For some of us, stress is not an occasional condition, but a way of life. When friends tell us to “just relax,” they might as well be telling us to be taller or shorter or somebody else. And when we become pregnant? Nothing changes. We are fiercely anxious: fat, under-slept, and cranky, awaiting every blood test and ultrasound with the avidity of lionesses, and dissecting the results with the stamina of coding geeks. When friends tell us to “just relax,” we lash out. My first-world stress is not going to hurt my baby, we insist. At this particular moment, when I cannot drink, smoke, have coffee, eat sushi, soft cheeses, or deli meats, when my life is about to change forever, let me at least have my stress, a relic of my former self.

How irritating, then, when science seems to agree with those friends: High levels of stress can hurt your baby. Maternal stress has long been known to be associated with pre-term delivery and low birth weight, and can lead to psychological problems in children later in life (an increased risk of schizophrenia, behavioral problems, and possibly low IQ). And a study published in March in the New England Journal of Medicine concludes what has long been suspected: Autism, too, has its roots in abnormal brain development in utero — abnormal development that could possibly be caused by stress.

But wait: just what kind of stress are we talking about here? Princeton molecular biologist Sam Wang argued recently that stress is “a highly underappreciated prenatal risk” for autism. But thankfully he went into a lot more detail in analyzing a variety of different factors, assigning each a “risk ratio” reflecting the increased risk represented by a particular environmental factor (a ratio of two means double the risk, of four means quadruple). Emigrating to a new country registers a ratio of 2.3, getting caught in a hurricane zone during the second half of a pregnancy registers about 3, as does maternal post-traumatic stress during pregnancy. (For comparison, premature birth has a ratio of about 5.3 and an injury to the baby’s cerebellum at birth close to 40.)

These are not, you might notice, garden-variety “stresses.” This is more like “trauma.” And while autism has long been considered a yuppie disease, rare among diseases for disproportionately afflicting the affluent and educated, that conventional wisdom is under reconsideration. A 2012 study in the Journal of the American Academy of Child and Adolescent Psychiatry measured rates of autism against socioeconomic status. To test whether the high rates of autism among the affluent in the U.S. were a results of selection bias — richer people see the doctor more and get their kids screened more frequently and thus receive more ASD diagnoses — researchers looked at a group of half a million kids in Stockholm, where free universal health care removes the imbalance between richer and poorer in the realm of diagnostic testing. They found higher rates of autism among the poorer families and among families whose parents were manual laborers. Which makes sense: Isn’t it, in fact, much more stressful to be poor than rich?

In 2012, after looking at large groups of pregnant women and their babies in England and Sweden, psychiatrists at the University of Bristol found no clear association between maternal stress and autism risk. They looked at Swedish women who, while pregnant, experienced a death in the family, had a serious accident, or received a diagnosis of a life-threatening disease. They looked at English women who, while pregnant, experienced one of 40 different stressful life events — everything from “you argued with your partner” to “you were in trouble with the law.” “We found no evidence of any relationships between prenatal life events and offspring autism spectrum disorders,” the authors wrote.  “The evidence supporting the relationship between psychological stress in pregnancy and ASD in human studies is limited and inconsistent.” The English questionnaire did not ask expectant mothers whether they had stress about their stress, but based on the evidence, the effects of such first-world overthinking is probably no more harmful to a developing fetus than the occasional mouthful of unapproved cheese.