The Great Lakes Fish Eaters Project (Eaters Project) was funded by Health Canada to identify
individuals consuming large amounts of Great Lakes fish (≥ 26 meals over the past 1 year
period) and to document relevant contaminant information to describe the potential health risks and benefits associated with the consumption of fish from the Great Lakes. Data were analyzed on 91 participants in the Eaters Project.
The Great Lakes are home to a wide variety of fish. In this study alone, the participants have
collectively eaten over 35 different species of Great Lakes fish.
The participants of this study were asked for their opinions on the environment and their health. In addition, blood work was completed on 89 participants. The laboratory results were collected on different contaminants. Seven contaminants were considered for modeling purposes since the laboratory value was above a pre-defined detection limit in greater than 75% of the participants in this study.
These seven contaminants included six Organochlorine Pesticides: p,p'-DDE (100% of the
participants above the detection limit), hexachlorobenzene (HCB) (100%), β-
hexachlorocyclohexane (β-HCH) (92.1%), mirex (78.7%), oxychlordane (92.1%),
transnonachlor (98.9%) and one Polychlorinated Biphenyl Congener, Aroclor 1260 (100%).
Organic mercury was also modeled since it contributed on average 77.2% to a maximum of
97.5% of the total mercury value.
A 2-stage modeling approach on linear and categorical variables was done as a hypothesis
building exercise. Each contaminant was age-adjusted at the start of this modeling exercise. The age-adjusted contaminant was then used as the response variable in the second stage of the modeling of each specific contaminant.
The gender (male and female) and country of birth (Canada/Europe/US and Asian) variables
were confounded with each other. The model was be stratified on country of birth or on gender if one of the variables were found to be the only significant variable left in the full model.
Not all the contaminants that were modeled had significant results. That is, there were levels of certain contaminants that the modeling approach could not explain.
The country of birth (Canada/Europe/US and Asian) variable was important in the modeling of
some of the contaminants. This suggested stratifying by the two country-of-birth groups. / Thesis / Master of Science (MSc)
| Identifer | oai:union.ndltd.org:mcmaster.ca/oai:macsphere.mcmaster.ca:11375/23779 |
| Date | 04 1900 |
| Creators | Sigouin, Christopher |
| Contributors | Macdonald, P.D.M., Statistics |
| Source Sets | McMaster University |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0021 seconds