Until last week, most women probably thought that checking their breasts monthly for lumps, bumps or changes was a good idea. The general theory was that breast self-examination saved lives. Apparently this is not true. Last week, Professor Malcolm Law of the Wolfson Institute of Preventive Medicine in London wrote in an editorial in the British Medical Journal that "breast self-examination and testicular self-examination are further examples of the failure to apply scientific rigour to screening. Both have been widely advocated on an assumption that they must be beneficial and cannot do harm; neither should be advocated".

Law argues that it is false logic to assume that any screening, or self-screening, for cancer must do some good. Some - like breast and testicular self-exams - have in fact been shown to do more harm than good. One recent large study found that women who examined their breasts regularly did not spot more genuine cancers than other women. They did, however, experience more anxiety and more unnecessary surgical biopsies. Cancer Research UK says it is important for women to understand the difference between self-screening and "breast awareness". More than 90% of breast tumours are spotted by women.

We need to be familiar with the normal feel, weight and appearance of our breasts (the same goes for men and their testicles) and report changes to the doctor. But systematically examining ourselves once a month is not the way to achieve this.

Medical screening these days is, however, on the up. Women are now routinely offered mammography and cervical smears; men are entitled to request a prostatic specific antigen (PSA) blood test for prostate cancer. Pilot studies are under way for screening for the sexually transmitted infection chlamydia; for bowel cancer and for aortic aneurysms. And most of us think this is great. We are living in an age of responsible, preventive health care. Even if we are not testing our own cholesterol levels or performing our own cervical smears using the new "DIY" self-testing kits, most of us would agree that screening, in general, is good.

Then again, most of us don't fully understand the uses and benefits, not to mention downsides, of screening. Screening is meant to diagnose a condition in a person with no symptoms. Yet some women, aware of a breast lump, will wait for their routine screening appointment before mentioning their findings to their doctor. But surely popping along for a screening test by a trained doctor would eradicate such subjectivity? Again, this is a public misconception. Screening tests are not as accurate as most of us think: "false positives" and "false negatives" are relatively common, and tests that are both sensitive and specific are extremely rare.

Hazel Thornton had a routine NHS breast screening test in 1991. "I went along thinking that these people knew what they were doing. It seemed to me the responsible thing to do."

She was diagnosed with a ductal carcinoma in situ - a type of breast cancer often picked up with screening, but one that has an uncertain natural course. Thornton might, in fact, have done equally well without ever having had the screening, the treatment or the worry. "In retrospect," she says, "it's difficult to know if I would have gone for that mammogram, knowing what I do now. It's not more information we need. It's better information."

Dr Tom Marshall, lecturer in public health in Birmingham University agrees. "The patient information provided by the UK breast screening programme could be seen as misleading," he says. "They say that one in nine women will develop breast cancer at some point in her life. This is true. But screening only occurs between the ages of 50 and 65 and most breast cancer deaths are in older women. It would be much more relevant if they pointed out the difference breast screening makes to a woman's chances of dying. At best estimate, if there were no breast screening, 20 out of every 1,000 women aged 50 would die from breast cancer by the age of 75. If every woman aged 50 diligently took part in the screening programme from the age of 50 to 65 [as is currently the plan], 14 women per 1,000 would die from breast cancer before the age of 75." In other words, according to Marshall, mammography may not be as effective as most of us are led to believe it will be.

The same may be said for prostate cancer tests. In the US, about a third of healthy men aged over 50 had the PSA blood test in the past two years. In Britain only a tiny percentage did. But pressure to change this in the UK may be misguided. "PSA testing does identify prostate cancers in men when they have not presented with symptoms," says Law. But there are no published trials yet showing that earlier detection of prostate cancer actually saves lives. As Law explains: "Not all prostate cancers are worth diagnosing. Many men develop prostate cancers of low grade [that is, a low level of malignancy] and these cancers may never trouble the men."

Treating them, however, might: "The treatments are often hazardous, causing incontinence or impotence. Only the trials will tell us whether any reduction in mortality is sufficient to outweigh this."

Of course, some would say it's better to go through a bit of unnecessary worry and even some nasty side effects in order to rule out even the smallest chance that your body is harbouring a malignant killer. Others would disagree. But the point is that if we are to make informed choices we need all the information. And that, too often, does not happen. "To make a truly informed choice," says Thornton, "we should not just be concerned with the potential for our lives to be saved, but the potential for our lives to be marred."

In Britain, all women between the ages of 25 and 64 are eligible for a free cervical smear test every three to five years. The government sets cervical screening targets for GPs - if they don't get certain numbers of women to have them, the practice gets less money. Doctors therefore tend to advocate screening rather than inform women of the complexities of all the risks and benefits.

There is no doubt that cervical screening saves lives (the Department of Health says it saves around 3,900 lives a year). But many women are ill-informed about the possible risks and benefits. Many of us, in the course of our routine screens, will undergo "false positive" tests which may lead to unnecessary further tests and panic. Many of us will face worrying recalls for repeat screenings. This is probably worthwhile if it saves lives.

But are we being allowed to decide for ourselves whether we want to participate? "People have the right to be told the truth in a non-misleading way," says Marshall. "If this leads them to make decisions that are different to the ones I would like them to take, that is their right. We have no right to force - or manipulate - competent adults into undergoing screening just because we think it is in the public interest."

"I certainly don't think all screening is good," says Professor Peter Sasieni, cancer screening expert for Cancer Research UK at Queen Mary's Hospital, London. "It's important to balance the pros and cons of any screening that is being introduced." It is also crucial, he says "to have quality control measures in screening programmes - checks on labs, good training". For instance, what may work in a carefully conducted randomised study may not work so well in fallible "real world" situations.

In future it may be possible to target those most at risk of a disease or condition and so save more lives, and avoid some unnecessary distress in low-risk people. Sudden cardiac death is one example here - between four and eight under-35-year-olds die every week from it. It could be picked up by screening "high risk" groups. (First-degree relatives can have a 50% chance of inheriting some of the conditions that cause sudden cardiac death.) Of course, screening and eradicating disease is an ideal. Screening will always be imperfect. But surely, the more we understand these imperfections, the better we will cope when our results come through - whatever they are.