Background: Observational studies consistently have identified that social support is negatively associated
with depression among adolescents. The causal connections between these two factors, however, are not
well understood. Does the relationship between social support and depression indicate that social support
protects against depression, or that characteristics of these less depressed adolescents make it easier for
them to obtain social support? To address this, this study examines whether social support predicts later
depressive symptoms, controlling for earlier clinical presentation.
Methods: The sample comprised adolescents (n=777) who were assessed as part of the National Survey of
Child and Adolescent Well-Being. Caregivers and youth provided information about children's clinical
characteristics and family background. Propensity scoring was used to adjust for baseline covariates when
examining the impact of social support at eighteen months post-baseline on depressive symptoms at three
years post-baseline.
Results: Children who received more social support at eighteen months post-baseline had fewer depressive
symptoms at three years post-baseline, even after adjusting for baseline covariates using the propensity
score; however, the adjusted effect size was small.
Conclusion: Previous attempts to study the impact of social support without adjusting for possible
confounders have likely inflated the independently predictive role of social support.