Federal analysts studied pregnancy risks. Until they were told to stop

‘Our program was unique’

Since 1987, thousands of new mothers have filled out surveys every month on what their experience of pregnancy was like. Did she use illicit drugs? Were she and her baby safe from violence at home? Was she able to afford the food she and her baby needed to grow together? Did her doctor talk to her about any of those potential dangers?

The future of the CDC’s PRAMS program is in jeopardy, as analysts have been on leave since April.

The surveys are collected by the Pregnancy Risk Assessment Monitoring System (PRAMS) team, an office at the Center for Disease Control.

The PRAMS team coordinates data collection from all over the country, cleans up the data, and then makes it available for outside researchers. A state public health department might turn to PRAMS to guide health care spending. An independent researcher might rely on PRAMS to assess whether treatment of hypertension is improving.

For Dr. Isaac Michael, it was pro-life work.

Until April 1, when he and the rest of the PRAMS team were placed on administrative leave, pending a final firing in June. While the Department of Health and Human Services undergoes a Trump-administration restructuring, the future of his work is unclear.

Michael, like many Americans, has mixed feelings on abortion.

“I’m somewhat pro-life,” he told me. He’s uncomfortable with abortion and opposes it personally in most cases, but says he loves people in his own life who have chosen to undergo abortions in hard circumstances.

Doing statistics at PRAMS seemed to him the least controversial kind of pro-life, pro-woman, pro-baby work he could imagine.

Michael joined PRAMS in June 2023 as the data weighting lead.

But when he and more than 10 colleagues were told in April to stop their work, he started emailing every pro-life leader he could find, asking for help preserving his program.

Michael said he hadn’t had the chance to make the case for PRAMS to higher-level federal administrators. No one spoke to him about his work, and, as far as he knew, no one else on the PRAMS team got the chance to explain their work before they were shelved.

As the presidential administration talked about wanting to raise the birth rate, PRAMS’s longitudinal data on women’s fertility hopes and contraceptive practices would seem relevant, Michael said.

And in the wake of Dobbs, with increased scrutiny on maternal mortality and the risk pregnant women were asked to bear, PRAMS offered the chance to identify higher risk women and target support.

There are questions that PRAMS doesn’t ask, which a pro-life administration might want to add to the next iteration of the survey.

The survey is sent to moms post-partum—it doesn’t ask whether they sought or were pressured to seek an abortion during their pregnancy. Questions along those lines would seem to complement to the existing questions about domestic violence or other dangers at home.

But with the whole team on administrative leave pending layoffs, Michael sees little hope of continuing the survey at all.

“Our program was unique,” he told me, “No one else was doing this. We were the only ones with this massive infrastructure and the relationships built up with the state departments for years to be able to do this.”

If a state governor decided to try to stand up an alternate survey, they’d need to start over from scratch. And every day Michael and his colleagues are on leave, the gap in the data grows.

With the office frozen, no data is being collected on births in the last three months of 2024. For mothers of babies born today, the data of their pregnancy experiences will disappear, unasked for, unrecorded.

Credit: Unsplash.

PRAMS is the quiet sort of lifesaving work, its defenders say.

It is not possible to point to specific babies or mothers who count as “saves” in the way that sidewalk counselors and PEPFAR clinic workers can.

But PRAMS works at the margins.

In Georgia, PRAMS informed the state’s 2025 allocation of its $17 million Title V Maternal and Child Health block grant. The persistent data collection operates like a kind of COMPSTAT for pregnancy and neonatal health, allowing the state to shift resources to address problem spots.

In New Jersey, PRAMS data guided a public health intervention for black mothers, who had the lowest rates in the state of getting post-partum checkups or successfully initiating breastfeeding.

The Department of Heath launched a five year plan to train doulas and hire community health workers to support vulnerable mothers through delivery and beyond. They wanted to connect women to other mothers in their community, and reduce the isolation that put them and their babies at risk.

When you do this kind of work, the saves are stochastic—they show up in the averages, but you can’t point to the specific individuals. The work has to be done in trust that connecting more women with doctors, lowering smoking rates, increasing breastfeeding support, etc. means that somewhere out there are mothers and babies snuggled together, who might have otherwise lost each other.

It takes a statistician to feel romantic about the numbered-but-nameless children who will grow up not knowing they were counterfactually in danger.

It's clear the numbers are deeply moving to Michael. When I asked him why he chose to work at PRAMS, when there will always be demand for data weighting, in public health and elsewhere, he said the program was his dream job.

Michael had gotten a PhD in theoretical mathematics, and accepted a post-doc at Louisiana State. As a side project to his main research, he got a master’s in Applied Statistics, and it fired him up to leave behind theory for practice.

“I wanted to do something important,” he told me. “I wanted to help people and I wanted to see the results.”

He’s all for triaging government spending, he explained.

For him, it makes sense to go beyond eliminating obvious waste, fraud, and abuse (where it exists) and take a critical look at state and federal spending, to prioritize higher impact programs over underperforming ones. He wanted to work at PRAMS, to uncover the opportunities to do good, and to do it better.

A baby requires a family to embrace uncertainty and risk. Michael’s job was to lower the danger that came as a twin to their hope. He wants to go back to work.