I have been a doctor for 10 years. I have treated many illnesses, but I have also spent many hours trying to stop people from getting ill in the first place. I support people in stopping smoking, in drinking sensibly, in shedding extra pounds - because prevention is better than cure, right? The rationale seems so sensible that it would be illogical - mad, even - to think otherwise. We immunise and stop measles, rubella and meningitis C in their tracks; we seek out high blood pressure and cholesterol and treat them before they lead to heart attacks and strokes. And we use mammography to screen for breast cancer.

Since 1988, the NHS Breast Screening Programme has been checking women aged between 50 and 64 for breast cancer. Every three years, women are invited to a clinic to have their breasts x-rayed, and those x-rays are analysed for signs of cancer. Some women will need a biopsy, using a needle. Some women will be told they have cancer. Most will not: the programme estimates that, since its launch, 17 million women have been screened and nearly 100,000 breast cancers have been detected.

But screening is often misunderstood. It is for women with no breast symptoms at all. A woman with a breast lump needs a fast diagnosis after assessment from her GP. Screening, when there are no symptoms, is a completely different situation. Misunderstanding this can be tragic: I know of one woman, aware that she had a lump, who waited patiently - too long - for a screening appointment. And many people have the impression that screening is a perfect test. It’s not. Even with the best staff, training and equipment, screening is never perfect. A "false positive" for example, can suggest that a problem exists, when in fact there isn’t one, causing needless alarm. The opposite - an all-clear where a real problem is missed.

Breast screening is also highly controversial, although most of the controversy takes place in medical journals, and seldom reaches the women who actually have the screening. Professor Mike Baum is emeritus professor of surgery and visiting professor of medical humanities at University College London, and with more than 30 years’ experience in the field of breast cancer, has been a critic of breast screening for many years. He is on record as saying that there is a "subtle deception encrypted within the invitations for mammographic screening".

Breast screening - dutifully wedging one’s bosoms in an x-ray machine, anxiety, recalls, biopsies - is only worthwhile if you gain more than you lose. But it is possible to lose.

Consider this. Over the last couple of years, it has become clear that teaching women to examine their own breasts actually does no good. Women, the study found, noticed cancerous lumps without needing to be taught to do a formal, regular breast check. Those who were encouraged into a routine of self-examination found more lumps - but died from no more cancers than women who were not taught the self-check. The bottom line was that teaching regular self-checks could lead to more harm - traumatic biopsies, lump removals, worry - than gain.

In the case of breast-cancer screening, it’s far more complex. Are we doing ourselves a favour by picking up early breast cancer? All breast cancers are not all the same. Not all are aggressive, and not all are lethal. A Norwegian study published in the British Medical Journal earlier this year illuminates the point: the researchers found that cases of invasive breast cancer in women eligible for screening had risen dramatically. However, after the age of 69, when women had "finished" being screened, there was no fall in the amount of breast cancer being diagnosed. Breast screening is meant to pick up cancer that you didn’t know about. As time goes on, and early cancers are picked up by screening, and treated, there should be fewer new cases diagnosed. But that didn’t happen. The researchers concluded that a third of the invasive cancers found by screening would never have become life threatening. In other words, there was "no point" in this group of women knowing that they had breast cancer. It was not going to interfere with their lifespan.

Should this concern us? Hazel Thornton was diagnosed with Ductal Carcinoma in Situ (DCIS) at a breast screening several years ago. Since then, she has become an honorary visiting fellow at the Department of Health Sciences at the University of Leicester. "Finding it early may not be a good thing," she says. "In fact, finding it ‘too early’ can lead you to a whole raft of uncertainties and problems and dilemmas."

The problem is that DCIS is common (one in five of the breast cancers found at screening is DCIS) yet the natural course and best treatment for it is uncertain. It is a not an aggressive cancer, but since some - thought to be around 30 per cent - do become invasive over the following decades, it can’t be dismissed. A woman in this position has to deal with knowing that she harbours an "in-between" cancer, with all the insurance, employment and family worries this brings - as Hazel Thornton discovered for herself.

As a doctor, I need to give people honest, impartial information. But clear information is difficult to come by. For example, the commonly used figure of a woman’s risk of breast cancer as being "one in nine" is misleading- It refers to a lifetime risk of breast cancer and is only true for a woman in her 85th year, younger women having a lower risk.

There is a huge debate about how many lives are saved by breast screening, because the treatment and the surgery itself are not risk-free. It is thought that the radiation from mammograms causes around one new breast cancer case for every 170 breast cancers detected by screening. As professor Mike Baum says, "You have to screen 1,200 women for 10 years to save one life from breast cancer - assuming that no life is lost as a result of the screening process."

Overall, Is breast screening a good thing? Julietta Patnick, director of the breast screening programme in England, says that it is. "There is no group of women who wouldn’t benefit from screening, as long as they are in good health and with a reasonable life expectancy. Each woman whose life is saved by breast screening will gain an average of 20 years, but it should be a personal decision." Even with an "over-diagnosis" of cancer? "You can’t tell if an individual woman will benefit or not. Some women gain nothing, others get 20 years," she comments.

So, if you’re invited for screening, what should you do? It’s a difficult decision to make. If screening were undoubtedly good, with no side-effects, I would feel justified in recommending it - just as I do when I recommend quitting smoking. But breast screening is different. We have to face the fact that the benefits are not so clear-cut. While the chance of having your life saved by breast screening exists, it is slender.

So, do I recommend it? I am by no means against it, but I do think that each woman should consider the pros and cons carefully. And that the debate about screening should not just be held in medical journals. Women need to know about the harms, limitations and consequences of breast screening, not just the benefits.