Recall (report) bias and reliability in the retrospective assessment of melanoma risk.

In a case-control study nested in the Nurses' Health Study cohort, the authors assessed recall bias in the ascertainment of two risk factors for melanoma: hair color and ability to tan. Participants reported on these risk factors in a 1982 questionnaire and in a subsequent case-control questionnaire or telephone interview. The test-retest reliability among controls was high for both questions (Spearman's r = 0.76). Among women diagnosed with melanoma after the first questionnaire and before the second, there was a substantial shift toward reporting a reduced ability to tan when participants were questioned after the diagnosis of melanoma (p = 0.035). No shift was noted for the hair color question (p = 0.8). The authors conclude that recall bias was observed among female nurses with cutaneous melanoma in the assessment of tanning ability, a major risk factor for melanoma.