A government target of a maximum 2-week wait for women referred urgently with suspected breast cancer was introduced in April 1999. We have assessed changes in the distributions of waiting times and the proportions of cases meeting proposed targets before and after this date, using clinical audit data on 5750 women attending 19 hospitals in southeast England during the period July 1997-December 2000, who were subsequently found to have breast cancer. The proportion of cases being seen within 2 weeks of referral rose from 66.0 to 75.2%, and the median wait to first appointment fell from 13.6 to 12.3 days, following the introduction of the government target. The proportion of cases waiting 5 weeks or less between first hospital appointment and treatment fell from 83.8 to 80.3%, and median waits for treatment increased from 21.4 to 24.1 days. We also examined the effects on waiting times of various sociodemographic and care related factors. A total of 85.7% of screening cases vs 67.9% of symptomatic cases were seen within 2 weeks, and 95.0% of cases treated with tamoxifen received treatment within 5 weeks, as opposed to 77.6% of cases treated with surgery, 81.2% of chemotherapy cases and 52.8% of radiotherapy cases. While waiting times from GP referral to first hospital appointment have improved since the introduction of the government target, times from first appointment to treatment have increased, and consequently total waiting times have changed little.