Skip to content

Pharmacogenomics

What a mental health DNA test reads, which genes have published prescribing guidance, what the depression trials found, and which genes have none.

By the GeneMatrix editorial teamUpdated 7 min read

A woman sitting in a window seat with a cup of tea, a knit blanket over her knees and plants on the sill.

In short

Mental health DNA tests are pharmacogenomic tests, and the evidence is mixed. They read genes, mainly CYP2D6 and CYP2C19, that affect how your body processes some psychiatric medicines. It can’t diagnose depression or anxiety, predict whether you’ll develop a condition, or say which medicine will work. In trials, using the results brought modest gains on some measures, not all.

On this page
  1. What these tests read, and what they don’t
  2. Which genes have published guidance for psychiatric medicines?
  3. Why do some tests report genes with no prescribing guidance?
  4. Do mental health DNA tests work? What the trials found
  5. Should you test before a first prescription, or after?
  6. What should you do with a result?
  7. What a PGx result can’t tell you about mental-health medicines
  8. How GenePGx’s mental-health view works
  9. What to do next

What these tests read, and what they don’t

This article weighs the evidence. For what the test is and how ours works, see mental health DNA test.

A mental health DNA test is a pharmacogenomic (PGx) test sold for mental health. It mainly reads a small set of genes that make the enzymes your body uses to process many psychiatric medicines, and sometimes a few immune-system genes linked with rare, serious reactions. The report describes how you are likely to process medicines that have published guidance.

You may see it called a mental health DNA test, a genetic test for antidepressants or a psychiatric pharmacogenomic test. They describe the same kind of test. Here is what it isn’t:

  • Not a diagnosis. No DNA test can diagnose depression, anxiety, bipolar disorder or any other mental-health condition. Diagnosis comes from a clinician who talks with you.
  • Not a risk score. A pharmacogenomic report doesn’t tell you whether you are likely to develop a mental-health condition.
  • Not a prescription. It doesn’t say which medicine to take, or how much of it.
Calm ripplesRings spreading slowly outward from a single point.
A pharmacogenomic result is one more piece of information for the conversation about mental-health medicines. It doesn’t replace that conversation.

Which genes have published guidance for psychiatric medicines?

The Clinical Pharmacogenetics Implementation Consortium (CPIC) publishes peer-reviewed guidelines on how a genetic result can be used for specific medicines. Three of its guidelines cover medicines prescribed for depression, anxiety and ADHD.[1]

CPIC guidelines for antidepressants and an ADHD medicine, with examples of level A pairs
The guideline coversGenes with recommendationsExamples of level A pairs
Antidepressants that act on serotonin reuptake (SSRIs, SNRIs and two related antidepressants)CYP2D6, CYP2C19 and CYP2B6[2]citalopram, escitalopram and sertraline with CYP2C19; paroxetine and vortioxetine with CYP2D6[3]
Tricyclic antidepressantsCYP2D6 and CYP2C19[1]amitriptyline with CYP2D6 and CYP2C19; nortriptyline with CYP2D6[3]
A non-stimulant ADHD medicineCYP2D6[1]atomoxetine with CYP2D6[3]

Pharmacogenomic results describe how your body is likely to process certain medicines. They do not tell you whether a medicine will work for you, and they are not dosing instructions. Never start, stop or change a medication because of a genetic result. Bring it to the prescriber who manages your medicines.

CPIC’s 2023 antidepressant guideline gives recommendations based on CYP2D6, CYP2C19 and CYP2B6 results.[2] In the FDA’s pharmacogenomic table, most of the entries for psychiatric medicines name CYP2D6.[4] These are cytochrome P450 enzymes, and your result for each is usually given as a metabolizer status, from poor to ultrarapid.

One more guideline matters for some people. A variant called HLA-B*15:02 is strongly associated with a greater risk of Stevens–Johnson syndrome and toxic epidermal necrolysis, two serious reactions, in people treated with carbamazepine or oxcarbazepine. A variant called HLA-A*31:01 is associated with a greater risk of several reactions to carbamazepine, including serious ones.[5]

Pharmacogenomic results describe how your body is likely to process certain medicines. They do not tell you whether a medicine will work for you, and they are not dosing instructions. Never start, stop or change a medication because of a genetic result. Bring it to the prescriber who manages your medicines.

Why do some tests report genes with no prescribing guidance?

Some mental health DNA tests also report genes that have been studied for links with how people respond to psychiatric medicines: SLC6A4 (the serotonin transporter), HTR2A (a serotonin receptor), COMT, MTHFR and BDNF. Here is where each stood with CPIC on September 28, 2026:

  • SLC6A4 and HTR2A. CPIC reviewed both for its 2023 antidepressant guideline and concluded that the data don’t support using them to prescribe antidepressants.[2] Every pair between these genes and those antidepressants is rated level C.[3]
  • COMT. In CPIC’s published guidelines, it appears only in the guideline on opioids, and every COMT pair there is level C.[3]
  • MTHFR and BDNF. No CPIC guideline uses either gene.[1][3]

CPIC doesn’t consider level C evidence strong enough for a prescribing recommendation.[3] If a report shows these genes, read them as research findings, not as reasons to choose one medicine over another.

Do mental health DNA tests work? What the trials found

First, the problem these tests try to help with. In STAR*D, a large US study of 2,876 outpatients with depression, 28% to 33% reached remission in the first step of treatment, depending on the symptom scale used.[6] So most people in that study didn’t reach remission with their first medicine.

Three large looks at test-guided antidepressant prescribing
StudyWho took partWhat it found
GUIDED, 20191,167 outpatients with depression who hadn’t responded well to at least one antidepressantNo significant difference in its main measure, symptom improvement at week 8 (27.2% against 24.4%). Significantly more people responded (26.0% against 19.9%) and reached remission (15.3% against 10.1%).[7]
PRIME Care, 20221,944 veterans with depression who were starting or changing a single antidepressant, at 22 VA medical centersPrescribers chose fewer medicines with a predicted drug–gene interaction. Remission across 24 weeks was slightly higher (a 2.8 percentage-point difference), but not significantly higher at week 24.[8]
Pooled analysis, 202213 trials with 4,767 patientsRemission was 1.41 times as likely with test-guided prescribing. The authors called it a modest but significant increase.[9]

So the honest answer is: on average, the gains were modest, and not on every measure. The tests in these trials also differed in the genes they read and the advice they gave, which the pooled analysis named as a source of uncertainty.[9] In PRIME Care, the authors described the effects on remission as small and not lasting.[8]

A genetic result can narrow the conversation. It can’t pick the medicine for you.

Should you test before a first prescription, or after?

Testing before any medicine is chosen is called pre-emptive testing. Testing after a medicine hasn’t helped, or has caused side effects, is often called reactive testing. The evidence so far comes from both.

  • What CPIC expects. CPIC assumes pre-emptive testing will become more common. It writes its guidelines so clinicians know how to use a result that is already on file, even one that was ordered with no particular medicine in mind.[1]
  • What the trials tested. GUIDED enrolled people who had already not responded well to an antidepressant.[7] PRIME Care enrolled people starting or changing one.[8] PREPARE, a large pre-emptive trial, tested a 12-gene panel in people starting any of a range of medicines, not only antidepressants.[10]
  • Who decides. Whether and when a test would help is something to talk through with your prescriber. The trials show modest gains, not a rule that everyone should test first.

What should you do with a result?

  1. 1Bring the whole reportShare it with the prescriber who manages your mental-health medicines, not just the summary page.
  2. 2Ask what appliesAsk which of your results matter for the medicines you take now or might take, and which have published guidance.
  3. 3Mention everything elseTell them about your other medicines and supplements. They can change how a medicine is processed too.
  4. 4Keep it on fileYour genes don’t change, so keep the report and share it with anyone who prescribes for you later.

What a PGx result can’t tell you about mental-health medicines

  • Whether a medicine will work for you. A result describes how you are likely to process a medicine, not whether it will help your symptoms.
  • Whether you have, or will develop, a condition. A pharmacogenomic test isn’t a diagnostic test or a risk test.
  • Anything about medicines without guidance. Many psychiatric medicines have no level A or B guidance from CPIC yet.[3]
  • How everything else affects you. Other medicines, age, and how your liver and kidneys are working can all change how you respond.
  • Rare variants. Genotyping looks for known variants chosen in advance, so a rare one can be missed.

How GenePGx’s mental-health view works

  1. One sample. You collect a saliva sample with the kit and send it back.
  2. Genotyped in the lab. The lab checks the variants on the panel by genotyping.
  3. One report, four views. Mental-health medicines sit beside everyday and pain medicines, with a supplement view.
  4. A plain summary first. Then the detail, written to go through with your prescriber.

The supplement section is never billed to insurance.

The trials above studied other tests, so their results aren’t a promise about this one. For what the test includes and how to order it, see our antidepressant DNA test page.

What to do next

Want the basics first? Read how pharmacogenomics works, or the 2026 evidence roundup. To see whether a medicine has published guidance, try the Medication Check.

Sources

  1. [1]CPIC guidelines (29 guidelines listed). Clinical Pharmacogenetics Implementation Consortium, on ClinPGx. Accessed .
  2. [2]CPIC guideline for CYP2D6, CYP2C19, CYP2B6, SLC6A4, and HTR2A genotypes and serotonin reuptake inhibitor antidepressants. Clinical Pharmacology & Therapeutics, 2023. Accessed .
  3. [3]Genes-drugs: CPIC levels for each gene and drug pair. Clinical Pharmacogenetics Implementation Consortium, on ClinPGx. Accessed .
  4. [4]Table of Pharmacogenomic Biomarkers in Drug Labeling (content current as of August 12, 2026). U.S. Food and Drug Administration. Accessed .
  5. [5]CPIC guideline for HLA genotype and use of carbamazepine and oxcarbazepine: 2017 update. Clinical Pharmacology & Therapeutics, 2018. Accessed .
  6. [6]Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. The American Journal of Psychiatry, 2006. Accessed .
  7. [7]Impact of pharmacogenomics on clinical outcomes in major depressive disorder in the GUIDED trial: a large, patient- and rater-blinded, randomized, controlled study. Journal of Psychiatric Research, 2019. Accessed .
  8. [8]Effect of pharmacogenomic testing for drug-gene interactions on medication selection and remission of symptoms in major depressive disorder: the PRIME Care randomized clinical trial. JAMA, 2022. Accessed .
  9. [9]Pharmacogenomic testing and depressive symptom remission: a systematic review and meta-analysis of prospective, controlled clinical trials. Clinical Pharmacology & Therapeutics, 2022. Accessed .
  10. [10]A 12-gene pharmacogenetic panel to prevent adverse drug reactions: an open-label, multicentre, controlled, cluster-randomised crossover implementation study (PREPARE). The Lancet, 2023. Accessed .

How we write and source these pieces: our editorial policy.

Related reading

The GeneMatrix kit, box open and the collection tube beside it.

Find the test that fits you.

Find your testSee all tests

Full refund within 30 days of ordering, as long as your results haven’t been delivered.

Please read

PGxPharmacogenomic results describe how your body is likely to process certain medicines. They do not tell you whether a medicine will work for you, and they are not dosing instructions. Never start, stop or change a medication because of a genetic result. Bring it to the prescriber who manages your medicines.

NoteThe supplement section is never billed to insurance.

LabOur genetic tests are laboratory-developed tests performed in our own CLIA-certified laboratory. New York is the exception: orders shipped there are completed through a physician order or a reference laboratory permitted in New York. Our genetic tests have not been cleared or approved by the FDA. Results describe inherited traits and risks and are meant to inform decisions you make with your clinician. They are not a diagnosis.