University of Michigan Hospitals, Ann Arbor, MI 48109-5485, USA.
Ann Epidemiol. 2010 Jul;20(7):539-46. doi: 10.1016/j.annepidem.2010.03.020.
In this study we examined spousal concordance for two aggregate measures of health risk status and compliance with preventive service recommendations among 9620 pairs of cohabitating, opposite-sex married couples.
Health risk appraisals were the primary data source to measure two outcome variables. Health risk status was compiled from 12 health risks and categorized into three levels (low-, medium-, and high-risk status). Overall preventive service compliance status was estimated by seven age-sex specific preventive service recommendations and dichotomized into lower and higher compliance status. For each of the husband and wife populations, we conducted proportional odds models and logistic regression models to assess spousal concordance for the two aggregate measures respectively. All models were adjusted for household income, one's characteristics (age, race, education, disease burden), and the same set of characteristics and the corresponding outcome variable from the spouse.
A positive correlation within spousal pairs was statistically significant for both health risk status and compliance status (p < .001) based on multivariate modeling. The odds ratios were similar in magnitude for the two spouse populations.
The analyses showed spousal concordance for aggregate measures of health behaviors. This study also provides some evidence for dominance of husband's education.
本研究通过对 9620 对同居的异性已婚夫妇进行调查，考察了两种健康风险状况综合指标以及对预防服务建议的依从性在配偶间的一致性。
健康风险评估是衡量两个结果变量的主要数据来源。健康风险状况由 12 种健康风险因素编制而成，并分为低、中、高风险状况三个等级。整体预防服务的依从性状况由七项特定于年龄和性别的预防服务建议来评估，并分为较低和较高的依从性状况。对于每对丈夫和妻子，我们分别进行了比例优势模型和逻辑回归模型来评估这两种综合指标的配偶间一致性。所有模型都调整了家庭收入、个体特征（年龄、种族、教育程度、疾病负担）以及配偶的相同特征和相应的结果变量。
基于多变量建模，配偶对之间在健康风险状况和依从性状况上的相关性具有统计学意义（p <.001）。对于两个配偶群体，优势比的大小相似。
分析表明，健康行为的综合指标存在配偶间的一致性。本研究还提供了一些证据表明丈夫的教育具有主导地位。
