Assessing the Amount of Unscheduled Screening (“Contamination”) in the Control Arm of the UK “Age” Trial

Abstract

The UK Age Trial of mammographic screening from age 40 has reported a nonsignificant 17% reduction in breast cancer mortality calculated on an "intention to treat" basis. High levels of ad hoc screening in the control arm could potentially have diluted the estimated effect. To estimate the level of unscheduled mammography in the control arm of the UK Age Trial. Data were obtained from questionnaires sent to a random sample of 3,706 women at five centers in the control arm of this trial. Questions included in the Office for National Statistics Omnibus Surveys about the timing of and reasons for any breast screening provided comparable data. The overall response rate was 58.8%. Overall, 24.9% (95% confidence interval, 23.0-26.8) of Age Trial controls responding reported ever having had a mammogram, 18.2% reported a mammogram for symptomatic reasons, and 8.4% reported unscheduled mammography. Overall, 4.0% and 1.8% of women reported symptomatic and unscheduled mammography, respectively, within the previous 12 months. Results from the Omnibus surveys were similar, 14.2% of women reported previous mammography for symptomatic reasons or follow-up after breast cancer and 6.8% reported unscheduled mammography. The level of contamination due to mammographic screening in the control arm of the Age Trial was low and will have had a minimal effect on the estimated reduction in mortality from breast cancer. Estimating the extent of screening in the control arm in randomized trials of screening is important to inform interpretation of the results.