Field notes
As of Oct 2026 · 4 studies · newest 2026 Antibiotics
New: Baldanzi 2026, 14,979 people
01
The short version
Here the evidence goes furthest: one study followed the same people through a course of antibiotics, and a large study saw differences years after people took them. Which of the differences matter for health isn't known.
02
What studies have seen
Each line is one finding. The word says how many separate groups of people showed it.Fewer kinds of gut microbes in people who took antibiotics.
2 of 2 cohorts · 17,488 people · Estonia, Sweden Baldanzi 2026Aasmets 2025In 12 healthy men given a 4-day course of three antibiotics, Bifidobacterium species fell and Enterococcus faecalis bloomed. Their gut microbes were back near where they started within 1.5 months, but 9 common species were still undetectable in most of them after 180 days.
followed the same people · 12 people · Denmark Palleja 2018The difference in kinds of gut microbes was largest for antibiotic courses in the year before, and still seen for courses 4 to 8 years before.
3 cohorts pooled · 14,979 people · Sweden Baldanzi 2026Clindamycin, fluoroquinolones and flucloxacillin taken 4 to 8 years before were associated with differences in 10–15% of the gut species studied; penicillin V with only a few.
3 cohorts pooled · 14,979 people · Sweden Baldanzi 2026Enterocloster bolteae and Flavonifractor plautii are higher in people who took clindamycin, fluoroquinolones or flucloxacillin.
3 cohorts pooled · 14,979 people · Sweden Baldanzi 2026
Bifidobacterium is lower in people taking antibiotics, in 1,883 people in the Netherlands.
3 cohorts pooled · 1,883 people · Netherlands Vich Vila 2020Of 41 kinds of medicine, antibiotics were among the four most tied to gut microbes, with proton pump inhibitors, metformin and laxatives, once the other medicines people took were accounted for.
3 cohorts pooled · 1,883 people · Netherlands Vich Vila 2020People take these medicines for reasons that can also go with different gut microbes.
Antibiotics can loosen or firm stools, and stool form and transit go with differences in gut microbes on their own.
A difference is an association: it doesn't mean antibiotics change these microbes, or that they affect health. Where a study followed the same people over time, its line says what changed.
03
Ideas being tested
What researchers think might explain it, and how far each idea has been tested.Idea · The authors suggest
Does this matter for health? In one of the three cohorts, the authors found that many of the species tied to these antibiotics also go with BMI, waist–hip ratio, triglycerides and C-reactive protein. It hasn't been tested.
How far it’s been tested ProposedIn a dishIn animalsSeen in peopleTested in a trial Baldanzi 202604
What nobody knows yet
Open questions, as the studies leave them.- Q1How much comes from the antibiotic and how much from the infection it was prescribed for: the prescription register doesn't record why it was given.
- Q2Whether it looks the same outside Sweden, where antibiotic use is notably restrictive, the authors note.
- Q3Whether the differences seen years later are lasting changes: the large study measured each person once.
05
Who's been studied
WhereDenmark · Estonia · Netherlands · Sweden
How many4 studies · 19,383 people
HowMostly measured once; some followed over time
06
The microbes
07
The studies
Each person measured once · Shotgun sequencing
Per-microbe tables open · CC BY 4.0 Competing interests Some authors own shares in companies that provide services to the health sector, sit on advisory boards for or received fees from pharmaceutical companies, or have funding from them, all stated as unrelated to this study.
Each person measured once · Shotgun sequencing
Per-microbe tables open · CC BY 4.0 Competing interests None declared.
Each person measured once · Shotgun sequencing
Per-microbe tables open · CC BY 4.0 Competing interests None declared.
Followed the same people · Shotgun sequencing
Cited in our own words · Publisher's terms
Registered at ClinicalTrials.gov · NCT01633762 ↗ Competing interests None declared. Authors list companies among their affiliations: Clinical-Microbiomics A/S (2 authors), BGI-Shenzhen (5 authors), iCarbonX (1 author).