Health hazard evaluation report: HETA-2003-0097-3018, Belle River Power Plant, China, Michigan.
On December 4, 2002, the National Institute for Occupational Safety and Health (NIOSH) received a management request for a health hazard evaluation (HHE) at the Detroit Edison (DTE) Belle River Power Plant (BRPP) in China, Michigan. The request asked for NIOSH assistance in determining if there was an excess of cancer cases among current and former workers at the power plant. The concern arose because there had been contamination of the potable water with hydrazine, which was used as an anti-corrosive additive in the boilers. An unauthorized cross-connection between the boiler and the potable water system had reportedly been made sometime between 1984 and 1989, and hydrazine was found in the water in August 1989, when the water was noted to be hot and to have a bad taste. DTE provided us with a list of persons working at BRPP between 1984 and 1989. The list was provided to the Michigan Cancer Surveillance Program in order to determine incident primary cancers in this population. The Michigan Cancer Surveillance Program provided us with a computer datafile of all matches for each primary cancer diagnosed. They also provided us with cancer rates for the time period 1988-2002 (the last year for which information was available) in 5-year intervals for Livingston, Macomb, Oakland, Ogemaw, Roscommon, St. Clair, Sanilac, Washtenaw, and Wayne counties, which are the counties surrounding the BRPP in which BRPP employees had been diagnosed with cancer. The incidence of cancer was analyzed with the NIOSH modified life table analysis system (LTAS) for a cohort of 712 DTE employees assigned to BRPP between 1984-1989. The ratio of observed to expected number of cancers was expressed as the standardized incidence ratio (SIR). We calculated 95% confidence intervals (95% CI) for the SIRs. We also calculated SIRs with a 5-year lag, restricting analysis to those workers who were working in one of the affected areas during 1989 (the year of documented contamination). Incidence of all cancers from 1988-2002 among BRPP employees was statistically significantly lower than the Michigan combined counties rate (SIR 0.62, 95% CI 0.46 to 0.82), as were lung cancer rates (SIR 0.33, 95% CI 0.09 to 0.85). Colon cancer rates were not significantly elevated (SIR 1.05, 95% CI 0.34 to 2.45) during that time period. The findings were similar with a lag of 5 years, which excludes cases diagnosed in the first 5 years after exposure. When the analysis was restricted to only those working in 1989, there was no statistically significant increase in colon cancer (SIR 1.79, 95% CI 0.58 to 4.19), or no decrease in lung cancer incidence (SIR 0.28, 95% CI 0.03 to 1.01) compared to the Michigan combined counties rate. Incidence for all cancers in this latter group was significantly lower than the Michigan combined counties rate (SIR 0.60, 95% CI 0.40 to 0.87). The NIOSH investigation did not find evidence of an excess of cancer among DTE employees assigned to BRPP between 1984 and 1989. NIOSH investigators recommend no further investigation of this issue.