Mediators and Moderators of Caregiving Strain in a Community-Based Sample of Caregivers of Youth with Mental Health Conditions

Background and Purpose: Approximately 22% of U.S. adolescents have experienced a mental health condition over the course of their lives. Caregivers of these youth experience a variety of struggles including significantly higher levels of depression, anxiety, and other forms of emotional distress. Caregiver emotional distress often takes the form of internalized (directed inward at oneself) and externalized (directed outward at the youth) caregiving strain. Although problematic youth behaviors are frequently identified as the largest contributor to caregiving strain, few studies have examined mediators and moderators of this link. Based on Pearlin’s Stress Process Model which describes the role of contextual factors, primary/secondary stressors, and moderators of these stressors, the aims of this study are to determine whether the association between youth behaviors and internalized and externalized strain is: 1) mediated by negative occurrences for caregivers (e.g., interruption of personal/family time, financial strain), and; 2) moderated by youth emotional and behavioral strengths.

Methods: Baseline data from a community-based sample of youth and caregivers (N = 218) were analyzed with the PROCESS macro for SPSS. Internalized and externalized caregiving strain were treated as the dependent variables in all models. Path coefficients, direct, and indirect effects for mediational models were estimated. Moderation effects were tested using interaction terms in linear regression models. Significant interactions were plotted at the 10th, 25th, 50th, 75th, and 90th percentiles of the moderator variables and Johnson-Neyman estimates were calculated to determine the precise values of the moderators at which the association between the independent and dependent variables became statistically significant.

Results: Negative occurrences for caregivers mediated the association between youth behaviors and externalized caregiving strain (c’: B = .018, p=.059). Higher caregiver age (B = .093, p = .006), biological relationship to youth (B = 1.722, p = .010), more youth behaviors (B = .029, p = .015), more negative occurrences (B = .274, p <.001), and fewer school functioning strengths (B = -.110, p = .028) significantly predicted higher internalized strain (R² = .64). Negative occurrences (B = .143, p <.001) and fewer family involvement strengths (B = -.250, p <.001) significantly predicted higher externalized strain (R² = .44). There were marginally significant interactions between youth behaviors, family involvement, and internalized strain (B = .002, p = .092), as well as between youth behaviors, affective strengths, and internalized strain (B = .003, p = .059), such that the associations between youth behaviors and strain were stronger at higher levels of youth strengths.

Conclusions and Implications: This study provides evidence that negative occurrences, not youth behaviors in and of themselves, are a primary contributor to externalized caregiving strain. We also identified unique and shared predictors of internalized and externalized strain, suggesting that practitioners should consider each outcome separately when working with this population. Finally, we unexpectedly found a stronger link between youth behaviors and internalized strain at higher levels of youth strengths, suggesting that caregivers with a more positive view of their children may be more susceptible to strain from youth behaviors.