New research finds evictions during pregnancy are tied to adverse birth outcomes

Babies born to people evicted during pregnancy are more likely to have lower birth weights and be born earlier or prematurely than those whose parents were evicted at other times, according to new research.

“It’s sort of this intergenerational transmission of disadvantage,” said Gracie Himmelstein, a Princeton researcher and author of the study, published in JAMA Pediatrics. Birth outcomes can serve as predictive measures not just for infant mortality, but for a person’s entire life, according to Himmelstein.

The study, which was co-authored by sociologist and “Evicted” author Matthew Desmond, comes just days after a federal judge in Texas ruled the government’s moratorium on evictions during the pandemic to be unconstitutional. There was no injunction, meaning the moratorium is still in place, but sociologists are more worried than ever about a massive wave of evictions once it runs out.

“Everybody is on the edge of eviction,” said Morgan Hoke, who researches the health effects of eviction but was not involved with this study. “Unless we do something about that, it will happen.”

Researchers used an algorithm to cross-reference birth records and eviction records from the state of Georgia between the years of 2000 and 2016 to investigate the association. The main analysis included eviction actions, when one is taken to court for a potential removal from the home, instead of eviction judgements, when the order is officially made. They compared the birth outcomes of infants whose mothers were evicted while pregnant with those who had been evicted outside of pregnancy.

The association was strongest when evictions happened in the second and third trimesters of pregnancy. Overall, infants weighed almost 1 ounce less, with 0.88% higher chance of being categorized as low birth weight; they were 1.14% more likely to be born prematurely and had gestational ages lowered by 15 hours. But in the second and third trimester, infants weighed roughly 1.2 ounces less, and were more likely to be categorized as low birth weight. Infants exposed during the second and third trimester also had gestational ages lowered by 20 and 24 hours respectively.

“It’s the same populations that are getting evicted that are at highest risk of poor birth outcomes to begin with,” said Himmelstein, pointing particularly to low-income women of color, who are most frequently evicted. Those compounding risks can make it hard to study eviction, which is why the team wanted to use a control group that experienced the same factors that might have otherwise accounted for any difference in birth outcomes.

“I’m super excited about this study because it’s really hard to link evictions to biological outcomes, to actual health outcomes,” said Hoke.

The researchers also compared outcomes of different pregnancies from the same person, where one sibling was exposed to eviction in the womb while the other hadn’t. The results proved consistent.

In an accompanying perspective piece, outside experts note two major caveats on the research.  First, that the decision to focus on eviction actions instead of eviction decisions may mean the research acts more as a study of “extreme prenatal social stress” than of eviction. They also questioned whether the results, while statistically significant, would translate to clinically meaningful health outcomes, given that the differences in birth weight and gestational age between the two groups “appear modest.”

“However, by virtue of the high prevalence of the exposure and the potential long-term consequences of the outcomes, the findings may have relevant implications at a population level,” they write.

Hoke said some of that limitation comes from how the study was designed. “Any time that you’re studying housing insecure populations and you’re relying on addresses to make claims about those populations, that can be a little bit difficult,” said Hoke. But typically, Hoke said, that approach makes it harder to tease out insights, “and so the fact that we find something is even more important.”

Hoke believes that the best chance to accurately capture the biological effects of eviction would be a study that follows people through the process of eviction in real time. Himmelstein also sees a trend in eviction literature toward “natural experiments,” and a greater need for understanding the effects of eviction at the individual level.

“It’s not just that you’re evicted and then you have some short-term health problems,” said Himmelstein. “You’re evicted, you have a low birth weight baby, and now that has consequences for your family for generations.”

With the pandemic putting millions more families at risk of eviction, many believe the evidence is more than adequate to start taking steps that could help protect families.

“It feels quite urgent to to start thinking about policies and ways that this can be addressed,” Himmelstein said.

Correction: A previous version of this story misstated Desmond’s role as lead author of the research. He is a co-author.