View report page

16
The influence of industry.
Two, then, of the three problems mentioned at the outset, (a) and (c), Migration and the Three
Types, have been examined. But there remains to be considered (b), Influence of Industry as illustrated
by the study of Miss Thompson's coefficients. Dr. Greenwood's Table 11 presents us with two striking
paradoxes. First, the partial correlations for males and females between tuberculosis and overcrowding,
at 15-25, are distinctly negative. This question is considered elsewhere, Part II. (p. 70). Then,
second, Table 11 shows a striking negative partial correlation between tuberculosis and factory employment
in females at 45-65. This point is dealt with by Dr. Greenwood on p. 341. The conspicuous drift
away from the towns of the older women may also have some influence in this connection.
Coming now to the ages 25-45, those mainly affected by industry, the fuller Table 13 shows
three high partial correlations, set out by Dr. Greenwood on p. 339, the first of them easy to interpret;
but there remain two rather, at any rate at first sight, paradoxical correlations. ,
(i.) "The female death-rate from tuberculosis is highly correlated with the death-rate
from all other causes in women ; there is no such substantial correlation in men.
(ii.) The male death-rate from tuberculosis is highly correlated with the index of overcrowding
; the female death-rate is not so correlated."
Now it should be noted that the high partial correlations which are associated with one
or the other sex are already ab initio observed in the total correlations ; the sex then showing the
lower correlation, in each instance, loses ground in one or both the corresponding first partial correlations,
and emerges with a very low coefficient in the second partial correlation.
As to the original total correlations, no difficulty arises. The coefficient for women for tuberculosis
and other causes might be expected to be higher than for men; for with the former the child-bearing
diseases and the higher birth-rate fit well with more poverty and more tuberculosis; with the latter
the results of accident and strain and of syphilis and alcoholism figure more largely in the " other causes,"
and these are less closely associated with poverty and tuberculosis than the special diseases of childbirth
are. Again, the fact that the total correlation between tuberculosis and overcrowding is higher
in men than in women need cause no surprise, having regard to the evidence already adduced as to the
social, or rather unsocial, habits of the immigrant males at these higher ages in London, leading to their
resort to lodging-houses and shelters, while the older women generally have a friend or friends and may
be perhaps presumed to live under conditions of less crowding as estimated by a " more than two to a
room " standard.
Coming now to the First Partials," the more severe exhaustion of the total correlation for tuberculosis
and other causes in the case of males than of females, when overcrowding is made constant,
needs no comment after what has already been said. The exhaustion of the total correlation for tuberculosis
and overcrowding is much more severe when either other causes or employment are made constant
in the case of women. Here, again, no further comment seems to be needed. In the light of these
facts, perhaps, the " Second Partials " are not so paradoxical after all.
While, therefore there may be some force in Dr. Greenwood's suggested solution of problem (6)—
viz., that industrial employment may act indirectly, sensitising the male to deleterious influence of bad
home conditions, a subject discussed by him very judicially on p. 142 of the paper on Tuberculosis and
Industry (Collis and Greenwood, Chapter VI.)—it is to be noticed that he concludes (p. 143) that the
inference as to this differential effect of home conditions (in men and women) cannot be accepted as proven.
At the same time he delivers a like verdict concerning the theory of the three types. On the other
hand he refers to the importance of migration, for, he says: "We find that with advancing age the
disadvantage of the town dwellers increases, and that the apparent advantage of young female city
inhabitants can be accounted for more or less completely by immigration from the country of healthy
females." He adds that at ages above 35 the excessive death-rates among town dwellers are " far
greater amongst males than amongst females."
The influence of employment has been more particularly considered in the case of young women
(15-25) and of the older men. As regards the former, see Annual Report Med. Off. Ministry of Health,
1919-20, pp. 143 to 145, and a report by Dr. Rosenfeld on Tuberculosis in Austria, where from 1917
onwards food shortage was very severely felt in certain urban communities. Dr. Rosenfeld finds that
industry was commonly accompanied by high mortality, but where the industry was not associated with
food shortage this influence was far less apparent.
In the case of the older males the high phthisis rates may be due not so much to factory-produced
infection, direct or indirect, as to urban influences of a more general character operating especially on
immigrant males who are already on the down grade. These phthisical men no doubt to some extent swell
the number of deaths among shoemakers, tailors, printers and bookbinders, as Dr. Greenwood justly
claims they do ; but the large majority of them have not seen the inside of a factory for many years, they
are made up from general labourers, casual labourers, and some of the porters and messengers of Dr.
Greenwood's Table 12, and include even more definitely characterisable " blind alley employments," such
as sandwichmen, street newspaper vendors, and the like. It may be observed that open-air blind alley
employments after middle age may not be so conducive to length of life as Dr. Greenwood hints they
possibly may be in adolescence. It is clear, too, that important influences in aggravating an already
acquired phthisis at these higher ages are alcoholism and exposure to cold and wet weather.
A word may perhaps be interpolated here concerning Dr. Greenwood's Table I., p. 129 (Collis and
Greenwood), apropos of the much greater decline since 1851-60 in the death rate of women than of men.
Here it is probable altered nomenclature is in considerable part responsible, chlorosis in young girls,
and cancer in older women often figured as " phthisis " 60 or 70 years ago. Again as regards the comparatively
light incidence at the higher ages upon women two factors must not be lost sight of. First,
as already hinted, alcoholism is less destructive in women than in men; and second, the harmful effect