Women who took a class of widely used antidepressants during their second and third trimesters of pregnancy were roughly twice as likely as those who did not to have a child who would later receive a diagnosis of autism spectrum disorder, says a new study.

The new research, published Monday in JAMA Pediatrics, is among the strongest findings linking antidepressant use in pregnancy to poor outcomes in the children born of those pregnancies, said experts. But it leaves many questions unanswered about the roots of autism, the prevalence of which appears to be surging.

The new findings emerged from a Canadian registry of 145,456 newborn children who were followed for an average of about six years. The medical records of the babies’ mothers were available for at least a year before their birth, allowing researchers to look at whether and when the babies’ pregnant mothers took selective serotonin reuptake inhibitors in medications marketed as Prozac, Zoloft and Lexapro.

In the population as a whole, 0.7 percent of the registered babies (1,054) later received an autism diagnosis. Among the 2,532 babies whose mothers took an SSRI during her second and/or third trimester of pregnancy, 31 infants (or 1.2 percent) would be diagnosed with autism some time in his or her first six years of life. Some 40 babies (1 percent) whose mothers took an SSRI in her first trimester of pregnancy were eventually diagnosed with autism.

A mother’s history of depression has long been suspected of raising a child’s autism risk.

The researchers compared rates of autism among babies born to women with a history of depression with autism rates among babies born to those who took antidepressants during pregnancy. They found that babies whose mothers took an SSRI were still about 75 percent more likely to get an autism diagnosis than were those whose mothers had a history of depression.

The study is notable for the fact that it does not rely on a mother’s recollection of her pregnancy habits after an autism diagnosis has occurred. In research gauging links between autism and a range of environmental factors (including air pollution, cigarette smoking, stress and poor nutrition), studies that rely on delayed recall are widely thought to overstate the effect of those factors, since women are more likely to remember problematic exposures in the wake of a child’s diagnosis.