What wastewater revealed
A study published in Nature Communications on 9 October 2026 suggests that sewage can reveal patterns in a community’s antibiotic use, but the timing of those signals depends on which bacteria carry resistance genes.
Researchers combined five years of outpatient antibiotic prescription records with a year of twice-weekly genetic sampling at one wastewater treatment plant. Their analysis found that some resistance indicators tracked antibiotic use closely, while others followed with a delay of one to three months.
The findings matter because antibiotic resistance is a growing public-health concern. Monitoring wastewater may offer a complementary way to understand resistance trends beyond information gathered from clinical testing.
However, the researchers caution that sewage is a complex microbial environment. A resistance gene detected in wastewater does not automatically reveal which person, infection or prescription produced it.
Why sewage could be a public-health resource
Wastewater collects material from homes, workplaces and other buildings. This makes it possible to sample biological traces from many people without asking each individual to provide a clinical specimen.
During the COVID-19 pandemic, wastewater surveillance gained attention as a way to follow changes in viral circulation. Extending that approach to antibiotic resistance introduces additional scientific challenges.
Viruses and resistance genes behave differently in sewer systems. Resistance genes can occur in many bacterial hosts, and some of those organisms can survive, grow or decline after entering wastewater.
The quantity of a resistance gene in a sewage sample therefore reflects more than human infections. It can also be influenced by bacterial ecology, temperature, wastewater chemistry and the physical conditions inside pipes.
The study asked whether these complicated signals could nevertheless reveal meaningful relationships with antibiotic prescribing in the surrounding community.
The research design
Connor Lee Brown and colleagues brought together two types of evidence that are often studied separately: clinical antibiotic-use data and measurements of microbial genes in sewage.
The prescription dataset covered five years of outpatient antibiotic use. This offered a longer record of community prescribing patterns than the wastewater sampling period alone.
The researchers also conducted a metagenomic survey of sewage twice a week for approximately one year at a single wastewater treatment plant.
Metagenomics involves sequencing genetic material from a mixed sample. Instead of growing one bacterial species in a laboratory, researchers can examine genetic signatures from many organisms at once.
The team looked for antibiotic resistance genes and considered the bacterial groups with which those genes were associated.
They then used statistical models to investigate how changes in resistance indicators corresponded to changes in outpatient antibiotic use.
The models accounted for sewage characteristics and seasonality when evaluating particular associations. These adjustments matter because seasonal and environmental conditions can affect the bacterial community independently of prescribing.
The study was observational. The researchers analysed naturally occurring patterns rather than randomly assigning communities different antibiotic prescriptions.
A difference between bacterial families
The most notable finding was that resistance indicators behaved differently depending on their bacterial hosts.
Genes associated with Enterobacteriaceae tracked antibiotic use relatively closely. This family includes bacteria commonly associated with humans and other animals.
By contrast, resistance indicators linked to Pseudomonadaceae showed correlations that lagged antibiotic use by approximately one to three months.
This is not a small technical distinction. A monitoring programme that expects every resistance signal to change immediately after prescribing changes could misread delayed signals.
The authors interpreted the different patterns as evidence that the ecology of bacterial hosts affects what wastewater measurements mean.
Some bacteria may be closely connected to human-derived material, while others can be strongly influenced by processes within wastewater systems.
The result supports a more targeted approach in which analysts examine not just whether a resistance gene is present, but which organisms appear to carry it.
What a one-to-three-month lag means
A lagged correlation means that changes in one measured series tend to be associated with changes in another series at a later time.
In this study, the lag for Pseudomonadaceae-associated indicators was reported as one to three months. It does not mean that a person necessarily develops resistance one to three months after taking an antibiotic.
Nor does it show that a particular course of treatment caused a specific resistant infection.
The delay describes the timing of statistical associations between aggregated prescribing records and wastewater signals.
Possible explanations include the persistence or growth of particular bacterial groups and the way resistance genes move through a complex sewage environment.
The research did not establish a single mechanism responsible for every observed delay.
For public-health interpretation, the important point is that a delayed signal should not automatically be dismissed as unrelated to current prescribing.
A closer look at a commonly used antibiotic combination
The investigators also examined associations involving sulfamethoxazole and trimethoprim, two antibiotics frequently used together.
Using nested statistical models, they linked changes in Pseudomonas metagenome-assembled genomes with lagged use of this antibiotic combination.
A metagenome-assembled genome is a computational reconstruction of genetic material associated with a microbial organism or population from mixed sequencing data.
This technique can provide more information about potential bacterial hosts than a simple count of resistance genes alone.
The reported association remained relevant after the researchers accounted for sewage characteristics and seasonal variation.
That result strengthens the case that host-specific analysis can clarify wastewater resistance signals.
It does not demonstrate that sulfamethoxazole-trimethoprim directly caused every observed change in Pseudomonas populations.
Why the quantitative details matter
The study’s five-year prescribing record and approximately one-year, twice-weekly wastewater survey offer two complementary time scales.
The longer clinical record provides context for changes in antibiotic use, while the repeated sewage sampling captures fluctuations in microbial communities.
Sampling twice a week can detect variation that might be obscured by monthly or occasional collections.
However, frequent sampling at one treatment plant is not equivalent to a national surveillance network.
The one-to-three-month delay is the study’s clearest reported timing result. The published abstract does not establish a universal percentage change in resistance for a given percentage change in prescribing.
Consequently, it would be misleading to claim that reducing antibiotic use by a fixed amount would produce an equally fixed decline in sewage resistance genes.
The findings instead demonstrate that relationships can be detected and that their timing depends on the bacterial group being examined.
What this could mean for hospitals and health departments
Clinical laboratories remain essential for identifying resistant infections and selecting appropriate treatment for individual patients.
Wastewater analysis could complement that work by revealing broad community-level patterns that do not depend on every infection being diagnosed or reported.
For example, repeated samples might help analysts investigate whether resistance indicators change alongside local prescribing trends.
Understanding the bacterial hosts could improve interpretation and reduce the risk of treating every detected resistance gene as an equivalent public-health signal.
These data might eventually inform antibiotic stewardship programmes, which seek to ensure that antibiotics are prescribed only when appropriate and used effectively.
Yet this study did not test a clinical intervention or show that wastewater monitoring alone improves patient outcomes.
Any operational use would require validation against clinical surveillance and careful attention to local sewer conditions.
Relevance beyond the study location
Antibiotic resistance affects countries with very different health systems, sanitation infrastructure and access to diagnostic testing.
That creates interest in environmental surveillance methods that could supplement traditional clinical reporting.
However, the reliability of sewage monitoring may vary according to wastewater coverage, industrial inputs, population movement and the treatment plant’s catchment area.
In places where many households are not connected to central sewers, wastewater samples may represent only a portion of the community.
The study’s findings therefore should not be applied directly to every city without additional local research.
South African public-health researchers, for example, would need to evaluate local sewer infrastructure and the representativeness of sampling sites before drawing population-wide conclusions.
The broader opportunity is to integrate environmental, clinical and prescribing information rather than rely on any single source.
Important limitations
The wastewater component was conducted at one treatment plant. Local bacterial ecology may differ in other cities, climates and sewer networks.
The metagenomic survey lasted approximately one year, which limits conclusions about longer-term patterns across multiple seasons and years.
Prescription records indicate antibiotic use in the outpatient setting but may not capture every exposure, including some hospital, agricultural or unrecorded uses.
Metagenomic methods detect genetic material. The presence of a resistance gene does not necessarily prove that a living bacterium can cause disease or that the gene is actively expressed.
Statistical association cannot, by itself, establish causation. Changes in antibiotic use and microbial abundance may also reflect other influences.
Although the researchers adjusted for selected sewage characteristics and seasonality, unmeasured factors could still affect the findings.
Future studies across multiple treatment plants and regions would help determine whether the same host-specific delays are reproducible.
Comparisons with clinical resistance measurements would also clarify which sewage indicators best reflect risks to human health.
The main takeaway
The study provides evidence that wastewater can capture patterns related to community antibiotic prescribing, provided researchers account for the bacteria carrying resistance genes.
Enterobacteriaceae-linked indicators followed antibiotic use more closely, while Pseudomonadaceae-linked signals were associated with delays of one to three months.
The contrast shows why a simple count of resistance genes may be insufficient for public-health interpretation.
Rather than treating sewage as a direct readout of resistant infections, the researchers suggest that its microbial ecology must be considered.
With additional validation, host-aware wastewater surveillance could become a useful complement to existing antibiotic resistance monitoring.
Source Information
Study: Resistance indicators in sewage microbiota mirror clinical antibiotic use with temporal lag linked to host
Authors: C. L. Brown, MA. Rumi, L. McDaniel and colleagues.
Journal: Nature Communications
Published: 9 October 2026
Methods: Five years of outpatient antibiotic prescription data, one year of twice-weekly wastewater metagenomic sampling at a single treatment plant, and nested statistical modelling.
