As a way of monitoring patients at risk of specific diseases, body screening is invaluable. But should it ever be used for a general check-up? Dr Margaret Mccartney reports

If you live in the US, you have money in your pocket and you want a little reassurance about the state of your health, what do you do? Why, you head for the nearest clinic that has a CT scanner, naturally. There, your heart, lungs, liver, kidneys, pancreas and maybe even your bowel will be X-rayed, electronically sliced, virtually reconstructed and then dissected by a sharp-eyed radiologist. Fingers crossed, you'll get a clean bill of health in exchange for a swipe of your credit card.

There are numerous health websites bulging with testimonials from recently scanned customers in good health. They praise the reassurance that their body scan has brought about: "peace of mind" and "weight off my shoulders" are phrases you will read again and again. Among them, the odd, guilty person will put up a hand and say that, having observed their internal workings in virtual form, they have finally vowed to give up smoking.

So is body screening a good idea? Would a scan encourage us all to take better care of ourselves, or would an "all clear" only serve to convince us that we are getting away with bad habits (smoking, overeating or drinking too much)? Is it, at just under £2,000 a pop, a rather extravagant pastime?

CT scans are typically done as a follow-up procedure for post-treatment cancer patients, or to investigate unidentified symptoms or unusual test results. However, some UK clinics are offering body scans to people with no symptoms at all but who, crucially, have enough money and the inclination to pay for such a screening. I telephoned a few of these clinics, including the Virtual Imaging Centre in London. On asking for a checkup, I was told that, for slightly less than £2,000, my heart, lungs, pelvis and abdomen could be scanned, and I would also have a thorough consultation with a GP.

Rob Wilson, the managing director of Virtual Imaging, is in no doubt as to the superiority of CT scanning as a screening tool. "This is the ultimate checkup," he says, "the gold standard." In addition, he says that demand has outstripped supply, and that they are planning to open more clinics in the near future.

At first glance, a CT screening test might sound like a proactive, healthy and desirable choice. However, screening, especially in cases such as this, is seldom a straightforward exchange of time - and money - for reassurance. Screening is a complex beast, inexorably tied up with probabilities, risks, and benefits. All of which should be considered before deciding to have a CT screening test.

In the 1960s, the World Health Organisation first set out the criteria for screening tests, which have since been refined. The purpose of a screening test is to pick up any diseases which usually present at a stage too advanced for a cure. A screening test is done to look for signs of very early disease, at a stage where the person has no symptoms of illness. This is quite different from a diagnostic test, where a person has a test to ascertain what is causing a particular symptom. A person being screened has no symptoms, and would not normally be seeking medical advice, because they feel well and are unaware of any problems.

In order for a screening test to meet WHO criteria, it has to do several things. The disease you are screening for should be significant, in terms of the harm it could do and whether or not the person being screened is in a high-risk category for that disease. It should have a period of "latency" - that is, a period during which the disease can be detected but is not yet causing either harm or symptoms.

Additionally, there should be good evidence that early treatment for the detected disease is better than late treatment. The screening test should be simple, accurate, and acceptable to the patient - internal examinations or invasive biopsies, for example, may be required, and some people prefer not to have them. The process of screening itself should be under continual review by central audit, to ensure that the adverse impact of the disease is decreasing in the population targeted. It should also be possible to say which test results are abnormal, and worthy of investigation, and which results are normal and can safely be ignored.

However, when it comes to CT scanning, deciding what is normal or abnormal is not as easy as it sounds. This is because CT scanners have not traditionally been employed for "screening" examinations. CT scanners are more commonly used to investigate symptoms or abnormal blood tests, or to track treatments for diseases such as cancer. A large and unexpected cancer, although rare, may be easy to spot, but there are many other smaller, minor abnormalities - for example, benign, non-cancerous tumours - which scans may reveal yet whose significance is uncertain.

Richard Hayward, consultant neurosurgeon at Great Ormond Street, recently wrote in the British Medical Journal about this problem. He coined the acronym "VOMIT" (victims of modern imaging technology) to describe those patients - and their relatives - concerned about an abnormality on a scan, which later turns out to be "no more than a red herring". The policy of using imaging like CT for screening tests, he says, will "produce a bumper harvest of both 'normal' and unanticipated 'abnormal' results."

Borderline results, individual quirks, or "innocent pathology", as Hayward calls it, are difficult to ignore. Yet it is never easy to decide on the best course of action for them, either. Minor deviations from scanned bodily perfection are common.

Ignore too many minor abnormalities and you may miss something of importance. Ignore too few, and you may subject your patients to unnecessary tests, biopsies and worry. Fear of litigation may also influence the decision to investigate a probable "red herring", just in case. This balance between sensitivity and specificity has been well-researched in the use of CT for cancer scanning, but little has been done to establish the body scan as a screening tool.

Additionally, there are problems with the test itself. A negative scan informs you only about a singular point in time. If body scanning were going to be an effective screening tool, it would need to be repeated, as mammograms and smear tests are, at some point in the future. Repeat the test too soon, and no problems will have developed yet; repeat it too late, and the very disease you were trying to pick up could be beyond cure. Because body scans are not an established screening tool, a reasonable time frame between tests is unknown. This is especially important when the test itself may have side effects.

Most clinics claim that the radiation from a body scan is the approximate equivalent of a year's worth of background radiation (the kind we're all exposed to in everyday life from our environment). The National Radiological Protection Board places the risk of developing a cancer from one year's background radiation at somewhere between 1 in 10,000 and 1 in 1,000. This small risk may be worth taking, however, if you are having the scan to investigate potentially serious symptoms.

The use of body scans for the asymptomatic population at large is a different matter, however. Professor Jamie Weir, former Vice-President of the Royal College of Radiologists, is concerned. "Even an additional year's background exposure [to radiation] places a significant burden on a general population whose benefit from having these scans is not proven."

If body scans were targeted towards at-risk groups of individuals, would that make a difference? "There may be a case for targeting some individuals for certain limited scans: for example, heart scanning may detect cardiac disease without the need for invasive tests," says Professor Weir. "But the research is still at the very experimental stage. We don't have enough good-quality evidence to say that this kind of screening would do more good than harm."

The National Screening Committee is usually responsible for instigating national screening programmes. Bowel-cancer screening, for example, has just had successful pilot studies extended. Careful analysis showed that testing stool samples for microscopic traces of blood was accurate enough to predict that a colonoscopy could usefully pick up abnormalities.

Some private CT clinics offer virtual colonoscopy, where the bowel is imaged with CT scanning. This test is perhaps more comfortable than a standard colonoscopy, but is it any better, or even as good?

"The problem is that up to 25 per cent of individuals aged between 50 and 70 have bowel polyps which would be detectable on CT colonography, leading to the requirement for further investigation. Of course, not all of these are going to be malignant and the patient will be subjected to further investigation, not always to his or her benefit," says Professor Weir. The next step for investigation of these polyps would usually be a biopsy, which would require a traditional colonoscopy.

Although the colonoscopy procedure is low-risk, rare complications, for example bowel perforation, are possible. Audits of the bowel-screening programme should enable patients to make an informed choice about having it, based on the likely risks versus benefits of further investigation after an initial positive stool sample. This detailed information is unlikely to be as forthcoming with screening using CT scanning, because it is currently not part of an audited study.

I spoke to a 57-year-old man from Kent who was very glad to have had a screening CT scan. "I had one because I was anxious about my family history of heart disease. The scan was very straightforward, and when I found out that I was low risk, I was extremely reassured."

However, on some health websites, there are rare tales from those who have made the unexpected discovery of an operable cancer. The suggestion to "gift" a scan to a loved one is made. Would I have one, or give one? Absolutely not. If I wanted to extend the lives of my family and friends, I would give, rather more affordably, nicotine patches to the smokers, and feed them all fruit and vegetables regularly, preferably after exercise, and before a glass of red wine.

It is, unfortunately, all too easy to descend through a downward spiral of investigations, spurred on by anxiety. It is not a pleasant road to go on, either physically or psychologically, and it is certainly not one that I myself would go down without extremely careful thought. The real problem is that the technology available is outstripping our knowledge about how useful as a screening instrument CT scanning is.

I was taught: first, do no harm. Until we do know exactly what diseases body scans will diagnose correctly and the ones that it will miss, until we know the risk of radiation versus the chance finding of a curable, unexpected disease, we are not in a position to know that a screening body-scan will do us any good at all.