A new meta-analysis published in the JNCI: Journal of the National Cancer Institute concludes that the mammogram overdiagnosis rate is likely below 5%, a figure that sits far beneath the 30% to 50% estimates that have shaped international screening debates for decades. The study, published on 14 September 2026 according to SciTechDaily, reanalysed all eight major randomised trials of mammography screening and argues that earlier high estimates were, in large part, a methodological artefact.
The claim matters. Overdiagnosis, the detection of a breast cancer that would never have caused symptoms or threatened a woman’s life, has long been cited as a reason for caution about population-wide screening programmes. If those estimates were inflated, the calculus changes.
What the mammogram overdiagnosis rate analysis actually found
The researchers behind the meta-analysis brought together results from all eight randomised controlled trials in this field: the New York Health Insurance Plan, Malmö, Two-County, Edinburgh, the Canadian National Breast Screening Study, Stockholm, Gothenburg, and UK Age. They then compared those trial results with data from Danish routine screening programmes, which offered an unusual real-world reference point.
According to News Medical, organised breast cancer screening was introduced in some Danish regions 17 years before others. That gap allowed researchers to observe how diagnosis rates shifted immediately after screening was introduced and how patterns changed over longer periods, something a short-term trial cannot easily replicate.
The team’s conclusion was that, once three specific methodological factors are properly accounted for, the apparent excess of cancers detected through screening shrinks considerably. Those factors were: whether women in control groups later underwent screening themselves; how many screening rounds each group received; and the length of follow-up available for each group.
Why timing distorts the numbers
The reasoning behind the correction is not complicated, but it is easy to miss. When mammography screening is introduced, more cancers are detected earlier than they would otherwise have been. That produces an initial rise in diagnosis numbers. Over time, that rise should be followed by a corresponding dip, as cancers that were brought forward in time are no longer available to be diagnosed later.
If a study ends before enough time has passed for that dip to appear, the early excess looks like overdiagnosis when it may simply be earlier detection of cancers that would eventually have surfaced anyway. Elsebeth Lynge, professor emerita at the Department of Public Health, University of Copenhagen, put it plainly: ‘When screening is introduced, the number of breast cancer diagnoses initially rises because cancers are detected earlier than they would have been without screening. Over time, this should be followed by a drop, as some of these cancers would otherwise have been diagnosed later.’
She added that results are further skewed when women in control groups obtain screening after a trial ends, which was common across the trials examined. Failing to account for that exposure artificially inflates the apparent overdiagnosis figure in the screened group.
Matejka Rebolj, Senior Epidemiologist at Queen Mary University of London, said the team believed ‘some previous high estimates of overdiagnosis, which influenced screening guidelines and communication, were based on evidence before trial data had fully matured. When interpreted in their full temporal context, randomised trial data are consistent with overdiagnosis of less than five percent, rather than with estimates nearing 50%.’
What the findings mean in practice
Sisse Helle Njor, professor at the University of Southern Denmark and Lillebælt Hospital and the study’s lead author, said the aim was to bring together evidence from all randomised controlled trials to get ‘a clearer picture of the extent of overdiagnosis in breast cancer screening’. She described the earlier interpretation of trial data as ‘not as straightforward as it may seem’.
On the practical implications for women considering whether to attend screening, Njor said: ‘Most women will not develop breast cancer, but with this study we can now be reassured that the benefits of detecting breast cancer early and preventing premature death will outweigh the small risk of unnecessary treatment.’ The team also examined both invasive breast cancer and ductal carcinoma in situ in their analysis.
The analysis does not claim overdiagnosis is zero, and the researchers’ language throughout is careful: rates ‘consistent with’ below 5%, not a proven precise figure. But the direction of the revision is clear, and the methodology used to reach the earlier high estimates is now in dispute. Funding for the work came from the Novo Nordisk Foundation (supporting Casper Urth Pedersen) and Cancer Research UK (supporting Matejka Rebolj). The study is published in full in the JNCI: Journal of the National Cancer Institute under DOI 10.1093/jnci/djag302.
