Crude beta-diversity differences between bovine milk consumers and plant-based beverage consumers were small and were not reproduced in propensity score-matched sensitivity analyses, suggesting that observed associations between milk consumption patterns and gut microbiota composition may be attributable to demographic and body composition confounding.
Key Findings
Results
A small but statistically significant beta-diversity difference between bovine milk consumers and plant-based beverage consumers was detected in crude (unmatched) analyses.
Beta diversity was assessed using Bray-Curtis distances and permutational multivariate analysis of variance (PERMANOVA).
The crude beta-diversity difference had an effect size of R2 = 0.0119 and p = 0.014.
The effect size (R2 = 0.0119) indicates the difference explained approximately 1.19% of variance in microbiota composition.
Marked group imbalances were present in crude analyses including differences in age, body mass index, fat mass, and impedance-derived parameters.
Results
The crude beta-diversity difference between bovine milk consumers and plant-based beverage consumers was not reproduced after propensity score matching.
Pairwise propensity score matching was applied as a sensitivity analysis to control for measured demographic, anthropometric, and body composition imbalances.
In the matched subset, pair-stratified PERMANOVA yielded R2 = 0.0214 and p = 0.506, indicating no statistically significant difference.
No matched comparison showed significant alpha-diversity differences.
The plant-based beverage consumer group had a limited sample size (n = 27), which constrained the matched subset size.
Results
No individual microbial taxa showed significant differences between consumer groups after correction for multiple comparisons.
Taxon-level analyses were corrected using the false discovery rate (FDR).
No taxa remained statistically significant after FDR correction in any comparison.
This applied across all three group comparisons: non-milk consumers, bovine milk consumers, and plant-based beverage consumers.
Results
No significant alpha-diversity differences were found between any consumer groups in matched analyses.
Alpha diversity was assessed using standard diversity indices.
In matched sensitivity analyses, no comparison showed significant alpha-diversity differences.
This finding held across all pairwise matched comparisons among the three consumer groups.
Methods
The study sample consisted of 265 participants divided into three consumer groups with notable baseline imbalances in body composition and demographic characteristics.
The sample comprised 99 non-milk consumers, 139 bovine milk consumers, and 27 plant-based beverage consumers.
Clinical, anthropometric, body composition, and microbiota data were integrated.
Groups differed markedly in age, BMI, fat mass, and impedance-derived parameters prior to matching.
The cross-sectional design used questionnaire-derived habitual milk consumption patterns as the exposure measure.
Conclusions
Dispersion differences rather than true centroid differences may partly explain the crude beta-diversity finding between bovine milk and plant-based beverage consumers.
The authors identified 'dispersion differences in crude analyses' as a specific limitation when interpreting the crude PERMANOVA result.
PERMANOVA can be sensitive to differences in within-group dispersion, which can produce significant results even without true centroid-level differences between groups.
This concern was listed alongside categorical exposure assessment, residual imbalance, limited plant-based beverage consumer sample size, and potential unmeasured confounding as reasons for cautious interpretation.
What This Means
This research examined whether the type of milk or milk alternative people habitually drink is associated with differences in their gut microbiome — the community of bacteria and other microorganisms living in the digestive tract. The study compared 265 healthy adults who were grouped as non-milk drinkers, cow's milk drinkers, or plant-based beverage drinkers (such as oat, almond, or soy milk). When the researchers first compared the groups without accounting for other differences between them, they found a small but statistically significant difference in the overall microbiome composition between cow's milk drinkers and plant-based beverage drinkers. However, these groups also differed considerably in age, body weight, and body fat levels — factors that are themselves known to influence the gut microbiome.
To address this problem, the researchers used a statistical technique called propensity score matching, which creates more comparable subgroups by pairing individuals who are similar in those background characteristics. When this more rigorous comparison was performed, the apparent difference in microbiome composition between cow's milk and plant-based beverage drinkers disappeared entirely and was far from statistically significant. Similarly, no differences in microbial diversity measures or in the abundance of any specific bacterial group survived statistical correction for multiple testing in any comparison.
This research suggests that earlier or simpler observational findings linking dairy or plant-based beverage consumption to gut microbiome differences may reflect confounding by age, body composition, or lifestyle factors rather than a direct effect of the beverages themselves. The study highlights the importance of carefully accounting for background differences between consumer groups before concluding that diet-microbiome associations are real. Limitations include the relatively small number of plant-based beverage consumers (only 27 people), the cross-sectional design which cannot establish cause and effect, and the reliance on questionnaires to assess dietary habits.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
La Placa G, Gualtieri P, Peluso D, Marcoccia D, Merra G, De Lorenzo A, et al.. (2026). Milk Consumption Patterns and Gut Microbiota Composition in Healthy Adults: Insights from a Propensity Score-Matched Cross-Sectional Study.. Nutrients. https://doi.org/10.3390/nu18193272