Who Actually Shapes Functional Medicine — and What Their Evidence Looks Like

Functional medicine is less a treatment than a way of organising a consultation: ask what preceded the illness, what triggered it, and what keeps it going, then intervene across diet, sleep, stress, environment and, where indicated, drugs. That framing came from Jeffrey Bland, a biochemist who named the field in 1990 and built the Institute for Functional Medicine around it. The framework is genuinely useful. It is also untested as a framework — no trial compares 'thinking in systems' to 'not thinking in systems'.

What can be tested is what practitioners do with it, and here the field's most-cited names diverge sharply in how much they have submitted to outside scrutiny.

The outcomes data that exists

Mark Hyman is the field's public face and co-founded the Cleveland Clinic Center for Functional Medicine in 2014. Its 2019 retrospective cohort study in JAMA Network Open remains the single strongest outcomes signal functional medicine has: patients seen in the functional medicine centre showed larger improvements in the PROMIS global physical health score at six months than matched primary-care patients. It is also retrospective, single-site, and unblinded, with substantial loss to follow-up — a reason to run a trial, not a substitute for one.

Hyman's newer venture, Function Health, sells membership access to more than a hundred blood biomarkers. Broad panels in people without symptoms are not endorsed by any major screening body: most abnormal results in healthy adults are statistical noise, and the downstream cost of chasing them is real. The counter-argument — that baselines and trends are worth more than single readings — is reasonable and, so far, unproven.

The clinicians who ran trials

Terry Wahls is the clearest example of the field doing this properly. Rather than selling her multiple sclerosis dietary protocol on the strength of her own recovery, she took it into funded randomised studies at the University of Iowa, where modified Paleolithic and low-saturated-fat diets both improved fatigue and quality of life in relapsing-remitting MS. Those are patient-reported outcomes in small samples over months; disability progression, the endpoint that matters most, remains unaddressed.

Dale Bredesen's ReCODE protocol for cognitive decline applies the same logic to Alzheimer's: treat the multiple contributors — insulin resistance, sleep apnoea, inflammation, toxin exposure — instead of a single amyloid target. A 2024 proof-of-concept precision-medicine trial reported improvement on cognitive measures in a small, closely supervised cohort. There was no randomised control arm, participants were self-selected and highly motivated, and the intervention bundles dozens of components, so no single element can be credited. Bredesen's mechanistic case has become more mainstream as the amyloid-only model has weakened; his clinical evidence has not kept pace with his claims.

The dividing line in this field is not conventional versus alternative. It is who publishes their failures.

The academics who made personalisation measurable

The most rigorous work adjacent to functional medicine is not branded as functional medicine at all. Michael Snyder's group at Stanford tracked the same individuals across genomics, proteomics, metabolomics and continuous wearable data for years, catching pre-diabetes and viral infections before symptoms. Tim Spector's PREDICT studies, which grew into ZOE, showed that two people eating an identical meal can have very different glucose and triglyceride responses, and that part of that variance tracks with the gut microbiome.

Both lines of work vindicate the core functional-medicine intuition — that population averages hide clinically meaningful individual variation — while illustrating the discipline the clinical field often lacks. Detecting variation is not the same as knowing what to do about it, and Spector in particular has been careful to say so.

How to read a practitioner

The useful questions are consistent across every name here. Is the specific protocol published, or only the philosophy? Is there a control group? Are the outcomes patient-reported over weeks, or hard endpoints over years? Does the practitioner sell the tests, the supplements, or the programme they are recommending? And when results are negative, do they say so in public? Functional medicine's better figures answer these comfortably. Its weaker ones change the subject to anecdote.

This article is general information, not medical advice. See our Editorial Standards.