A study of more than 2,500 people has found that gut microbiome medication history leaves detectable traces long after a prescription ends, and that the effect is not confined to antibiotics. Antidepressants, beta-blockers, acid-reducing drugs, and benzodiazepines were all associated with lasting microbial changes, with some differences still measurable years after the last dose.
The research was led by investigators at the University of Tartu Institute of Genomics and published in mSystems Journal on 22 October 2025. Participants were drawn from the Estonian Biobank, as part of the Estonian Microbiome cohort. Stool samples were matched against prescription records to look for associations between drug exposure and the composition of gut bacteria.
What the researchers found about gut microbiome medication history
The gut microbiome (the community of bacteria and other microorganisms in the digestive tract) is already known to be sensitive to antibiotics. What the study adds is evidence that a far wider range of drugs may shape that community, and that the shaping can outlast the treatment itself.
Proton pump inhibitors, which reduce stomach acid and are commonly prescribed for acid reflux, were among the drugs associated with lasting microbial differences. So were selective serotonin reuptake inhibitors, a widely used class of antidepressant, and beta-blockers, typically used for conditions such as high blood pressure. Benzodiazepines, prescribed for anxiety and related disorders, showed associations with the microbiome that were, the researchers say, comparable in magnitude to those seen with broad-spectrum antibiotics.
That last finding is worth pausing on. Broad-spectrum antibiotics are designed to act against many different types of bacteria and are already known to produce substantial disruption to gut microbial communities. The suggestion that anxiety medications may produce effects of a similar scale to a gut microbiome has not, to this point, been a central assumption in microbiome research.
‘Most microbiome studies only consider current medications, but our results show that past drug use can be just as important as it is a surprisingly strong factor in explaining individual microbiome differences,’ said Dr Oliver Aasmets, lead author.
Individual drugs differ, not just drug classes
The study also found that drugs within the same class did not necessarily affect the microbiome in the same way. Diazepam and alprazolam, for instance, differed in how strongly they appeared to disrupt gut microbes despite both being benzodiazepines. That distinction matters because microbiome studies have often grouped drugs by class rather than treating them individually. The results suggest that approach may obscure real differences between specific compounds.
A follow-up component of the study examined stool samples from a smaller group of participants at a second time point, allowing the team to observe microbial shifts when people started or stopped certain medications. The changes were described as predictable, offering some support for the idea that the medications themselves (rather than confounding factors) are driving at least part of what is observed. Persistent effects were confirmed for proton pump inhibitors, selective serotonin reuptake inhibitors, and specific antibiotic groups including penicillins in combination and macrolides.
The research team at the Institute of Genomics, Estonian Genome Centre, University of Tartu, included Mait Metspalu, Andres Metspalu, Lili Milani, and Tõnu Esko, alongside lead author Dr Aasmets and corresponding author Professor Elin Org. The study drew on real-world medical health records, which the researchers describe as a strength of the design: the sample was not a clinical trial population but a biobank cohort, giving access to prescription histories as they actually unfolded in practice.
Ethics approval was granted by two bodies: the Ethics Committee of the University of Tartu and the Estonian Committee on Bioethics and Human Research, according to a medRxiv preprint of the research.
Why prescription history may need to enter microbiome research
‘This is a comprehensive systematic evaluation of long-term medication effects on the microbiome using real-world medical health records,’ said Professor Org. ‘We hope this encourages researchers and clinicians to factor in medication history when interpreting microbiome data.’
The practical implication is pointed. Studies that look for microbiome patterns associated with a disease may be picking up signals left by treatments taken months or years earlier. A person enrolled in a study may no longer be taking a drug, but their gut bacteria may still carry its signature. Without accounting for that history, researchers risk misattributing those microbial differences to the condition under study rather than to past prescriptions.
The study is listed under the journal reference mSystems, Volume 10, Number 10, with the DOI 10.1128/msystems.00541-25. The authors call on both researchers and clinicians to treat a patient’s prescription history as a variable in its own right when interpreting microbiome data, rather than focusing only on what a person is currently taking.
