likelihood of death due to CHD.

But let's also look at the absolute figures, the 'percentage points', the actual numbers of people who die because of CHD. In this case, it is evident that death from a heart attack was prevented in only a small number of treated individuals. This figure was highest in the trials that included patients with CHD, but was only a trivial 0.12 per cent in the AFCAPS/TexCAPS trial, which included healthy individuals with normal cholesterol levels.

Put another way, the chances of not dying from a heart attack over four to six years for patients with CHD and high cholesterol is about 92 per cent without treatment and increases to 95 per cent with statin treatment.

For healthy individuals, the figures are even less impressive. In the WOSCOP trial, for instance, the chances of not dying from a heart attack during the five years of the study was 98.4 per cent without treatment and 98.8 with treatment. In the AFCAPS/TexCAPS trial, the chances of surviving was 99.55 per cent without treatment and 99.67 with treatment.

Let's compare these figures with another kind of treatment-for instance, treatment of urinary tract infections. Nine out of ten patients with a urinary tract infection will recover immediately if treated with an antibiotic for a few days and at the cost of a few dollars for each treatment. But, in the 4S trials, 289 patients had to be treated for five

years to prevent one fatal heart attack.

So, while one of the patients benefited from the treatment, the others took the drug in vain because they would have survived anyway. To prevent one fatal heart attack in healthy people, 235 individuals with high cholesterol and 826 with normal cholesterol would have to take a statin drug for four to five years.

The implications of such reasoning is that, to add as many years as possible to longevity, more than half of mankind should be taking a statin drug every day from an early age to the end of life.

Statins and cancer

In 1996, Drs Thomas Newman and Stephen Hulley, from San Francisco, published the results of a meticulous review of what we currently know about cancer and cholesterol-lowering drugs. They found that clofibrate, gem-fibrozil and all the statins stimulate cancer growth in rodents (JAMA, 1996; 275: 55-60).

Newman and Hulley asked themselves how it was that these drugs had been approved by the Food and Drug Adminstration at all. The answer was that the dosages used in the animal experiments were much higher than those recommended for clinical use. But, as Newman and Hulley commented, it is more relevant to compare blood levels of the drug. Their review showed that the blood levels that caused cancer in rodents were close to those seen in patients taking the statin drugs.

Because the latent period between exposure to a carcinogen and the incidence of clinical cancer in humans may be 20 years or more, the absence of any controlled trials of this duration means that we do not know whether statin treatment will lead to an increased rate of cancer in the coming decades.

Thus, millions of healthy people

are being treated with medications, the ultimate effects of which are not yet known. Newman and Hulley therefore recommended that the new statins be used only for patients at very high risk of CHD and avoided for those with life expectancies of more than 10-20 years. Healthy people with high cholesterol as their only risk factor belong to the latter category. Yet, these are the very people targeted for cholesterol-lowering drugs.