Imagine two patients sitting in the same clinic. They have the same diagnosis, are being considered for the same medication, and have the same clinical reason for drug-gene (also known as pharmacogenomic, or PGx) testing. For one patient, the test is covered. For the other, it is not. The difference may have little to do with the clinical evidence. It may come down to their insurance.
That is the access problem emerging as PGx becomes more relevant to everyday prescribing. PGx examines how inherited genetic differences affect a person’s response to medications, including whether a drug is likely to work, whether a different dose may be needed, or whether treatment could carry a higher risk of toxicity.
These genetic differences are not rare. In a six-year study of more than 7.7 million Veterans Health Administration pharmacy users, researchers projected that 99 percent would carry at least one actionable PGx variant. That does not mean nearly everyone needs PGx testing or a medication change. An actionable variant becomes clinically relevant when a patient encounters a drug affected by it. What is striking is how often that happens. More than half of the veterans in the study had received at least one medication for which strong evidence shows genetics can inform prescribing.
There is another feature of PGx that makes access particularly important. Unlike a potassium level or an HbA1c, inherited genetic variants generally do not change. Results obtained to help with one prescribing decision may become useful again years later when a different medication is considered. Recommendations for how to use those results may evolve as new research emerges, but the underlying genetic information can remain relevant throughout a patient’s life.
Real physician voices, twice a week
Free, and one click to unsubscribe.
Of course, prevalence alone does not establish clinical utility. The more important question is whether using PGx information improves care.
The PREPARE trial offers some of the strongest prospective evidence to date. Nearly 7,000 patients across seven European countries participated in a study of genotype-guided prescribing using a 12-gene panel. Among patients with an actionable result for the medication being prescribed, clinically relevant adverse drug reactions occurred in 21.0 percent when genetic information was used to inform treatment, compared with 27.7 percent with standard care, indicating a roughly 30 percent reduction in the odds of an adverse reaction. That does not mean every PGx test or every drug-gene interaction provides the same benefit. It does show that when genetic information is connected to evidence-based prescribing recommendations, it can improve medication safety.
And this is where the access problem becomes difficult to explain to patients. A 2025 analysis in the American Journal of Managed Care examined coverage policies for 65 clinically relevant drug-gene pairs across major U.S. insurers, laboratory benefit managers, and Medicare’s MolDX program. In the policies available as of February 2024, MolDX covered all 65 drug-gene pairs evaluated. Eight of the 12 policies reviewed covered 10 or fewer. Why such a large difference?
It is tempting to conclude that some insurers simply accept the evidence and others do not. The reality is more complicated. Drug-gene pairs supported by Clinical Pharmacogenomics Implementation Consortium (CPIC) guidelines or Food and Drug Administration (FDA) pharmacogenetic information were more likely to be covered. Society guidelines were also the most frequently cited evidence in the policies studied.
Insurers have legitimate questions to answer before covering a test. Is it analytically valid? Is the drug-gene relationship clinically meaningful? Will knowing the result change treatment? Is testing appropriate for this specific patient? Those questions help distinguish clinically useful PGx from testing with limited evidence or little impact on prescribing.
The problem is not that those questions are being asked. It is that different approaches to answering them can leave clinically similar patients with very different access. The authors of that 2025 analysis reached a similar conclusion, finding substantial variability in both policy determinations and the evidence used to support them.
Medicare and the Department of Veterans Affairs (VA) illustrate two ways of approaching the issue. Medicare’s MolDX program uses defined criteria to determine when PGx testing is reasonable and necessary, and the applicable MolDX pharmacogenomics local coverage determination (LCD) remains in effect in 2026. The VA, as an integrated health system rather than simply an insurer, has gone further in incorporating PGx into clinical care. Multiple VA health systems currently offer testing at no cost to veterans.
States are beginning to address access from another direction. More states have enacted biomarker testing coverage laws, but these laws vary in what testing they address, what evidence is required, and which health plans they reach. Some state insurance requirements may not apply to self-funded employer plans. Even when coverage expands, providers may still encounter prior authorization, differences in medical necessity documentation, and variation in payer implementation.
Patients do not need PGx testing for every medication. But, when a drug has an established genetic variation-related effect, clinicians can help determine whether testing is likely to change prescribing and whether genetic results may already be useful. Because coverage varies by plan and indication, checking coverage may also be part of that conversation.
So the answer is not simply to mandate coverage of every PGx test. Nor is it to test every patient for every possible drug-gene interaction. Evidence is stronger for some clinical scenarios than for others, and coverage should reflect those differences. CPIC itself makes an important distinction: Its guidelines explain how available genetic results should be used to guide drug therapy, rather than whether a test should be ordered in the first place. What is harder to justify is a system in which patients who meet the same evidence-based clinical criteria can face substantially different access depending on their insurance.
As PGx becomes part of medication management, we need greater transparency about what level of evidence is sufficient for coverage, and greater consistency in applying those standards. Physicians should be able to recognize when a test is clinically actionable. Payers should still be able to distinguish well-supported testing from low-value use. And patients who meet the same clinical criteria should have a more predictable path to access. Coverage should also account for the fact that PGx results can have value beyond a single prescribing decision. A multigene test ordered today may inform treatment choices years later as new medications are considered. The goal is to have coverage approaches that recognize the lasting value of PGx validated results.
A patient’s biology does not depend on who provides their insurance. Neither should access to evidence-supported PGx testing.
Houda Hachad is a health care executive.
