Accuracy of family history of cancer as reported by men with prostate cancer.

OBJECTIVES: To report on the accuracy of probands providing information on specific cancer types in their families and the ability of investigators to document these reports. Accurate information on the health status of family members is critical when studying familial patterns of diseases. However, collecting these data require significant resources.
METHODS: We identified 143 patients with prostate cancer from the University of Texas M. D. Anderson Cancer Center who had reported at least 1 first-degree relative with cancer. There were 263 first-degree relatives identified, for whom we confirmed diagnoses using medical records, death certificates, and verbal confirmation. The data are reported in summary statistics and compared with chi-square analysis.
RESULTS: We documented 78% of the reports, with an accuracy rate of 81.6%. We found that accuracy was highly related to the site reported. Accuracy and documentation levels were not related to the age or income of the proband. The education level was significantly associated with the ability to document cancer, but not with the accuracy of the report. The accuracy and documentation differed by the relationship of the first-degree relative to the proband.
CONCLUSIONS: Proband reporting of cancer in first-degree relatives varies widely by site, with common metastatic sites the most inaccurate. No reliable demographic factors were found that would reasonably predict the ability to document the accuracy of the report. We found a significant proportion of proband-reported prostate cancer was, in reality, benign prostatic hyperplasia. We propose a strategy of targeting male relatives and reports of cancer in common metastatic sites for aggressive follow-up.