Table 3 Study Characteristics and Findings—Secondary Analyses Testing PS as a Predictor, Moderator, or Mediator of Treatment Response

Becker-Weidman et al. [70]

USA

439 ^a

12–17 (14.6)

MDD

Clinical

1. Fluox

2. CBT

3. Fluox + CBT

4. Placebo

Pred / Mod

Baseline PPO, NPO, AS, ICS, RPS,

Depression severity

(CDRS-R)

NPO, PPO, AS = Pred**

Kennard et al. [71]

USA

166 ^b

12–18 (16.0)

MDD (TR)

Clinical

Switch to:

1. Diff SSRI

2. Diff. SSRI + CBT

3. Ven

4. Ven + CBT

Pred

Receipt of PS training as part of CBT vs. no receipt

Treatment response (CDRS-R & CGI-I)

PS training receipt = Pred*

Kaufman et al. [72]

USA

93 ^a

13–17 (15.1)

MDD & CD

Community

1. CWD-A

2. LST

Med

Δ self-reported PS in specific situations

Depression severity

(BDI-II & HDRS)

No significant effect

Dietz et al. [73]

USA

63 ^a

13–18 (15.6)

MDD

Clinical

1. CBT

2. SBFT

3. NST

Med

Δ observed interpersonal PS interactions between youth and mothers

MDD remission

(K-SADS & BDI).

Δ in PS interactions associated with outcome in CBT** and SFBT* but test for formal mediation not significant