Simultaneous modeling of detection rate and exposure concentration using semi-continuous models to identify exposure determinants when left-censored data may be a true zero.
Simultaneous modeling of detection rate and exposure concentration using semi-continuous models to identify exposure determinants when left-censored data may be a true zero.
复制标题
DOI：
10.1038/s41370-021-00331-7
复制
发表时间：
2021-11
影响因子：
4.5
通讯作者：
Zhu B
中科院分区：
文献类型：
--
作者：
Friesen MC;Choo-Wosoba H;Sarazin P;Hwang J;Dopart P;Russ DE;Deziel NC;Lavoué J;Albert PS;Zhu B
大多数处理左删减数据的方法都假定分析物存在但未被量化。如果分析物不存在，则可能导致有偏差的估计，使得未观察到的数据代表混合暴露分布，其中未知比例聚集在零处。我们使用半连续模型来识别52,457个OSHA检查铅样结果的时间和行业趋势。半连续模型的第一个分量预测检测浓度≥0.007 mg/m3的概率（最高估计检出限，62%的测量值）。第二个分量预测测量值的中位数浓度≥0.007 mg/m3。两个组成部分都包括行业的随机效应和年份、行业组、分析方法和其他变量的固定效应。我们使用这两个组成部分一起预测中位数行业和时间特定的铅浓度。检测到浓度的概率和检测到浓度的中位数逐年下降；对于分析多种金属（相对于一种金属）和通过电感耦合等离子体（相对于原子吸收光谱）分析的测量值，两者也都较低。协方差为0.30（标准误差=0.06），证实两成分相关。我们确定了60%以上左删减数据中暴露的决定因素，同时考虑了相关关系，而没有假设删减数据的分布。
Most methods for treating left-censored data assume the analyte is present but not quantified. Biased estimates may result if the analyte is absent such that the unobserved data represents a mixed exposure distribution with an unknown proportion clustered at zero. We used semi-continuous models to identify time and industry trends in 52,457 OSHA inspection lead sample results. The first component of the semi-continuous model predicted the probability of detecting concentrations ≥0.007 mg/m3 (highest estimated detection limit, 62% of measurements). The second component predicted the median concentration of measurements ≥0.007 mg/m3. Both components included a random-effect for industry and fixed-effects for year, industry group, analytical method, and other variables. We used the two components together to predict median industry- and time-specific lead concentrations. The probabilities of detectable concentrations and the median detected concentrations decreased with year; both were also lower for measurements analyzed for multiple (vs. one) metals and for those analyzed by inductively-coupled plasma (vs. atomic absorption spectroscopy). The covariance was 0.30 (standard error=0.06), confirming the two components were correlated. We identified determinants of exposure in data with over 60% left-censored, while accounting for correlated relationships and without assuming a distribution for the censored data.
登录
查看更多内容
影响因子：
2.3
作者：
Kreif N;Gruber S;Radice R;Grieve R;Sekhon JS
通讯作者：
Sekhon JS
影响因子：
10.4
作者：
Lubin JH;Colt JS;Camann D;Davis S;Cerhan JR;Severson RK;Bernstein L;Hartge P
通讯作者：
Hartge P
影响因子：
--
作者：
Tran Huynh;Ramachandran, Gurumurthy;Stewart, Patricia A.
通讯作者：
Stewart, Patricia A.
影响因子：
--
作者：
Tran Huynh;Quick, Harrison;Stewart, Patricia A.
通讯作者：
Stewart, Patricia A.
影响因子：
--
作者：
Lavoue, J.;Friesen, M. C.;Burstyn, I.
通讯作者：
Burstyn, I.