Related Experiment Video
Updated: Jul 23, 2026

Long-term Culture of Human Breast Cancer Specimens and Their Analysis Using Optical Projection Tomography
Published on: July 29, 2011
Results from 10 years of breast screening in Wales
H Fielder1, C Rogers, K Gower-Thomas
1Breast Test Wales/Cervical Screening Wales, 18 Cathedral Road, Cardiff CF11 9LJ, UK. hilary.fielder@velindre-tr.wales.nhs.uk
Objective:
To present results from 10 years of breast screening in Wales, including rates of interval cancers.
Setting:
All women aged 50-64 years in Wales; approximately 250,000 in each screening round.
Methods:
Breast Test Wales (BTW) began mammographic screening in 1989 as part of the National Breast Screening Programme. Two view mammography was introduced at the inception of the Welsh programme for all prevalent screens. Single view mammography was subsequently performed for incident screens. Interval cancers were identified by matching details from the BTW administrative screening database with the BTW database of all incident breast cancers for residents in Wales.
Results:
For the first and second round prevalent screens, the cancer detection rate was 8.6 per 1000 and 7.4 per 1000 screened, respectively. The interval cancer rates following round one were 4.9 per 10,000 in the first 12 months, 9.0 per 10,000 between 12 and 24 months, and 11.6 per 10,000 between 24 and 36 months after screening. For the second round incident screens the detection rate was 4.6 per 1000 and the standardised detection ratio was 0.89, but the interval cancer rates in the first year after incident screens were not statistically different from those following prevalent screens. There was no change in the proportion of invasive breast cancers which were < 15 mm in diameter, or in the grades of invasive cancer between the first and second rounds, prevalent or incident screens.
Conclusions:
Breast Test Wales has achieved the standards set by the National Health Service Breast Screening Programme. Taking two views at the prevalent screen gave high sensitivity and may have resulted in the low standardised detection ratio at subsequent incident screens.

