Variations in the maternal age effect on twinning rates: the Nordic experience.

Maternal age is the most important nongenetic factor influencing the twinning rate. Every study of the effect of other factors should consider the age distribution of the mothers. Besides standardizing techniques, the age-specific twinning rates are analyzed. Especially, the slope of the regression line is considered when the age-specific twinning rates are regressed against maternal age. How well the maximum age-specific twinning rate measures the variation in the maternal age effect is also studied. The two measures, which are strongly correlated, give consistent results. The proposed methods are applied to data from Finland (1866 to 2003) and Denmark (1855 to 2000). Comparisons with Sweden (1861 to 2000) and Norway (1855 to 2003) are also made. In general, marked decreasing trends in the twinning rates are discernible up to the middle of the 20th century. In Sweden the decline starts around the beginning of the 20th century and is stronger than in Denmark and Norway. The decrease in the twinning rate for Finland, which up to World War II was rather isolated, sets in around the middle of the 20th century and is not as strong as in other Nordic countries. After the minima around 1980, strong increasing trends are noted. This is mainly caused by the influence of the artificial reproduction technologies and particularly the use of fertility-enhancing drugs on the commonly noted dependence between maternal age and the twinning rate. Similar temporal trends can be observed in the slopes of the regression line and in the maximum age-specific twinning rates.