Five Developments Moving Functional Medicine Toward Actual Evidence
Continuous glucose data, microbiome-guided nutrition, multi-omics baselines, GLP-1 integration and standardised outcome measures — what is maturing, and what is still marketing.

Functional medicine has spent thirty years arguing that care should be personalised, root-cause oriented and measured over time. The argument was always plausible; the tools to test it were not available. Several of them now are. Here is where the field is genuinely advancing, and where the claims still outrun the data.
i. Continuous glucose monitoring outside diabetes
Over-the-counter sensors have made minute-by-minute glucose data available to people without diabetes. In insulin resistance and prediabetes, seeing a postprandial curve changes behaviour in ways a fasting number does not, and the underlying variability is real. What has not been shown is that CGM in metabolically healthy people improves any outcome — and normal glucose excursions after a meal are frequently misread as pathology.
ii. Microbiome-guided nutrition
The PREDICT programme established that responses to identical meals differ substantially between individuals and that microbiome composition explains part of that difference. A subsequent randomised trial found that personalised dietary advice outperformed generic guidance on short-term markers. That is a real result. It is short-term, in motivated participants, on surrogate endpoints — not evidence that microbiome testing prevents disease.
iii. Multi-omics baselines
Deep longitudinal phenotyping — proteomics, metabolomics, immune profiling, wearables — is falling in price fast enough to reach clinics. Research cohorts have shown it can catch metabolic and infectious change before symptoms. The unresolved problem is interpretation: with thousands of analytes, false positives are guaranteed, and most functional medicine practices have no validated decision rules for what to act on.
The bottleneck is no longer measurement. It is knowing which measurements deserve a response.
iv. GLP-1 drugs inside integrative practice
The most interesting shift is cultural. A field historically hostile to pharmaceuticals for metabolic disease is now prescribing semaglutide and tirzepatide alongside nutrition, resistance training and sleep work — partly to protect lean mass, which the trials show is a genuine concern. This is where functional medicine's strengths are most defensible: the drug is not in dispute, and the surrounding programme addresses what the drug does not.
v. Standardised outcome measures
The quietest development matters most. Clinics adopting validated instruments such as PROMIS, and reporting them at fixed intervals, make their results comparable to conventional care for the first time. The Cleveland Clinic cohort was possible only because someone recorded outcomes systematically. Until that is normal practice rather than exceptional, the field will keep having the same argument with its critics.
Read together, these are the conditions under which functional medicine could become an evidence-based discipline rather than a philosophy with testimonials. None of them is a breakthrough on its own. Collectively, they are the difference between a claim and a measurement.
Sources & evidence
- Berry, Spector et al., Human postprandial responses to food and potential for precision nutrition, Nature Medicine (2020)
- Bermingham et al., Effects of a personalized nutrition programme on cardiometabolic health (METHOD randomised controlled trial), Nature Medicine (2024)
- Chen, Snyder et al., Personal Omics Profiling, Cell (2012); Nature Medicine follow-ups on wearable-based early detection
- Wilding et al., Once-Weekly Semaglutide in Adults with Overweight or Obesity, NEJM (2021); body composition sub-analyses
- Beidelschies et al., JAMA Network Open (2019), on PROMIS-based outcome reporting in functional medicine care
- US Preventive Services Task Force and ADA statements on CGM use outside diabetes
This article is general information, not medical advice. See our Editorial Standards.



