View report page

General death-rates, however, do not reveal the whole picture for which a detailed
examination of the rates for various sex and age groups is necessary. Table A on page 153
sets out the number of deaths from lung cancer in London Administrative County
by age and sex for each year from 1931, and Table B on page 154 reproduces the same
information in the form of rates. It should be mentioned that during the period 1939
to 1949 inclusive, the Registrar-General did not give an age break-down of the London
population aged 15 and over and only a partial one in 1950—hence the figures on
which the death-rates for these years are based are the result of interpolation ; another
complication arises from changes in 1938 and 1948 in the code numbers of the international
classification of causes of death—lung cancer has been taken as Nos. 162 and 163
in the current hst (i.e., malignant neoplasm of trachea, and of bronchus and lung) and
by interpolation cancer of the larynx and mediastinum has been isolated and removed
from earlier and less specific classifications.
[ Effects of age
From Table B, it will be seen that lung cancer is of negligible importance until the
35— age group is reached ; that from then on its incidence increases steadily with age
up to and including the 65— age group and that over the period there has been an
increasing incidence in both sexes although the male rates are well in excess of the
female. For ease of comparison graphs of the rates for each sex in the age groups 35—,
45—, 55— and 65— are set out on page 151; the aged 75 and over group is excluded
merely because its rates come between those of the 55— and 65— groups and would
render the graphs less clear. The effect on the 35— male group of the removal of the
fitter members during the war years is noticeable, as are the fluctuations due to small
numbers in the rates for the 35— and 45— females.
The graphs have been plotted on semi-logarithmic paper which is equivalent to
plotting the logarithms of the death rates on ordinary paper. This type of graph has
the effect of flattening the curve but it has been adopted in order to focus attention on
the proportional rate of increase rather than the absolute—a straight line representing
a constant proportional increase. Broadly they show a steady proportional increase in
the rates with a higher rate of increase in the male than in the female and in the older
males than in the younger. The all-male and all-female rates in 1954 on the basis of
the 1931 age structure would have been 0-76 and 0-13 respectively—increases of 262 per
cent, and 160 per cent. It is interesting to note that in New York State (Symposium
(1952) ) the age-standardised mortality from cancer of the lung between 1930 and 1950
increased by 385 per cent, for males and 68 per cent, for females.
In order to ascertain which of the sexes has the greater rate of increase the sex-ratio
of the deaths from age 55 onwards has been extracted for the period. For comparative
purposes, the ratios for all causes, all cancer (International List Nos. 140-205) and
bronchitis (International List Nos. 500-502) have also been extracted and the numbers,
rates and sex ratios are shown in Table C (page 155). It is appreciated that there was an
apparent increase in bronchitis from 1940 onwards owing to the abandonment of the
former method of selection in multiple causes of death and to the revision of the International
List of Causes of Death, but it was felt that the sex-ratio should not thereby be
unduly affected, and it was desired to compare lung cancer with another disease which
might be attributable in whole or in part to the inhalation of irritants. Crudely there has
been an increase in the ratio over the period of 25 per cent, for all causes ; 36 per cent,
for all cancer, presumably a reflection of the changes in lung cancer incidence ; 73 per
cent, for lung cancer and 164 per cent, for bronchitis—for ease of comparison the rates
have been plotted (also on semi-logarithmic paper) and are shown on page 152. The
graphs are by no means as regular as might have been anticipated even in the ' normal'
years from 1931-38, particularly in the case of lung cancer although the irregularities of
the war years are understandable in the light of migration and (perforce) estimated
populations. The general similarity in the slopes of the lung cancer and bronchitis
graphs since the end of the war is noticeable although their earher courses are different :
the close association of the ' all causes ' and ' all cancer * graphs is also noticeable though
Effects of sex
147
K2