Epidemiology of Ageing and Neurodegenerative Diseases, Université de Paris, Inserm U1153, Paris, France
Department of Epidemiology and Public Health, University College London, London, UK.
J Epidemiol Community Health. 2020 Oct;74(10):799-805. doi: 10.1136/jech-2019-213014. Epub 2020 Apr 17.
Personality traits have been liked to cognitive outcomes such as dementia, but whether these associations are robust to the effects of third variables remains the subject of debate. We examined the role of socioeconomic status, depression (history and depressive symptoms), health behaviours and chronic conditions in the association of the big-5 personality traits with cognitive performance, cognitive impairment and incidence of dementia.
Data on 6135 persons (30% women), aged 60-83 years in 2012/13, are drawn from the Whitehall II Study. Participants responded to the 26-item Midlife Development Inventory to assess personality traits (openness, conscientiousness, extraversion, agreeableness and neuroticism), underwent cognitive testing in 2012/13 and 2015/16 and were followed for incidence of dementia (N=231) until 2019.
Logistic regression, adjusted for sociodemographic factors, suggested a cross-sectional association with cognitive impairment for four of the five traits but only neuroticism was associated with incident cognitive impairment. All associations were completely attenuated when the analyses were adjusted for depression. Cox regression (mean follow-up: 6.18 years) adjusted for sociodemographic variables showed higher conscientiousness (HR per SD increment=0.72; 95% CI 0.65 to 0.81) and extraversion (HR=0.85; 95% CI 0.75 to 0.97) to be associated with lower dementia risk; higher neuroticism (HR=1.32; 95% CI 1.17 to 1.49) was associated with increased risk. Further adjustment for depression led to only conscientiousness retaining an association with dementia (HR=0.81; 95% CI 0.69 to 0.96), which was robust to adjustment for all covariates (HR=0.84; 95% CI 0.71 to 0.91; P=0.001).
Our results show that only conscientiousness has an association with incidence of dementia that is not attributable to socioeconomic status or depression. The association of neuroticism with dementia was explained by depression.
人格特质与认知结果（如痴呆）有关，但这些关联是否能抵抗第三变量的影响仍存在争议。我们研究了社会经济地位、抑郁（病史和抑郁症状）、健康行为和慢性疾病在大五人格特质与认知表现、认知障碍和痴呆发生率之间的关联中的作用。
数据来自 2012/13 年参加 Whitehall II 研究的 6135 名年龄在 60-83 岁的人（30%为女性）。参与者用 26 项中年发展量表（MIDAS）来评估人格特质（开放性、尽责性、外向性、宜人性和神经质），并在 2012/13 年和 2015/16 年进行认知测试，随访至 2019 年痴呆的发生情况（N=231）。
经过社会人口统计学因素的调整，逻辑回归显示，五个特质中的四个与认知障碍有横断面关联，但只有神经质与认知障碍的发生有关。当分析调整为抑郁时，所有的关联都完全减弱。调整社会人口统计学变量的 Cox 回归（平均随访时间：6.18 年）显示，尽责性更高（每标准差增加的 HR=0.72；95%CI 0.65 至 0.81）和外向性（HR=0.85；95%CI 0.75 至 0.97）与痴呆风险降低相关；神经质更高（HR=1.32；95%CI 1.17 至 1.49）与风险增加相关。进一步调整抑郁，只有尽责性与痴呆仍有关联（HR=0.81；95%CI 0.69 至 0.96），这在调整所有协变量后仍然稳健（HR=0.84；95%CI 0.71 至 0.91；P=0.001）。
我们的结果表明，只有尽责性与痴呆的发生有关，而且这种关联不能归因于社会经济地位或抑郁。神经质与痴呆的关联可以用抑郁来解释。
