View report page

25
As has been shown in connection with births, an underestimation
of the population artificially improves the birth-rate.
In the same way it gives an unfairly bad idea of the health of the
district if that idea is based on the death-rate, and therefore
certain factors must be considered before any accurate deductions
can be made.
In calculating the death-rate for the year for any district,
information on four points is required: (a). the total population
at the middle of the year; (b). the number of deaths registered
during the year; (c). the number of deaths of non-residents and
(d). the number of deaths of residents registered in other places
during the year. Accurate facts in connection with the last three
points is ohtained from the registrar's returns, but the first depends
entirely on the particular method of calculating the population
adopted by the Medical Officer of Health.
It may seem to some that too much notice is taken of this
matter, but it is of real importance when comparisons are made
between the health of different districts and it is assumed that
one is healtber than another because it has a higher birth or
lower death-rate. Such comparisons can only be correct when
the rates are calculated on actual census returns, and points b, c,
and d, all properly taken into account. To show the effect of
these various data in this district, the following table has been
prepared. It will be seen that six different death-rates for the
year may be quite correctly calculated and all the results obtained
may very likely be wrong. Two estimates of the population are
taken, and of course the actual number of deaths used in each
group are the same:—
1. Death-rates based on official estimate of population
(a). Including all deaths registered in district 22.9
(b). Deaths of residents only registered 15.1
(c). Deaths of residents registered in district plus
deaths of residents registered outside 15.9