Dietary and Lifestyle Correlates of Gut Microbiota Diversity and Composition in Healthy Adults: A Cross-Sectional Pilot Study Informing Longitudinal Research.
In 20 healthy adults, no association between Mediterranean diet adherence, ultra-processed food frequency, physical activity, or sleep and gut microbiota outcomes survived false-discovery-rate correction, but effect estimates were not uniformly negligible and imprecise, so the pilot does not establish absence of relationships and instead provides planning benchmarks of 275-465 analyzable participants for adequately powered studies.
Key Findings
Results
No diet or lifestyle exposure was associated with any microbiome outcome after false-discovery-rate correction.
Twenty healthy adults (10 females, 10 males) provided a fresh stool sample and completed the questionnaire on the same day.
Exposures were Mediterranean diet adherence, ultra-processed food (UPF) consumption frequency, physical activity, and sleep characteristics.
Outcomes included OTU-level Shannon diversity (primary), Pielou's evenness, genus-level Bray-Curtis and Aitchison distances, CLR coordinates of eight prespecified genera, and the log Prevotella/Bacteroides ratio.
Analyses used age- and sex-adjusted linear regression and PERMANOVA with FDR control; complete exposure and microbiome data were available for all participants.
Results
Observed effect sizes for alpha diversity were not negligible despite the lack of statistical support.
The maximum partial R2 across the alpha-diversity models was 0.140.
Estimates were described as imprecise and participant-sensitive.
Influence diagnostics and robustness analyses did not alter the overall interpretation.
Results
The largest single effect estimate was for the association between UPF consumption and Akkermansia.
The maximum partial R2 for the UPF-Akkermansia association was 0.210.
This association was not supported after FDR correction.
Results
The Prevotella/Bacteroides log-ratio showed modest, non-negligible effect estimates.
The maximum partial R2 for the natural-log Prevotella/Bacteroides ratio was 0.108.
No association was supported after FDR correction.
Results
Overall community composition explained little variance for any exposure.
The maximum marginal R2 for overall community composition (PERMANOVA) was 0.0649.
Beta diversity was assessed with genus-level Bray-Curtis and Aitchison distances.
Results
Sample-size scenarios indicate that a much larger cross-sectional study would be needed to detect modest associations.
Scenarios assumed partial R2 values of 0.03-0.05 in age- and sex-adjusted cross-sectional linear regression.
These yielded 275-465 analyzable participants.
Assuming 10-20% unusable observations, enrollment of 306-582 participants would be needed.
Calculations were restricted to cross-sectional regression models, not longitudinal designs.
Conclusions
The pilot does not establish the absence of diet-microbiota or lifestyle-microbiota relationships and supports a framework for hypothesis prioritization and longitudinal design.
Authors recommend repeated, temporally aligned exposure and microbiome measurements in future studies.
The study is a small, single-timepoint cross-sectional pilot, with exposures and stool collected on the same day.
What This Means
This small pilot study looked at whether four everyday factors, Mediterranean diet adherence, ultra-processed food intake, physical activity, and sleep, were linked to the makeup and diversity of gut bacteria in 20 healthy adults. After statistical correction for testing many comparisons, none of the links were supported.
However, the study was very small, and some estimated effects were not trivial. For example, ultra-processed food intake appeared to explain a fair share of variation in one gut bacterium, Akkermansia, and some diversity measures showed similar moderate estimates. These estimates were imprecise and sensitive to individual participants, so they could easily be chance findings. The authors stress that the results do not show these factors are unrelated to gut bacteria.
The main practical value is for planning future research. The authors calculated that a cross-sectional study aiming to detect modest associations would need roughly 275-465 analyzable participants, or about 306-582 enrolled. They also suggest future studies should measure diet, lifestyle, and gut bacteria repeatedly over time rather than once. This research suggests that conclusions about diet and gut bacteria need larger, better-timed studies.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Guerra M, Brandolini M, Dionisi L, De Pascali A, Ingletto L, Colosimo C, et al.. (2026). Dietary and Lifestyle Correlates of Gut Microbiota Diversity and Composition in Healthy Adults: A Cross-Sectional Pilot Study Informing Longitudinal Research.. Nutrients. https://doi.org/10.3390/nu18203325