Rhodes Serum Dioxin Study
Rhodes serum dioxin study: Mid Point Summary Report
By Adam Craig
Rhodes Serum Study Coordinator 2007/8
Public Health Officer Trainee
Background
The Rhodes Serum Dioxin Study's aim is to determine if there is a change in the serum dioxin levels of the Rhodes community relative to a control group (recruited from the Clarence Street Red Cross Blood Donation Centre) during the period of remediation of the Rhodes peninsula.
Blood samples have been collected from the Rhodes community and controls at baseline (before the remediation started in 2004/5) and at mid-point of remediation (in 2007). A third sample will be taken at the completion of remediation, envisaged to be in 2009. A description of the study methodology can be found in the Technical Report: Rhodes Serum Dioxin Study: Methodology and Baseline Data available at
http://www.health.nsw.gov.au/PublicHealth/environment/hazards/env_rhodes_serum.asp
At baseline, 60 Rhodes residents and 594 controls provided blood samples for this study. Each of the Rhodes residents provided a 50ml sample while the controls provided a 5ml sample. At mid-point sample collection 52 (87%) of the Rhodes community group and 243 (41%) of the control group provided their second sample. Table 1 summarises the number of samples collected from each group at baseline and mid-point sample collections.
All samples have been stored at -20 degrees Celsius in specified freezers at the National Measurement Institute (NMI) in Pymble. In 2008 NMI extracted a portion of the samples of those participants who had given blood at both sample collection time points (52 Rhodes and 243 controls). These portions were combined into four pooled samples based on group and sample collection time points (2004 Rhodes, 2007 Rhodes, 2004 controls & 2007 controls) and were subsequently analysed by the Eurofins Analytik GmbH laboratory in Germany for the standard suite of dioxins and polychlorinated biphenyls (PCBs).
At baseline, Rhodes residents tended to be slightly older compared to the control group (mean age of Rhodes residents = 56; mean age of controls = 51.4). When excluding participants that have not provided mid-point samples (and hence have not been included in the analysis) the difference in mean age between the residents and control group reduces from 4.6 to 3.5 years. At baseline, controls were more likely to be regular smokers than Rhodes residents; this was reversed at mid-point sampling. Other comparison characteristics such as education levels, dietary intake of fish or dairy, and consumption of home grown vegetables or home reared meat/eggs tend to be similar between the groups (Table 1).

