Melodic Intonation Therapy (MIT) is a prominent rehabilitation program for individuals with post-stroke aphasia. Our meta-analysis investigated the efficacy of MIT while considering quality of outcomes, experimental design, influence of spontaneous recovery, MIT protocol variant, and level of generalization. Extensive literature search identified 606 studies in major databases and trial registers; of those, 22 studies—overall 129 participants—met all eligibility criteria. Multi-level mixed- and random-effects models served to separately meta-analyze randomized controlled trial (RCT) and non-RCT data. RCT evidence on validated outcomes revealed a small-to-moderate standardized effect in noncommunicative language expression for MIT—with substantial uncertainty. Unvalidated outcomes attenuated MIT's effect size compared to validated tests. MIT's effect size was 5.7 times larger for non-RCT data compared to RCT data (g̅_{case report} = 2.01 vs. g̅_RCT = 0.35 for validated Non-Communicative Language Expression measures). Effect size for non-RCT data decreased with number of months post-stroke, suggesting confound through spontaneous recovery. Deviation from the original MIT protocol did not systematically alter benefit from treatment. Progress on validated tests arose mainly from gains in repetition tasks rather than other domains of verbal expression, such as everyday communication ability. Our results confirm the promising role of MIT in improving trained and untrained performance on unvalidated outcomes, alongside validated repetition tasks, and highlight possible limitations in promoting everyday communication ability.

中文翻译：

---

失语症的旋律语调治疗：随机对照试验和个体参与者数据的多层次荟萃分析

旋律语调疗法 (MIT) 是一项针对中风后失语症患者的重要康复计划。我们的荟萃分析调查了 MIT 的功效，同时考虑了结果质量、实验设计、自发恢复的影响、MIT 协议变体和泛化水平。广泛的文献检索在主要数据库和试验登记库中确定了 606 项研究；其中，22 项研究（总共 129 名参与者）符合所有资格标准。多级混合和随机效应模型用于分别对随机对照试验 (RCT) 和非 RCT 数据进行荟萃分析。关于经过验证的结果的 RCT 证据表明，麻省理工学院的非交流语言表达具有小到中等的标准化效果——具有很大的不确定性。与经过验证的测试相比，未经验证的结果削弱了麻省理工学院的效应量。g̅_案例报告= 2.01 vs. g̅ _RCT = 0.35 对于经过验证的非交流语言表达措施）。非 RCT 数据的效应量随着中风后月数的增加而减少，表明自发恢复造成混淆。偏离最初的麻省理工学院协议并没有系统地改变治疗的益处。验证测试的进展主要来自重复任务的收益，而不是其他语言表达领域，例如日常沟通能力。我们的结果证实了麻省理工学院在提高训练和未训练的未验证结果表现方面的有前途的作用，以及经过验证的重复任务，并强调了在提升日常沟通能力方面可能存在的局限性。