Background: We sought to determine the minimum survival benefits that patients judged sufficient to make adjuvant chemotherapy for early colon cancer worthwhile, factors associated with these judgments; and, to compare a self-administered questionnaire with a validated, scripted interview. Patients and methods: One twenty three subjects who completed adjuvant chemotherapy for early colon cancer 3-60 months earlier completed a questionnaire; 97 were randomised to complete an interview before or after the questionnaire. Preferences were elicited by the time trade-off method in 4 hypothetical scenarios. Concordance between the interview and questionnaire was assessed with the intraclass correlation coefficient (ICC). Results: Median age was 65 years (range 19-86), 52% were female and 74% had involved lymph nodes. Over 60% of patients judged an additional 1 month beyond life expectancies of 5 years or 15 years, and an additional 1-2% beyond 5-year survival rates of 85% or 65%, sufficient to make chemotherapy worthwhile. Subjects with tertiary education (p = 0.003) or aged 75 years or less (p = 0.02) judged larger benefits necessary to make chemotherapy worthwhile. Concordance between the interview and questionnaire was high (ICCs 0.71-0.82). Conclusions: Most subjects judged small survival benefits sufficient to make adjuvant chemotherapy worthwhile. A self-administered questionnaire was a valid and acceptable way of eliciting preferences.