Objective: To examine the trends in the proportion of annual live birt
hs that were male in Canada and to compare the trends with those in th
e United States.Design: Analysis of census data. Setting: Canada as a
whole and 4 main regions (West, Ontario, Quebec and Atlantic). Subject
s: All live births from 1930 to 1990. Outcome measures: Sex ratio (exp
ressed as the proportion of total live births that were male [male pro
portion]) overall and by region. Results: The male proportion in Canad
a decreased significantly after 1970 (p < 0.001); this represented a c
umulative loss of 2.2 male births per 1000 live births from 1970 to 19
90. Although a decrease was observed in all four regions studied, only
that in the Atlantic region was significant (p < 0.001), representing
a cumulative loss of 5.6 male births per 1000 live births from 1970 t
o 1990. A significant decrease in the male proportion was also observe
d in the United States from 1970 to 1990 (p < 0.001), although to a le
sser degree than that observed in Canada, and represented a cumulative
loss of 1.0 male births per 1000 live births. Conclusions: The decrea
sed sex ratio in Canada adds to the growing debate over changes in bio
logical markers and their potential causes. In addition,the study illu
strates the potential use of the sex ratio as a widely available, unam
biguous measure of the reproductive health of large populations.