Improvements in coparenting conflict and child adjustment following an online program for relationship distress

Abstract

Many children never receive treatment for their mental health symptoms-and those that do often receive it only after years of delay. Given that relationship and parenting conflict is an identified mechanism of child mental health symptoms, reducing distress in the parents' romantic relationship may help reduce this unmet need. In the current study, 213 couples with 1 or more children between the ages of 3 and 17 (inclusive) were randomized to receive the web-based OurRelationship program or to a 2-month waitlist condition. Intervention couples were also assessed in the year following the program. Couples in the OurRelationship program experienced a significant decrease in coparenting conflict during the intervention (Cohen's d = -0.27) but also a significant increase in coparenting conflict in the following year, reducing the long-term effects of the intervention (within-group d = -0.20 at 1-year follow-up compared with baseline). Additionally, parent-reported children's externalizing (within-group d = -0.40) and internalizing (within-group d = -0.27) symptoms significantly decreased from baseline to the 1-year follow-up. As hypothesized, improvement in relationship satisfaction during the program was significantly associated with a decrease in coparenting conflict which, in turn, was associated with reduction in both externalizing and internalizing symptoms in the children. These results indicate that online relationship-focused interventions offer an important, adjunctive approach to meet society's need for reducing children's mental health symptoms. Furthermore, the improvements in child functioning strengthen the evidence suggesting the cost-effectiveness of these relationship-focused interventions. (PsycINFO Database Record (c) 2020 APA, all rights reserved).

Trial registration: ClinicalTrials.gov NCT03292692.

Figures

Figure 1:

The left shows the pre to post latent change score model used for Aim 1. The right shows the pre to follow up latent growth curve model used for Aim 1; all loadings on the intercept were set to 1 but omitted from figure for simplicity. Linear and quadratic slopes were constrained equal across mothers and fathers to present an average effect.

Figure 2:

The two group design employed for Aim 2. All loadings on the intercepts were set to 1 but omitted here for simplicity. Intercepts were constrained across mothers and fathers in the Same group; Slopes were constrained across mothers and fathers as well as across Same and Different groups. * = loadings left free to vary.

Figure 3:

Within-Group Changes Over Time