Prizer Lindsay Penny, Smith Matthew Lee, Housman Jeff, Ory Marcia G
a Department of Health Promotion and Behavior, College of Public Health , The University of Georgia , Health Science Campus, Wright Hall, Athens , GA , USA.
b Department of Health and Human Performance , Texas State University , San Marcos , TX , USA.
Aging Ment Health. 2016;20(1):13-21. doi: 10.1080/13607863.2015.1021748. Epub 2015 Mar 20.
This study aims to examine the relationship between middle-aged and older adults' depressive symptomology and anti-depressant use and the frequency of falls within the previous 12 months, controlling for sociodemographic variables, health indicators, and health behaviors.
From the 2010 National Social Life, Health, and Aging Project, 2338 cases were examined. Falls were categorized into a binary variable, comparing zero falls with one or more falls. An unadjusted model was run to examine the relationship between independent and dependent variables. Potential covariates were added into the model, and backward elimination was used among independent variables with a univariate P < 0.05 to identify the covariates with the strongest association with falls. This final adjusted binary logistic regression model was then used to examine the relationship between falls and the independent variables.
In the adjusted model, anti-depressant use was positively associated with falls (P = 0.001), as was being female (P < 0.001), having diabetes (P = 0.018), and having increased limitations in daily activities (P < 0.001). The relationship between depressive symptomology and anti-depressant prescription was also significantly associated with falls (P = 0.006).
While findings confirm that a relationship between depressive symptomology and anti-depressant use are associated with falls among middle-aged and older adults, additional studies are needed that simultaneously examine the influence of these two risk factors.
本研究旨在探讨中老年人群抑郁症状与抗抑郁药物使用之间的关系，以及过去12个月内的跌倒频率，并对社会人口统计学变量、健康指标和健康行为进行控制。
从2010年全国社会生活、健康与老龄化项目中选取了2338个案例进行研究。跌倒被分类为一个二元变量，即比较无跌倒与有一次或多次跌倒的情况。运行一个未调整的模型来检验自变量与因变量之间的关系。将潜在的协变量纳入模型，并在单变量P<0.05的自变量中采用向后剔除法，以确定与跌倒关联最强的协变量。然后使用这个最终调整后的二元逻辑回归模型来检验跌倒与自变量之间的关系。
在调整后的模型中，使用抗抑郁药物与跌倒呈正相关（P=0.001），女性（P<0.001）、患有糖尿病（P=0.018）以及日常活动受限增加（P<0.001）的情况也与跌倒呈正相关。抑郁症状与抗抑郁药物处方之间的关系也与跌倒显著相关（P=0.006）。
虽然研究结果证实了抑郁症状和抗抑郁药物使用之间的关系与中老年人群的跌倒有关，但仍需要进一步的研究同时考察这两个风险因素的影响。
