Mechanisms underlying the associations of maternal age with adverse perinatal outcomes: a sibling study of 264 695 Danish women and their firstborn offspring

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

The mechanisms underlying the association between maternal age (both young and older maternal age) and adverse perinatal outcomes are unclear.
Methods We examined the association of maternal age at first birth with preterm birth (<37 weeks gestation) and small for gestational age (SGA) in a cohort of 264¿695 Danish women, each of which had at least one sister in the cohort (n¿=¿121¿859 sibling groups). We compared cohort analyses with sister-controlled analyses. The sister-controlled analyses control for all observed and unobserved characteristics that are identical or very similar between sisters, such as childhood socio-economic characteristics—a confounder we hypothesized would exaggerate the young maternal age–adverse outcomes association but mask the older maternal age–adverse outcome association.
Results There was a U-shaped association of maternal age with risk of preterm birth (lowest risk age 24–30 years) and SGA (lowest risk age 26–30 years) in cohort analyses. In analyses with sister control, there was a J-shaped association of maternal age with preterm birth, with a monotonic increase in risk across the maternal age range from 24 years of maternal age. For SGA, risk increased across the age range in sister-controlled analyses, being lowest at age 15 years and highest at age 45 years (thought with wide confidence intervals at the extremes of the age distribution).
Conclusions Our findings suggest that different mechanisms underlie the association of younger and older maternal age with adverse perinatal outcomes. Socio-economic position and other characteristics shared by sisters appear to explain most of the association of young maternal age with adverse perinatal outcomes, but the association of older maternal age with preterm birth, and SGA is not explained by this confounding and may even be masked by it.
OriginalsprogEngelsk
TidsskriftInternational Journal of Epidemiology
Vol/bind40
Udgave nummer5
Sider (fra-til)1205-1214
Antal sider10
ISSN0300-5771
DOI
StatusUdgivet - 2011

ID: 40286969