Papers

Unexplained differences in sex ratios at birth in Europe and North America

BMJ 2002; 324 doi: https://doi.org/10.1136/bmj.324.7344.1010 (Published 27 April 2002) Cite this as: BMJ 2002;324:1010
  1. Victor Grech, consultant (victor.e.grech{at}magnet.mt)a,
  2. Charles Savona-Ventura, consultantb,
  3. P Vassallo-Agius, consultant paediatricianc
  1. a Paediatric Department, St Luke's Hospital, Guardamangia MSD 09, Malta
  2. b Department of Obstetrics and Gynaecology, St Luke's Hospital
  3. c Medical School, University of Malta, Msida MSD 06, Malta
  1. Correspondence to: V Grech
  • Accepted 4 October 2001

In mammals, male live births exceed female ones. In humans, the ratio of male births to total births is expected to be 0.515. In Europe during 1990-5 this differed significantly with increasing geographical latitude.1 We analysed and compared the male to female ratio in Europe and North America over 50 years.

Methods and results

We obtained annual data on male and female live births from the World Health Organization (WHO) for the North American continent for 1958-97 and for European countries for 1950-99. Overall <3% of data were missing.

Numbers of male, female, and total births, with calculated male:female ratios for countries included in study of sex ratios at birth in Europe and North America

View this table:

European countries were banded by latitude. Southern countries (latitude 35-40°) included Bulgaria, Greece, Italy, Malta, Portugal, and Spain. Central Europe (40-55°) included Austria, Belgium, Czech Republic, France, Germany, Hungary, Ireland, Luxembourg, Netherlands, Poland, Romania, Switzerland, and the United Kingdom. Nordic countries (>55°) include Denmark, Finland, Iceland, Norway, and Sweden. North America was divided by latitude into Canada (>50°), the United States (30-50°), and Mexico (<30°).

We analysed contingency tables using Embedded Image2 and Embedded Image2 for trend and obtained 95% confidence intervals for ratios by using the Fleiss equations. PEmbedded Image0.05 was taken as significant.

Results

Significantly more boys were born in southern countries (table) than in central Europe (Embedded Image2=57, P<0.0001) or the Nordic countries (Embedded Image2=8.8, P=0.003; Embedded Image2 for trend=120, P<0.0001). The difference between central Europe and the Nordic countries was not significant. All had a male to female ratio <0.515, with a resultant male birth deficit of 12 744 in the Mediterranean, 212 780 in central Europe, and 13 169 in the Nordic countries (total deficit of male births 238 693).

A low male to female ratio was found in Mexico, a higher ratio in the United States, and an even higher ratio in Canada (Embedded Image2 for trend=57, P<0.0001). All had a male to female ratio <0.515, with a resultant male birth deficit of 21 993 in Canada, 410 932 in the United States, and 521 789 in Mexico (total deficit 954 714).

In the two continents, the total male birth deficit was 1 193 407 live births.

Central Europe and the southern countries span latitudes equivalent to the United States. The male to female ratio for the total number of births of these two European regions was significantly higher than that for the United States (Embedded Image2=499, P<0.0001).

The Nordic countries span latitudes equivalent to Canada. Although the male to female ratio in these countries was higher than in Canada, this difference was not significant.

Comment

In Europe, significantly more male babies were born in southern latitudes than in northern latitudes, whereas the reverse was found in North America. We are unable to explain these findings, which do not support a temperature related effect.

Acknowledgments

We thank Mie Inoue, World Health Organization; Garry Mac Donald, Statcan; Catherine Vella, National Statistics Office, Malta; and Pablo Aviles Hernandez, Mexican National Statistical Office, Mexico.

Footnotes

  • The male:female ratio at birth by latitude in Europe is the reverse of that in North America. A temperature related effect cannot account for this

  • Funding None

  • Competing interests None declared

References

  1. 1.
View Abstract