View report page

51
officer, and it is impossible not to recognise a trend of opinion in the direction of regarding the
phthisical person as a public danger. It becomes, therefore, very necessary to examine the grounds
upon which a demand is made which, if fully acted upon, would have the effect of depriving many
thousands of persons in London of their liberty.
These grounds consist in the main of observed instances of association of the sufferer with houses
previously invaded or with other persons known to be suffering from the malady. The number of such
instances is sufficiently large to create the impression that the association of the sufferer with such houses
or persons has been the cause of his illness. It should, however, be borne in mind that this number
would necessarily be large, even if infection played no part in the matter, and the question needs to be
considered whether the number greatly differs from the number which would obtain if chance alone
determined the observed association.
A table contained in one of the annual reports shows that of 232 deaths from phthisis occurring
in 190(5 in the district, 70, or 30.1 per cent., occurred in houses in which one or more deaths from the
same disease had occurred in the preceding ten years. There is no information as to the
length of time during which the persons dying in 1906 had resided in the houses in question, or
whether, in fact, they had taken up their residence before or after their fatal illness first manifested
itself, and the assumption must be that this point was not unconsidered, and that the 70 persons
had resided in these houses sufficiently long to warrant the opinion that the disease had
its origin in the houses in which they subsequently died or from which they were removed
to institutions where they died. The question, therefore, which needs consideration is, in what proportion,
as a matter of arithmetical probability, would the deaths of 1906 occur in houses in which previous
deaths from this disease had occurred in the preceding ten years. The annual average number of deaths
in those years may be taken as 224 and the number of houses in the district may be taken as 8,500.
The result arrived at on purely arithmetical grounds is that 54 deaths out of the 232 recorded in 1906,
or 23.3 per cent., is the most probable number which would have occurred in houses in which one
or more deaths took place in the ten years 1896-1905. This is a near approach to the 30 per cent.
actually recorded, and, indeed, the difference between the two percentages is within the range of
"probable erro" owing to the smallness of the figures on which the calculation is based. Further,
it must be recollected that if malnutrition, overcrowding, ill-ventilation and similar concomitants of
poverty play an important part in the causation of phthisis it might be expected that quite apart from
the question of the infectivity of the disease more deaths would occur in houses occupied by poor
persons than in houses occupied by persons who are well to do, and that if it had been possible to
make allowance for this the proportion of deaths occurring in houses previously invaded would have
been still less conspicuous. Examination of the figures certainly suffices to negative the view that
in themselves they afford any reliable evidence of the infectivity of the disease.
Another method of investigation which is frequently adopted in connection with endeavours to
throw light on the question of infectivity of phthisis is to inquire of a phthisical person in what way he has
been associated with other cases of phthisis, and having found such association, before his illness was
conspicuous, to assume that the disease was thus contracted. The inference is that but for this association
the phthisical person in question would have remained free from his disease, and in this way a foundation
is being laid for the claim to isolate persons suffering from this malady. For precise inferences a comparison
of the relative incidence of disease upon persons thus exposed and others not so exposed is
necessary. In this connection Dr. G. B. Longstaff's paper published in the Collective Investigation
Record of 1883 is of especial value. Dr. Longstaff calculated the probability of the accidental and fatal
incidence of phthisis upon both husband and wife, and he found that during ten years 1871 -80, in
England and Wales, assuming marriage to have no influence upon phthisis, by selection, infection, or
otherwise, it would happen 4,363 times that both husband and wife would die of that disease, and that
of 4,363 husbands dying in the ten years 829 wives would die within one year of the husband's death,
742 within one year and under two years, 654 within two years and under three years, and so on in
diminishing proportions. There was no evidence that this number is actually exceeded.
Again, Professor Karl Pearson has studied the subject. He had for his use the facts relating to
384 families in which cases of pulmonary tuberculosis had occurred. The records were supplied by
Dr. W. C. Rivers from the Crossley Sanatorium, Frodsham, and the patients were almost wholly of
the lower middle and working classes from Manchester and Liverpool. Professor Pearson found
that the number of tuberculous married persons who had a tuberculous mate was not sufficiently
great to require an appeal to infection to account for this number, but what he did find was
that the condition which governed the liability to attack was heredity of the diathesis of pulmonary
tuberculosis. "A theory of infection does not," he writes, "account for the facts. I am inclined to
think that the risks run, especially under urban conditions, are for tuberculosis, as for a number of
other infectious diseases, so great that the constitution or diathesis means almost everything for the
individual whose life cannot be spent in self-protection;" and again, " I feel fairly certain that for the
artisan class the inheritance factor is far more important than the infection factor, because in a very
large proportion of cases it does not lie in the power of the individual to maintain in the stress of
urban life a wholly safe environment."
The acceptance of the view that under the conditions which exist in urban populations the susceptibility
of the individual and not the exposure to known cases of phthisis governs the probability
of attack does not itself negative the theory of the infectivity of tubercular phthisis. It does, however,
raise serious question as to the practicability of reducing the death-rate of phthisis mortality by efforts
to segregate persons suffering from this disease, even if the theory of infection from person to person
is accepted as the common way in which susceptible persons are infected. However this may be, it may
14911
g 2