Radiologists from seven independent practices in Germany (average number of cases 3,500 in 2001) combined to form a mammography quality control group. Documentation and follow up were standardised for all cases that resulted in a referral for open biopsy. A quantitative analysis of the data would be expected to clarify the quality achieved in ambulatory care and to show further possibilities for quality assurance measures. The first overall analysis of 2,500 referrals for biopsy from 1998 to 2001 shows differences between practices as regards the frequency of referral for biopsy and the proportion of referrals for suspected malignancy, which are based on practice-specific conditions (referral behaviour, age structure of patients) and individual case management. Practices are very close in relation to positive predictive value (ppv) and detection rate. The biopsy rate in 2001 slightly exceeded 3%, the overall detection rate was 1.7%, and ppv 81%. For asymptomatic women detection rate and ppv were 0.5% and 70%, respectively. Quality parameters for all mammography referrals in total exceed those of screening programmes. When restricted to asymptomatic women, results have been achieved that are comparable to those of established screening programmes. Naturally the patients involved here are those in whom mammography was indicated. Nonetheless, these results demonstrate a high quality of detection on the part of those performing mammography in routine ambulatory care, and can be used as a reference for other types of care. With consistent quality assurance in relation to technical considerations and personnel, a performance quality can be achieved in independent practices which to date has not been surpassed by specialist mammography centres.