Date Awarded


Document Type


Degree Name

Doctor of Philosophy (Ph.D.)


Virginia Institute of Marine Science


Michael C. Newman


The use of community-based participatory research (CBPR) methods to conduct environmental exposure assessments provides valuable insight about disparities in fish consumption and contaminant exposure. Ninety-five community-specific fish consumption surveys were administered to low-income African American women (ages 16--49) residing in the Southeast community of Newport News, Virginia, USA, in 2008. The mean fish consumption rate for the women surveyed was 147.8 g/day (95% CI: 117.6-185.8), a rate substantially higher than the mean fish consumption rate reported for U.S. women (1.8 g/day 95% CI: 1.51-2.04). Through collaborative partnerships established between current researchers and The Moton Community House (a local community center), African American women (ages 16--49 yrs) from the same community were surveyed in 2010 to assess the reproducibility and consistency of fish consumption patterns (ingestion rates, exposure frequencies, weight, and fish consumption rates), and the reliability of the survey responses. Fish consumption patterns were reproducible and the survey responses were reliable. Comparison between years revealed that fish consumption patterns remained consistent over time. In addition, the high fish consumption rate estimated in 2008 was reaffirmed in 2010 with a rate (134.9 g/day; 95% CI: 88--207 g/day) not materially different and still considerably higher than mean fish consumption rates reported for U.S. women. Daily mercury intake rates were estimated using consumption data from 2008 and three consumption scenarios (canned white, canned light, and no tuna) due to confirmed differences in mercury concentration between canned white and light tuna. Arithmetic mean daily mercury intake rates were 0.284 ug/kg-bw/day (95% CI: 0.229 - 0.340 ug/kg-bw/day) using canned white tuna, 0.212 ug/kg-bw/day (95% CI: 0.165 - 0.259 ug/kg-bw/day) using light tuna, and 0.197 ug/kg-bw/day (95% CI: 0.151 - 0.243 ug/kg-bw/day) using no tuna. Probabilistic estimations of dietary mercury exposure for African American women (ages 16--49) from the Southeast Community were generated and compared to point estimates. Four different consumption scenarios were assumed, representing 1) no, 2) light, 3) both light and white, and 4) white tuna consumption. The probabilistic models generated lower dietary mercury intake rates than the point estimations, under these consumption scenarios. Arithmetic mean daily mercury intake rates (95% Cl) for the probabilistic models were 0.149 (+/-0.003), 0.148 (+/-0.003), 0.172 (+/-0.004), and 0.202 (+/-0.004) ug/kg-bw/day, respectively for no, light, both, and white tuna consumption. Reducing the amount of fish consumed in probabilistic models resulted in lower dietary mercury exposures for each consumption scenario. at a rate that was a quarter of what was normally consumed, the percentages of exposures that exceeded the US EPA's oral RfD for mercury were 14%, 13%, 18%, and 25% respectively for no, light, both, and white tuna consumption. In this community we learned that even though African American women in Southeast Newport News, Virginia are not subsistence fishers, they consume seafood at a subsistence fisher rate. In addition, estimates of dietary mercury exposure were high enough to warrant concern.



© The Author