Two new studies. Two different research teams. Two different outcomes. One increasingly difficult question: could everyday glyphosate exposure during pregnancy be affecting processes that regulators have not adequately examined?
We’ve asked a version of this question before. This time, the evidence is bigger, newer — and, as you’ll see below, more complicated than a single tidy conclusion.
Have you had questions about glyphosate exposure during pregnancy that no one could answer? We’d like to hear from you — get in touch.
The New Study: 1,450 Women, One Specific Window
Just published in Environmental Pollution, a team from NYU Langone Health has released one of the largest studies to date examining glyphosate exposure and preterm birth in humans.
Researchers analysed data from 1,450 women, using urine samples collected at two stages of pregnancy rather than relying on a single measurement — something few studies before it have done. The association appeared only in the mid-pregnancy measurements, taken between 18 and 25 weeks: higher glyphosate concentrations at that stage were associated with a greater likelihood of spontaneous preterm birth, after accounting for several other factors that can influence preterm-birth risk. The same association did not appear in samples taken earlier in pregnancy, and no link was found with medically indicated preterm birth.
These women were recruited through New York City hospitals, rather than from an agricultural worker cohort. The study didn’t identify each participant’s exposure source — the researchers noted that everyday exposure can occur through food containing residues and through the wider environment.
If exposure is measurable in a large urban population like this one, it’s reasonable to ask what New Zealand biomonitoring might find here.
A Second Study, Looking at Hormones Instead
While that study was making headlines, a second, quieter one appeared in the Journal of Exposure Science & Environmental Epidemiology. This one didn’t look at birth timing — it looked at hormones, across 752 pregnancies in Puerto Rico’s PROTECT cohort.
Each interquartile-range increase in AMPA (glyphosate’s primary environmental breakdown product) was associated with 10.6% lower estriol; a similar increase in glyphosate itself was associated with 8.3% lower estriol. AMPA was also associated with higher T3. In later-pregnancy sensitivity analyses, AMPA was linked to higher TSH and glyphosate to higher CRH—a hormone involved in the body’s stress response and in biological signalling associated with labour.
The authors describe this as the first evidence of associations between glyphosate exposure and changes across several maternal hormonal pathways. It does not establish that these hormonal shifts caused the preterm birth pattern seen in the New York study — the authors themselves are explicit that this is an open question for future research, not a finding they tested here.
Could those hormonal changes eventually help explain why some studies report differences in gestational length and spontaneous preterm birth? We don’t yet know. But it’s now a question researchers have evidence to investigate.
What About the Study That Found No Association?
Here’s where an honest accounting matters: not every study agrees.
The Canadian MIREC study, one of the largest of its kind, analysed first-trimester urine samples from 1,765 singleton births and found no association between glyphosate or AMPA and preterm birth, spontaneous preterm birth, or gestational age at all. Its own conclusion was direct: neither compound was associated with these outcomes.
That’s not a reason to dismiss the newer findings. It may be a clue. MIREC measured exposure only in the first trimester — exactly the window where the New York study also found no association. The pattern only showed up in mid-pregnancy. Timing, in other words, may be the reason different studies have reached different conclusions, rather than any one of them being simply wrong.
That’s a more interesting story than “the evidence all agrees” — and a more honest one.
We Asked a Version of This Question Before
Long-time readers might remember our earlier look at glyphosate and preterm birth, built around a single 2021 study of roughly 250 women in Puerto Rico. At the time, we asked why New Zealand was moving to raise allowable glyphosate residue limits on food while that evidence already existed.
Now we have a cohort nearly six times the size, in a different population and using a different measurement design, reporting a broadly consistent association — alongside a hormone study raising related questions, and a large null result that sharpens rather than settles the picture. This isn’t one isolated red flag anymore. It’s an active, unresolved scientific question, while New Zealand has neither generated comparable exposure data nor publicly explained how this emerging pregnancy research is being weighed.
The Data New Zealand Doesn’t Have
Here’s what should trouble every expecting parent reading this: New Zealand does not currently offer routine clinical or public-health biomonitoring for glyphosate. A pregnant woman cannot currently obtain a glyphosate urine test as part of standard GP or maternity care. There is no national programme measuring exposure in the wider population, at any life stage.
Every study above exists because someone, somewhere, collected urine samples and measured them. Without that data, none of this research would be possible — and without equivalent data here, nobody knows whether New Zealand exposure levels look anything like New York’s, Puerto Rico’s, or Canada’s.
Why are international research teams able to measure glyphosate exposure across pregnancy, while New Zealand has no comparable population data of its own?
What Would Actually Close the Gap
The honest answer isn’t a single test any expecting parent can book tomorrow. It’s population-level biomonitoring — repeated urine sampling across pregnancy, in a properly designed study, so regulators can find out whether exposure is occurring here, at what levels, at which stage of pregnancy, and from which likely sources.
That’s the same kind of data that made every study above possible in the first place. Until New Zealand begins collecting comparable information, any discussion about glyphosate exposure during pregnancy is taking place without knowing what exposure levels actually look like in New Zealand.
New Zealand is currently collecting none of the data needed to ask, let alone answer, the question these international studies keep raising.
Further Reading
If these findings raised questions for you, here’s what happens when you follow the thread further. This isn’t settled science, and treating it as though it were would do the topic a disservice — the value here is in an unresolved question worth taking seriously, not a unanimous verdict.
Exposure to Weed Killer During Pregnancy May Increase Risk of Preterm Birth — NYU Langone Health The original release on the study discussed above, including detail on the cohort and why the association appeared only in mid-pregnancy measurements.
Glyphosate Exposure and Hormonal Disruption in Pregnancy: Evidence from a Birth Cohort in Puerto Rico — Journal of Exposure Science & Environmental Epidemiology (2026) The full peer-reviewed hormone study referenced above, detailing associations between glyphosate/AMPA and estriol, thyroid hormones, and CRH across 752 pregnancies.
Gestational Urinary Concentrations of Glyphosate and AMPA in Relation to Preterm Birth: The MIREC Study — Journal of Exposure Science & Environmental Epidemiology (2024) This large Canadian study found no association between first-trimester urinary glyphosate or AMPA concentrations and preterm birth. Its findings differ from those of several US and Puerto Rican studies, raising a real question about whether exposure timing helps explain the difference.
Count Down: How Our Modern World Is Threatening Sperm Counts, Altering Male and Female Reproductive Development, and Imperiling the Future of the Human Race*
Shanna H. Swan
A wider-lens book on how everyday environmental chemical exposure, glyphosate included, may be reshaping human reproduction across generations — useful context for readers wanting to understand why researchers keep circling back to pregnancy as a critical exposure window.
*For your convenience, we provide links to Amazon.com. If you choose to purchase through these links, we may receive a small commission — at no additional cost to you. Your support helps us continue our work.
Three research teams. Three countries. Three studies asking the same broad question from different angles. New Zealand still has no comparable data of its own.
Image Source & Attribution
We’re grateful to the talented photographers and designers whose work enhances our content. The feature image on this page is by Gustavo Fring on pexels.com.
