Coverage, Comprehension, and Engagement: Intervention Fidelity and Six-Month Relational Outcomes for Sibling Dyads in Foster Care

Background

This study sought to determine if intervention fidelity moderated improvements in sibling relationship quality for participants of a randomized controlled relationship enhancement intervention trial for sibling dyads in foster care. The research questions guiding this investigation were:

1. Are individual, sibling dyad, and placement characteristics associated with intervention fidelity?
2. Is intervention fidelity associated with changes to sibling relationship quality over a six-month period of time?

Methods

160 youth nested within 80 dyads from the RCT treatment condition provided the study data. The sample was comprised of slightly greater number of female (51.8%) and non-white (51.2%) participants. Average youth age was 12 (sd=2.0) years, dyads were separated an average age range of 2.4 (sd=1.0) years. Over half of dyads were the same gender (58.3%). At baseline siblings were primarily living together (70.2%) and in non-relative care (40.7%). Placement changes during observation occurred infrequently (x[bar]=.25, sd=.50, range=2).

Fidelity was assessed following each intervention session. Interventionists completed a fidelity form which assessed coverage, youth’s engagement, and comprehension of content across seven intervention domains. Items were rated on a six-point scale from 0=no coverage/engagement/comprehension to 5=complete coverage/engagement/comprehension. An average fidelity score for each dimension was created across total intervention sessions attended.

Relationship quality was assessed using the MAC-SRQ, a multiple-method, multiple informant construct (Authors, 2017). Changes to MAC-SRQ were modeled by subtracting baseline scores from term scores. Individual and dyad characteristics were collected at baseline assessment, placement changes were recorded as they occurred. A correlation matrix was computed to answer research question #1,  multilevel modeling procedures were employed to answer question #2.

Results

Bivariate associations suggest differential comprehension and engagement with intervention content based on sibling living situation, youth age, and experiences with placement change.  Sibling separation was associated with decreased comprehension of activity planning (r=-.26, p<.01), managing feelings (r=-2.16, p<.05), and cooperation (r=-.14, p=.08) content. Older youth age was associated with increased comprehension of managing feelings (r=.21, p<.01) and building adult allies (r=.14, p=.08), but negatively associated with problem solving content (r=-.15, p=.05). Placement change was associated with increased engagement with managing feelings content (r=.18, p<.05) but negatively associated with total intervention coverage.

HLM results identified an association between the sibling dyad position, content comprehension, and changes to sibling relationship quality. Specifically, differential comprehension of cooperation (YS: Y₀₂=-1.57, SE=.48, p<.01), and managing feelings (OS: Y₀₂=-1.01, SE=.52, p=.05, YS: Y₀₂=1.08, SE=.48, p<.05) was associated with changes to MAC-SRQ. The dyad accounted for 60% of variance in this outcome.

Discussion

Content comprehension was one of the primary operating mechanisms through which the intervention influenced MAC-SRQ, although type and direction of associations differed for older and younger siblings. This may be a key consideration for the development and implementation of dyadic interventions for siblings in foster care.