The United Kingdom's National Health Service offers breast cancer screening every three years to women aged 50 to 70. That interval, in place since the programme's expansion in the early 2000s, has become a subject of ongoing discussion in epidemiology. Some researchers argue that shortening it to two years would catch more cancers at an earlier stage and reduce mortality. Others counter that the harms of more frequent screening—overdiagnosis, false positives, and cumulative radiation—already outweigh the benefits at the current interval. The debate, long simmering in academic journals, has spilled into policy discussions as the UK National Screening Committee prepares for its next review.

The 3-Year Interval That Divides the Field

The current three-year schedule stems from the 2012 Marmot review, which concluded that the NHS Breast Screening Programme prevents roughly 1,300 deaths per year while causing about 4,000 overdiagnosed cases—cancers that would never have caused symptoms. The review deemed the interval reasonable given the balance of benefits and harms. But new modelling from Queen Mary University of London, published in 2024, challenges that equilibrium. The model, which simulates tumour growth and detection over time, suggests that a two-year interval could reduce breast cancer mortality by an additional 8–12% compared with three-year screening. However, the model also predicts a 15–20% increase in overdiagnosis.

The disagreement hinges on how one weighs a prevented death against an unnecessary diagnosis. Epidemiologists who prioritise mortality reduction tend to favour shorter intervals. Those who emphasise patient-centred outcomes—avoiding anxiety, unnecessary biopsies, and overtreatment—argue that the current interval already pushes the limits of acceptable harm.

The NHS programme screens around 2 million women annually. A shift to two-year intervals would require roughly 50% more mammograms, straining radiology capacity and increasing wait times for diagnostic workups. The cost implications are substantial, though estimates vary widely. The UK National Screening Committee has historically been cautious about changing intervals without clear evidence of net benefit.

Overdiagnosis: The Harms That Rarely Make Headlines

Overdiagnosis refers to the detection of a cancer that would never have progressed to cause symptoms or death during a woman's lifetime. In breast cancer, estimates of overdiagnosis range from 1% to 10% of screen-detected cases, depending on the definition and study methodology. The Harding Center for Risk Literacy at the University of Potsdam has flagged the uncertainty in these estimates, noting that different modelling assumptions produce widely divergent figures.

When a woman is diagnosed with breast cancer through screening, she typically undergoes surgery, often followed by radiation and endocrine therapy. If the cancer is overdiagnosed, those treatments provide no benefit—only harm. The consequences include pain, scarring, fatigue, and long-term side effects such as cardiac toxicity from radiation or bone loss from aromatase inhibitors. The psychological impact is harder to quantify but equally real: the anxiety of living with a cancer diagnosis, the impact on family and work, and the altered body image.

Proponents of longer screening intervals argue that overdiagnosis is more common with slower-growing, low-risk tumours, which are more likely to be detected by screening. A three-year interval may allow some of these indolent cancers to be missed, which is arguably a benefit, not a harm. Critics counter that this reasoning assumes women are willing to accept a higher risk of interval cancers—cancers that appear between screens—in exchange for avoiding overdiagnosis.

The NHS Breast Screening Programme provides information leaflets that mention overdiagnosis, but studies suggest that most women still overestimate the benefit of screening and underestimate the potential for harm.

To illustrate the real-world impact, consider the case of a 58-year-old woman in Manchester who was diagnosed with ductal carcinoma in situ (DCIS) through routine screening. DCIS is a non-invasive condition that may or may not progress to invasive cancer. She underwent a lumpectomy and radiation, experiencing fatigue and breast tenderness for months. A decade later, she still worries about recurrence, though the original lesion may never have posed a threat. Such stories are common but rarely captured in mortality statistics.

Why Two Years Might Not Be Better

Shortening the screening interval would increase cumulative radiation dose from mammography. While the risk from a single mammogram is small—roughly equivalent to a few months of natural background radiation—the lifetime risk from repeated exposure is not negligible, particularly for women with dense breasts who may also undergo supplemental imaging. The International Agency for Research on Cancer classifies ionising radiation as a carcinogen, and modelling studies suggest that annual screening from age 40 could cause one radiation-induced cancer per 100,000 women screened.

Recall rates—the proportion of women called back for additional imaging—are higher with shorter intervals. In the NHS programme, the recall rate is roughly 4–5% after a three-year screen. A two-year interval would increase the cumulative probability of a false positive over a decade of screening. False positives lead to anxiety, additional imaging, and often biopsy, with no cancer found. The psychological impact can persist for months.

Interval cancers—cancers diagnosed between scheduled screens—occur even with annual screening. In the UK, about 25–30% of breast cancers in screened women are interval cancers. The proportion is higher in women with dense breasts and in younger women. A shorter interval reduces the rate of interval cancers but does not eliminate them. The UK Age Trial, which randomised women to annual mammography from age 40 or to usual care (three-year screening starting at 50), found no mortality benefit from earlier and more frequent screening after 10 years of follow-up.

Cost-effectiveness analyses consistently favour a three-year interval in the UK context. The NHS has a fixed budget, and money spent on more frequent mammography is money not spent on other interventions, such as improving treatment access or expanding screening for other cancers. The trade-off is explicit in health technology assessments, but it rarely features in public debate.

A counter-argument from advocates of shorter intervals is that the cost-effectiveness models underestimate the value of earlier detection in reducing the need for aggressive chemotherapy. A 2023 analysis from the Institute of Cancer Research suggested that biennial screening could reduce the proportion of women requiring mastectomy by around 5–10%, which would offset some of the added screening costs through lower treatment expenses. However, the analysis relied on assumptions about tumour growth rates that are not universally accepted.

The Nottingham Study That Shifted the Debate

A 2020 study led by Stephen Duffy at Queen Mary University of London, using data from the NHS Breast Screening Programme, examined the relationship between screening interval and tumour characteristics. The study found that cancers detected at a three-year interval were, on average, larger than those detected at a two-year interval. Tumour size at detection is a known predictor of prognosis, so this finding seems to support shorter intervals. However, the study also found no significant difference in stage distribution or survival between the two interval groups.

Critics pointed out that the study was observational, not randomised, and thus vulnerable to confounding. Women with a family history of breast cancer, for example, might be more likely to attend screening regularly and might also have different tumour biology. Lead-time bias—the apparent prolongation of survival due to earlier diagnosis rather than delayed death—could also explain the lack of survival difference. The study adjusted for some confounders but could not eliminate all.

Despite its limitations, the Nottingham study has been influential. It provided real-world data from a large, organised screening programme, which is more generalisable than trial populations. The finding that interval length affects tumour size but not survival has been cited by both sides: those who favour shorter intervals argue that size matters for treatment options and quality of life, while those who favour longer intervals argue that survival is the more important endpoint.

The study also highlighted the heterogeneity of breast cancer. Some tumours are fast-growing and will present as interval cancers regardless of screening frequency. Others are slow-growing and may be overdiagnosed even with triennial screening. A one-size-fits-all interval may never be optimal.

Another important piece of evidence comes from the Swedish Two-County Trial, which compared screening intervals of roughly two years (in the intervention group) with no screening (in the control group). The trial reported a 30% reduction in breast cancer mortality with screening, but the interval was not directly compared with a longer interval. Subsequent observational analyses from Sweden suggested that women screened every 18–24 months had similar mortality to those screened every 24–30 months, raising questions about the marginal benefit of shorter intervals.

Risk-Stratified Screening as a Way Out

Risk-stratified screening offers a potential path beyond the interval debate. Instead of a uniform three-year interval, women would be assigned to different screening schedules based on their individual risk, as determined by breast density, family history, genetic variants, and lifestyle factors. The PROCAS study, led by Gareth Evans at the University of Manchester, demonstrated the feasibility of this approach in a primary care setting. Women were invited to complete a risk questionnaire and, if eligible, received a mammogram and breast density assessment. High-risk women were offered annual or biennial screening; low-risk women were offered screening every four to five years.

The results showed that risk stratification could maintain the same rate of cancer detection while reducing the number of screens by up to 30% in low-risk groups. This would free up capacity for more frequent screening in high-risk groups without increasing overall programme costs. The NHS has begun piloting risk-based pathways, with a planned rollout in 2025–2027. The pilots will evaluate the logistics of implementing risk assessment at scale, including the use of polygenic risk scores, which combine many small-effect genetic variants into a single risk estimate.

Critics worry that risk stratification could increase health inequalities. Women with lower health literacy may be less likely to complete risk questionnaires or attend screening if they are deemed low-risk. There is also the risk of overmedicalising low-risk women who are told they need less screening, which could cause anxiety. The ethical implications are still being debated.

Risk stratification does not eliminate the interval debate; it reframes it. For high-risk women, the question becomes: should they be screened annually, or every two years? For low-risk women, can a four-year interval be justified? The evidence base for these tailored intervals is thin, relying largely on modelling rather than randomised trials. The UK National Screening Committee has called for more research before adopting risk-based intervals nationally.

A separate but related development is the use of digital breast tomosynthesis (DBT), which provides 3D imaging and may reduce recall rates. The NHS is gradually introducing DBT in some screening centres, but the technology is more expensive and increases radiation dose. Whether DBT would allow longer intervals to be safe is an open question. A 2022 modelling study from the University of Warwick suggested that DBT every three years might achieve similar outcomes to standard mammography every two years, offering another potential compromise.

What Clinicians Actually Face in the Consultation

In a typical GP consultation, a 52-year-old woman might ask, "Should I go for screening?" The GP knows that the number needed to screen to prevent one breast cancer death over 20 years is roughly 200 women. The absolute risk reduction is about 0.5%. The GP also knows that about 5% of screened women will have a false positive, and that some will be overdiagnosed. Explaining these numbers in a way that respects the patient's values is difficult.

Shared decision-making is the ideal, but it requires time and communication skills that are in short supply. Many GPs report discomfort with the uncertainty inherent in screening advice. They may default to a recommendation to attend screening because it is the official policy and because missing a cancer feels more blameworthy than causing a false positive.

The Harding Center's work on risk communication has shown that patients systematically overestimate the benefit of screening and underestimate the harm. When presented with absolute risks, many women still choose screening, but they do so with a more accurate understanding. The challenge is implementing this in a 10-minute appointment.

Some clinicians argue for a more paternalistic approach: the evidence supports screening, so recommend it. Others favour a more nuanced discussion that acknowledges the trade-offs. The interval debate adds another layer of complexity. Should a GP tell a patient that the current interval is contested? Most would say no, because it would confuse patients without clear guidance. But as the evidence evolves, that position may become harder to defend.

Consider the experience of a GP in Leeds who participated in a shared decision-making training programme. She now uses a decision aid that shows women their absolute risk of dying from breast cancer with and without screening, as well as the risk of overdiagnosis. She reports that most women still choose screening, but they ask more informed questions. A few women have declined screening after seeing the numbers, and she respects their choice. The aid takes about five extra minutes per consultation, which is a significant time investment in a busy practice.

The Unresolved Question for Policy Makers

The UK National Screening Committee reviews the evidence for each screening programme at regular intervals. The breast cancer programme was last reviewed in 2020, and the committee concluded that the three-year interval should remain unchanged. However, the modelling from Queen Mary University and the ongoing risk-stratification pilots have prompted calls for an earlier review. The next comprehensive review is expected after the UK Cancer Research data release in 2027, which will include updated mortality and overdiagnosis estimates.

Policy makers face political pressure to appear proactive on cancer outcomes. Breast cancer is the most common cancer in women in the UK, and any change to screening policy attracts intense media scrutiny. A move to two-year screening would be portrayed as a life-saving intervention; a move to risk-based screening could be seen as rationing. The resource trade-off with other screening programmes—bowel and cervical—is a constant constraint.

The debate is unlikely to be resolved by a single study. It reflects deeper differences in how epidemiologists conceptualise benefit and harm, and how societies value early detection versus avoiding unnecessary treatment. The absence of consensus is not a failure of science; it is a reflection of the irreducible uncertainty in preventive medicine. Until better biomarkers distinguish lethal from indolent cancers, the interval question will remain a matter of judgment, not proof.

This article is for informational purposes only and does not constitute personalised medical advice. Screening decisions should be made in consultation with a healthcare provider based on individual risk factors and preferences.