Malhotra Rahul, Chei Choy-Lye, Østbye Truls, Chan Angelique, Matchar David B
Program in Health Services and Systems Research, Duke-National University of Singapore Graduate Medical School, 8 College Road, Singapore 169857, Singapore.
Program in Health Services and Systems Research, Duke-National University of Singapore Graduate Medical School, 8 College Road, Singapore 169857, Singapore.
Arch Gerontol Geriatr. 2014 Mar-Apr;58(2):269-77. doi: 10.1016/j.archger.2013.10.004. Epub 2013 Oct 24.
We assess for the mediation of the association between older person cognitive impairment and caregiver depressive symptoms through older person BPS and functional limitations, and whether the mediation varies by caregiver-older person relationship (spouse/adult child). Data for 1111 older person (aged 75+ with activity of daily living (ADL) limitation)-caregiver dyads from Singapore were used. The outcome variable was dichotomous (caregiver clinically significant depressive symptoms [CSDS]: yes/no) in the primary analysis and continuous (caregiver depressive symptoms score) in the sensitivity analysis. The causal steps approach assessed for the mediation of the association between older person cognitive impairment (yes/no) and the outcome variable through the two potential mediators. A bootstrapping approach calculated point estimates and confidence intervals (CIs) of the indirect (∼mediated) effects. Variation of the indirect effects by caregiver-older person relationship was also assessed. In the primary analysis, the causal steps approach supported older person BPS and functional limitations as mediators. The bootstrapping approach confirmed both as significant mediators, though BPS (indirect effect odds ratio (OR) 1.32 [95% bootstrap CI 1.19,1.48]; %mediation: 70.6%) was a stronger mediator than functional limitations (1.04 [1.01,1.11]; %mediation: 11.5%). Variation of the indirect effects by caregiver-older person relationship was not supported. Results of the sensitivity analysis confirmed these results. We conclude that while caring for an older person with cognitive impairment is detrimental for the caregiver's mood, management of associated BPS and functional limitations, especially the former, among such older persons may reduce depressive symptoms among their caregivers. Spouse as well as adult child caregivers benefit.
我们通过老年人的行为和心理症状（BPS）及功能受限情况，评估老年人认知障碍与照料者抑郁症状之间关联的中介作用，以及这种中介作用是否因照料者与老年人的关系（配偶/成年子女）而异。我们使用了来自新加坡的1111对老年人（年龄在75岁及以上且有日常生活活动（ADL）受限）-照料者二元组的数据。在主要分析中，结果变量是二分变量（照料者具有临床显著抑郁症状[CSDS]：是/否），在敏感性分析中是连续变量（照料者抑郁症状评分）。因果步骤法通过两个潜在中介变量评估老年人认知障碍（是/否）与结果变量之间关联的中介作用。自抽样法计算间接（~中介）效应的点估计值和置信区间（CI）。还评估了间接效应因照料者与老年人关系的差异。在主要分析中，因果步骤法支持将老年人的BPS和功能受限作为中介变量。自抽样法证实二者均为显著中介变量，不过BPS（间接效应比值比（OR）为1.32[95%自抽样CI为1.19,1.48]；中介作用百分比：70.6%）比功能受限（1.04[1.01,1.11]；中介作用百分比：11.5%）的中介作用更强。不支持间接效应因照料者与老年人关系的差异。敏感性分析结果证实了这些结果。我们得出结论，虽然照料认知障碍老年人对照料者的情绪有害，但管理此类老年人的相关BPS和功能受限情况，尤其是前者，可能会减轻其照料者的抑郁症状。配偶以及成年子女照料者均能从中受益。
