Neonatal outcomes from a quasi-experimental clinical trial of Family Integrated Care versus Family-Centered Care for preterm infants in U.S. NICUs

Table 5 Per protocol analysis of intervention components

Infant outcome Intervention component	Adjusted group difference^b or OR^c (95% CI)	p-value
Weight gain during 3-week study period (n = 250)
Paired with a parent mentor
Yes	reference
No	-0.128 (-0.227, -0.030)	0.011
Participating in clinical team rounds
12 or more rounds	reference
2–11 rounds	0.062 (-0.018, 0.142)	0.13
0–1 rounds	-0.084 (-0.154, -0.015)	0.018
Attending parent group educational classes
3 or more classes	reference
1–2 classes	-0.000 (-0.090, 0.090)	> 0.99
0 classes	0.009 (-0.067, 0.085)	0.82
Using the mFICare app
12 or more app logins	reference
4–11 app logins	0.031 (-0.045, 0.106)	0.42
0–2 app logins	0.050 (-0.031, 0.130)	0.23
Infection during NICU hospitalization (n = 253)
Having a parent mentor
Yes	reference
No	7.9 (0.7, 66.7)	0.059
Participating in clinical team rounds
2 or more rounds^a	reference
0–1 rounds	2.9 (1.0, 9.1)	0.043
Attending parent group educational classes
1 or more classes^a	reference
0 classes	3.8 (1.2, 14.3)	0.016
Using the mFICare app
12 or more app logins	reference
4–11 app logins	2.1 (0.5, 8.1)	0.30
0–3 app logins	1.8 (0.6, 5.6)	0.30

P-values < 0.05 are in bold. Other secondary outcomes were not associated with use of the intervention components
^aDue to small cell counts in the infection models, low and high doses were combined both for rounds and for parent group classes, and exact logistic regression was used
^bWeight gain models use intervention doses during the 3-week study period and are reported as the group difference in weight gain (change in weight z-scores) adjusted for site, gestational age at birth, small for gestational age, necrotizing enterocolitis, and morbidity count (a proxy of clinical course complexity, defined as the total of five common preterm morbidities during the NICU stay [nosocomial infection, BPD, ROP, necrotizing enterocolitis, and intraventricular hemorrhage], trichotomized as 0, 1 or 2–5)
^cInfection models (logistic) use intervention dose from study enrollment to hospital discharge and are reported as odds ratios adjusted for site, gestational age at birth, and any ventilation

ISSN: 1471-2431