Currently there are proposals to extend the range of people who will be offered statins on the grounds that this will reduce the risks of developing heart disease. There are serious doubts about the justification of this strategy. Many of the researchers who have evaluated the evidence have concluded that treatment of those who do not have heart disease with statins does not result in any reduction in all-cause mortality. At present the NHS spends about £300 M annually on statins so obviously this figure will increase substantially if there is a big increase in the numbers treated. Furthermore many of those who are on the statin therapy will experience side-effects which can be unpleasant and may actually damage the health of some people.

On the other hand there is now extensive convincing evidence that much of the poor health is directly caused by a lack of physical activity and therefore significant improvements could be achieved by encouraging more people to take regular exercise. The purpose of this blog is to focus on some of the research which demonstrates that those who maintain their physical fitness have a much better standard of health when compared with those who lead a sedentary existence.

In a study in the USA 10,224 men and 3120 women were monitored over a period of about 8 years. During this time 240 men and 43 women died (1).

The results are shown in Table 1.

Table 1 Relationship between fitness level and all-cause mortality in men and women

                Death rates per 10,000 person   years
Men Women
Least fit 64 39.5
Most fit 18.6 8.5


The differences in the all-cause mortality rates are absolutely huge. With men those who are least fit are over 3 times more likely to die than those who are sedentary while the difference with women is almost 5. It is important to stress that these differences persisted even when adjustments were made to allow for other factors including smoking habits, blood cholesterol levels, blood pressure, fasting blood glucose and parental history.

In a study conducted in Norway, 1428 men aged 40-60 years at the outset were monitored for their physical fitness on 2 separate occasions between 1972 and 1975 and again between 1980 and 1982 (2). By the end of 1994 there had been 238 deaths of which 120 were attributed to cardiovascular disease, 75 to cancer and 43 to other causes. The results are shown for 4 equal groups which vary according to fitness level with Q1 being the lowest and Q4 the highest (Table 2). It is evident that those who are unfit have almost 3 times the death rate of those who are extremely fit. However it is noteworthy that there is quite a big difference between Q1 and Q2 which suggests that there are significant benefits with a moderate level of fitness. Furthermore there is virtually no difference between Q3 and Q4 indicating that there is little benefit from very high fitness levels.

Table 2. Relationship between physical fitness and all-cause mortality in Norwegian men

Fitness levels Number of deaths
Q1 97
Q2 65
Q3 39
Q4 37


What is even more impressive in this study is the insight gained by considering the impact of changes in fitness levels between the initial assessment and the second one (Table 3).

Table 3 Changes in fitness levels between the 2 assessments and the effect on the mortality ratios.

Q1 (PF1) Q2 (PF1) Q3 (PF1) Q4 (PF1)
                              Changes in   fitness levels
Q1 (PF2/PF1) 61 62 62 62
Q2(PF2/PF1) 85 84 85 81
Q3 (PF2/PF1) 103 99 98 92
Q4 (PF2/PF1) 136 124 118 110
                             All-cause   mortality ratios
Q1 (PF2/PF1) 1.22 1.19 0.87 0.73
Q2(PF2/PF1) 0.80 0.77 0.62 0.46
Q3 (PF2/PF1) 0.72 0.33 0.60 0.17
Q4 (PF2/PF1) 0.47 0.43 0.40 0.17


At first sight this information may seem rather daunting but it is certainly worthwhile grappling with it. Each column provides information on the fitness level as determined at the first assessment. In the top half of the table the information is broken down in accordance with how the fitness level changed when the second assessment was done. A value below 100 means the fitness level had deteriorated while a value above 100 means there had been an improvement. The values in the lower half show how the death rates for each of the different groups. It is evident a change in fitness level is associated with a corresponding change in relative mortality. When the fitness level drops there is an increase in the death rate and similarly an improvement in fitness level is associated with a reduction in death rate.

The results of this study are in line with those of many other investigations which repeatedly confirm the critical importance of physical fitness in determining a person’s general health and life expectancy. In addition this research demonstrates that raising the level of fitness has a definite impact on life expectancy but also that if the fitness level falls it will be accompanied by an increased risk of death.

The fact that major changes in mortality occur in response to changes in fitness levels comes across very convincingly from this work. This knowledge has enormous implications for public health. If more people can be persuaded of the improvements that can be achieved to their own personal health by taking a moderate amount of exercise then it would follow that they would be less likely to suffer from major diseases and to extend their life span. From a policy perspective, This approach has much to commend it compared with the blanket use of drugs such as statins. First of all, the effect on reducing all-cause mortality is much greater than anything that can be achieved with drugs. Secondly it would be much less expensive. Finally there would be much less undesirable side-effects.


  1. S N Blair et al (1989) Journal of the American Medical Association 262 (17) pp 2395-2401
  2. G E Erikssen et al (1998) The Lancet 352 (9130) pp759-762