208 THE ALCOHOL PROBLEM 
of the occupational groups are too small to yield reli- 
able data, so in the Table of results recorded below all 
groups are ignored in which the number of deaths from 
alcohol was less than ten. 
The best basis of comparison of death rates is to 
take the * comparative mortality figure,” which shows 
the number of deaths per year in a standard population 
of 71,005 men, aged twenty-five to sixty-five. In the 
1900-02 census this number of men suffered on an 
average exactly 1,000 deaths, but by 1910-12 the 
expectation of life had so greatly improved that the 
deaths fell to 790. This is the average mortality 
figure, the healthiest occupational groups showing one 
of less than 400, and the unhealthiest (the barmen), one 
of 1,724. For purposes of classification, the various 
occupational groups have been averaged according as 
their comparative mortality figure was under 600, or 
ranged from 600 to 699, 700 to 799, and so on. 
The 790 deaths experienced by the average standard 
population (of England and Wales) included 4 deaths 
from “alcoholism ”” and 13 from “ cirrhosis of the 
liver,” and for our purpose it is best to combine these 
two numbers, and say that there were 177 deaths attrib- 
utable to ‘““ alcohol.” From the data in the Table we 
see that of the various occupational groups of skilled 
and semi-skilled workmen (which numbered 47 in 
all), those with a comparative mortality figure of less 
than 6oo suffered only 7 deaths from alcohol, those 
groups with a mortality figure of 600 to 69g suffered 
10 deaths, and those in the succeeding divisions 
suffered 13, 15, and 16 deaths respectively. That is 
to say, the alcohol mortality increased steadily as the 
general mortality increased. Expressed as a per-