P2.58: Assessment and quantification of the exposure to xenoestrogens in a population‐based case‐control study on testicular cancer

Abstract

Endocrine disrupting agents are widely discussed as risk factors for several tumor sites. For testicular cancer only one study supports this hypothesis (e.g. Ohlson and Hardell, 2000). Between 1995 and 1997 a population-based case-control study on testicular cancer was conducted on 269 cases and 797 controls aged 15 to 69 years. Data on endocrine disruptors were not raised specifically. A quantification concept for the exposure was developed with xenoestrogens to be applied to personal interviews including detailed working histories. Occupations comprising the contact with known xenoestrogens were identified. Exposures to hormones, alkyl phenols, phthalates, PCB, pesticides, bisphenol A, dioxin, phenol resin, and others were considered. Supplementary questionnaires describing in detail the particular work processes were used to obtain information about frequency, intensity and duration of the respective processes and possible occupational safety measures that were taken. A manual was developed that describes in detail the semi-automatic algorithm for the individual assessment and quantification of the agents listed above. New variables were introduced for exposure dimensions that had to be rated manually, like the type, the probability, and the intensity of the exposure and the confidence in the manual rating. Exposure dimensions were classified into 4 to 6 different categories. Manual classification was done twice by independent raters. Both estimates were generally in good agreement. Weighted Kappa values were 0.9 (95% Confidence Interval: 0.87 -0.93) for type, 0.74 (95% CI: 0.70 -0.79) for intensity, 0.74 (95% CI: 0.69 -0.79) for probability, and 0.99 (95% CI: 0.97 -1.00) for frequency of exposure, respectively. The newly developed procedure for assessment and quantification of the exposure to xenoestrogens proved to be sufficiently reliable to be used for the epidemiological analysis in the case-control study.