BACKGROUND: There remains disagreement on the long-term effect of mammographic screening in women aged 40-49 years. OBJECTIVES: The long-term follow-up of a randomised controlled trial that offered annual mammography to women aged 40-49 years. The estimation of the effect of these mammograms on breast cancer and other-cause mortality, and the effect on incidence, with implications for overdiagnosis. DESIGN: An individually randomised controlled trial comparing offering annual mammography with offering usual care in those aged 40-48 years, and thus evaluating the effect of annual screening entirely taking place before the age of 50 years. There was follow-up for an average of 23 years for breast cancer incidence, breast cancer death and death from other causes. We analysed the mortality and incidence data by Poisson regression and estimated overdiagnosis formally using Markov process models. SETTING: Twenty-three screening units in England, Wales and Scotland within the NHS Breast Screening Programme. PARTICIPANTS: Women aged 39-41 years were recruited between 1990 and 1997. After exclusions, a total of 53,883 women were randomised to undergo screening (the intervention group) and 106,953 women were randomised to have usual care (the control group). INTERVENTIONS: The intervention group was invited to an annual breast screen with film mammography, two view at first screen and single view thereafter, up to and including the calendar year of their 48th birthday. The control group received no intervention. Both groups were invited to the National Programme from the age of 50 years, when screening is offered to all women in the UK. MAIN OUTCOME MEASURES: The main outcome measures were mortality from breast cancers diagnosed during the intervention phase of the trial (i.e. before the first National Programme screen at 50 years), mortality from all breast cancers diagnosed after randomisation, all-cause mortality, mortality from causes other than breast cancer, and the incidence of breast cancer. RESULTS: There was a statistically significant 25% reduction in mortality from breast cancers diagnosed during the intervention phase at 10 years' follow-up (relative rate 0.75, 95% confidence interval 0.58 to 0.97;  = 0.03). No reduction was observed thereafter (relative rate 0.98, 95% confidence interval 0.79 to 1.22). Overall, there was a statistically non-significant 12% reduction (relative rate 0.88, 95% confidence interval 0.74 to 1.03;  = 0.1). The absolute benefit remained approximately constant over time, at one death prevented per 1000 women screened. There was no effect of intervention on other-cause mortality (relative rate 1.02, 95% confidence interval 0.97 to 1.07;  = 0.4). The intervention group had a higher incidence of breast cancer than the control group during the intervention phase of the trial, but incidence equalised immediately on the first National Programme screen at the age of 50-52 years. LIMITATIONS: There was 31% average non-compliance with screening and three centres had to cease screening for resource and capacity reasons. CONCLUSIONS: Annual mammographic screening at the age of 40-49 years resulted in a relative reduction in mortality, which was attenuated after 10 years. It is likely that digital mammography with two views at all screens, as practised now, could improve this further. There was no evidence of overdiagnosis in addition to that which already results from the National Programme carried out at later ages. FUTURE WORK: There is a need for research on the effects of modern mammographic protocols and additional imaging in this age group. TRIAL REGISTRATION: Current Controlled Trials ISRCTN24647151. FUNDING: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in ; Vol. 24, No. 55. See the NIHR Journals Library website for further project information. Other funding in the past has been received from the Medical Research Council, Cancer Research UK, the Department of Health and Social Care, the US National Cancer Institute and the American Cancer Society.