Technical Note

Children with higher live-microbe intake had lower odds of hypertension — but look at what that metric actually counts

A cross-sectional analysis of NHANES 2009–2023 found that among 9,342 children and adolescents aged 8–17, those in the high dietary live-microbe group had an odds ratio for hypertension of 0.70 (95% CI 0.50–0.98). The “live microbe intake” metric is worth a careful look from a formulator's point of view — starting with what it measures.

Published 15 October 2026 · 7 min read

What the analysis did

9,342 children and adolescents aged 8–17, of whom 433 had hypertension. Logistic regression by dietary live-microbe intake group:

Main effect. High versus low intake: OR = 0.70 (0.50–0.98), P = 0.039; after further adjustment for overweight/obesity, OR = 0.71 (0.51–0.99), P = 0.046.
Joint effect. Against a reference group of low intake + overweight/obesity: high intake + overweight/obesity OR = 0.60 (0.39–0.92); low intake + not overweight OR = 0.29 (0.20–0.42); high intake + not overweight OR = 0.28 (0.16–0.49). Trend P = 0.019.

The authors' conclusion is appropriately restrained: high live-microbe consumption was associated with reduced odds of hypertension, and intake and weight status showed a joint effect.

First, how “live microbe intake” is calculated

This is the section to read first, because the name of the metric invites misreading. It comes from the classification system published by Marco et al. in the Journal of Nutrition in 2022, which assigns foods to three tiers by live microbes per gram:

Tier	Live microbes	Typical foods
Lo	< 10⁴ CFU/g	Pasteurised or cooked foods, peeled produce
Med	10⁴–10⁷ CFU/g	Fresh fruit and vegetables eaten unpeeled
Hi	> 10⁷ CFU/g	Unpasteurised fermented foods, probiotic supplements

The children's analysis groups by MedHi — Med and Hi combined. A few numbers show what is in that pocket: of the 9,388 food codes in the original classification, 8,954 were Lo; of the codes assigned to Med or Hi, over 85% contained vegetables, fruit, or milk and dairy, and the Hi tier is primarily fermented dairy.

So the “high live-microbe intake” group is, to a large extent, the group that eats more unpeeled fresh produce and fermented dairy. It is not the group that takes more probiotics. Reading this paper as “probiotics lower blood pressure” mistakes a dietary-pattern metric for supplement intake.

How far this evidence reaches

Three limits worth naming

Cross-sectional: diet and blood pressure were measured at the same time. Neither reverse causation nor confounding can be excluded. Households that eat more fermented dairy and fresh produce tend to have better overall diet quality, more activity and higher socioeconomic status — all of which lower blood pressure and none of which adjusts away cleanly.
The effect size only just clears the bar. OR 0.70 with the upper confidence bound at 0.98, grazing 1; P = 0.039. This is not a result robust enough to quote repeatedly — a different adjustment set could push it across 1.
In the joint model, weight status does most of the work. Against low intake + overweight: not overweight + low intake is already at 0.29, and adding high intake moves it only to 0.28 — almost nothing. The one place intake appears to matter is within the overweight group (0.60), which could equally be residual confounding. The largest signal in this dataset is body weight, not live-microbe intake.

So what it supports is this: getting fermented dairy and fresh produce into a child's everyday diet is associated with lower odds of hypertension. What it does not support: that supplementing live microbes lowers blood pressure, or that intake offsets the risk carried by excess weight.

What this means for brand owners and formulators

1. The real value of this metric: it turns a number on the spec sheet into a citable dietary exposure

The Lo/Med/Hi boundaries are drawn in CFU/g — the very quantity formulation and release testing already control. That makes “live microbe intake” a dietary exposure metric that is published, citable and dimensionally compatible with the spec sheet.

Talking about dietary exposure and claiming efficacy are two different things in regulatory terms. The first says how many live microbes a food category carries per gram and where a population's intake sits; only the second needs efficacy evidence. It is a safer position to argue from, and one that holds up under scrutiny.

2. But do not stretch this paper onto a supplement — the magnitudes do not match

The classification does list probiotic supplements in the Hi tier, but the exposure in this NHANES analysis is overwhelmingly food. A single probiotic capsule carries orders of magnitude more CFU than any food, placing it outside the range this tiering can resolve. Using a food-epidemiology odds ratio to support a capsule crosses both a magnitude gap and a study-design gap.

3. To use the metric, state what is in it

Over 85% of the “live microbe food” pocket is produce and dairy, and the Hi tier is mostly fermented dairy. The narrative it supports is “get fermented dairy and fresh produce into the daily diet”, not “supplement microbes to lower blood pressure”. The second version takes about a minute to take apart — and children plus blood pressure is exactly the pairing that attracts that scrutiny.

4. The foundation of this narrative is whether the CFU number is real

If a product's story rests on live-microbe intake, its foundation is whether the release count and the end-of-shelf-life count both hold up — and those two are often not the same number. Building a story on a metric scaled in CFU only works if the ruler itself is accurate.

If you are considering this metric

We can supply strain-level documentation, release and end-of-shelf-life viability data, and a view on blending feasibility — the baseline data any “live microbe intake” narrative needs underneath it.

We will not use this analysis to support an efficacy claim for any product. It is a cross-sectional association, and the largest signal in it comes from body weight, not from live-microbe intake.

References

Li X, Li X, Han Y, Huang X. Relationship between dietary live microbial consumption and hypertension risk in children and adolescents: results from the NHANES 2009-2023. Jornal de Pediatria, 1 October 2026. doi:10.1016/j.jped.2026.101630. PubMed 42822535
Marco ML, Hutkins R, Hill C, et al. A classification system for defining and estimating dietary intake of live microbes in US adults and children. Journal of Nutrition, 2022. doi:10.1093/jn/nxac074

This note summarises and interprets published research in order to explain the definition and limits of the dietary live-microbe intake metric. It is not a claim about the efficacy of any product and is not medical advice. The findings are cross-sectional associations and apply only to the population and variable definitions described in the original paper. Blood pressure concerns in children should be discussed with a physician.