In 1980 the EORTC-BCCG initiated a multicentre randomised clinical trial comparing modified radical mastectomy (MRM) with breast-conserving therapy (BCT) in stage I and II breast cancer. The main endpoint of the trial was survival. A brief quality of life (QoL) questionnaire consisting of two multi-item scales (body image and fear of recurrence) and two single items (satisfaction with treatment and cosmetic result) was included in the trial. A cosmetic evaluation of the breast after conservative surgery was also performed. This report concentrates on the results of the QoL study and the cosmetic evaluation. Multitrait scaling analysis was employed to verify the hypothesised scale structure of the questionnaire. Treatment comparison of the QoL scores at 2 years post-treatment was performed using a stratified Wilcoxon rank sum test. Both patients' and doctors' ratings of the cosmetic result were documented on the clinical follow-up form. A multivariate analysis was performed to identify which factors influenced the cosmetic outcome. In total, 127 patients in the MRM arm and 151 in the BCT arm completed a QoL questionnaire at approximately 2 years after randomisation (months 25-36). The Cronbach's alpha coefficients were 0.79 and 0.73 for the body image and fear of recurrence scales, respectively. Significant benefit in body image and satisfaction with treatment was observed in the BCT patients. No significant difference was observed in fear of recurrence between the two groups. Ratings of cosmetic results decreased with time, in line with clinical observations of long-term side-effects of radiotherapy. Wide excision appeared to be the most important predictive factor for poor cosmetic result. In this multicentre randomised study, BCT helped to maintain the patients' body image, resulted in higher satisfaction with treatment and yielded no significant difference from MRM with respect to fear of recurrence.