The purpose of this study is to test an educational intervention for first-time fathers of late preterm (34 to 36 weeks gestation) infants.

Background: Preterm birth (PTB) is associated with 75% of morbidity in neonates. Although mortality and morbidity is higher for children with early PTB (< 32 weeks' gestation), late PTB (between 34 to 36 weeks gestation) is more common and creates a serious impact on public health. Indeed, late PTBs constitute 74% of all PTBs, and even modest increases in the rate of late PTB have dramatic effects on health care costs. PTB rates are rising across Canada and in Alberta at an even faster rate: between 1996 and 2005, the rate of PTB in Alberta increased by 23%. Late PTBs accounted for the majority of this increase. While it is clear that early PTB and being small for gestational age (GA) are risk factors for later developmental challenges even into adulthood, the risks and the negative effects of late PTB on child development are emerging. In spite of evidence that heavier weight (and usually older gestation) infants benefit more from early interventions than their lighter weight counterparts, late preterm infants are ineligible for early intervention programs unless they have other health problems. Most programs are designed for mothers, yet fathers contribute to their child's development by providing interactions that are more vigorous, stimulating, and state-disruptive than mothers. Given their increased risk for delays, it is important that both mothers and fathers of late preterm infants are skilled in parent-child interactions to support development. A new intervention for fathers using positive individualized feedback on parent-child interactions has shown promise in improving interactions with healthy infants. However, it is unknown whether this intervention will be effective for fathers of late preterm infants. The purpose of this study is to test an educational intervention for first-time fathers of late preterm (34 to 36 weeks gestation) infants.

Hypothesis: Compared to a comparison group, first-time fathers of late preterm infants who receive an educational intervention will have more positive father-infant interactions when the infant is 8 months old (adjusted age to account for prematurity) as measured during structured play using the Parent Child Interaction Teaching Scale (PCITS) (also known as the Nursing Child Assessment Teaching Scale [NCATS]) scored by raters who are blind to group assignment.

Secondary Research Question: What is the effect of the intervention on fathers' perceptions of parenting stress as measured by the Parenting Stress Index (PSI) at 4 and 8 months adjusted age? Method: In this pilot randomized controlled trial, data will be collected from 148 English speaking, first time fathers of healthy, late preterm, singleton infants. Fathers in the intervention group will receive two home visits when their infant is 4 and 6 months old (adjusted age) and one web-based booster session. Fathers in the comparison group will receive a home visit and discuss a list of age appropriate toys. For all fathers, baseline interactions and parenting stress will be measured at 4 months with outcomes at 8 months.

Measures: The PCITS is an observational measure used to measure changes in father-child interaction after intervention and includes a Parent Domain score composed of sensitivity to cues, response to distress, cognitive growth fostering, and socio-emotional growth fostering. The PSI is a self-report instrument that provides a Parent Domain score reflecting a parent's experience of stress as a parent. The Child Domain score reflects the parental perceptions of the child's temperament and behaviour. All measures have established reliability and validity and inter-rater reliability will be assessed.

Analyses: The socio-demographic and baseline variables will be compared between study groups using descriptive statistics (mean or median when appropriate, standard deviation or interquartile range when appropriate, and proportions). Correlations will be used to determine relationship between the dependent variable (interaction) and any known confounders that may need to be used as covariates. To test our primary hypothesis that differences will exist between groups on father-infant interaction a linear model adjusting for covariates will be used. The same approach will be used to test overall treatment effect on parenting stress in the Parent and Child Domains. We will include subgroup analyses to test for interaction effects between baseline variables and group, and between fathers who did and did not use the web-based booster.

Significance: The results will help in the design of educational interventions to mitigate risks, and optimize developmental outcomes for late preterm infants through enhancing fathers' interactions with their children, and complementing the effects of mothers' contributions to their children's development.

Inclusion Criteria:

• Fathers - first-time father, 18 years or older, speak and read English, live with the child's other parent, live within one hours' drive of the university
• Infants - born between 34 weeks and 0 days and 36 weeks and 6 days gestation, singleton

Exclusion Criteria:

• Fathers - known addictions
• Infants - one of a multiple birth, required assisted ventilation, had culture-proven sepsis, major surgical procedure, identified major congenital anomalies, chromosomal anomalies, intracranial hemorrhage, neurologic disorder, or perinatal asphyxia.

Sociodemographic and health data were collected with the first home visit.

Challenges recruiting first-time fathers of late preterm infants; recruiting challenges led to a smaller sample than planned.

Limited inter-rater reliability.