Pharmacogenomic Testing for Psychiatric Medications: Coverage, Limits and When It Helps

Pharmacogenomic (PGx) testing looks at genes that affect how your body processes certain medications. For psychiatric drugs, the most useful results are about metabolism, mainly the liver enzymes CYP2D6 and CYP2C19, and they can help a prescriber choose a dose or avoid a drug that is likely to build up or cause side effects. They do not tell you which antidepressant will work, and large trials have found only modest, inconsistent benefits on depression outcomes.

Coverage depends on why the test is ordered. Medicare's regional coverage policies cover PGx testing when a clinician has already narrowed treatment to specific medications with a recognized, actionable gene-drug interaction, not simply because you have a diagnosis. Commercial plans set their own rules, so the honest answer to "is it covered?" is: sometimes, for specific reasons, and you have to ask.

What a pharmacogenomic test actually measures

Most psychiatric PGx panels report your versions of drug-metabolizing genes and translate them into a predicted phenotype, such as poor, intermediate, normal, rapid or ultrarapid metabolizer. A poor metabolizer clears certain drugs more slowly, so standard doses can produce higher blood levels; an ultrarapid metabolizer may clear them faster.

Some panels also include genes linked to serious drug reactions, such as HLA-B*15:02, which is associated with severe skin reactions to carbamazepine. Others add genes tied to serotonin signaling. The most recent antidepressant guideline from the Clinical Pharmacogenetics Implementation Consortium (CPIC) reviewed two of those, SLC6A4 and HTR2A, and concluded that existing data do not support their clinical use in antidepressant prescribing.

Your germline genes do not change, so a valid result can stay relevant for life. That is also why one Medicare contractor's policy treats repeat testing of the same germline content as not medically necessary.

The FDA's table of pharmacogenetic associations

The FDA maintains a public Table of Pharmacogenetic Associations, last updated in September 2026 at the time of writing. It sorts gene-drug pairs into three sections: associations where the data support therapeutic management recommendations; associations with a potential impact on safety or response; and associations that affect drug levels only, where an impact on safety or response has not been established.

The FDA states plainly that including a gene-drug pair in the table does not necessarily mean it advocates pharmacogenetic testing before prescribing, unless the test is a companion diagnostic. The table is a map of where genetics matters, not an instruction to test everyone.

Where common psychiatric medications sit

The table is not a prescribing guide, and doses are always set by your prescriber from the full label and your situation. Its value for you is perspective: for several widely used antidepressants, the FDA lists either a pharmacokinetic-only effect or no entry at all.

What CPIC guidelines add

CPIC publishes peer-reviewed guidelines that tell clinicians how to use a genetic result once they have it. Its 2023 guideline for serotonin reuptake inhibitor antidepressants gives recommendations based on CYP2D6, CYP2C19 and CYP2B6 results for SSRIs, SNRIs, vilazodone and vortioxetine.

Some of its conclusions are reassuring rather than dramatic. For example, it found that existing data do not support adjusting starting doses of citalopram, escitalopram or sertraline for CYP2C19 intermediate metabolizers, although a slower titration and lower maintenance dose may be warranted. It gives no gene-based dosing recommendation for fluoxetine or duloxetine, because the evidence that CYP2D6 status changes outcomes for those drugs is not strong enough.

CPIC also adds a caveat that matters for anyone thinking about testing after the fact: patients on stable, effective doses without significant side effects may not benefit from dose changes based on these genes, and test results are one of many pieces of clinical information. In other words, a "red" result for a medication that is working well is a conversation with your prescriber, not a reason to stop it.

What the trials show: GUIDED and PRIME Care

Two large randomized trials are the ones most often cited, and both are more modest than marketing suggests.

GUIDED

The GUIDED trial enrolled 1,167 outpatients with major depression who had an inadequate response to at least one antidepressant, and randomized them to usual care or care in which clinicians saw a commercial combinatorial PGx report. At week 8, the primary outcome, symptom improvement, was not significantly different between groups (27.2% versus 24.4%). Secondary outcomes were better in the guided group: response was 26.0% versus 19.9%, and remission 15.3% versus 10.1%.

PRIME Care

PRIME Care was a pragmatic trial at 22 Veterans Affairs medical centers involving 676 clinicians and 1,944 patients who were starting or switching a single antidepressant. Clinicians who received test results were far more likely to prescribe a drug with no predicted drug-gene interaction (59.3% versus 25.7%). Remission over 24 weeks was modestly higher with testing (risk difference 2.8%), but the difference was not significant at week 24, and the authors described the effect on remission as small and nonpersistent.

Expert reviews

A 2018 review in the American Journal of Psychiatry concluded that there were insufficient data to support widespread use of combinatorial pharmacogenetic testing in clinical practice, while noting that it may be informative in some situations, particularly in predicting side effects. That remains a fair summary: the strongest case for testing is safety and dosing for specific drugs, not picking a winner.

Where testing can genuinely help

• You have had unusual side effects at low doses, or no effect at high doses, across more than one medication metabolized by the same enzyme.
• Your prescriber is considering a specific drug that has an FDA section 1 entry or a CPIC recommendation, and a result would change the dose or the choice.
• A medication with a serious genetic safety association, such as carbamazepine, is being considered.
• You already have results from an earlier test, and your prescriber can apply them to a new decision.

The guide to what a psychiatrist considers when two antidepressants haven't worked puts testing in context alongside dose checks, adherence, diagnosis review and augmentation.

Single-gene test or broad panel

PGx testing comes in two broad forms. A single-gene test answers one focused question, such as an HLA-B*15:02 check before carbamazepine is considered. A multi-gene or combinatorial panel tests many genes at once, and a combinatorial panel adds a proprietary algorithm that turns the results into drug recommendations or warnings.

Panels are convenient, and a result for CYP2D6 or CYP2C19 can inform many future prescriptions. The trade-off is that panels often include genes without demonstrated clinical utility, and Medicare policies cover only the components that meet their actionability standard. A practical approach is to ask your prescriber which genes would change the decision in front of you, then check whether a narrower test answers that question.

What results cannot do

• They cannot predict which medication will relieve your symptoms. In the large trials, test-guided prescribing changed remission only modestly, and a major review found insufficient data to support widespread use of combinatorial testing.
• They do not capture drug interactions. CPIC notes that an inhibitor or inducer of an enzyme can make your actual metabolism differ from your genetically predicted phenotype, so a normal metabolizer may behave like a poor one on certain other medications.
• Laboratory oversight under CLIA covers analytical accuracy, but CMS's CLIA program does not address the clinical validity of any test. A report's color-coded categories may reflect a company's own algorithm rather than an FDA or CPIC recommendation.
• They are not a reason to stop or switch a medication on your own. See switching antidepressants for how changes are made safely with a prescriber.

If you are thinking about suicide or are in crisis, call or text 988 (the 988 Suicide & Crisis Lifeline) or go to the nearest emergency room.

Coverage: Medicare, commercial plans and self-pay

This section is general information about coverage rules, not legal or tax advice. Your own plan documents and the lab's billing office are the final word.

Medicare

Medicare coverage for PGx testing is set by regional contractors through local coverage determinations (LCDs), so the policy that applies depends on where the test is billed. Several contractors use the MolDX program's policy, which says testing is reasonable and necessary only once a treating clinician has narrowed options to specific medications, and those medications have a gene-drug interaction shown to be actionable by the FDA or by CPIC guidelines at level A or B. The same policy states that testing is not reasonable and necessary merely because a patient has a particular diagnosis.

Another contractor's LCD, revised in April 2026, uses similar criteria: a condition requiring a medication with a known gene-drug interaction, and a test whose actionability is supported by CPIC level A or B or by the FDA's tables or labeling. For panels that include both actionable and unproven content, these policies cover only the components with demonstrated actionability.

If you have Medicare and see an out-of-network or opted-out clinician, lab orders and billing can get more complicated; the guide to Medicare and out-of-network psychiatrists explains the basics.

Commercial insurance

Commercial insurers write their own medical policies, and their criteria for PGx testing vary. Before testing, ask the lab and your plan:

1. Which genes are on the panel, and does my plan cover all of them or only some?
2. Is prior authorization required, and who submits it?
3. If the claim is denied, what will the lab bill me, and does it offer a self-pay price in writing?
4. Will a single-gene test meet my prescriber's need at lower cost than a large panel?

Genetic privacy

The Genetic Information Nondiscrimination Act (GINA) bars health insurers from using genetic information for eligibility or premium decisions and bars employers from using it in employment decisions, but GINA does not apply to employers with fewer than 15 employees. Its health insurance protections do not cover life insurance, disability insurance or long-term care insurance. If you are applying for those products, factor that into the timing of any genetic test.

Questions to ask before you test

• Which specific medication decision will this result change?
• Which genes in this panel have FDA or CPIC-supported recommendations for the drugs we are considering?
• If my result says a medication that is working for me is "use with caution," what will we do?
• Are any of my other medications likely to affect the same enzymes?
• Who will interpret the report, and will the results go into my medical record so future prescribers can use them?

Readers comparing first-line options may also find how a first antidepressant is chosen useful, since most of that decision rests on factors other than genes.

How the practice handles lab testing and medication decisions

Dr. Lewis's practice provides psychiatric evaluation and medication management alongside lab testing as part of ongoing care. Lab costs are separate from visit fees, and the practice is out of network with all insurance plans, including Medicare and Medicaid, providing a superbill you can submit for possible reimbursement. Whether PGx testing makes sense for you is a question for the prescriber who knows your medication history. The practice's articles on how genetic testing guides medication choices and precision psychiatry describe its approach, and payment details are on the insurance and payment page. For the bigger picture, return to the Choosing Psychiatric Care guide.

Frequently asked questions

Can a cheek swab test tell me which antidepressant will work for me?

No. Pharmacogenomic results mainly predict how quickly you process certain drugs and your risk of some side effects. Trials of guided prescribing have shown modest, inconsistent effects on depression remission, so results are one input among many.

My report puts my current medication in the red category. Should I stop it?

Do not stop or change it on your own. Talk with your prescriber, especially if the medication is working and side effects are manageable. CPIC notes that patients stable on effective doses may not benefit from changes based on these genes.

Does a direct-to-consumer DNA test count?

Sometimes the raw data include relevant genes, but consumer reports are not designed for prescribing. Ask your prescriber whether a clinical-grade result is needed before acting on it.

Is MTHFR testing part of pharmacogenomic testing?

Some commercial panels include it, but it is not among the genes the CPIC antidepressant guideline uses to make dosing recommendations (CYP2D6, CYP2C19 and CYP2B6). Ask what decision an MTHFR result would change before paying for it.

Can I use an HSA or FSA for pharmacogenomic testing?

Medical diagnostic tests are often eligible, but rules depend on your plan. See using an HSA or FSA for psychiatry and lab tests and confirm with your plan administrator.

Will my results be useful for other medications later?

Often, yes. Germline genes do not change, so results for enzymes such as CYP2D6 and CYP2C19 can inform future prescribing for many drugs. Keep a copy and share it with each new prescriber.

Sources

1. Table of Pharmacogenetic Associations. U.S. Food and Drug Administration, 2026. fda.gov
2. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6, CYP2C19, CYP2B6, SLC6A4, and HTR2A Genotypes and Serotonin Reuptake Inhibitor Antidepressants. Clinical Pharmacology & Therapeutics 114(1):51-68 (Bousman et al.), 2023. pubmed.ncbi.nlm.nih.gov
3. Impact of pharmacogenomics on clinical outcomes in major depressive disorder in the GUIDED trial. Journal of Psychiatric Research 111:59-67 (Greden et al.), 2019. pubmed.ncbi.nlm.nih.gov
4. 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 328(2):151-161 (Oslin et al.), 2022. pubmed.ncbi.nlm.nih.gov
5. Clinical Implementation of Pharmacogenetic Decision Support Tools for Antidepressant Drug Prescribing. American Journal of Psychiatry 175(9):873-886 (Zeier et al.), 2018. pubmed.ncbi.nlm.nih.gov
6. LCD L38294: MolDX: Pharmacogenomics Testing. Centers for Medicare & Medicaid Services, Medicare Coverage Database (Palmetto GBA), 2025. cms.gov
7. LCD L39995: Pharmacogenomic Testing. Centers for Medicare & Medicaid Services, Medicare Coverage Database (Wellpoint Federal), 2026. cms.gov
8. LDT and CLIA FAQs. Centers for Medicare & Medicaid Services, 2013. cms.gov
9. Genetic Discrimination. National Human Genome Research Institute, 2022. genome.gov

Educational information, not medical, tax or legal advice. Decisions about diagnosis, medication and treatment belong with a clinician who knows your history.