Bandolier
is keen on evidence on healthy living. One of the excellent sources of evidence is
the US nurses' health study, started in 1976, when about 122,000 female nurses in the
30 to 55 age group were enrolled. Questionnaires were filled in then, and other
information was collected in 1980, '82, '86, '88 and 1992. This information showed
the relationship between folate and multivitamin use and reduced colon cancer
incidence (
Bandolier 60
).
A new analysis [1] sought to pin down the relationship between coronary heart
disease and exercise in women. In Japanese men, a study showed that regular walking
decreased mortality (
Bandolier 50
).
Study
Detailed information on physical activity was first collected in 1986 and updated in
1988 and 1992. This included various forms of vigorous and non-vigorous exercise,
plus a walking history that included their walking pace: easy or casual (less than 2
mph), average (2.0 to 2.9 mph) or brisk (more than 3 mph) [note that 1 mph = 1.6
kph]. A weekly metabolic-equivalent (MET) was calculated for the various forms of
activity. Total MET hours per week was then calculated.
|
Activity
|
MET per hour
|
| Vigorous |
more than 6 |
| Nonvigorous |
less than 6 |
| Walking |
2.5 to 4.5, depending on pace |
| a MET is the caloric need per Kg per hour of
activity divided by the caloric need per Kg at rest |
Outcomes
The primary endpoint was coronary events defined as nonfatal myocardial
infarction or death due to coronary disease occurring after 1986 and before
mid-1994. There were 645 coronary events. The physical activity assessed in 1986
was used as the baseline.
Results
Information was grouped into quintiles of MET hours per week, with about 14,000
to 15,000 women in each quintile. The distribution of physical activity between
the quintiles is shown in Figure 1. The mean of 15 MET hours per week in quintile
4, for instance, implies that a woman would walk briskly for about four hours per
week. Women in the high MET quintiles were less likely to smoke, were leaner and
had lower prevalences of diabetes, hypertension and hypercholesterolaemia than
those in the lowest quintiles.
Figure 1: MET hours per week for each quintile
|
|
When adjustment was made for these factors there remained a significant
association between higher levels of total physical activity and reduced risk of
a coronary event (Figure 2). There was a 34% reduced risk for the highest total
activity quintile.
Figure 2: Relative risk of coronary event for each quintile
Significant reduction in quintiles 4 and 5
|
In women who took no vigorous exercise, brisk walking had a similar effect on
reducing the risk of a coronary event (Figure 3). Using women who walked at a
casual or easy pace as a baseline, women who walked briskly reduced their risk by
36%. Compared with sedentary women, women who walked briskly for 1 to 3 hours a
week had a risk reduced by 30% (95% CI 5 to 49%) after allowing for other
factors.
Figure 3: Relative risk of coronary event by walking pace for women who did
not take vigorous exercise
|
|
Comment
This is another study linking increased physical activity to better health.
Importantly this study is one of the biggest to examine the effects of exercise
in women. The bottom line is that women who walk briskly or exercise vigorously
for three hours a week or more can reduce their risk of heart disease.
ImpAct 3
examined a visionary scheme for prescribing walking exercise.
Bandolier 50 had reports of walking reducing coronary and all cause mortality in
retired Japanese men in Hawaii, and of exercise reducing death rates in a Finnish
twin study. The benefits of moderate exercise have been known for some time, and
the evidence continues to accumulate. The size of the effects continues to
impress, and just shows that evidence-based healthy living is something upon
which we should concentrate more.
Reference:
- JE Manson et al. A prospective study of walking as compared with vigorous
exercise in the prevention of coronary heart disease in women. New England
Journal of Medicine 1999 341: 650-8.
|
previous
or
next
story in this issue
