Three years ago, a little-known panel, the U.S. Preventive Services Task Force, elbowed its way into the national conversation with a startling recommendation that most women under 50 could forgo the annual mammogram.

The finding stuck a finger in the eye of conventional wisdom. A generation or more of women had come of age believing regular screening is the best way to protect oneself from dying of cancer. Arriving in the midst of President Barack Obama's health care reform effort, the recommendation fed into opponents' conviction that government was trying to force its way into life-or-death decisions. Members of Congress blasted the task force. Doctors continued to recommend mammograms, and women kept having them.

But the findings, and the science behind them, did not go away. For more than a decade, multiple studies have advanced the argument that routine screening for some forms of cancer doesn't prevent premature death and may do more harm than good. The findings of a large study released on Thanksgiving Day added more weight.

Research in The New England Journal of Medicine looked at more than 30 years of U.S. health statistics. It found that mammograms are effective at detecting early stage breast tumors, but they don't reduce the number of women diagnosed with advanced stage cancers.

The conclusion was that early detection doesn't save lives. Rather, it prompts millions of women to undergo expensive, intrusive treatment for tumors that would never become life-threatening.

Similar mixed messages have been sounded to men about screenings for prostate cancer. Studies have found that early detection and aggressive treatment of that disease may be unnecessary, ineffective and risky.

This is the third rail of health care - the science that no one knows how to come to grips with. Not doctors, not patients, not politicians.

Is it time to rethink the value of preventive screenings, and cut back on the costly diagnosis and treatment options that come with early detection? Doctors and politicians can quibble with the accuracy and methods of individual studies. It becomes harder to push against their combined weight.

And yet four decades of doctors' warnings and public information campaigns will not be easily undone. A vast industry of equipment, technology, expertise and advocacy built around screening and early detection won't be easily dismantled, either, no matter how earnestly we talk about putting the brakes on health care costs.

Personally, I would love to skip the annual mammogram. But unless or until my doctor advises me to do so, I'm not sure I will. Studies are remote; my long-time internist is real.

The screen-or-not-screen question is actually the easy part. The real dilemma comes if a mammogram or a PSA test turns up something suspicious. Research indicates that often a low-tech treatment called watchful waiting is as effective as more aggressive interventions. But that's not how most patients or the U.S. medical system are wired.

We are at the place where technology, science and data collide. But those three elements are also the way forward.

"I'm still looking for the data that's going to give me a better basis on which to weigh in," said Tarris Rosell, a scholar of medical ethics on staff at the Center for Practical Bioethics in Kansas City. "Good ethics starts with good facts."

Barbara Shelly is a columnist for the Kansas City Star. Readers can email her at bshelly@kcstar.com.