文献速递慢病防控与筛查卫生政策与健康促进

行为危险因素监测系统与其他全国健康调查的估计值比较(2011—2016)

Comparisons of Estimates From the Behavioral Risk Factor Surveillance System and Other National Health Surveys, 2011-2016.
PubMed 文献 · Am J Prev Med · 2020-01-01

该研究比较了2011—2016年BRFSS与NHANES或NHIS的10个关键常用变量,发现BRFSS与NHANES在9个变量中有8个的患病率或均值时间趋势相似,BRFSS与NHIS在5个变量的趋势比较中未发现调查与时间段的交互作用,尽管许多估计值存在统计学显著差异,但跨年度的绝对差异较小,从公共卫生监测角度看多数无实际意义。

要点

  • BRFSS与NHANES在9个变量中有8个的患病率或均值线性时间趋势相似。
  • BRFSS与NHANES在自报健康状况一般/差、曾诊断糖尿病、曾诊断高血压的患病率上无显著差异。
  • 尽管许多估计值存在统计学显著差异,但跨年度绝对差异较小,从公共卫生监测角度多无实际意义。

核心结论

  • BRFSS与NHANES、NHIS的患病率或均值时间趋势大多相似,尽管许多估计值存在统计学显著差异,但绝对差异多数较小且从公共卫生监测角度看无实际意义。

研究设计与样本

  • 研究比较了2011—2016年BRFSS与NHANES或NHIS的10个关键且广泛使用的变量,数据分析于2018年进行。

主要发现

  • BRFSS与NHANES在9个研究变量中有8个的患病率或均值线性时间趋势相似。
  • BRFSS与NHANES在自报健康状况一般/差、曾被告知患糖尿病、曾被告知患高血压的患病率上无显著差异。
  • BRFSS与NHIS的趋势比较中,当前吸烟、自报健康状况一般/差、曾被告知患糖尿病、自报身高和体重5个变量未发现调查与时间段的交互作用。
  • 尽管BRFSS与NHANES或NHIS之间许多估计值存在显著差异,但跨年度的绝对差异较小。

临床怎么用

  • BRFSS的时间趋势可用于公共卫生监测,与基准调查的估计值差异在多数情况下较小且无实际意义。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

INTRODUCTION: The Behavioral Risk Factor Surveillance System (BRFSS) is composed of telephone surveys that collect state data from non-institutionalized U.S. adults regarding health-related risk behaviors and chronic health conditions. A new design was implemented in 2011 to include participants on cellular telephones. It is important to validate estimates since 2011. METHODS: A total of 10 key and widely used variables between BRFSS and the National Health and Nutrition Examination Survey (NHANES) or National Health Interview Survey (NHIS) in 2011-2016 were compared. Data analysis was conducted in 2018. RESULTS: Between BRFSS and NHANES, similar linear time trends of prevalences or means were found for 8 of 9 studied variables. There were no significant differences in the prevalences of the following variables: self-reported fair/poor health, ever told have diabetes, and ever told to have hypertension. In trend comparison of BRFSS versus NHIS, interactions of prevalence between survey and time period were not found for 5 variables: current smoking, self-reported fair/poor health, ever told have diabetes, and self-reported height and weight. Although there were significant differences in many estimates between BRFSS and either NHANES or NHIS, the absolute differences across years were rather small. CONCLUSIONS: Comparing BRFSS time trends with those of 2 national benchmark surveys in 10 key and widely used variables suggests that the trends of prevalences (or means) from BRFSS, NHANES, and NHIS are mostly similar. For many variables, despite statistically significant differences in the prevalences (or means) between surveys, absolute differences in most cases were small and not meaningful from a public health surveillance perspective.

来源:Am J Prev Med,查看原文。版权归原作者所有。

引用

  1. Hsia J, Zhao G, Town M, et al. Comparisons of Estimates From the Behavioral Risk Factor Surveillance System and Other National Health Surveys, 2011-2016. Am J Prev Med. 2020. PMID: 32444008. doi:10.1016/j.amepre.2020.01.025

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。