View report page

The pattern of male mortality for the metropolitan boroughs is also shown in Figure 3
on page 14. It will be seen that London is characterised by high ratios, except for a ' wedge '
of low ratios running from Hampstead, Paddington and Kensington in the north.west
through St. Pancras, Chelsea, Westminster and the City of London to the river. (The ratio
for the City of London, based on only 11 deaths in the four years, is not a very reliable
measure and in this context and in what follows should be ignored.) On the south side of
the river there are comparatively low ratios in the boroughs towards the periphery of the
county—Wandsworth, Lewisham, Greenwich and Woolwich. The pattern of female
mortality is similar. As regards the other items in the table, social class has been included
because the Registrar General's census volume on Occupational Mortality* shows, in
general, a steeply rising gradient of male mortality from bronchitis as social class descends
from class I to class V.
Strictly, the social class indicator relates only to the male population, since it is based
on the census analysis of male occupations. Its use in relation to female mortality can
probably be justified on the basis that wives and unmarried daughters, whether or not
they are of the same social class as their husbands or fathers, will for the most part live
in the same household and share the same social conditions. The percentage born in London
is given as a very rough indication of duration of exposure to the London atmosphere
and atmospheric pollution.
The indices of atmospheric pollution which are taken from the twenty.seventh report
of the Department of Scientific and Industrial Research are, as I have indicated before,
somewhat unsatisfactory. To ensure as much homogeneity as possible readings from
deposit gauge stations have been used—these relate to the ten years ended March, 1954,
but only nine of the 51 stations operated over the whole ten years and 24 of them were in
operation for one year only.
However, the figures given in table (x) are the best available.
The interaction of these various indices is perhaps best summarised by the calculation
of the product.moment correlation coefficient (r) and these are given below for the standardised
mortality ratios and the other indices listed (figures for the City of London have
been excluded from the calculations).
Correlation coefficients for standardised mortality ratios for bronchitis, London boroughs
(excluding the City of London) and:
For both males and females the two first are highly significant (P<.001). Such a highly
significant relationship with percentage in social classes IV and V is to be expected because
of the known social class gradient of mortality from bronchitis; the high correlation with
percentage born in London conforms with the known increase in mortality from bronchitis
with increase in the degree of urbanisation, though the index used here is only a crude
measure of exposure to the atmosphere of various parts of London and takes no account
of movement within the county for work, pleasure or residence. The poor correlation
with the two indices of atmospheric pollution is not altogether surprising in view of the
differing quantity and quality of the measurements, only the coefficient for males with
solid deposit attains significance (P<.05). However, the percentage in social classes IV
and V is also highly correlated with the percentage born in London (r = 0.702). It is therefore
pertinent to enquire whether percentage born in London is a reflection of social class
or whether duration of London residence has a separate effect—and, if the latter be true,
which of the two factors is the more important. This can be done by a process of partial
* Registrar General's Decennial Supplement (1951)—Occupational Mortality.
Male Female
Percentage in social classes IV and V
Percentage born in London
Solid deposit
Sulphate ions (SO4)
+0.780 +0.673
+0.825 +0.850
+0.380 +0.293
+0.213 +0.311
13