Two nurses greeted me as I entered my clinic at Abbott Northwestern Hospital last Tuesday morning, faces tense. "Have you seen the article in the Trib? What does this mean? Our patients are already calling."

JAMA Pediatrics had just released a study suggesting that prenatal exposure to antidepressants may be associated with autism spectrum disorders. The Star Tribune published a summary from the Los Angeles Times ("Drugs for depression are linked to autism," Dec. 15). My inbox was quickly flooded with messages from concerned patients and OB-GYNs.

"I saw the article in the paper today. I know we talked about this, but I'm really worried about autism and wondering if I should stop my Zoloft?" one woman asked. "This patient is struggling, and I think she needs an antidepressant," an OB-GYN messaged. "Probably not going to happen now with this autism thing. Hoping she can see you."

As a perinatal and reproductive psychiatrist, I care for women who experience mental health conditions during pregnancy and the postpartum period. For mild to moderate depression, approaches like psychotherapy and lifestyle modification may be effective treatment. Women with moderate or severe depression, however, generally require a combination of antidepressants and other interventions to get and stay well.

I help my patients balance the risks of untreated psychiatric illness with the risks and benefits of medication use in pregnancy. Together, we create a treatment plan that best fits each woman's unique situation. Without this type of individualized assessment, pregnant women and their physicians may focus on medication risks and underestimate the impact of untreated depression and the benefits of treatment.

These patients are desperate to feel better and consumed with worry that taking antidepressants will harm their baby. Nothing contributes to this anxiety more than media reports that emphasize medication risks without providing the necessary context.

The article that appeared in the Star Tribune led with the most alarming finding from the JAMA Pediatrics study: Women who used selective serotonin reuptake inhibitor (SSRI) antidepressants in late pregnancy were twice as likely to have a child with autism as women who did not take SSRIs. Although a mother's history of depression was noted as a risk factor for having a child with an autism spectrum disorder, the study is credited with separating the impact of depression from the impact of medication. The article highlighted the strengths of the study design and closed with a pediatrician characterizing a mother's antidepressant use as a possible factor in her baby's "suboptimal outcome."

One pediatrician's "suboptimal outcome"? Every pregnant woman's biggest worry. But the JAMA Pediatrics findings and the ensuing media coverage tell only part of the story.

This study had significant limitations. The authors did not account for risk factors for autism, like family history of autism spectrum disorders, paternal age, or a mother's use of alcohol, tobacco, or illicit drugs. They did not confirm whether women actually took the antidepressants doctors prescribed. And when the authors considered only the children whose diagnosis of autism was made by a neurologist or a psychiatrist, the association between SSRI use and autism spectrum disorders disappeared.

Most important, the media coverage missed the crucial point: The study's authors did not adequately account for severity of depression in the mothers studied, so there is no way to tell if the higher rates of autism spectrum disorders were due to use of SSRIs or to maternal depression. This is particularly relevant because several other large studies either found no association between SSRI use and autism spectrum disorders or concluded that the slightly increased rates the autism spectrum disorders identified were due to maternal depression, not the medications used to treat it.

We don't yet know what causes autism, but it likely results from a complex interaction of genetic and environmental factors. Patients, doctors and journalists all can agree on the critical need for more research. In contrast, the risks of untreated maternal depression — poor prenatal care, adverse obstetric outcomes, use of alcohol and tobacco, suicide, increased risk of cognitive, emotional and behavioral problems in children, and rarely, infanticide — are well-documented.

So when our patients call? We give them a balanced view of the data. And a recommendation to avoid unnecessary exposure — not to antidepressants, but to oversimplified and misleading media reporting.

Dr. Elizabeth M. LaRusso is director of the Mother Baby Mental Health Program at Allina Health/Children's Hospitals and Clinics of Minnesota.