Bias in the assessment of family history of melanoma and its association with dysplastic nevi in a case-control study.

Family history of melanoma is an important risk factor for both melanoma and, it is thought, dysplastic nevi. However, assessment of family history of melanoma in epidemiologic investigations has typically been limited to interview of the proband. As part of a case-control study of dysplastic nevi, we attempted to confirm family histories. We disproved about half of the reported family histories of melanoma among first-degree relatives, and confirmed them by medical records in only 17%. Few family histories pertaining to other relatives could be confirmed. We documented the association of melanoma family history with dysplastic nevus risk, and we further documented a substantially greater odds ratio for this association when history was based on confirmation by medical records, compared to confirmation by proband interview only. The bias thus documented must be considered in evaluating the many published epidemiologic studies of melanoma and related disorders. Future research should attempt to confirm family histories of melanoma whenever possible, despite the practical difficulties.