Cognitive interventions may improve cognition, delay age-related cognitive declines, and improve quality of life for older adults. The current meta-analysis was conducted to update and expand previous work on the efficacy of cognitive interventions for older adults and to examine the impact of key demographic and methodological variables. EBSCOhost and Embase online databases and reference lists were searched to identify relevant randomized-controlled trials (RCTs) of cognitive interventions for cognitively healthy or mildly impaired (MCI) older adults (60+ years). Interventions trained a single cognitive domain (e.g., memory) or were multi-domain training, and outcomes were assessed immediately post-intervention using standard neuropsychological tests. In total, 279 effects from 97 studies were pooled based on a random-effects model and expressed as Hedges’ g (unbiased). Overall, results indicated that cognitive interventions produce a small, but significant, improvement in the cognitive functioning of older adults, relative to active and passive control groups (g = 0.298, p < .001, 95% CI = 0.248–0.347). These results were confirmed using multi-level analyses adjusting for nesting of effect sizes within studies (g = 0.362, p < .001, 95% CI = 0.275, 0.449). Age, education, and cognitive status (healthy vs. MCI) were not significant moderators. Working memory interventions proved most effective (g = 0.479), though memory, processing speed, and multi-domain interventions also significantly improved cognition. Effects were larger for directly trained outcomes but were also significant for non-trained outcomes (i.e., “transfer effects”). Implications for future research and clinical practice are discussed. This project was pre-registered with PROSPERO (#42016038386).

中文翻译：

---

认知健康，轻度受损和老年人混合样本的认知干预：随机对照试验的系统评价和荟萃分析

认知干预可以改善认知，延缓与年龄有关的认知能力下降，并改善老年人的生活质量。进行了当前的荟萃分析，以更新和扩展有关老年人认知干预功效的先前工作，并检查主要人口统计学和方法学变量的影响。搜索EBSCOhost和Embase在线数据库和参考列表，以识别针对认知健康或轻度障碍（MCI）老年人（60岁以上）的认知干预措施的相关随机对照试验（RCT）。干预训练了单个认知领域（例如，记忆）或进行了多领域训练，干预后立即使用标准的神经心理学测试评估结局。总共，g（无偏）。总体而言，结果表明，相对于主动和被动对照组，认知干预对老年人的认知功能产生了微小但显着的改善（g  = 0.298，p  <.001，95％CI = 0.248-0.347）。这些结果通过调整研究中效应大小嵌套的多级分析得到证实（g =  0.362，p  <.001，95％CI = 0.275，0.449）。年龄，教育程度和认知状态（健康vs. MCI）不是重要的主持人。工作记忆干预被证明是最有效的（g = 0.479），尽管记忆，处理速度和多域干预也显着提高了认知度。对于直接训练的结果，其影响较大，但对于未经训练的结果，其影响也很显着（即“转移效应”）。讨论了对未来研究和临床实践的影响。该项目已预先在PROSPERO（＃42016038386）注册。