A sweeping new review of prenatal antidepressant use underscores a finding that has surfaced repeatedly throughout the last decade: While parental depression is strongly linked to child neurodevelopmental disorders, taking antidepressants during pregnancy does not appear to significantly increase a child’s risk of autism.

In an analysis of 37 separate studies covering more than 25 million pregnancies, a research team from the University of Hong Kong found that children born to women who took antidepressants while pregnant were indeed more likely to later be diagnosed with autism or attention-deficit/hyperactivity disorder (ADHD).

But when the researchers took into account confounding factors such as a family history of neurodevelopmental disorders or mothers’ preexisting mental health conditions, the correlation disappeared.

The data showed that children born to women with a history of depression were more likely to be diagnosed with autism or ADHD, regardless of whether their mother took psychiatric medication. Children were also more likely to be diagnosed with autism and ADHD if their fathers took antidepressants during their gestation, even if their mothers did not — an association that suggests a genetic link, not a pharmacological one.

The results were published this month in the journal the Lancet.

“Our findings are consistent with current clinical guidelines, which generally support continuing antidepressant treatment during pregnancy when it is clinically indicated,” said Dr. Wing-Chung Chang, a psychiatry professor at the University of Hong Kong and the paper’s senior author. “Our findings do not provide strong evidence that prenatal antidepressant exposure causes neurodevelopmental disorders.”

The possibility that antidepressant use in pregnancy may play a role in neurodevelopmental conditions has been a source of anxiety for many expectant parents since at least 2015, when a much-publicized Canadian study observed that women who took certain antidepressants later in pregnancy were about twice as likely to have an autistic child than women who did not take the drugs.

Multiple studies since then have also identified a correlation between a woman’s use of antidepressants during pregnancy and her child’s later diagnosis of autism, and to a lesser extent, ADHD.

But ending the analysis there overlooks a crucial distinction, researchers say: the possibility that the association actually is between the neurodevelopmental disorders and depression, not the medication.

Autistic people of all ages are significantly more likely than their neurotypical peers to be diagnosed with mental health conditions such as depression or anxiety. Large-scale population studies have found that autistic adults are up to three times as likely to have depression compared with non-autistic people.

The reasons for mental health symptoms in autistic people are varied and complex, and the challenges of navigating a world designed for a different way of thinking may play an important role. But research has also identified multiple genetic profiles and biological pathways common to autism and mood disorders, and it’s likely that both conditions are at least partially the result of family genetics.

“The mental health of your family tree is in some way statistically associated with your risk of autism,” said Brian K. Lee, a professor of epidemiology and biostatistics at Drexel University.

Neither depression nor autism causes the other. Lee compared their frequent co-occurrence to the pairing of fiery red hair and pale, sunburn-prone skin: two highly heritable traits that can easily occur independently in a given individual, but that often travel together through family trees.

“What the literature has shown us so far is that while there does, at face value, appear to be an association of slightly increased risk of autism in mothers who take antidepressant medications, when you control for the underlying depressive disorder that risk goes away,” said Dr. Kathryn Erickson-Ridout, a senior psychiatrist for the Permanente Medical Group and research scientist with the Kaiser Permanente Division of Research. “This evidence shows us that most likely, the biological pathways that are disrupted in major depression are also important for autism.”

Erickson-Ridout compared the chilling effect of the 2015 Canadian study on psychiatric care for pregnant women with the anxiety around vaccines sparked by Andrew Wakefield’s since-retracted 1998 paper inaccurately linking autism to the mumps, measles and rubella shot.

The Canadian study did not contain major errors as Wakefield’s paper did, though some critics argued at the time that it didn’t sufficiently control for confounding factors such as maternal depression.

But its media coverage often failed to make clear both the low overall risk of autism — 1.2% of babies born to women who took selective serotonin reuptake inhibitors during their second or third trimester were later diagnosed with autism, compared with 0.7% of babies in the general population — or weigh the risk of antidepressant use against the risks of untreated depression.

Its effects persist today. Last year, the U.S. Food and Drug Administration convened a controversial panel on prenatal SSRI use. Nine of the panel’s 10 members were researchers, doctors or psychologists who have previously questioned the drugs’ safety or criticized antidepressant use in general. Among them was Anick Berard, an epidemiologist and lead author of the 2015 Canadian paper.

Suicide is the second-leading cause of maternal mortality in the U.S., with homicide being the first.

Any discussion of the risks of antidepressant medications has to be weighed against the potential harms of abruptly ceasing or refusing to treat a potentially life-threatening mental health condition, said Dr. Katie Unverferth, a reproductive psychiatrist and medical director of UCLA’s Maternal Mental Health Program.

“Pregnancy is such an anxious time at baseline — so many new things are happening, and your body’s changing, and you want to make sure you’re doing the right thing for yourself and your developing baby,” Unverferth said. “This study just provides additional reassuring data.”

After six days of trying to avoid an overheating chemical tank erupting into a giant fireball or spilling thousands of gallons of toxic substances at an aerospace facility in Garden Grove, Orange County leaders announced Tuesday that the risk of catastrophic explosion had largely been eliminated.

Local authorities lifted a large section of the evacuation zone surrounding GKN Aerospace and allowed tens of thousands of residents to return.

Firefighters sprayed more than 9 million gallons of water onto a piping-hot tank of flammable methyl methacrylate (MMA), drastically bringing down the vessel’s temperature — but not before the high temperatures resulted in high pressure and a crack in the side of the tank, which acted as a relief valve.

Interim Orange County Fire Authority Chief TJ McGovern indicated in a Tuesday afternoon community meeting that evacuation zones might soon shrink further. He noted that crews had stopped spraying water onto the tank and were in the process of assessing whether the vessel’s temperature had stabilized.

“Once we know that temperature is stabilized, we will be taking the fire risk off the table,” he said. “If there’s no fire risk, our evacuation zones are going to shrink.”

McGovern said, about 5 p.m., that officials were “hoping that we’re going to have a very positive outcome very soon.” He asked the community for continued patience over the next few hours as crews worked to validate the initial data they were seeing regarding temperature stabilization.

The Orange County health officer and fire officials have insisted there have been no vapors or chemical leaks over the course of the six-day crisis. The U.S. Environmental Protection Agency said its air monitors surrounding the facility had not detected methyl methacrylate or other toxic airborne chemicals (known as volatile organic compounds).

But environmental experts remained skeptical that no toxic substances had been released. Andrew Whelton, a Purdue University professor who studies environmental disasters, said the ruptured chemical tank would have acted similar to a soda can with a hole punched in it.

“I find it hard to believe you can heat up a tank with a [chemical] like methyl methacrylate, see that it clearly cracked under pressure and think that nothing came out it,” Whelton said. “That defies logic.”

It’s possible, Whelton said, that spraying copious amounts of water on the tank had effectively suppressed much of the toxic vapors and the airborne risk.

Fire officials had previously said that the tank of MMA was experiencing thermal runaway, a chain reaction resulting in an uncontrollable spike in temperatures. They said the situation was likely to end in an explosion or chemical spill.

Whelton said an explosion is still possible.

To guard in the event of a spill, authorities set up sandbag barriers to block the chemical from storm drains that lead to the ocean.

The Orange County Fire Authority said it was also testing water that had been hosed onto the tank to ensure it didn’t contain elevated levels of contaminants.

Jane Williams, executive director of California Communities Against Toxics, warned that MMA is just one of the chemicals being stored on the site. She fears there is a danger from other chemicals.

The company in 2024 reported that, in addition to MMA, it had released thousands of pounds of flammable chemicals, including methyl ethyl ketone and methanol n-butyl alcohol, according to records from the South Coast Air Quality Management District.

GKN Aerospace had previously been cited for failing to disclose flammable chemicals at other facilities.

In 2007, the U.S. EPA alleged the company stored about 8,000 pounds of hydrofluoric acid and 34,000 pounds of nitric acid at a Kent, Wash., facility — but neglected to report these stockpiles to the appropriate government agencies.

A year earlier, the company settled with the EPA over allegations that it had improperly stored ignitable hazardous waste at a facility near San Diego.

“For me, this is not about MMA,” Williams said. “You have a company with a bunch of chemicals, and it lost containment, and it’s across from residences. I do not trust this company to disclose what else is on their site. I do not trust them with first responders. I do not trust them with my health.”