Perinatal mortality rates differ markedly between countries in Europe. If population characteristics, such as maternal age, parity or multiple births, contribute to these differences, standardised rates may be useful for international comparisons of health status and especially quality of care. This analysis used aggregated population-based data on fetal and neonatal mortality stratified by maternal age, parity and multiple birth from 12 countries participating in the EURO-PERISTAT project to explore this question. Adjusted odds ratios were computed for fetal and neonatal mortality and tested for inter-country heterogeneity; standardised mortality rates were calculated using a direct standardisation method. There were wide variations in fetal and neonatal mortality rates, from 3.3 to 7.1 and 2.0 to 6.0 per 1000 total and livebirths, respectively, and in the prevalence of mothers over 35 (7-22%), primiparae (41-50%) and multiple births (2-4%). These population characteristics had a significant association with mortality, although results were less consistent for primiparity. Odds ratios for older mothers and primiparae showed significant inter-country heterogeneity. The association between maternal age and fetal mortality declined as the prevalence of older mothers in the population increased. Standardised rates did not substantially change inter-country rankings and demographic characteristics did not explain the higher mortality observed in some countries. Our results do not support the use of mortality rates standardised for age, parity and multiple births for international comparisons of quality of care. Further research should explore why the negative effects of older maternal age decrease as delayed childbearing becomes more common and, in particular, whether this is due to changes in the social characteristics of older mothers or in health care provision. © 2009 Blackwell Publishing Ltd.