View report page

13
tions have been made in arriving at these rates, (1) that all unmarried mothers are
primiparae, (2) that the mortality among unmarried mothers is twice that of married,
and (3) that nearly all unmarried mothers are under 25 years of age. These assumptions
enable the effects of illegitimacy upon the maternal mortality rate to be
approximately eliminated.

The fact that so many assumptions have had to be made to arrive at a statement
of the elementary facts relating to maternal mortality shows the difficulty with which
statistical consideration of the subject is attended owing to the incompleteness of our
records. In order to define and direct measures for the improvement of maternal health
it is essential that greater attention should be given to records bearing upon maternal
conditions.
The point of particular interest in the rates shown in the table is that, while
the mortality per 1,000 married women in the period 1911-20 showed the
continued decrease apparent in preceding years, the mortality calculated per 1,000
births increased during this period ; and the rates for 1921-27 show insignificant
decrease upon those for 1901-10. In short, the period of least improvement in
maternal mortality is that in which the decline in fertility has been greatest.
Figures for the more recent periods, selected according to fertility as shown by
the number of births, may serve to illustrate the inverse relation of fertility and
maternal mortality per 1,000 births. Omitting the years of great influenzal epidemics
and selecting those of high and low fertility since 1910, the deaths in childbirth from
causes other than puerperal fever per 1,000 live births in London are as follows:—
Births Deaths from accidents of child birth
Period. (annual average). (per 1,000 births).
1911-15 109,184 1.56
1916-17 89,943 1.70
1920-21 110,184 1.58
1926-27 76,044 1.67
It will be seen that as the birth-rate falls maternal mortality, stated as a rate per
1,000 births, tends to increase. Herein lies in large part the explanation of the
apparent lack of improvement in maternal mortality, as judged from such rates, in
spite of the ever-widening scope of administrative effort to improve the conditions
of childbirth.
The deaths from puerperal fever have been excluded in the above table because
the mortality from this cause is subject to periodical fluctuations which are closely
related to the variations in the mortality of erysipelas, and, though in a less degree,
of scarlet fever. See page 23.
It is impossible to state accurately the rates of maternal mortality, or, what is more
important, the changes in the rate of fertility of women, without a knowledge of the age of
the mother at each confinement. There is no requirement on registering or notifying
a birth that the age of the mother should be stated ; this defect in the records was
referred to by Dr. T. H. C. Stevenson in his evidence before the National
14679 b