Background
Children with callous-unemotional (CU) traits are at high lifetime risk of antisocial behavior. It is unknown if treatments for disruptive behavior disorders are as effective for children with CU traits (DBD+CU) as those without (DBD-only), nor if treatments directly reduce CU traits. Separate multilevel meta-analyses were conducted to compare treatment effects on DBD symptoms for DBD+CU versus DBD-only children and evaluate direct treatment-related reductions in CU traits, as well as to examine moderating factors for both questions.
Methods
We systematically searched PsycINFO, PubMed, Cochran Library (Trials), EMBASE, MEDLINE, APA PsycNet, Scopus, and Web of Science. Eligible studies were randomized controlled trials, controlled trials, and uncontrolled studies evaluating child-focused, parenting-focused, pharmacological, family-focused, or multimodal treatments.
Results
Sixty studies with 9,405 participants were included (M
age = 10.04, SD
age = 3.89 years, 25.09% female, 44.10% racial/ethnic minority). First, treatment was associated with similar reductions in DBD symptoms for DBD+CU (SMD = 1.08, 95% CI = 0.45, 1.72) and DBD-only (SMD = 1.01, 95% CI = 0.38, 1.64). However, DBD+CU started (SMD = 1.18, 95% CI = 0.57, 1.80) and ended (SMD = 0.73, p < .001; 95% CI = 0.43, 1.04) treatment with more DBD symptoms. Second, although there was no overall direct effect of treatment on CU traits (SMD = .09, 95% CI = −0.02, 0.20), there were moderating factors. Significant treatment-related reductions in CU traits were found for studies testing parenting-focused components (SMD = 0.21, 95% CI = 0.06, 0.35), using parent-reported measures (SMD = 0.16, 95% CI = 0.04, 0.28), rated as higher quality (SMD = 0.26, 95% CI = 0.13, 0.39), conducted outside the United States (SMD = 0.19, 95% CI = 0.05, 0.32), and with less than half the sample from a racial/ethnic minority group (SMD = 0.15, 95% CI = 0.002, 0.30).
Conclusions
DBD+CU children improve with treatment, but their greater DBD symptom severity requires specialized treatment modules that could be implemented alongside parenting programs. Conclusions are tempered by heterogeneity across studies and scant evidence from randomized controlled trials.