Oral contraceptives taken just before or during pregnancy do not increase the risk of birth defects, according to a large-scale study published Wednesday.

Examining records for nearly 900,000 live births in Denmark, researchers found that even women who used the pill after becoming pregnant were no more likely to have babies with serious defects than mothers who had never used it.

"We confirmed that there wasn't any association between oral contraceptives and major birth defects," lead author Brittany Charlton, a scientist at the Harvard T.H. Chan School of Public Health, told AFP by phone.

"Our findings were especially reassuring given that we were able to use a different approach."

Most previous research, she explained, is based on "case controlled" studies, which begin with a fairly rare outcome -- a birth defect -- and then work backwards, looking for a cause.

Several of these studies, some decades old, had found a link between use of the hormone-based contraceptives and defects, even if most did not.

"We were able to leverage prescription registries and thus eliminate any bias from women inaccurately recalling their use" of the pill, Charlton said.

Drawing from Danish national health records from 1997 to 2011, Charlton and colleagues divided the women into four groups.

A fifth -- some 176,000 women -- had never used the pill, while more than two-thirds stopped at least three months before becoming pregnant.

Eight percent discontinued use within three months of conceiving, while one percent -- well above a statistically significant 10,000 women -- used oral contraceptives after becoming pregnant.

For all categories, the ratio of normal birth to those with major defects was exactly the same -- 25 per 1,000 live births.

This ratio remained consistent across all groups even with the inclusion of pregnancies that ended in stillbirths or induced abortions.

The study was published in the medical journal BMJ.