The gut, and what it is reasonable to believe

Few areas of medicine have moved as fast or been commercialised as aggressively as gut health, and separating the established from the speculative has become genuinely difficult for a reader.

What is well established

The gut contains an enormous and diverse microbial community that ferments dietary fibre, produces short-chain fatty acids used by the colon, synthesises certain vitamins, influences immune development and interacts with the nervous system.

Composition varies substantially between individuals, is influenced by diet, medication, geography and early life, and changes measurably within days of a dietary change.

Antibiotics disrupt it, with recovery over weeks to months and not always completely.

Low diversity is associated with several disease states, though whether it is cause or consequence is frequently unresolved.

What is not established

That there is a defined optimal microbiome to aim for.

That a stool test can tell you what to eat, which is the basis of a substantial industry and which is not supported by validated evidence.

That specific probiotic strains reliably treat most conditions they are sold for.

That leaky gut, as popularly described, explains a broad range of unrelated symptoms — intestinal permeability is a real phenomenon studied in specific diseases and is not the general-purpose explanation it is marketed as.

And that supplements can substitute for the dietary pattern that actually shapes the microbiome.

Fibre, which is the actual intervention

The single most evidence-supported thing anyone can do for gut health, and the least profitable.

Most populations consume substantially less than recommended intakes.

Higher fibre intake is associated in large prospective studies and meta-analyses with reduced cardiovascular disease, colorectal cancer, type 2 diabetes and all-cause mortality, with a dose relationship.

Diversity of plant foods appears to matter alongside quantity, which is the basis of the widely repeated suggestion to eat a wide range of different plants each week — a rule of thumb from observational work rather than a validated prescription, but a harmless and useful one.

Fermented foods have supporting evidence from a small number of controlled trials showing effects on microbial diversity and inflammatory markers.

Increase fibre gradually and with fluid, since a sudden increase produces exactly the symptoms people are trying to fix.

Probiotics

Where the evidence is strain-specific and generally weaker than the packaging implies.

There is reasonable evidence for particular strains in particular situations: antibiotic-associated diarrhoea, infectious diarrhoea in children, and some benefit in irritable bowel syndrome.

There is little good evidence that general-purpose probiotics benefit healthy people, and effects do not transfer between strains, so a product's evidence base is not shared across the category.

Colony counts on packaging tell you nothing about survival through the stomach or colonisation.

They are also not risk-free in immunocompromised or seriously ill people.

Irritable bowel syndrome

Common, real, and frequently dismissed.

Diagnosis is clinical, after excluding conditions with overlapping symptoms — coeliac disease, inflammatory bowel disease, and cancer where alarm features are present.

Alarm features that require investigation rather than dietary experimentation: bleeding, unexplained weight loss, anaemia, a family history of bowel cancer or inflammatory bowel disease, onset over the age of about fifty, and nocturnal symptoms.

Treatments with evidence include a low-FODMAP diet delivered properly by a dietitian in three phases including reintroduction, soluble fibre, certain antispasmodics, peppermint oil, gut-directed hypnotherapy and cognitive behavioural therapy — the last two having surprisingly good trial evidence via the gut–brain axis.

The low-FODMAP diet is not intended to be permanent, and remaining on the restriction phase indefinitely reduces microbial diversity, which is precisely counterproductive.

Gluten and dairy

The two most commonly self-eliminated foods.

Coeliac disease is an autoimmune condition requiring lifelong strict gluten avoidance, and it must be tested for while still eating gluten — stopping beforehand invalidates the test, which is the single most common practical error.

Non-coeliac gluten sensitivity is described and contested, with some evidence pointing towards fructans rather than gluten as the trigger in a proportion of cases.

Lactose intolerance is common, varies substantially between populations, and is dose-dependent rather than absolute — most people with it tolerate some lactose, particularly in fermented forms.

The reasonable position

Eat a lot of different plants, get enough fibre, include fermented foods if you like them, do not take antibiotics unnecessarily, and be sceptical of anything that promises to optimise your microbiome for a subscription fee.

And investigate persistent symptoms properly rather than eliminating foods one by one, which produces a narrower diet and rarely produces an answer.

General information only, not medical advice. Consult a qualified clinician about persistent digestive symptoms, and get tested for coeliac disease before removing gluten.