An independent charity for science and integrity in healthcare

Read the latest HealthWatch newsletter:  Issue 106, Autumn 2017


One of the things that many people like about screening and check-ups - especially those who can afford the extra time available in a private clinic — is a chance for a discussion of their health and lifestyle with a doctor of nurse who seems interested and keen to reassure them and to give advice of various kinds. Modern technology and the detailed thoroughness of various examinations may also please and impress them. Simple practical instruction, especially in the elderly, may also be useful in conditions that are not life threatening, yet needlessly impairing quality of life.

Unfortunately much advice, for example diet to make cancer less likely, is based on evidence that is suggestive but far from conclusive. Almost the only really convincing evidence, relevant to common conditions, concerns smoking cigarettes and the big effect this has on the risk of developing certain cancers (it has no effect at all on others) and also on certain diseases of the heart, blood vessels and lungs.


Can the finding and treating of a condition at a very early stage prevent the development of established disease of the type that causes symptoms and threatens life? Note that with earlier diagnosis, survival from the time of diagnosis will automatically increase, so merely reporting longer survival after treatment (as is often done) provides no evidence of benefit.

With a life-threatening condition it is fewer deaths from this cause that needs to be demonstrated. And experience shows that a comparison of this kind can lead to false conclusions unless the groups being compared are assessed concurrently and are alike apart from whether or not they have been screened. By far the most reliable way is to randomise either individual patients or groups of patients, so that it is a matter of chance whether or not they are screened. Merely trying to match people to make the groups as alike as possible has been shown to be much less reliable. And when mortality is less in the years that follow a screening campaign, we still cannot be sure that this is due to the screening. There may be other factors.

A randomised comparison is ethical when nobody knows which group will come out best. The screening may in the end turn out to have done enough good to justify its disadvantages (see below) or it may turn out to be doing more harm than good.


Unfortunately, when screening has been assessed in this way the results have often been disappointing. For example, extensive checkup - costing several hundred pounds each - though popular with many executives (especially if their employers foot the bill) show little, if any, benefit apart from the detection of high blood pressure.

The greatest interest has been in checkups for cancer - especially cancer of the breast or cervix. When mammography is done to look for early breast cancer, not all studies show any benefit, but several large randomised trials have shown a 20 - 30% lower breast cancer mortality in women between the ages of 50 and 65 in the screened group. Women under 50 (who are often the most keen to be screened) show no benefit.

Sometimes, though death certificates may show a fall in deaths due to the condition being screened, overall death rates (deaths from all causes at various ages) may fall much less, or not at all. Preventing the development of a particular condition may have helped quality of life — and made deaths from that condition less common — but other causes of death have meant that life has not been prolonged.

The relative risk reduction may look much more impressive than the absolute risk reduction. The individual who is regularly screened for some condition may be only marginally less likely to suffer or die from this condition at some time in the future. However, with some early cancers, treatment can be less drastic and there is a better chance of at lease avoiding recurrence at the original site.

In cervical cancer (where screening consists of taking vaginal smears) there is much suggestive evidence that mortality is reduced, but no randomised trials have been carried out; and now, with so many convinced, it would be difficult to mount a proper study.

The results of screening for other cancers, for example, bowel, prostate, or ovary are so far not encouraging. This follows the finding in 1974, ten years after mass chest X-rays and sputum examinations had successfully detected many lung cancers at a much earlier state, that deaths due to lung cancer remained just as numerous.


1. Anxiety and its effect on quality of life. Many of those given a complete medical check up will now be labelled as unhealthy in some way. If no cure is available may this not sometimes adversely affect the peace of mind and quality of life of people who previously thought of themselves as healthy? As for cancer if the result of a smear or mammogram is doubtful (as is quite common) intense anxiety may be caused. Further tests will need to be done and distressing doubt may continue for weeks or even months.

"False negatives" are equally common. Screening may show no abnormality, but - though it would be foolish to stress this at the time and thus risk spoiling the good news that all is normal - this can never guarantee that no cancer is present. There is room for a million cancer cells in a cubic millimetre. There is currently no way of revealing such a small area of abnormality, so no method of screening can be guaranteed effective.

2. Expense. Though it is hoped that at least a small part of the cost of treating established disease is being saved, in the UK private sector alone screening is now swallowing up something like £l00 million a year and the time of many skilled workers, quite apart from the cost of campaigns to improve recruitment. The cost of each "life saved" can easily come to something of the order of £100,000.

3. Other disadvantages. We know that many early cancers will cause no trouble if left alone. With many people mere knowledge of having had a cancer treated will damage the quality of their future life. There may also be adverse effects on life insurance, pensions and employment. And many who decline screening will experience anxiety as a result.


Screening, though it takes time and money away from other needs, has real (though sometimes quite modest) benefits in certain situations. But HealthWatch stands for informed choice and does not think it right to give people an exaggerated view of the benefit to the individual, purely in the hope of improving community statistics. Screening should be a matter of choice for the informed citizen who  unlike the situation when immunisation controls infectious disease  does no harm to others should he or she prefer not to join a campaign. And here it is interesting to note that many doctors do not seem to believe in screening or checkups for themselves.

This position paper by Thurstan Brewin

Endorsed by the Executive Committee, June 1996