Multiplicative factor analysis with a latent mixed model structure for air pollution exposure assessment

A primary objective of current air pollution research is the assessment of health effects related to specific sources of air particles, or particulate matter (PM). Because most PM health studies do not observe the activity of the pollution sources directly, investigators must infer pollution source contributions based on a complex mixture of exposure. Methods such as source apportionment and multivariate receptor modeling use standard factor analytic techniques to estimate the source-specific contributions from a large number of observed chemical components. In the interest of a more flexible source apportionment, we propose a multiplicative factor analysis with a latent mixed model structure on the latent source contributions. A factor analysis with multiplicative errors serves to maintain the non-negativity of the measured chemical concentrations. A mixed model on the latent source contributions provides for systematic effects on source contributions as well as an adjustment for residual correlation in the source-specific exposures. In a simulation study, we examine the impact of (1) accounting for meteorological covariates and (2) adjusting for temporal correlation in the exposures on the estimation of the source profiles and the source contributions. Finally, we explore the influence of meteorological conditions on source-specific exposures in an analysis of PM exposure data from an animal toxicology study. Copyright © 2010 John Wiley & Sons, Ltd.

[1]  John D. Spengler,et al.  Source apportionment of ambient particles in steubenville, oh using specific rotation factor analysis , 1987 .

[2]  Peter Guttorp,et al.  Multivariate Receptor Modeling for Temporally Correlated Data by Using MCMC , 2001 .

[3]  P. Koutrakis,et al.  Source Apportionment of PM10 and PM25 in Five Chilean Cities Using Factor Analysis , 2001, Journal of the Air & Waste Management Association.

[4]  K. Bandeen-Roche.  Resolution of Additive Mixtures into Source Components and Contributions: A Compositional Approach , 1994 .

[5]  W. Stahel,et al.  Linear Unmixing of Multivariate Observations , 2005 .

[6]  Edward J. Bedrick,et al.  Bayesian Binomial Regression: Predicting Survival at a Trauma Center , 1997 .

[7]  Ronald C. Henry,et al.  Bilinear estimation of pollution source profiles and amounts by using multivariate receptor models , 2002 .

[8]  J. Aitchison,et al.  The multivariate Poisson-log normal distribution , 1989 .

[9]  P. Catalano,et al.  Inhaled concentrated ambient particles are associated with hematologic and bronchoalveolar lavage changes in canines. , 2000, Environmental health perspectives.

[10]  B. Coull,et al.  Concentrated ambient air particles induce vasoconstriction of small pulmonary arteries in rats. , 2002, Environmental health perspectives.

[11]  P. Hopke,et al.  Environmental policy analysis: modeling atmospheric particulate matter. , 1999, Environmental Science and Technology.

[12]  P. Hopke,et al.  Modeling atmospheric particulate matter , 1999 .

[13]  David D Billheimer,et al.  Compositional receptor modeling , 2001 .

[14]  R. Christensen,et al.  A New Perspective on Priors for Generalized Linear Models , 1996 .

[15]  Paul J Catalano,et al.  An informative Bayesian structural equation model to assess source-specific health effects of air pollution. , 2007, Biostatistics.

[16]  P. Hopke,et al.  Analysis of ambient particle size distributions using Unmix and positive matrix factorization. , 2004, Environmental science & technology.

[17]  William F. Christensen,et al.  Accounting for Dependence in a Flexible Multivariate Receptor Model , 2002, Technometrics.

[18]  Bradley P. Carlin,et al.  Bayesian measures of model complexity and fit , 2002 .