背景:出生队列研究提供了一个机会,通过前瞻性收集环境样本来进一步了解环境因素对儿童健康和发育的影响。
方法:我们在国家儿童研究的初始试验阶段评估了环境样本收集方法的可行性和信息价值,计划中的美国环境出生队列研究。从2009年1月至2010年9月,在多达三次家访中收集了环境样本:孕前(n=306),怀孕(n=807),和产后6个月(n=117)。收集物包括颗粒物质≤2.5µm(PM2.5)的空气,二氧化氮,臭氧,挥发性有机化合物(VOCs),和羰基化合物;用于过敏原/内毒素的真空粉尘;用于挥发性有机化合物的水,三卤甲烷(THMs),和卤代乙酸(HAAs);擦拭农药样品,半挥发性有机物,和金属。我们使用样品收集率和时间以及使用分析物检测频率(DF)的信息价值来表征可行性。
结果:在1230次家庭访问中,所有样本类型的环境样本收集率都很高(平均值=89%);除空气PM2.5样本外,所有样本的收集时间均<30分钟.水VOC(中位数DF=0%)和农药地板湿巾(中位数DF=5%)的信息值较低。空气样品的信息值适中(DF中位数=35%),水THM和HAAs的信息值较高(DF中位数=91%和75%,分别)。
结论:虽然采集环境样本是可行的,一些样本(例如,擦拭农药和水VOC)产生的信息有限。这些结果可以与其他研究设计考虑因素结合使用,例如目标人口规模和感兴趣的假设,为未来环境健康出生队列研究的方法选择提供信息。
BACKGROUND: Birth cohort studies provide the opportunity to advance understanding of the impact of environmental factors on childhood health and development through prospective collection of environmental samples.
METHODS: We evaluated the feasibility and informative value of the environmental sample collection methodology in the initial pilot phase of the National Children\'s
Study, a planned U.S. environmental birth cohort
study. Environmental samples were collected from January 2009-September 2010 at up to three home visits: pre-pregnancy (n=306), pregnancy (n=807), and 6-months postnatal (n=117). Collections included air for particulate matter ≤2.5 µm (PM2.5), nitrogen dioxide, ozone, volatile organic compounds (VOCs), and carbonyls; vacuum dust for allergens/endotoxin; water for VOCs, trihalomethanes (THMs), and haloacetic acids (HAAs); and wipe samples for pesticides, semi-volatile organics, and metals. We characterized feasibility using sample collection rates and times and informative value using analyte detection frequencies (DF).
RESULTS: Among the 1230 home visits, environmental sample collection rates were high across all sample types (mean=89%); all samples except the air PM2.5 samples had collection times <30 min. Informative value was low for water VOCs (median DF=0%) and pesticide floor wipes (median DF=5%). Informative value was moderate for air samples (median DF=35%) and high for water THMs and HAAs (median DF=91% and 75%, respectively).
CONCLUSIONS: Though collection of environmental samples was feasible, some samples (e.g., wipe pesticides and water VOCs) yielded limited information. These results can be used in conjunction with other
study design considerations, such as target population size and hypotheses of interest, to inform the method selection of future environmental health birth cohort studies.