By admin 
For decades, the exact frequency of overdiagnosis has been a contentious issue. Estimates derived from randomized controlled trials (RCTs), considered the gold standard in medical research, have varied dramatically, fueling the controversy. Some influential studies and meta-analyses suggested that a substantial proportion, ranging from 30% to an alarming 50% of breast cancers detected through screening, might fall into the overdiagnosed category. These high figures have profoundly influenced international discussions about the overall efficacy and ethics of population-based breast cancer screening programs. They have led some countries to reconsider their screening guidelines, prompted widespread public apprehension, and made it challenging for women to weigh the benefits against the potential harms when invited for screening. The discrepancy in these estimates created a significant challenge for healthcare providers trying to provide clear, evidence-based guidance.
"The aim of our study was to bring together the evidence from all randomized controlled trials to get a clearer picture of the extent of overdiagnosis in breast cancer screening," says Sisse Helle Njor, professor at University of Southern Denmark and Lillebælt Hospital, a lead researcher in this groundbreaking new analysis. She emphasizes the urgency of clarifying this long-standing debate. "Randomized trials have often been cited as evidence that overdiagnosis is a substantial problem, with some interpretations suggesting that up to half of all screen-detected cancers might be harmless. Our study shows that this interpretation is not as straightforward as it may seem, and that a more nuanced, temporally complete understanding of the data paints a very different picture."
The researchers’ comprehensive re-analysis revealed a stark contrast to many previous interpretations. They found that the additional breast cancer cases detected in randomized trials, when examined over a sufficiently long period and with careful consideration of various methodological factors, closely resembled patterns seen in Denmark. In Denmark’s well-documented national screening programs, overdiagnosis associated with screening is estimated to be below 5%. This dramatic difference, from potentially 50% down to less than 5%, has profound implications for how the benefits and harms of breast cancer screening are perceived and communicated.
A New Look at Mammography Trials: Synthesizing Decades of Data
To achieve this clearer understanding, the international research team embarked on a rigorous methodological journey. They combined and meticulously reanalyzed results from all eight major randomized trials of mammography screening ever conducted. These trials represent a foundational body of evidence in breast cancer research, spanning decades and involving hundreds of thousands of women across different countries. The trials included in their analysis were the New York Health Insurance Plan (HIP) study, Malmö, Two-County, Edinburgh, the Canadian National Breast Screening Study (CNBSS), Stockholm, Gothenburg, and UK Age. Each of these trials had its own design, follow-up period, and reporting methods, contributing to the complexity of synthesizing their findings.
Crucially, the team also compared these re-analyzed trial results with real-world data from Denmark. Denmark offered a particularly useful reference point because its organized breast cancer screening programs were introduced in different regions at different times, with a significant 17-year gap between the earliest and latest implementations. This staggered introduction created a natural experiment, allowing researchers to track precisely how breast cancer diagnoses changed immediately after screening was introduced in a population, and how those patterns evolved over much longer periods. By examining both invasive breast cancer and ductal carcinoma in situ (DCIS), a non-invasive form of breast cancer often considered a precursor to invasive disease, the study provided a comprehensive view of diagnostic shifts.
"When screening is introduced, the number of breast cancer diagnoses initially rises because cancers are detected earlier than they would have been without screening. This phenomenon is known as ‘lead time bias’," explains Elsebeth Lynge, professor emerita at the Department of Public Health, University of Copenhagen, and a co-author of the study. She elaborates, "Over time, this initial surge should be followed by a drop in diagnoses in the screened population, as some of these cancers would otherwise have been diagnosed later anyway. This pattern can also be significantly affected if women in either the intervention or control group continue to undergo screening after the trials had officially ended, which was a common occurrence and a significant methodological challenge for earlier analyses. If researchers do not take these lead time effects and subsequent screening exposure factors into account, the initial increase in diagnoses can be mistakenly interpreted as a much higher rate of overdiagnosis than actually exists."
The researchers meticulously compared breast cancer incidence rates at matching points in time, not only between the intervention and control arms of the randomized trials but also between the real-world Danish screening programs and their unscreened counterparts. This allowed them to assess whether the temporal patterns of diagnosis were similar across these diverse settings and, critically, what those similarities might reveal about the true scale of overdiagnosis when viewed through a complete temporal lens.
"Taken together, we believe some previous high estimates of overdiagnosis, which have significantly influenced screening guidelines and public communication strategies globally, were based on evidence collected before trial data had fully matured," states Matejka Rebolj, Senior Epidemiologist at Queen Mary University of London and another key researcher. "When interpreted in their full temporal context, accounting for the natural evolution of diagnosis rates over time, randomized trial data are remarkably consistent with overdiagnosis of less than five percent, rather than with estimates nearing 50%. This fundamental shift in understanding has profound implications for the ongoing debate."
Facts: Overdiagnosis
Overdiagnosis occurs when screening detects a breast cancer that would never have become life-threatening or caused symptoms during a woman’s lifetime. Without screening, the woman would never have known that the cancer was present, and thus would not have undergone potentially unnecessary and harmful treatments. This is distinct from lead time bias, which refers to the earlier detection of a cancer that would eventually have become clinically apparent.
The definition can also encompass women who die from another cause shortly after receiving a breast cancer diagnosis through screening. In these cases, screening may have offered little to no benefit because the woman’s poor overall health or limited life expectancy meant that finding and treating the breast cancer was unlikely to improve her health outcomes or extend her life. Such cases highlight the complex ethical considerations surrounding screening in older or frail populations.
Why Timing Can Change the Numbers: The Dynamics of Screening
A key issue underlying the historical discrepancies in overdiagnosis estimates is the dynamic way screening changes when a cancer is diagnosed. When mammography is first introduced to a population, it acts like a net, catching many cancers that are already present but have not yet become symptomatic. This leads to an immediate and noticeable rise in the number of breast cancer diagnoses – the "prevalence screen effect" or "lead time effect." Some of those cancers would eventually have been found even without screening, just at a later stage, while others would have remained dormant and harmless throughout the woman’s life.
If a study or analysis concludes before enough time has passed for this initial surge to be balanced by a subsequent decline in diagnosis rates (as the pool of "prevalent" cases is cleared), researchers may incorrectly attribute a larger portion of the early increase to overdiagnosis. Essentially, if you only look at the first few years, it appears many more cancers are being found, but without a long-term perspective, it’s impossible to distinguish between cancers simply found earlier (lead time bias) and those that would never have progressed (true overdiagnosis).
Furthermore, estimates can be significantly distorted when women in control groups (who were initially not offered screening) later receive screening themselves outside the trial protocol. This "contamination" blurs the distinction between the screened and unscreened populations, making it harder to accurately assess the unique impact of the organized screening program. If the control group eventually gets screened, their cancer detection rates will also increase, making the difference between the intervention and control groups appear smaller, potentially masking the true benefits of screening and complicating the measurement of overdiagnosis.
The new analysis meticulously accounts for these critical timing effects, including the duration of follow-up, the extent of screening exposure in both groups, and the natural evolution of incidence rates over time. By doing so, the researchers were able to demonstrate that these factors, when properly addressed, can substantially change estimates of how often mammography identifies cancers that would otherwise never have caused a problem. Their refined methodology provides a much more robust and accurate picture of overdiagnosis.
What Overdiagnosis Means for Women: Reassurance and Informed Choice
Understanding both the benefits and potential downsides of screening is paramount for women deciding whether to participate in breast cancer screening programs. The historical uncertainty and high estimates of overdiagnosis have contributed to widespread confusion and anxiety, sometimes leading women to forgo screening despite its proven mortality benefits.
"Most women will not develop breast cancer in their lifetime, but for those who do, detecting it early significantly improves prognosis and can prevent premature death. With this study, we can now offer substantial reassurance that the benefits of detecting breast cancer early and preventing premature death will comfortably outweigh the small risk of unnecessary treatment due to overdiagnosis," affirms Sisse Helle Njor. She highlights the ethical imperative of providing accurate information.
"With this robust evidence in mind, we hope this study will provide a clear framework for a more realistic interpretation of the evidence surrounding breast cancer screening. This, in turn, should help us better inform women when they are invited for screening, enabling them to make truly informed decisions based on a balanced and accurate understanding of the risks and benefits," Njor concludes. This clearer picture is vital for public health campaigns aiming to promote screening uptake and for individual physicians counseling their patients. Reducing the perceived risk of overdiagnosis may encourage more women to participate in screening, ultimately saving more lives.
Facts: About the Study
The researchers performed a comprehensive new analysis of existing mammography screening research, synthesizing data from the most important studies in the field.
Their analysis included all eight randomized controlled trials (RCTs) dedicated to mammography screening:
- The New York Health Insurance Plan (HIP)
- Malmö
- Two-County
- Edinburgh
- The Canadian National Breast Screening Study (CNBSS)
- Stockholm
- Gothenburg
- UK Age
Two well-established regional screening programs in Denmark were used as a crucial real-world reference, providing contemporary data on how screening influences incidence over time in a population setting. The researchers rigorously examined the incidence of both invasive breast cancer and ductal carcinoma in situ (DCIS), offering a complete view of screen-detected pathologies.
When reassessing the earlier trials and integrating them with the Danish data, the team specifically focused on three critical methodological factors that can significantly influence estimates of overdiagnosis:
- Length of follow-up: Ensuring that studies had sufficient time for the lead time effect to manifest fully and for any subsequent decline in incidence to be observed, thereby distinguishing true overdiagnosis from early detection.
- Differential screening exposure: Accounting for the fact that women in control groups might have received screening outside the trial, which can dilute the observed differences and skew overdiagnosis estimates.
- Temporal patterns of incidence: Analyzing how breast cancer diagnosis rates evolve over time in both screened and unscreened populations, paying close attention to the initial surge followed by a potential decline.
After meticulously taking these differences in screening exposure and follow-up duration into account, the researchers arrived at their robust conclusion: overdiagnosis may be substantially less common than earlier, often-cited estimates suggested. This new synthesis provides a much-needed clarification in a long-standing scientific and public health debate.
Funding
This important research was made possible through dedicated support for the contributing scientists. Casper Urth Pedersen is supported by the Novo Nordisk Foundation (reference: NNF22OC0076184), a leading philanthropic organization focused on improving people’s health and society. Matejka Rebolj is supported by Cancer Research UK (reference: C8162/A29083), the world’s largest independent cancer research charity, highlighting the significance of this work in the broader fight against cancer. The collaborative effort across institutions and national borders underscores the global importance of accurately assessing the impact of breast cancer screening.
