Numerous studies have shown that the symptoms of angina pectoris are more prevalent in women than men, than other manifestations of coronary heart disease, with the greatest discrepancies at younger ages. Variation in symptom reliability between genders could be a potential explanation for these differences. The Lipid Research Clinics Prevalence Study included two standardized Rose Questionnaire interviews, allowing evaluation of the relationship between reliability and prevalence of angina pectoris. Analyses of the results of two interviews in 2348 men and 2085 women who were at least 30 years old in 1972-1976 showed that women generally had lower reliability estimates, but that the differences were minimal for age groups where prevalence differences were greatest. Furthermore, for interviews less than 1 month apart differences in symptom reliability were quite small (kappa = 0.65 for men and 0.58 for women). In addition, if consistently positive interviews were used to define angina, gender differences in prevalence decreased but did not disappear. In the LRC population, reporting unreliability did not explain the higher prevalence of angina pectoris in young women.