How much of holistic health is actually backed by evidence? It is the hardest question in the field to answer honestly, and the one we built Gyfts to address. This page publishes the numbers from our reviewed map of the holistic and complementary landscape: 641 modalities, 261 health conditions, and 1,496 individually reviewed relationships between them, each carrying an evidence context label assigned through our editorial review process. Nobody else publishes this data. It expands the evidence chapter of our full report, Holistic Health Statistics.
Key evidence statistics
- •Fewer than one in three holistic modalities (32%) carry a strong or moderate evidence label (Gyfts, September 2026).
- •Only 47 of 641 mapped modalities, about 1 in 14, meet the bar for a strong evidence label.
- •The largest single group, 193 modalities (30%), rests on traditional use: long cultural histories, little formal clinical research.
- •Only 13% of reviewed condition-to-modality relationships (197 of 1,496) are rated as having strong evidence for that specific condition.
- •Less than 1% of global health research funding goes to traditional medicine, which explains much of the gap. Source: WHO
How the evidence labels break down
Every modality on our map carries one of seven evidence context labels. As of September 2026, the 641 published modalities divide as: strong 47 (7%), moderate 157 (25%), mixed 40 (6%), emerging 84 (13%), limited 92 (14%), traditional use 193 (30%), and contemporary practice 26 (4%).
Read together, the shape of the landscape is clear. The practices with substantial research behind them, the strong and moderate groups, make up just under a third of the field. The single largest group is traditional use: practices with centuries of cultural history that formal clinical research has largely never examined. Given that less than 1% of global health research funding reaches traditional medicine, absence of evidence here is mostly absence of research, which is not the same thing as evidence of absence, and not a substitute for it either.
Evidence is specific to purpose
A single evidence label per practice would still be too blunt, because a practice can be well studied for one purpose and unstudied for another. Our reviewers have documented 1,496 relationships between the 261 conditions on the map and the modalities people explore for them, an average of 5.5 potentially relevant modalities per condition. Only 197 of those relationships, 13%, are rated as having strong evidence for that specific condition.
This is why blanket questions like "does holistic health work" have no honest answer. The answerable question is what the evidence context looks like for a specific practice, for a specific purpose, and that is the question our Living Map exists to answer.
The MACH category split
Using the MACH framework (Metaphysical, Alternative, Complementary, Holistic), the mapped modalities divide as: Metaphysical 217, Alternative 179, Complementary 128, and Holistic 116. The categories describe the nature and philosophy of each approach and how it relates to conventional care, not a rating. Metaphysical practices work with energy, spirit, and consciousness. Alternative approaches are complete healing systems with centuries of history, such as Traditional Chinese Medicine. Complementary approaches are used alongside existing care, and holistic approaches work with mind, body, and lifestyle as a whole picture. Metaphysical practices form the largest single category at roughly a third of the map, and every modality page on Gyfts states its category and evidence context so nobody has to guess.
About this data and how to cite it
The statistics on this page are original data from the Gyfts reviewed content system as of September 2026, and they are updated as the map grows. External figures are linked at the point of use. You are welcome to cite this page with attribution:
Source: Gyfts, Holistic Health Evidence Statistics (2026), gyfts.io/blog/holistic-health-evidence-statistics
If you are writing about this space and need a number we have not published, ask us. If we have it and can share it honestly, we will.