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

Auteurs

Melissa Friesen, National Cancer Institute
Hyoyoung Choo-wosoba, NIHFollow
Philippe Sarazin, IRSSTFollow
Jooyean Hwang, Western Kentucky UniversityFollow
Pamela Dopart, NIHFollow
Daniel E. Russ, NIHFollow
Nicole C. Deziel, Yale UniversityFollow
Jérôme Lavoué, Université de MontréalFollow
Paul S. Albert, National Cancer InstituteFollow
Bin Zhu, National Cancer InstituteFollow

Type de document

Études primaires

Année de publication

2021

Langue

Anglais

Titre de la revue

Journal of Exposure Science & Environmental Epidemiology

Première page

1047

Dernière page

1056

Résumé

Background

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.

Objective

We used semi-continuous models to identify time and industry trends in 52,457 OSHA inspection lead sample results.

Method

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.

Results

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.

Significance

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.

Mots-clés

Sol, Ground, Sable, Sand, Caisson, Aluminium, CAS 7429905, Acier, Steel, Argile, Clay, CAS 1302870, Essai des matériaux, Material testing

Numéro de projet IRSST

2018-0018

Citation recommandée

Friesen, M. C., Choo-Wosoba, H., Sarazin, P., Hwang, J., Dopart, P., Russ, D. E., . . . Zhu, B. (2021). 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. Journal of Exposure Science & Environmental Epidemiology, 31(6), 1047-1056. https://doi.org/10.1038/s41370-021-00331-7