Table 4 Callous unemotional trait outcomes for adolescent self-report measures, Caregiver adolescent-report measures, and Study limitations

Butler et al. [5]

Psychopathic traits decreased pre-post treatment (d = -0.05), however these decreases were not statistically significant between MST and TAU

Psychopathic traits decreased in the MST group pre-post treatment* (d = -0.44)

Unclear process of change

Insufficient power

Sample included fewer chronic and violent offenders than in the US

Merits of the TAU condition not described

Fonagy et al. [20]

Callous unemotional traits decreased pre-post treatment (d =– 0.12), at 52– week follow-up (d =– 0.11), and at 78– week follow-up** (d =– 0.27). Statistically significant decreases were only found at 78-week follow-up for MST

Compared to TAU, MST resulted in decreases in callous unemotional traits pre-post treatment*** (d =– 0.37), pre-52-week follow-up (d =– 0.06), and pre-78-week follow up (d =– 0.07). These differences were only statistically significant pre-post.

For adolescents with low CU traits at baseline: MST was detrimental compared to TAU

For adolescents with high CU traits at baseline: High CU scores at baseline did not moderate the effect of MST

Heterogenous TAU group

MST not as flexible as TAU

Some scales were not internally consistent

Hogan [32]

Psychopathic traits did not significantly change pre-post treatment (d = 0.90)

Treatment developed from theory based on a case study

Exercises not suitable for residential treatment facility

Small sample size (n = 16)

No long-term follow-up

Teacher reports not completed

No random sampling

Group heterogeneity

Findings not generalizable

Lui [44]

Callous unemotional traits decreased in the EPST group pre-post treatment (d =– 0.44), pre-6-week follow-up* (d =– 0.84) and pre-12-week follow-up (d =– 0.49). These decreases were only statistically significant at 6-week follow-up.

Callous unemotional traits did not decrease pre-12-week follow-up* (d = 0.86)^a

Higher self-reported CU traits at baseline were positively correlated with higher self-reported externalizing problems (r = 0.38) and poorer emotional recognition (r = 0.43). Higher self-reported CU traits at baseline were negatively correlated with peer-reported isolation (r = − 0.30), suggesting higher CU traits are associated with greater isolation. Parent reported CU traits were not significantly related to any outcomes.

CU traits did not moderate changes in affective perspective taking, empathy, or externalizing problems.

Small magnitude of change

Small sample size (n = 56)

Predominantly male sample

High attrition

No random assignment

Manders et al. [48]

Callous unemotional traits (d =– 0.21), narcissism (d =– 0.09), and impulsiveness (d =– 0.19) decreased pre-post treatment in the MST group, but none were statistically significant

For adolescents with lower psychopathic traits: MST was more effective than TAU in reducing post-treatment externalizing problems. This finding was consistent across both adolescent self-report and parent report.

For adolescents with higher psychopathic traits: there was no significant differences in the effectiveness of MST versus TAU in reducing externalizing problems. This finding was consistent across both adolescent self-report and parent report.

No adolescent self-report for psychopathic traits

US scale norms cannot be generalized with the sample

Low power

No follow-up data

Muratori et al. [53]

Callous unemotional traits decreased in the treatment group pre-post treatment* (d =– 0.86)

No random assignment

Small sample size (n = 55) Heterogenous sample

Norlander [55]

Psychopathic traits (PCL: YV) (d =– 0.24), and interpersonal (d =– 0.22), affective (d =– 0.34), lifestyle (d =– 0.32), and antisocial features (d =– 0.10) decreased in the CBT group pre-post treatment, but none were statistically significant.

There was a decrease in callous unemotional traits (d =– 0.24), impulsivity conduct problems (d =– 0.12), and psychopathic traits (APSD) (d =– 0.26) in the CBT group, but none were statistically significant.

There was a decrease in psychopathy in the CBT group (SALE), pre-post treatment (d =– 0.12), but it was not statistically significant

Adolescents in the higher psychopathy group showed positive changes in their attitudes towards treatment with moderate increase in Readiness to Change Index scores (d = 0.67). Changes in readiness to change scores were less pronounced in the lower psychopathy group.

Adolescents with higher psychopathy scores had a significant reduction in their psychopathy scores across all PCL-YV subscales (d = 0.35 to 0.69). PCL-YV scores increased among participants in the lower psychopathy group. Despite these increases, the treatment group showed smaller increases compared to the comparison group.

Post-test PCL: YV was not masked to group membership

Adolescents unexpectedly released

Short post-assessment (8 weeks after treatment start)

No follow-up

Small sample size (n = 72)

Sample had lower levels of psychopathy than other populations

Limited generalizability

Thøgersen et al. [65]

The FFT group did not result in larger decreases in callous unemotional traits from pre-post treatment (d = 0.07), but FFT did have a larger decrease pre-78-week follow-up (d =– 0.13).

Note: Statistical significance not reported on separate groups (full-sample analyses)

There was a significant short-term decrease (mean change of– 3.45 scale points; SE = 1.31, p = 0.008) in CU traits for youth who scored above the normative cutoff score. However, the long-term change was not statistically significant, suggesting an immediate reduction in CU traits after treatment, but it was not sustained over the long-term.

No adolescent self-report

Did not include measures for other psychopathy dimensions (narcissism and impulsivity)

Did not take into consideration CU trait typologies

Behavioral problem heterogeneity