Table 6.

Intervention arm: Are grouped usage patterns (super user and consistent tracker groups combined) associated with gestational weight gain after adjustment for demographics?

Usage patternsa	Excessive total GWGb			Excessive GWG rate			GWG
	RRc	95% CI	P value	RR	95% CI	P value	Estimate	95% CI	P value
Stratum 1: normal range BMId and lower income (n=179)e	0.92	0.59 to 1.44	.71	0.93	0.69 to 1.26	.65	−0.32	−2.08 to 1.43	.71
Stratum 2: normal range BMI and higher income (n=319)	0.64	0.45 to 0.90	.01	0.72	0.57 to 0.90	.004	−1.49	−2.44 to −0.54	.002
Stratum 3: overweight or obese BMI and lower income (n=172)	1.28	0.98 to 1.68	.07	1.14	0.94 to 1.38	.19	0.74	−1.56 to 3.05	.53
Stratum 4: overweight or obese BMI and higher income (n=228)	0.87	0.73 to 1.02	.09	0.94	0.84 to 1.06	.31	−2.17	−3.58 to −0.76	.003

^aAll models have been adjusted for age, race, parity, BMI, and timing variables.

^bGWG: gestational weight gain.

^cRR: relative risk.

^dBMI: body mass index.

^eRelative risk of excessive GWG and mean GWG estimates (kg) are relative to subjects who were either medium users, inconsistent trackers, or nonusers (reference group).