 (84.2%). Excluded participants: African American, n = 3 (12.5%); Native Hawaiian/Other Pacific Islander, n < 3; >1 race, n = 4 (14.3%); other race, n < 3; Asian, n < 3; and White, n = 17 (70.8%); 6 (15.0%) included and 5 (17.9%) excluded participants were identified as Spanish/Hispanic/Latino ethnicity. Three participants (2 included and 1 excluded) were missing race data but identified with a Spanish/Hispanic/Latino ethnicity. ⁴Sample sizes differed for dietary intake owing to missing data.