Every few months, pregnant women hear another scare story about antidepressant use: The drugs have been linked to birth defects, autism and lung problems in babies. These studies make it seem like giving up antidepressants during pregnancy is the only option, even though it might cost women their health and ability to function.

In reality, the jury’s still out. And often, women are being offered only the worst-case scenario, rather than a balanced view of the scientific debate.

I’ve spent more than 20 years working with pregnant women with psychiatric conditions such as depression, anxiety and addiction. I’ve also studied the consequences of maternal antidepressant use on pregnancy outcomes.

Here are some of the facts:

Depression and anxiety disorders affect 10 to 20 percent of pregnant women. Many use some kind of medicine as part of their therapy. And despite what the news media suggests, the decision to ease off antidepressant treatment is not the same as “weaning” oneself off caffeinated coffee.

Moms-to-be should not approach the decision casually. My research shows that more than half of women undergoing treatment before conceiving discontinue their medication during pregnancy. Some women know they will be able to manage their moods and anxiety temporarily and are able to stop treatment in pregnancy. But not all women will stay well. If a woman has severe recurrent depression, stopping antidepressant treatment in pregnancy increases her risk of relapse by a factor of five. There is also evidence that depression itself is linked with problems such as preterm birth. My work shows that a number of anxiety disorders, which also are typically treated with antidepressants and often accompany depression, can have a greater impact on pregnancy outcomes than antidepressants. This suggests is that some pregnancy complications are brought about by the conditions that antidepressants are designed to treat as well as the medication itself.

And research hasn’t conclusively established that antidepressants are harmful to fetuses. For every paper that finds a problem, such as preterm birth among babies exposed to antidepressants in utero, there are well-designed studies that do not replicate those risks.

One also needs to consider the magnitude of the risk. Problems such as preterm birth or physical malformations that may occur after in-utero exposure to antidepressants are relatively rare.

For example, 1 percent of babies are born with heart-valve defects. If paroxetine increases the risk, it is only by 2 percent. If the use of antidepressants causes shorter pregnancies, this translates to just three to four fewer days of pregnancy. It is also entirely possible that these results are not caused by antidepressant use but by other factors such as the use of cigarettes, alcohol or illicit drugs. In our work, we considered such complicating factors when we assessed the possible impact of antidepressants on pregnancy outcomes. The possible effect of antidepressants was reduced to nearly within the margin of error.

In fairness to those who have concerns about antidepressant use in pregnancy, many women are able to stop antidepressants in pregnancy and not relapse. Also, because research in this area remains inconclusive, physicians cannot say for sure that antidepressant use in pregnancy is risk-free to women and their children. There are only a few studies that assess long-term outcomes among children whose mothers took antidepressants in pregnancy. Although the results were favorable, the research is rather lean.

Physicians and their patients need to be cautious; medication treatment should be reserved for women who do not respond to other depression treatments, such as psychotherapy. But doctors don’t recommend that pregnant women with diabetes stop insulin or discontinue blood pressure medication despite some evidence that these treatments are associated with unfavorable birth outcomes. Instead, they weigh the maternal benefits and risk to the fetus and consider the bigger picture.

Why is antidepressant treatment different? Many people refuse to accept that illnesses such as depression are biologically based and not simply a function of character weakness or flaws. They fail to appreciate that in many cases, there is a need for psychiatric medication.

In my experience as a physician, pregnant women with a psychiatric illness know their bodies and they usually know what they can and cannot do in terms of their treatment. We need to listen to them, provide them with information and support them in their decisions rather than judge them for their health-care choices.