使用结构方程模型分析多种生物标志物
Analysis of Multiple Biomarkers Using Structural Equation Modeling.
作者信息
Cao Wenhao, Hecht Stephen S, Murphy Sharon E, Chu Haitao, Benowitz Neal L, Donny Eric C, Hatsukami Dorothy K, Luo Xianghua
机构信息
Wenhao Cao, Master of Science Student, Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN. Stephen S. Hecht, Professor, Masonic Cancer Center, University of Minnesota, Minneapolis, MN. Sharon E. Murphy, Professor, Masonic Cancer Center, University of Minnesota, Minneapolis, MN. Haitao Chu, Professor, Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN. Neal L. Benowitz, Professor, University of California, Department of Medicine, San Francisco, CA. Eric C. Donny, Professor, Wake Forest School of Medicine, Department of Physiology and Pharmacology, Winston-Salem, NC. Dorothy K. Hatsukami, Professor, Masonic Cancer Center and Department of Psychiatry, University of Minnesota, Minneapolis, MN. Xianghua Luo, Associate Professor, Division of Biostatistics School of Public Health and Masonic Cancer Center, University of Minnesota, Minneapolis, MN.
出版信息
Tob Regul Sci. 2020 Jul;6(4):266-278. doi: 10.18001/trs.6.4.4.
OBJECTIVES
When examining the relationship between smoking intensity and toxicant exposure biomarkers in an effort to understand the potential risk for smoking-related disease, individual biomarkers may not be strongly associated with smoking intensity because of the inherent variability in biomarkers. Structural equation modeling (SEM) offers a powerful solution by modeling the relationship between smoking intensity and multiple biomarkers through a latent variable.
METHODS
Baseline data from a randomized trial (N = 1250) were used to estimate the relationship between smoking intensity and a latent toxicant exposure variable summarizing five volatile organic compound biomarkers. Two variables of smoking intensity were analyzed: the self-report cigarettes smoked per day and total nicotine equivalents in urine. SEM was compared with linear regression with each biomarker analyzed individually or with the sum score of the five biomarkers.
RESULTS
SEM models showed strong relationships between smoking intensity and the latent toxicant exposure variable, and the relationship was stronger than its counterparts in linear regression with each biomarker analyzed separately or with the sum score.
CONCLUSIONS
SEM is a powerful multivariate statistical method for studying multiple biomarkers assessing the same class of harmful constituents. This method could be used to evaluate exposure from different combusted tobacco products.
目的
在研究吸烟强度与毒物暴露生物标志物之间的关系以了解吸烟相关疾病的潜在风险时，由于生物标志物存在内在变异性，单个生物标志物可能与吸烟强度没有很强的关联。结构方程模型（SEM）通过一个潜在变量对吸烟强度与多个生物标志物之间的关系进行建模，提供了一个有力的解决方案。
方法
来自一项随机试验（N = 1250）的基线数据用于估计吸烟强度与一个汇总五种挥发性有机化合物生物标志物的潜在毒物暴露变量之间的关系。分析了两个吸烟强度变量：自我报告的每日吸烟量和尿液中的总尼古丁当量。将结构方程模型与对每个生物标志物单独分析或对五个生物标志物的总分进行分析的线性回归进行比较。
结果
结构方程模型显示吸烟强度与潜在毒物暴露变量之间存在很强的关系，并且这种关系比分别对每个生物标志物或对总分进行线性回归时的对应关系更强。
结论
结构方程模型是一种用于研究评估同一类有害成分的多个生物标志物的强大多变量统计方法。该方法可用于评估来自不同燃烧烟草制品的暴露情况。
Zotero 插件