to an industrial and increasingly information-based society. As a simple example, a tele-
phone company in the USA has suggested that in the course of one year an extension
phone saves an individual approximately one mile of walking, which could be the equiva-
lent of 2–3 lb of fat or up to 10,500 kcals (Stern 1984). Further, at present only 20 per
cent of men and 10 per cent of women are employed in active occupations (Allied
Dunbar National Fitness Survey 1992) and for many people leisure times are dominated
by inactivity (Central Statistical Office 1994). Although data on changes in activity levels
are problematic, there exists a useful database on television viewing which shows that
whereas the average viewer in the 1960s watched 13 hours of television per week, in
England this has now doubled to 26 hours per week (General Household Survey 1994).
This is further exacerbated by the increased use of videos and computer games by both
children and adults. It has therefore been suggested that obesity may be caused by
inactivity. In a survey of adolescent boys in Glasgow in 1964 and 1971, whereas daily
food diaries indicated a decrease in daily energy intake from 2795 kcals to 2610 kcals,
the boys in 1971 showed an increase in body fat from 16.3 per cent to 18.4 per cent.
This suggests that decreased physical activity was related to increased body fat (Durnin et
al. 1974). To examine the role of physical activity in obesity, research has asked ‘Are
changes in obesity related to changes in activity?’, ‘Do the obese exercise less?’, ‘What
effect does exercise have on food intake?’ and ‘What effect does exercise have on energy
expenditure?’ These questions will now be examined.
Are changes in obesity related to changes in activity? This question can be
answered in two ways: first using epidemiological data on a population and second using
prospective data on individuals.
In 1995, Prentice and Jebb presented epidemiological data on changes in physical
activity from 1950 to 1990, as measured by car ownership and television viewing, and
compared these with changes in the prevalence of obesity. The results from this study
suggested a strong association between an increase in both car ownership and televi-
sion viewing and an increase in obesity (see Figure 15.4). They commented that ‘it
seems reasonable to conclude that the low levels of physical inactivity now prevalent in
Britain must play an important, perhaps dominant role in the development of obesity
by greatly reducing energy needs’ (Prentice and Jebb 1995). However, their data was
only correlational. Therefore, it remains unclear whether obesity and physical activity
are related (the third factor problem – some other variable may be determining both
obesity and activity) and whether decreases in activity cause increases in obesity or
whether, in fact, increases in obesity actually cause decreases in activity. In addition,
the data is at the population level and therefore could miss important individual differ-
ences (i.e. some people who become obese could be active and those who are thin could
be inactive).
In an alternative approach to assessing the relationship between activity and obesity
a large Finnish study of 12,000 adults examined the association between levels of
physical activity and excess weight gain over a five-year follow-up period (Rissanen et al.
1991). The results showed that lower levels of activity were a greater risk factor for
weight gain than any other baseline measures. However, although this data was pro-
spective it is still possible that a third factor may explain the relationship (i.e. those with
OBESITY AND CORONARY HEART DISEASE 359