G x I Effects of a Parenting Program for Military Families

Introduction:

Recent advances in examining intervention moderators (i.e. answering questions such as for whom is a preventive intervention more or less beneficial) have focused on gene x intervention (GxI) interactions. GxI analyses have revealed ways in which preventive interventions might confer specific protections for individuals carrying genetic vulnerability for a specific poor psychosocial outcome and have corresponding implications for the tailoring of these interventions.

The current study examined data from a randomized controlled trial of a parent training program (After Deployment, Adaptive Parenting Tools/ADAPT; Gewirtz, DeGarmo, & Zamir, 2018) targeting families in which a parent had been deployed to war in Afghanistan or Iraq. Previous intent-to-treat analyses found the ADAPT program to have beneficial intervention effects on parenting, child and parental adjustment, including direct effects on mothers’ PTSD symptoms and indirect effects on fathers’ PTSD symptoms. This study tests a GxI differential susceptibility hypothesis by examining whether assignment to the ADAPT intervention might particularly benefit mothers and fathers with genetic vulnerability to PTSD (assessed with a polygenic risk score).

Methods:

The sample consists of 336 families, including 282 fathers and 314 mothers recruited from a midwestern state. All families included a target child ages 5-12 and at least one parent had deployed to the wars in Iraq and/or Afghanistan. Families were randomized (60/40) to a multi-family 14 week group-based parenting intervention, After Deployment, Adaptive Parenting Tools/ADAPT, or a literature (web and print) control condition. The intervention was a 2 hour per week mindfulness-informed parent training intervention based upon the Parent Management-Training Oregon model, with a strong focus on emotion regulation and emotion socialization skills. Multi-method, multi-informant assessments of parenting (parent-child interactions), parent, and child adjustment, were gathered at four timepoints over two years.

PTSD severity was measured using the PTSD Checklist (PCL-M for deployed fathers and mothers, and PCL-C for non-deployed mothers) at baseline, 12, and 24 months post baseline. Intervention effect was accounted for by a grouping variable indicating whether participants were randomly assigned to the intervention group or the control group. Genetic vulnerability was indexed with a polygenic risk score. This approach (compared to a candidate gene approach) reflects the evidence that there are multiple loci involved in behaviors (e.g., Duncan, Pollastri, & Smoller, 2014).

Results and Discussion:

We conducted a series of ANCOVAs to examine GxI effects. Results indicate that while the intervention operates to directly reduce PTSD symptoms in mothers (though not in fathers), the intervention also appears to buffer genetic vulnerability to PTSD symptoms.

Implications are discussed with regard to identifying potential tailoring variables such as genes which would allow for more efficient intervention delivery options. We plan to extend ANCOVA based models to latent growth models of polygenic risk for PTSD.