DNA-Based Diets: What Trials Show

What a DNA-based diet is

A DNA-based diet uses results from a genetic test, often a home saliva kit, to suggest what to eat. The report may name foods, nutrients or a diet style that are said to suit your genes. This is sometimes called nutrigenetics. This page looks at what trials show. It is information only and not advice about your own results.

How we label evidence here

We use plain labels: established, promising but limited, mixed, insufficient evidence, no good evidence of benefit and marketing claim. A biomarker change is not the same as a health outcome. More on this in How We Label Evidence in Biohacking.

What the best trials show

Weight loss and genotype: no good evidence of benefit. The DIETFITS trial randomized 609 adults to a healthy low-fat or a healthy low-carbohydrate diet for 12 months. [1] It also tested whether a gene pattern said to predict which diet works better made a difference. There was no significant interaction between diet and genotype pattern for weight loss. [1]

Personalized advice in general: mixed. The Food4Me trial randomized adults in seven European countries to standard advice or to advice based on diet, then diet plus body measures and blood markers, then diet plus those plus five diet-related gene variants. [2] The trial reported that personalized advice produced larger and more appropriate changes in dietary behaviour than a conventional approach. [2] A later scientific review summarizing the field found that most studies showed no clinical evidence for gene-based personalized nutrition, and that Food4Me found no difference in weight loss between the levels of personalization. [3]

Pooled trials of personalization: promising but limited. A 2026 meta-analysis of 15 randomized trials in adults with overweight or obesity found modest reductions in body weight and body fat with personalized nutrition. [4] It found no effects on waist circumference, lipids or blood sugar markers, and the certainty of evidence ranged from moderate to very low. [4] Those trials used many kinds of personalization, not only DNA. A second review found that none of 17 trials defined what "personalized nutrition" meant, and certainty was low for body weight. [5]

Telling people their genetic risk: no good evidence of benefit. A Cochrane-method review of 18 studies found that communicating DNA-based risk estimates did not change diet, physical activity or smoking. [6] A 2026 review of 27 trials of polygenic risk score disclosure found no significant effects on any of 22 outcomes tested in two or more trials. [7]

Company programs without a control group: insufficient evidence. One retrospective study of 426 people in a nutrigenetic weight-loss program reported weight loss at 12 months. [8] It had no comparison group, and its authors said randomized trials are needed to confirm cause and cost-effectiveness. [8] Weight loss in a program with dietitian contact cannot be credited to the DNA test.

What the tests can and cannot tell you

• Different companies test different variants, so reports can differ. [9]
• A 2026 review found 104 companies selling 204 nutrition-related panels, and only 54% publicly disclosed the genes used. [10]
• The same review described the nutrigenetics market as unregulated, with high variability in what is offered. [10]
• The FDA says tests for general wellness or low-risk purposes are generally not reviewed by the FDA before they are offered. [9]

Who should be careful

Do not use a DNA report to cut out whole food groups or to change how you take prescribed medicine. The FDA advises talking with a health care provider before making any dietary or health decisions from test results. [9] Extra caution applies if you have diabetes, an eating disorder history, kidney disease, or are pregnant. See Restrictive-Diet Risks From Personalized Results.

How to evaluate a gene-based diet claim

1. Ask whether the claim comes from a randomized trial in people like you, not from a gene study or an animal study.
2. Ask whether the outcome is weight, a health event or a lab number.
3. Ask which genes are used and whether the company says so.
4. Ask who profits if you buy the plan.

For the wider picture, see Genetics as a Nutrition Input, Consumer Genetic Tests for Wellness and Claim Check: "Personalized" Means Better. Privacy questions are covered in Genetic and Location Data Privacy.