Why finding more breast cancers doesn't always save lives
Identification of the Fraction of Indolent Tumors and Associated Overdiagnosis in Breast Cancer Screening Trials
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Surprising Findings
The fraction of indolent tumors cannot be precisely estimated from standard screening data — even with advanced mathematical models — because the data lacks enough information.
People assume that with enough data and fancy math, we can solve any medical mystery. But this study proves that even the best models fail if the data doesn’t capture the full timeline — a fundamental limit, not a technical flaw.
Practical Takeaways
If you’re a patient, ask your doctor: 'Is this tumor likely to be indolent? Do we have enough follow-up data to know?'
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Surprising Findings
The fraction of indolent tumors cannot be precisely estimated from standard screening data — even with advanced mathematical models — because the data lacks enough information.
People assume that with enough data and fancy math, we can solve any medical mystery. But this study proves that even the best models fail if the data doesn’t capture the full timeline — a fundamental limit, not a technical flaw.
Practical Takeaways
If you’re a patient, ask your doctor: 'Is this tumor likely to be indolent? Do we have enough follow-up data to know?'
Publication
Journal
American Journal of Epidemiology
Year
2018
Authors
M. Ryser, R. Gulati, M. Eisenberg, Yu Shen, E. Hwang, Ruth Etzioni
Related Content
Claims (10)
Finding and treating very slow-growing cancers through aggressive screening doesn’t help people live longer overall — it just finds cancers that wouldn’t have hurt them anyway.
Some cancers found during routine screening might never hurt you — they grow so slowly that you’d die of something else before they ever became a problem.
When doctors stop screening for breast cancer, they can't tell exactly how many of the tumors found would never have caused harm, even with fancy math, because the data they have just isn't detailed enough to give clear answers.
If you try to use a simple model that only accounts for fast-growing cancers to analyze data that also includes slow-growing ones, you’ll wrongly think the fast-growing cancers start earlier and last longer than they really do.
Scientists have shown that if new breast cancers develop slower than existing ones grow, then a certain math model can tell the difference between slow-growing and fast-growing cancers.