Birth weight of infants and mortality in their parents and grandparents: the Uppsala Birth Cohort Study

Methods We used the Uppsala birth cohort to examine all-cause and circulatory disease mortality among 14 129 mothers, 13 863 fathers, 3992 maternal and 3910 paternal grandmothers and 3896 maternal and 3798 paternal grandfathers. Follow-up time lasted up to 2002 with median of 21.6 years for parents and 47.8 years for grandparents. Results Hazard models controlling for socio-demographic characteristics indicated an inverse linear association between offspring's birth weight and maternal circulatory disease mortality [hazards ratio (HR) = 0.74; 95% confidence interval (CI): 0.56-0.99]. No association was observed for fathers. The association between birth weight of grandchildren and grandparental mortality varied. No association was detected for maternal grandmothers for the whole period examined; only among grandmothers whose grandchild was born prior to 1977 was a U-shaped association observed for circulatory disease mortality. There was a U-shaped association between birth weight of grandchildren and maternal grandfather's overall and circulatory disease mortality (quadratic term: HR = 1.05; 95% CI: 1.01-1.09). An inverse linear relationship was found for circulatory disease mortality in paternal grandmothers (HR = 0.93; 95% CI: 0.85-1.00). For paternal grandfathers there was an inverse association between grandchildren's birth weight and overall mortality; for circulatory disease mortality an effect was found only for grandfathers whose grandchild was born prior to 1977. Conclusions We showed, for the first time, associations between fetal growth in one generation and mortality in the previous two generations. Genetic and/or environmental pathways accounting for these associations should be further explored.