Genetic Test False Positives and Clinical Confirmation

Why a genetic result from a consumer test or raw data file can be wrong, what studies found about false positives for rare variants, and why clinical confirmation matters before any decision.

Content last updated October 22, 2026. This page has not been medically reviewed.

Why this page exists

A genetic result can look precise and still be wrong. This page explains how false positives happen, what studies measured, and why a clinical laboratory should confirm any surprising result before anyone acts on it. It is information only and not medical advice. It does not tell anyone to change a medicine or a screening plan.

Common versus rare variants

Many consumer tests read DNA with a "SNP chip," which checks a fixed list of known spots. A large UK study compared SNP-chip results with sequencing in 49,908 UK Biobank participants. For 108,574 common variants, sensitivity, specificity, positive predictive value and negative predictive value were all above 99%. [1] But the chance that a positive result is real fell sharply as variants became rarer. For very rare variants, only 16% of 4,757 heterozygous genotypes from the chips were confirmed by sequencing. [1] Positive predictive value means the share of positive results that are true.

For rare pathogenic variants in BRCA1 and BRCA2, the study reported sensitivity of 34.6%, specificity of 98.3% and positive predictive value of 4.2%. [1] In a small group of 21 people who bought consumer tests and shared their data, 20 had at least one false-positive rare pathogenic variant. [1]

Raw data and third-party reports

Some companies give customers a raw data file. A clinical laboratory reviewed 49 samples sent for confirmation after earlier results appeared in direct-to-consumer raw data. It found that 40% of the variants reported in those files were false positives. [2] Some variants labeled "increased risk" in raw data or by a third-party service were classified as benign by several clinical laboratories and are common in public population databases. [2] See raw genetic data interpretation.

What "confirmation" means

Confirmation means that a clinical laboratory repeats the finding on a new sample with a method and classification process suited to clinical use, interpreted by a clinician. This is different from running the same raw file through another website. Variant labels can differ between labs, as covered in variant interpretation and uncertain significance.

How to evaluate a result

Ask whether the result is a common variant or a very rare one, since rare results are more likely to be wrong.
Ask who performed the test, whether it was a clinical laboratory, and who interpreted it.
Treat a surprising "high risk" result as a question for a clinician or genetic counselor, not as a diagnosis.
A negative result on a consumer test does not rule out risk either. The BRCA sensitivity above shows a chip can miss most true carriers. [1]

Frequently asked questions

Can a consumer genetic test give a false positive?

Yes. In one laboratory review, 40% of variants reported in direct-to-consumer raw data were false positives. [2] In a UK study, only 16% of very rare heterozygous chip genotypes were confirmed by sequencing. [1]

Are common results also unreliable?

Much less so. For 108,574 common variants, chip accuracy measures were all above 99%. [1]

What should I do with a worrying result?

Take it to a clinician or genetic counselor, and expect confirmatory testing in a clinical laboratory where appropriate. [3] This site cannot say what any individual result means.

Does a negative consumer result mean I am safe?

No. For BRCA1 and BRCA2 pathogenic variants the chips had sensitivity of 34.6% in the UK Biobank comparison. [1]

More details: evidence, limits and cautions

The details behind this page: cautions, limits of the evidence and where sources disagree. Open any note to read it.

How to read the evidence labels

Labels used on this page: established (what studies measured), mixed (how often errors occur) and marketing claim (any promise of "lab-grade" accuracy). The full set is explained in how we label evidence.

What regulators say

In the US, the FDA says a positive direct-to-consumer genetic health risk test does not indicate that a person will develop a disease or condition. It advises consumers to talk with a health care provider about next steps, including confirmatory genetic testing if applicable. [3] The FDA also notes that laboratory developed tests, created and performed in a single laboratory, are generally not reviewed by the agency when offered only on a clinician's prescription. [3] The FDA's May 2024 final rule on laboratory developed tests was vacated by a federal court on March 31, 2025, and on September 19, 2025 the FDA reverted the regulation to its earlier text. [4] This describes the US only, laws change, and it is not legal advice.

Safety note

Do not start, stop or change any treatment, and do not schedule or cancel screening, because of an unconfirmed consumer result. See also direct-to-consumer genetic tests and false positives, multiple testing and incidental findings.

References

Weedon MN, et al. Use of SNP chips to detect rare pathogenic variants: retrospective, population based diagnostic evaluation. BMJ. 2021. PMID 33589468. Read 2026-10-22.
Tandy-Connor S, et al. False-positive results released by direct-to-consumer genetic tests highlight the importance of clinical confirmation testing for appropriate patient care. Genet Med. 2018. PMID 29565420. Read 2026-10-22.
U.S. Food and Drug Administration. Direct-to-Consumer Tests. Read 2026-10-22.
U.S. Food and Drug Administration. Laboratory Developed Tests. Read 2026-10-22.

Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.