To Mammogram or Not to Mammogram: That Is the Question

PerfeQTion™ Imaging & Wellness · October 21, 2026

A Position Paper on Screening Mammography

Jenn Simmons, MD

For more than 50 years, women have been told that mammograms save lives.

I believe it is time to ask whether the evidence actually supports that promise.

Mammography unquestionably finds breast cancer. It finds more breast cancer the more frequently we use it.

But a cancer found is not the same thing as a life saved.

That distinction is the foundation of my position.

After decades of randomized mammography screening, there has never been a demonstrated overall-survival benefit from screening mammography. Women assigned to mammographic screening have not been shown to live longer overall than women who were not assigned to screening.

And yet that is almost never what women are told.

They are told that because a mammogram found their cancer, the mammogram saved their life.

Those are not the same thing.

A Cancer Found Is Not a Life Saved

This is the fundamental flaw in the way we talk about mammography.

When a mammogram finds a cancer, we assume that woman has become a life saved.

That is simply not true.

There are two major reasons screening can make survival statistics look dramatically better without actually extending life: lead-time bias and overdiagnosis.

Lead-time bias

Suppose a woman’s cancer becomes clinically apparent at age 65 and she dies at 70.

Her survival after diagnosis is five years.

Now suppose mammography finds exactly the same cancer at age 60.

She still dies at 70.

Her survival after diagnosis has doubled to ten years.

But her actual life has been extended by:

zero days.

Screening moved the date of diagnosis.

It did not move the date of death.

That is lead-time bias.

And that is why longer survival after a mammographic diagnosis cannot be used as evidence that mammography saved a life.

But overdiagnosis creates an even more profound distortion.

Overdiagnosis

Screening creates a new population of breast-cancer patients: women with cancers that would never have become clinically apparent or threatened their lives.

That is overdiagnosis.

These are not false positives.

The pathology is real.

But the disease was never going to harm the woman.

And something very important happens when we add these women to breast-cancer statistics.

They become breast-cancer survivors.

If a mammogram finds a cancer at age 50 that never would have affected that woman’s life, and she lives to 85, she becomes a:

35-year breast-cancer survivor.

But mammography did not give her 35 additional years of life.

It gave her 35 years of knowing she had breast cancer.

Her excellent survival is then added to the survival statistics for screen-detected breast cancer, making screening appear extraordinarily successful.

The National Cancer Institute describes precisely this statistical distortion. In its illustrative example, adding overdiagnosed cancers increases five-year cancer survival from:

40% to 80%

while the number of people who die remains:

exactly the same.

Think about what that means.

SURVIVAL DOUBLED. DEATHS DID NOT CHANGE.

That is why survival after diagnosis cannot prove that mammography saves lives.

Screening can make survival statistics look better in two entirely different ways without saving a life:

It can start the survival clock earlier.

And:

It can add women to the breast-cancer population whose lives were never threatened by their cancer in the first place.

This is why the statement:

“Women whose cancers are found by mammography survive longer.”

does not answer the question:

“Does mammography make women live longer?”

Those are fundamentally different questions.

The endpoint that matters is not how long women survive after we label them with breast cancer.

The endpoint that matters is whether screening changes when women die.

A cancer found is not a life saved.

And a woman turned into a breast-cancer survivor by overdiagnosis is not evidence that screening saved her life.

The Canadian Trial: More Cancer Found. The Same Number of Women Died.

The Canadian National Breast Screening Study randomized 89,835 women aged 40–59.

During the screening period, mammography found substantially more invasive cancers:

Mammography: 666

Control: 524

So mammography worked exactly as intended.

It found more cancer.

Now look at what actually matters.

After up to 25 years:

Mammography: 500 breast-cancer deaths

Control: 505 breast-cancer deaths

Hazard ratio:

0.99

Meaning-

There was essentially no difference in breast-cancer mortality.

Meanwhile, cancers continued to accumulate disproportionately in the screened group. The investigators estimated that 22% of screen-detected invasive breast cancers were overdiagnosed—approximately one overdiagnosed invasive cancer for every 424 women screened.

Miller AB, Wall C, Baines CJ, Sun P, To T, Narod SA. BMJ. 2014;348. PMID 24519768.

The Canadian trial has been criticized. Fine. Debate its randomization. Debate the machines used in the 1980s.

But do not pretend the mortality result does not exist:

500 versus 505.

And saying that today’s machines find more cancer does not answer that result.

It simply tells us today’s machines find more cancer.

The question is whether finding more changes who dies.

So far the answer to that is a resounding no.

Cochrane: No Demonstrated Overall-Survival Benefit

The Cochrane Collaboration reviewed seven randomized mammography trials involving approximately 600,000 women.

Among the trials Cochrane judged adequately randomized:

Breast-cancer mortality

RR 0.90; 95% CI 0.79–1.02

Not statistically significant.

All-cause mortality

RR 0.99; 95% CI 0.95–1.03

No demonstrated benefit.

Total cancer mortality, including breast cancer

RR 1.02; 95% CI 0.95–1.10

Again, no demonstrated benefit.

Historical and adequately randomized screening trials have never demonstrated that women assigned to mammography live longer overall.

That is very different and somewhat less catchy than the slogan:

“Mammograms save lives.”

And Cochrane found something else women are rarely told.

Women invited to screening underwent more breast surgery, not less.

Lumpectomy and mastectomy combined:

RR 1.31

Mastectomy alone:

RR 1.20

So when women are told mammography “saves breasts,” that claim also deserves scrutiny.

The American College of Physicians Is Finally Acknowledging This in 2026

This is not simply an argument about old studies.

In April 2026, the American College of Physicians published its updated guidance:

Qaseem A, Harrod CS, Balk EM, Etxeandia-Ikobaltzeta I, Crandall CJ. Screening for Breast Cancer in Asymptomatic, Average-Risk Adult Females: A Guidance Statement From the American College of Physicians (Version 2). Ann Intern Med. 2026;179:842–856.

ACP continues to recommend biennial mammography for average-risk women aged 50–74.

That needs to be stated clearly.

But the evidence discussion is equally important.

The guidance reports that the USPSTF evidence review found no difference in all-cause mortality regardless of age when mammography was compared with no screening or usual care.

So why does the American College of Physicians continue to recommend screening mammography despite the evidence for no difference in all-cause mortality?

It’s a good question. I wish I had a good answer.

There Is a Reservoir of Breast Cancer That Never Becomes Clinical Disease

Overdiagnosis is not merely a mathematical theory.

Autopsy studies have repeatedly demonstrated that women can harbor malignant breast pathology without ever developing clinical breast cancer.

Nielsen et al., British Journal of Cancer, 1987

Researchers extensively examined the breasts of 110 women aged 20–54 undergoing medicolegal autopsy.

They found malignant breast pathology in:

22 women — 20%.

Only one woman had known clinical invasive breast cancer.

The remaining findings included:

• 2% invasive carcinoma
• 18% microfocal in-situ carcinoma.

In other words:

When investigators looked intensively enough, approximately one in five women had malignant breast pathology, despite almost none having clinical breast cancer during life.

Nielsen et al., Cancer, 1984

Among 77 women without a previous clinical breast-cancer diagnosis, researchers discovered:

• 1 occult invasive cancer
• 14 occult in-situ cancers.

That is:

15 of 77 women — 19.5%.

These women died with these abnormalities.

They did not die from them.

Systematic Review and Meta-analysis, 2017

A systematic review of 13 autopsy studies from 10 countries involving 2,363 autopsies found incidental cancer or precursor lesions in approximately:

19.5%.

That included invasive cancer, in-situ cancer and atypical hyperplasia.

And the investigators found something particularly important:

The more thoroughly pathologists examined the breast, the more disease they found.

That is the biological reservoir into which increasingly sensitive screening reaches.

The harder we look, the more abnormalities we will find.

But:

Finding more disease is not the same thing as preventing more death.

Overdiagnosis Is Not a Footnote

Depending upon the population, methodology and denominator used, published estimates of breast-cancer overdiagnosis vary considerably.

The National Cancer Institute discusses estimates extending as high as approximately 50%, while randomized studies generally produce lower estimates.

The Canadian randomized trial estimated:

22% of screen-detected invasive breast cancers

were overdiagnosed.

Other population analyses have produced estimates around 30%.

The exact percentage can be debated.

The existence of overdiagnosis cannot.

And now put the autopsy evidence beside the screening evidence.

We know there is a reservoir of occult breast pathology in otherwise healthy women.

We know increasingly sensitive screening finds more abnormalities.

We know some of those abnormalities would never become clinical disease.

And we know those women are subsequently counted as breast-cancer survivors.

This is exactly why:

MORE CANCER DETECTED ≠ MORE LIVES SAVED

Where Are the Missing Advanced Cancers?

If screening is working primarily by intercepting dangerous cancers before they progress, then the enormous increase in early-stage cancer should eventually be accompanied by an enormous decrease in late-stage cancer.

That is not what happened.

Bleyer and Welch examined U.S. breast-cancer incidence after widespread mammography.

Early-stage cancer increased by:

122 cases per 100,000 women.

Late-stage cancer declined by:

8 cases per 100,000 women.

Put those numbers beside each other:

+122 EARLY

−8 LATE

That discrepancy is enormous.

If those additional early cancers predominantly represented lethal cancers intercepted before progression:

Where are the missing advanced cancers?

The decline is nowhere close to proportional to the increase in early detection.

And that is exactly what we would expect if screening were detecting a substantial reservoir of disease that was never destined to become advanced disease.

Better Mammography Still Has to Prove Better Outcomes

The predictable response is:

“But today’s mammography is better.”

Of course it is.

Modern digital mammography and tomosynthesis find more abnormalities than the machines used decades ago.

But once again:

BETTER DETECTION ≠ BETTER OUTCOMES

Modern technology does not get to substitute cancer-detection rates for mortality.

If a newer technology finds even more of the reservoir of biologically inconsequential disease, higher sensitivity alone could increase diagnosis without proportionately reducing advanced disease or death.

That is precisely why contemporary technology must be judged on contemporary outcomes.

Does it reduce advanced disease?

Does it reduce mortality?

Does it improve overall survival?

Those are the endpoints that matter.

A better camera is not automatically a better screening program.

Mammography Is Not Harmless

Then there is the issue that receives far too little attention.

Mammography uses ionizing radiation.

Ionizing radiation is carcinogenic.

Breast tissue is radiation sensitive.

Therefore, the number of mammography-induced breast cancers cannot rationally be zero.

Miglioretti and colleagues modeled annual mammography from age 40 through 74.

For every 100,000 women undergoing that screening regimen, they estimated:

125 radiation-induced breast cancers

and:

16 resulting breast-cancer deaths.

For women near the 95th percentile of radiation exposure:

246 induced breast cancers per 100,000.

For women with large breasts requiring additional views:

266 radiation-induced breast cancers

and:

35 resulting deaths per 100,000 women.

Miglioretti DL, Lange J, van den Broek JJ, et al. Ann Intern Med. 2016;164:205–214. PMID 26756460.

These are modeled estimates.

They are not zero.

And this is a screening program.

We are deliberately exposing healthy, asymptomatic breast tissue to a known carcinogen in order to screen for the very cancer that exposure itself is capable of causing.

I do not believe we should dismiss that concern because the modeled absolute number is considered small.

A screening test should be safe.

If safer non-ionizing technologies can accomplish the goal, medicine has an obligation to investigate and validate them.

And providers have an obligation to use them. 

Waiting until we have the perfect study for alternative screening modalities that are safe, like QT or the Tears Test, comes at a cost. There is a cost in continuing to use mammogram

It Is Time to Put Mammography Back to the Test

Mammography has become so embedded in medicine that its continued use is treated as the starting assumption.

I think that is backward.

If mammography were presented to us today as a new population-screening technology, I would want it put through the same scrutiny we should demand of any intervention being offered repeatedly to tens of millions of healthy women.

I would want to know:

Does it make women live longer?

Does it meaningfully reduce advanced breast cancer?

Does it save breasts?

How much overdiagnosis does it create?

How many unnecessary biopsies, surgeries, radiation treatments and systemic therapies result?

How many breast cancers does its cumulative ionizing radiation create?

And how does it compare head-to-head with non-ionizing alternatives?

Those are not radical questions.

They are the questions we should ask of every screening test offered to healthy people.

A screening test should never be protected from scrutiny simply because it has become familiar.

Familiarity is not validation.

Standard of care is not synonymous with proof.

We should not ignore the facts and the data just because this is the way that it has “always been done”. 

My Position

Mammography finds cancer.

That has never been the question.

My position is that:

A cancer found is not a life saved.

Longer survival after diagnosis does not prove that a woman lived longer.

Lead-time bias can increase measured survival without adding one day to life.

Overdiagnosis creates breast-cancer patients and breast-cancer survivors out of women whose lives were never threatened by their disease.

There is a substantial reservoir of occult breast pathology in otherwise healthy women.

The rise in early disease has not been matched by a proportional disappearance of advanced disease.

Historical randomized mammography trials have never demonstrated that women assigned to screening live longer overall.

Mammography increases breast surgery rather than demonstrating that it “saves breasts.”

And mammography exposes healthy breast tissue to ionizing radiation that is predicted to cause a small but real number of breast cancers.

So no, I do not believe mammography deserves a permanent pass because it became the standard of care decades ago.

I believe it is time to put mammography to rest.

Not because women deserve less screening.

Because women deserve better screening.

Screening that is safe.

Screening that identifies disease that matters.

Screening that minimizes overdiagnosis.

Screening that does not itself contribute to breast cancer.

And screening that demonstrates improvement in the outcomes women actually care about—not simply improvement in how many abnormalities we can find.

After more than half a century, “we’ve always done it this way” is not enough.

Women deserve nothing less than the complete evidence and the freedom to decide what they want to do with it.

Jenn Simmons, MD

References

1. Miller AB, Wall C, Baines CJ, Sun P, To T, Narod SA. Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. BMJ. 2014;348. doi:10.1136/bmj.g366. PMID 24519768.
2. Gøtzsche PC, Jørgensen KJ. Screening for breast cancer with mammography. Cochrane Database Syst Rev. 2013;(6). doi:10.1002/14651858.CD001877.pub5.
3. Qaseem A, Harrod CS, Balk EM, Etxeandia-Ikobaltzeta I, Crandall CJ. Screening for Breast Cancer in Asymptomatic, Average-Risk Adult Females: A Guidance Statement From the American College of Physicians (Version 2). Ann Intern Med. 2026;179:842–856. doi:10.7326/ANNALS-25-05116. PMID 41996183.
4. Nielsen M, Thomsen JL, Primdahl S, Dyreborg U, Andersen JA. Breast cancer and atypia among young and middle-aged women: a study of 110 medicolegal autopsies. Br J Cancer. 1987;56:814–819. doi:10.1038/bjc.1987.296.
5. Nielsen M, Jensen J, Andersen J. Precancerous and cancerous breast lesions during lifetime and at autopsy: a study of 83 women. Cancer. 1984;54:612–615. PMID 6744199.
6. Thomas ET, Del Mar C, Glasziou P, Wright G, Barratt A, Bell KJL. Prevalence of incidental breast cancer and precursor lesions in autopsy studies: a systematic review and meta-analysis. BMC Cancer. 2017;17. PMID 29197354.
7. Welch HG, Black WC. Using autopsy series to estimate the disease “reservoir” for ductal carcinoma in situ of the breast: how much more breast cancer can we find? Ann Intern Med. 1997;127:1023–1028. PMID 9412284.
8. Bleyer A, Welch HG. Effect of three decades of screening mammography on breast-cancer incidence. N Engl J Med. 2012;367:1998–2005. doi:10.1056/NEJMoa1206809.
9. Welch HG, Prorok PC, O’Malley AJ, Kramer BS. Breast-cancer tumor size, overdiagnosis, and mammography screening effectiveness. N Engl J Med. 2016;375:1438–1447. doi:10.1056/NEJMoa1600249.
10. Miglioretti DL, Lange J, van den Broek JJ, et al. Radiation-induced breast cancer incidence and mortality from digital mammography screening: a modeling study. Ann Intern Med. 2016;164:205–214. doi:10.7326/M15-1241. PMID 26756460.
11. Trentham-Dietz A, Chapman CH, Jayasekera J, et al. Collaborative modeling to compare different breast cancer screening strategies: a decision analysis for the US Preventive Services Task Force. JAMA. 2024;331:1947–1960. doi:10.1001/jama.2023.24766.
12. National Cancer Institute. Breast Cancer Screening (PDQ®)—Health Professional Version.