Time trends of persistent organic pollutants in 30 year olds sampled in 1986, 1994, 2001 and 2007 in Northern Norway: measurements, mechanistic modeling and a comparison of study designs

Abstract

Background

Human biomonitoring studies have demonstrated decreasing concentrations of many persistent organic pollutants (POPs) in years after emission peaks. Objectives

To describe time trends of POPs in blood using four cross-sectional samples of 30 year olds from Tromsø, Norway across 1986–2007, and to compare the measured concentrations of polychlorinated biphenyl 153 (PCB-153) to model-estimated values. A second objective was to compare the repeated cross-sectional time trends with those observed in our previous longitudinal study using repeated individual measurements in older men from the same surveys. Methods

Serum from 45 persons aged 30 years in each of the following years: 1986, 1994, 2001, and 2007 was analyzed for 14 POPs. Further, predicted concentrations of PCB-153 in each sampling year were derived using the emission-based CoZMoMAN model. Results

The median decreases in summed serum POP concentrations (lipid-adjusted) in 1994, 2001, and 2007 relative to 1986 were − 71%, − 81%, and − 86% for women and − 65%, − 77%, and − 87% for men, respectively. The overall time trend in predicted PCB-153 concentrations demonstrated agreement with the observed trend although model predictions were higher than the measured concentrations at all time points. Compared to our previous longitudinal study of repeated individual measurements in older men, similar although more prominent declines were observed in the younger cross-sectional samples. Discussion

Observed declines in serum concentrations from 1986 to 2007 were substantial for legacy POPs in men and women at reproductive ages in Northern Norway and are generally consistent with previous longitudinal biomonitoring efforts in the study population. The measured concentrations and observed declines likely reflect a combination of recent and historic exposures. Small differences in time trends observed between the studies could be attributed to different study designs (i.e. the chosen age group or sex and cross-sectional versus repeated individual measurement sampling).