Differences of TF-CBT treatment effects using various outcome measures: a meta-analysis

<b>Background:</b> Diagnostic criteria of posttraumatic stress disorder in children and adolescents and corresponding instruments have undergone significant changes over time. However, the impact of different outcome measures on treatment effects in the context of posttraumatic stress symptoms (PTSS) has not yet been explored. <b>Objective:</b> TF-CBT is a well-researched first-line treatment for PTSS among children and adolescents and thus, an ideal candidate to examine the potential influence of different outcome measures by meta-analysis. <b>Method:</b> A comprehensive literature search was conducted in December 2023 using seven databases. Studies included RCTs as well as non-controlled studies examining the effects of TF-CBT on pediatric PTSS. We extracted treatment effects and investigated whether there were systematic differences in the effects based on the outcome measures and their underlying DSM version. <b>Results:</b> In total, 76 studies (35 RCTS) met the eligibility criteria. Hedges g effect sizes with 95% confidence intervals (CI) were computed and high-risk of bias studies were excluded. No significant difference was observed between DSM-IV and DSM-5 based instruments. Individual outcome measures were found to be comparable overall, with some appearing somewhat more sensitive to change. Although a small but significant difference in true effect sizes for individual outcome measures was found, this only concerned the UCLA PTSD (<i>g</i> = 1.06) and the CPSS (<i>g</i> = 1.61) with the effect most likely being due to chance or confounding variables. TF-CBT showed large effect sizes on PTSS in within-study comparison (<i>g</i> = 1.32) and medium between-studies effect sizes (<i>g</i> = .57). <b>Conclusions:</b> While we could not establish equivalence, there seems to be no difference regarding the measurement of treatment effects based on outcome measure and underlying DSM version. The updated TF-CBT effect size confirmed it as an effective treatment for PTSS and secondary outcomes in children and adolescents. No difference between outcome measures for posttraumatic stress symptoms in children and adolescents and their underlying DSM-criteria could be established.TF-CBT has again been confirmed TF-CBT as a treatment of first choice for PTSS in children and adolescents. No difference between outcome measures for posttraumatic stress symptoms in children and adolescents and their underlying DSM-criteria could be established. TF-CBT has again been confirmed TF-CBT as a treatment of first choice for PTSS in children and adolescents.

<b>背景：</b>儿童青少年创伤后应激障碍（posttraumatic stress disorder, PTSD）的诊断标准及其配套评估工具随时间推移发生了显著变化。然而，在创伤后应激症状（posttraumatic stress symptoms, PTSS）的研究语境中，不同结局测量工具对治疗效应的影响尚未得到探索。<b>目的：</b>创伤聚焦认知行为疗法（Trauma-Focused Cognitive Behavioral Therapy, TF-CBT）是针对儿童青少年PTSS的一线循证治疗方案，因此是通过元分析检验不同结局测量工具潜在影响的理想研究对象。<b>方法：</b>本研究于2023年12月针对7个数据库开展了全面的文献检索。纳入研究涵盖检验TF-CBT对儿童青少年PTSS疗效的随机对照试验（randomized controlled trial, RCT）与非对照研究。我们提取了治疗效应数据，并基于结局测量工具及其所依据的《精神障碍诊断与统计手册》（Diagnostic and Statistical Manual of Mental Disorders, DSM）版本，分析效应值是否存在系统性差异。<b>结果：</b>最终共有76项研究（含35项RCT）符合纳入标准。本研究计算了带有95%置信区间（confidence interval, CI）的赫奇斯g（Hedges g）效应量，并排除了偏倚风险较高的研究。基于DSM-IV与DSM-5的评估工具之间未观察到显著差异。整体而言，各类单一结局测量工具的表现具有可比性，部分工具对症状变化的敏感性略高。尽管部分单一结局测量工具的真实效应量存在微小但显著的差异，但仅涉及加州大学洛杉矶分校创伤后应激障碍量表（UCLA PTSD）（g=1.06）与儿童创伤后应激障碍量表（CPSS）（g=1.61），该差异极有可能由偶然因素或混淆变量导致。在研究内比较中，TF-CBT对PTSS的效应量较大（g=1.32）；而研究间效应量则为中等水平（g=0.57）。<b>结论：</b>尽管本研究未证实评估工具间的等效性，但基于结局测量工具及其所依据的DSM版本的治疗效应测量并未表现出显著差异。更新后的TF-CBT效应量证实，其作为儿童青少年PTSS及继发症状的有效治疗方案具有可靠性。本研究未发现儿童青少年PTSS的结局测量工具与其所依据的DSM诊断标准之间存在差异，再次证实TF-CBT是儿童青少年PTSS的一线治疗方案。