 What was the impact of the 2009 shift in recommendations to delay routine screening until age 50? Ironically, mammography rates of women in their 40s may have actually gone up. The thought is that all the media attention may have just reminded women about it, underscoring the need to better translate evidence into practice. The new recommendations bring the US closer to European standards, mammograms over a few years starting at age 50. In 2015, the American Cancer Society split the difference and recommended starting at 45 annually, then switching to every other year at 55, suggesting this would decrease the lifetime risk of dying from breast cancer from 2.7% down to under 2%. Based in part on a systematic review performed by the Cochrane Collaboration, a highly respected bastion of evidence-based medicine. The authors of the Cochrane Review, however, wrote in to say they used the wrong number, and that if you look at the studies they considered were adequately randomized, there did not appear to be any significant mortality benefit from mammograms at all, and that the data certainly do not support the popular idea that breast cancer screening saves lives, or choosing the American Cancer Society of being more of a political organization with financial ties to the multibillion-dollar mammogram industry. Here's the Cochrane Review they're talking about, which concluded that the studies which provide the most reliable information evidently show that screening did not reduce breast cancer mortality. If that's true, then that changes everything. They conclude that the time is therefore come to reassess whether routine mammograms should be recommended for women of any age, which is what the Swiss Medical Board did. They were struck by how non-obvious it was that the benefits of mammography screening outweighed the harms. It's easy to promote mammography screening when the majority of women believe that it prevents or reduces the risk of getting breast cancer and saves many lives through early detection of aggressive tumors. I mean, if those beliefs were valid, they'd be all for it. Unfortunately, they concluded they are not, and so believe that women need to be told so. From an ethical perspective, a public health program that does not clearly produce more benefits than harms is hard to justify. Their report, not surprisingly, caused an uproar. Scientists argued that the report unsettled women, but we wonder how to avoid unsettling women given the available evidence. If you ask women what they perceive to be the benefits of regular mammogram screening, they think it cuts the risk of dying from breast cancer in half. That's the orange box saving the lives of about one in 12 women, but in reality, the numbers appear to look like this. First of all, the risk of dying from breast cancer regardless is smaller than most women think, and the reduction in risk. The differences in size of the orange boxes is much smaller, maybe only 1 in 1,000 women benefits. But hey, doesn't saving the life of even 1 in 1,000 women make it all worth it? Imagine if you were in that 1 in 1,000 families whose mom was saved. But that may not be true. 1 in 1,000 women's screen may not die from breast cancer, so there's no evidence suggests that overall mortality was affected, which would mean no lives are actually saved. There's been 10 randomized trials of mammogram screening and not one has ever showed an overall mortality benefit. Wait, how does that make sense? If mammograms prevent 1,000 women from dying from breast cancer, then the only way no lives are saved is if mammograms ended up somehow leading to the deaths of 1,000 healthy women. That's preposterous, right? Let me introduce the concept of overdiagnosis, the fact that some of the tiny tumors picked up on mammograms may have never progressed, or even disappeared on their own. And so had they not been picked up, the woman would have been none the wiser and would have never been affected by it, or even known they had it. But once you pick up a cancer on a mammogram, you have to treat it since you don't know what it's going to do. But in some cases, the overdiagnosed cases where it would never hurt you, you're treating the breast cancer unnecessarily. How common is that, though? For every life saved by mammography, as many as 2 to 10 women are overdiagnosed, meaning turned to the breast cancer patients unnecessarily, along with all the attended harms of chemo, radiation, or surgery without the benefits. Horms can include death. Imagine being in the family whose mom was killed. The concern is that unnecessary radiation treatments may kill as many as are saved, hence the no evidence of net mortality benefit. Radiation treatments to the chest increase the risk of dying from heart disease and lung cancer. Now those may be acceptable risks. If you actually have breast cancer, that would otherwise kill you. Breast that are beneficial for real patients can be lethal for those who never should have been treated in the first place. Even if mammograms don't save your life, might they save your breast? I mean, if you catch a tumor early, maybe you can avoid a mastectomy. The opposite may actually be true. The Cochrane researchers explained that that's why they published their report. They thought it was important for women to know that screening may increase their risk of losing a breast. Basically, mammograms have been promoted to the public with three simple promises that all appear to be wrong. Screening does not seem to make the women live longer. Instead, may unnecessarily increase mastectomies and cancers are not caught early. It may take decades for a tumor to grow large enough to be picked up on a mammogram. And even when they are, they may not grow any further. That's the concern we're catching too many. There's so much overdiagnosis that if a woman really doesn't want to become a breast cancer patient, maybe they should avoid mammogram screening altogether. But if you have breast cancer, don't you want to know? The small probability that a woman may avoid breast cancer death must be weighed against the more likely scenario that she may have a false positive or false negative result, or most critically, a diagnosis and treatment of cancer that would otherwise not have threatened her health or even come to her attention.