Educational gradients in the prevalence of medically assisted reproduction births in a comparative perspective.

Objective: To study educational gradients in births after medically assisted reproduction across 5 countries with different institutional arrangements.

Design: We use logistic regression and compute predicted probabilities to estimate the association between education and giving birth after assisted reproduction, before and after adjustment for maternal age at delivery and marital/partnership status, using an overall sample of about 3.9 million live births in 5 countries.

Settings: Not applicable.

Patients: This study includes survey or register data containing information on births in 5 countries: N = 61,564 for Denmark, N = 37,533 for France, N = 12,889 for Spain, N = 17,097 for the United Kingdom, and N = 3,700,442 for the United States.

Intervention (for Rct) Or Exposure (for Observational Studies): None.

Main Outcome Measures: The probability of a child being born after medically assisted reproduction for mothers with a university degree relative to those having less than a university degree.

Results: University-educated mothers are more likely to give birth after assisted reproduction compared with mothers with lower levels of education. After adjustment for socio-demographic characteristics, educational differences disappear in the United Kingdom and to some extent Spain, whereas they attenuate but persist in the other countries. The United States seems to show a larger educational gradient.

Conclusion: The results suggest that the institutional setting around assisted reproduction may moderate the gradient. A possible explanation may be access to treatments, as the United States - the context with the lowest subsidization - seems to show larger educational gradients than other contexts. In the context of global postponement of childbearing to older ages, mothers with lower levels of socioeconomic resources might find it more difficult to fully realize their fertility intentions in countries with less generous subsidization of treatments.