Centre for Mental Health Research, The Australian National University, Building 63 Eggleston Road, Canberra, Australian Capital Territory 0200, Australia.
Age Ageing. 2014 Jan;43(1):84-90. doi: 10.1093/ageing/aft116. Epub 2013 Aug 8.
very few studies have examined the association between intra-individual reaction time variability and subsequent mortality. Furthermore, the ability of simple measures of variability to predict mortality has not been compared with more complex measures.
a prospective cohort study of 896 community-based Australian adults aged 70+ were interviewed up to four times from 1990 to 2002, with vital status assessed until June 2007. From this cohort, 770-790 participants were included in Cox proportional hazards regression models of survival. Vital status and time in study were used to conduct survival analyses. The mean reaction time and three measures of intra-individual reaction time variability were calculated separately across 20 trials of simple and choice reaction time tasks. Models were adjusted for a range of demographic, physical health and mental health measures.
greater intra-individual simple reaction time variability, as assessed by the raw standard deviation (raw SD), coefficient of variation (CV) or the intra-individual standard deviation (ISD), was strongly associated with an increased hazard of all-cause mortality in adjusted Cox regression models. The mean reaction time had no significant association with mortality.
intra-individual variability in simple reaction time appears to have a robust association with mortality over 17 years. Health professionals such as neuropsychologists may benefit in their detection of neuropathology by supplementing neuropsychiatric testing with the straightforward process of testing simple reaction time and calculating raw SD or CV.
很少有研究探讨个体内反应时变异性与随后死亡率之间的关系。此外，简单的变异性衡量指标预测死亡率的能力尚未与更复杂的衡量指标进行比较。
对 896 名 70 岁以上的澳大利亚社区成年人进行了前瞻性队列研究，从 1990 年到 2002 年，最多进行了四次访谈，截至 2007 年 6 月评估了存活状态。从该队列中，有 770-790 名参与者被纳入 Cox 比例风险回归模型中进行生存分析。使用存活状态和研究时间进行生存分析。在简单和选择反应时任务的 20 次试验中，分别计算了个体内反应时的平均反应时和三个个体内反应时变异性衡量指标。模型调整了一系列人口统计学、身体健康和心理健康指标。
简单反应时的个体内变异性越大，如原始标准差（raw SD）、变异系数（CV）或个体内标准差（ISD）评估，在调整后的 Cox 回归模型中与全因死亡率的风险增加强烈相关。平均反应时与死亡率没有显著关联。
简单反应时的个体内变异性与 17 年以上的死亡率之间似乎存在稳健的关联。神经心理学家等健康专业人员可能会受益于通过简单反应时测试和计算原始 SD 或 CV 来补充神经心理测试，从而更轻松地检测神经病理学。
