Tobacco and Public Health - TCSC Indonesia

8
EVOLUTION OF KNOWLEDGE OF THE SMOKING EPIDEMIC
lung cancer. Similar conclusions were reached, over the next 3 years, by a study group
on Smoking and Health (1957) appointed jointly by the US National Cancer and Heart
Institute and the US Public Health Service (Burney 1959), and by specially appointed
committees in The Netherlands and Sweden, and the World Health Organization
(see Doll 1998).
Impact on total mortality
A few years later, when the Royal College of Physicians (1962), in England, and the
Surgeon General (1964), in the United States, issued reports on smoking, it had
become clear that its total impact was greater than had at first been conceived, and that
the incidence of more diseases might be affected. With the passage of time, more and
more diseases were found that were linked in some way to smoking. The total number
now believed to be caused in part by smoking is at least 35, 22 of which can be recognized
in the cohort study of British doctors and in the massive study of a million men
and women subsequently undertaken by the American Cancer Society (Table 1.1).
Other harmful effects caused in part by smoking are listed in Table 1.2. These are
mostly relatively rare and less lethal, and evidence relating to them has often had to be
obtained from case-control studies or surveys, or, occasionally, from cohort studies in
which special enquiries have been made about the condition of interest. That so many
conditions are affected by smoking should not be surprising, as tobacco smoke contains
some 4000 different chemicals and many of the diseases are caused by similar
mechanisms.
Even the 35 smoking-related diseases mentioned above do not complete the list,
because a few others that are primarily associated with smoking through confounding
are also probably caused by it in part, including cancer of the liver (Doll 1996) and
death from conflagration—two doctors in our study, for example, set fire to themselves
by smoking in bed.
Of course, confounding does account for some of the excess mortality from all
causes of death. However, detailed investigation has shown that the amount is relatively
small and, in some populations, may not exist at all, for confounding can
operate in both directions, and certainly does in elderly populations in countries in
which there is a high mortality from ischaemic heart disease. Here, confounding
with alcohol both causes the excess for cirrhosis of the liver and reduces the mortality
from ischaemic heart disease.
It now appears that the excess mortality from smoking is greater than was long suspected,
for it has increased with time, as the smoking epidemic has matured and old
people have come to be smoking cigarettes throughout their smoking lives. This is
shown by the data in Table 1.3 from both the British doctors’ study, which has continued
for over 40 years with periodic updates on changes in smoking habit, and the studies of
the American public carried out by the American Cancer Society over two different
periods. Regular cigarette smoking since youth is now seen to double mortality,