Can You Hack Your Gut Microbiome (Economist)

Summary

A Well Informed column on whether direct-to-consumer (D2C) stool-testing firms can improve on generic "eat a varied, partly fermented diet" advice. The verdict: diversity is the only well-supported marker of gut health, there is no reference "right" microbiome to personalise against, and the firms' microbial censuses do not even agree with each other. Closing line: "Perhaps better to save your money and invest in a jar of kimchi." Same editorial posture as Electrolytes You Probably Don't Need (Economist) — a wellness product category whose marketing runs ahead of its evidence.

What the gut microbiome does (as stated)

• Digests complex carbohydrates into fatty acids that supply about 10% of an individual's calories.
• Synthesises vitamin K (blood clotting) and several B vitamins; tunes the immune system; suppresses pathogens; regulates signalling molecules such as serotonin on the gut–brain axis.

Key claims and the studies behind them

1. Dietary diversity → microbial diversity. A 2018 American citizen-science project (the American Gut Project) found people eating 30 types of plants a week had significantly more diverse microbiomes than those eating fewer than 10. Observational, self-reported diet.
2. Fermented foods diversify the microbiome. Stanford University work published in 2021 (Wastyk et al., Cell) found fermented foods such as kimchi and sauerkraut increase microbiome diversity.
3. No reference microbiome exists. Microbiomes "can differ quite a bit from each other while still being perfectly effective"; the only robust red flag is low biodiversity, which a varied diet corrects. The idea that adding or removing a particular bug perceptibly improves health "has scant published scientific basis", and many D2C firms will not share the proprietary data behind their claims.
4. Zoe's own research is modest. Tim Spector (King's College London; co-founder of the D2C firm Zoe) feeds customer data into peer-reviewed research; so far it suggests individually tailored diets "do not have huge effects, but may help at the margins".
5. The censuses are unreliable. Researchers at the US National Institute of Standards and Technology (NIST) prepared 21 sets of homogenised, identical faeces and sent three to each of seven D2C firms. Results published in February 2026 (Servetas et al., Communications Biology; NIST publication page) showed notable differences between firms and, in one case, within a single firm's three replicates. Secondary coverage of the paper (GenomeWeb) adds that firms used different sequencing methods (16S vs shotgun metagenomics) and different definitions of a "healthy" microbiome, and that one firm labelled two replicates of the same sample "healthy" and the third "unhealthy".
6. Jonathan Eisen (UC Davis, evolutionary microbiologist): "Most of the claims [in this area] are bogus."

Flagged as unproven

• Any health benefit from bug-by-bug tinkering (targeted probiotics, "bug-friendly" supplements) — no reference standard, scant published evidence.
• Whether personalised diets beat generic diversity advice — Zoe's data say at best marginal.
• The 30-plants and fermented-food findings show diversity, not downstream clinical outcomes; the article treats diversity as a proxy for health, which it calls "widely thought" rather than proven.

Connections

• Gut Microbiome — concept page.
• Electrolytes You Probably Don't Need (Economist) — parallel Well Informed verdict on a supplement category.
• Food Fads — D2C microbiome tests sit in the same wellness-spend wave as the protein craze.
• The Economist

Captured-by trigger

Source file: raw/processed/Economist 2026-08-15 Science---Techn - Can you hack your gut microbiome.md (full text).