文章基本信息

标题：Clustering of Midlife Lifestyle Behaviors and Subsequent Cognitive Function: A Longitudinal Study
本地全文：下载
作者：Emmanuelle Kesse-Guyot ; Valentina A. Andreeva ; Camille Lassale 等
期刊名称：American journal of public health
印刷版ISSN：0090-0036
出版年度：2014
卷号：104
期号：11
页码：e170-e177
DOI：10.2105/AJPH.2014.302121
语种：English
出版社：American Public Health Association
摘要：Objectives. We examined the association between individual and clustered lifestyle behaviors in middle age and later in cognitive functioning. Methods. Middle-aged participants (n = 2430) in the Supplémentation en Vitamines et Minéraux Antioxydant study self-reported their low physical activity, sedentary behavior, alcohol use, smoking, low fruit and vegetable consumption, and low fish consumption. We assessed cognition 13 years later via 6 neuropsychological tests. After standardization, we summed the scores for a composite cognitive measure. We estimated executive functioning and verbal memory scores using principal component analysis. We estimated the mean differences (95% confidence intervals [CIs]) in cognitive performance by the number of unhealthy behaviors using analysis of covariance. We identified latent unhealthy behavior factor via structural equation modeling. Results. Global cognitive function and verbal memory were linearly, negatively associated with the number of unhealthy behaviors: adjusted mean differences = −0.36 (95% CI = −0.69, −0.03) and −0.46 (95% CI = −0.80, −0.11), respectively, per unit increase in the number of unhealthy behaviors. The latent unhealthy behavior factor with low fruit and vegetable consumption and low physical activity as main contributors was associated with reduced verbal memory (RMSEA = 0.02; CFI = 0.96; P = .004). No association was found with executive functioning. Conclusions. Comprehensive public health strategies promoting healthy lifestyles might help deter cognitive aging. Noncommunicable diseases with notable lifestyle components are the leading causes of death worldwide. 1,2 There is also growing evidence of the critical role of different midlife health and risk behaviors in cognitive aging. 3–7 Because lifestyles are inherently modifiable and no treatment of cognitive decline is available, such findings argue for the paramount importance of prevention. 8,9 Current data support a deleterious effect of alcohol abstinence or abuse (compared with moderate alcohol consumption), 10 smoking, 7 low fruit and vegetable intake, 11 low fish intake, 12 and low physical activity (PA) levels 13 on cognitive aging. However, it has been widely documented that lifestyle factors are strongly correlated with each other, forming a cluster of healthy or unhealthy behaviors. 14 Traditionally, such interrelations have been accounted for by statistical adjustment; however, it is of major public health interest to consider the cumulative and combined effect of the various lifestyle behaviors on health by using multidimensional strategies. 14 Research that examines the combined effect of lifestyle factors on mortality is plentiful, and data have been colligated in a recent meta-analysis. 15 These authors reported a 66% reduction in mortality risk by comparing adherence to 4 or more healthy lifestyle behaviors versus engagement in any number of unhealthy behaviors. The combined effect of lifestyle factors has also been explored in relation to cardiovascular diseases, 16–18 cancer, 18–22 diabetes, 18,23 memory complaints, 24 and dementia 25–27 ; however, very few studies have reported findings regarding cognition. 28,29 Despite heterogeneity in the definition of a healthy lifestyle, study design, and residual confounding, available, but scarce, data support a critical, protective role of healthy lifestyles in cognitive health through their beneficial properties via oxidative, inflammatory, vascular, and other neuroprotective pathways. 30–33 Our objectives in this study were to examine the association between individual and clustered lifestyle behaviors and later cognitive functioning. We employed traditional and innovative techniques (structural equation modeling) in our epidemiological pursuit.