Genetic Testing Could Reshape Emergency Prescribing, but the Evidence Is Not There Yet

More than 1.5 million people are treated in United States emergency departments every year for adverse drug events, and roughly 500,000 of those encounters end in hospitalization. A new perspective article published in Frontiers in Pharmacology argues that pharmacogenomic testing, the practice of using a patient’s genetic profile to guide drug selection and dosing, could help reduce this burden in the emergency department, where nearly one in five visits involves a medication for which actionable pharmacogenomic guidance already exists. Yet the authors, a team from the University of Florida and Ohio State University, are equally clear about the limits of the current evidence: no randomized controlled trial has ever tested the clinical utility of pharmacogenomic testing in an emergency department population, and the field rests almost entirely on feasibility studies and observational data.

The scale of the opportunity is what makes the emergency department such an attractive target. Analysis of National Hospital Ambulatory Medical Care Survey data from 2010 to 2015 estimated that more than 148 million ED visits in the United States, nearly 20 percent of all visits during that period, involved medications with available pharmacogenomic guidance. Those medications include commonly used drugs such as ondansetron, proton pump inhibitors, opioids, and warfarin, the latter two classified as high-alert medications by the Institute for Safe Medication Practices. Beyond prescribing patterns, population-based studies indicate that more than 90 percent of U.S. adults carry at least one pharmacogene variant associated with altered drug response. In an acute care setting where clinicians must make rapid prescribing decisions with little patient history, the authors suggest that preemptive genetic data could accelerate and improve the safety of time-sensitive care.

The perspective highlights several feasibility studies suggesting that testing can function within emergency department workflows. In the acute cardiovascular setting, point-of-care CYP2C19 genotyping using rapid buccal swab-based testing has been shown to work within the real-world time constraints of acute care, enabling timely guidance of antiplatelet therapy selection in patients with acute coronary syndromes. In pediatrics, a study of 100 children aged 3 to 18 presenting to the ED with acute psychiatric or behavioral crises found that 73 percent were taking at least one medication with a gene-drug interaction, and among children presenting with depression or suicidal ideation, those prescribed pharmacogenomically concordant medications showed significantly better clinical improvement scores on parent-rated Clinical Global Impression assessments than those on discordant medications. A separate pediatric asthma feasibility study found that integrating buccal swab DNA collection and spirometry into acute care workflows was well accepted, with nearly 100 percent of physicians, 90 percent of nurses, and 90 percent of respiratory therapists reporting no significant negative impact on their clinical activities.

Observational evidence also points to clinically meaningful associations between genetic variation and adverse drug events presenting to the emergency department. The ADRED study, which analyzed 776 adverse drug reaction cases presenting to EDs, reported that genotype-predicted CYP2C19 activity was significantly associated with bleeding events in patients taking clopidogrel; higher CYP2C19 activity was associated with a nearly five-fold increased odds of clopidogrel-related bleeding adverse drug reactions. The same study identified a trend between phenprocoumon risk profiles and related adverse reactions. The investigators concluded that the impact of pharmacogenomics on serious adverse drug reactions may be highest for drugs that cannot be easily monitored or that do not provoke mild symptoms quickly. The authors of the perspective caution, however, that the observational design and narrow drug scope of such studies are not sufficient on their own to drive clinical implementation.

CYP2D6-dependent opioid prescribing emerges as another area of active investigation. A large retrospective cohort study found that patients taking hydrocodone, tramadol, or codeine who were concurrently exposed to CYP2D6-inhibiting antidepressants had 75 percent higher odds of pain-related ED visits compared with those on non-inhibiting antidepressants, with similar findings for oxycodone. A subsequent national cohort study of more than 31,000 patients from the All of Us Research Program essentially confirmed these results, reporting that CYP2D6 intermediate or poor metabolizers, as predicted by both genotype and drug interactions, experienced almost 20 percent increased risk of pain-related ED visits compared with those predicted to have normal or increased CYP2D6 activity. But the picture is not uniformly positive: a recent randomized pragmatic trial in post-operative acute pain patients found that using CYP2D6 genotyping to inform opioid prescribing did not improve pain control or reduce opioid consumption compared with usual care, suggesting that pharmacogenomics may not benefit all pain populations, particularly those using multimodal pain management.

That nuance matters because genotype alone does not tell the whole story in the emergency department. The authors emphasize that phenoconversion, driven by concomitant medications such as CYP2D6 inhibitors, polypharmacy, or acute inflammatory states, can alter functional enzyme activity independently of genotype. Emergency department pain management frequently incorporates multimodal analgesic strategies, and co-prescribed medications can change a patient’s actual metabolic capacity at the moment of prescribing. For this reason, the perspective argues that genotype alone is insufficient to guide opioid selection in the ED and calls for clinical decision support tools that incorporate concomitant medications and other patient-specific factors capable of modifying opioid response in real time.

Older adults may represent a particularly high-yield population. A pilot study of 44 middle-aged and older adults discharged from the ED with opioid prescriptions for acute pain demonstrated the feasibility of recruiting socioeconomically disadvantaged and racially diverse older populations for pharmacogenomic research in this setting. More recently, a prospective study of 25 geriatric ED patients, with a median age of 81, who sustained fall-related injuries found that 56 percent possessed potentially significant pharmacogenomic gene-drug interactions, and 24 percent had one or more potentially serious interactions based on evidence-based guidelines and FDA-approved drug label information. These patients were prescribed a median of eight medications, with antihypertensives, statins, anticoagulants, and antiplatelet agents among the most common. The authors stress that larger prospective studies are needed to determine whether pharmacogenomic-guided prescribing in this population actually improves clinical outcomes.

Even if the evidence matures, substantial barriers stand between current practice and routine ED testing. Logistically, genetic tests are not routinely ordered in emergency departments, ordering and collection processes often differ from standard laboratory workflows, and buccal swab collection is unfamiliar to many ED staff. Turnaround times for genetic tests typically range from a few days to a week or more, far longer than a typical ED stay, which means patients are rarely still under the ED clinician’s care when results arrive. The authors propose strategies to work around this, including starting with point-of-care CYP2C19 testing, which has an approximate one-hour turnaround and is currently the only such option available, or collecting a DNA sample in the ED for use at a later follow-up visit, since genetic results remain valid for a patient’s lifetime. They also note that EDs already have systems for following up on laboratory results that arrive after discharge, such as bacterial cultures, which could be adapted for pharmacogenomic results, potentially with support from a pharmacogenomics consult service.

Financial and knowledge barriers compound the logistical ones. Despite evidence that more than 90 percent of patients harbor at least one pharmacogene variant, third-party payers show little appetite for reimbursing widespread preemptive testing, often citing the need for more data on cost-effectiveness and patient outcomes. The authors suggest focusing implementation on patients most likely to benefit in the near future and note that multi-gene panels cost only marginally more than single-gene tests while providing results for several additional genes, making panels more cost-efficient. On the knowledge side, clinicians often report discomfort interpreting pharmacogenomic results and explaining them to patients, a deficit that likely depresses test ordering rates. Resources from the Clinical Pharmacogenetics Implementation Consortium, which provides standardized phenotype categories and prescribing recommendations, along with pharmacogenomics consult services and electronic health record-based clinical decision support, are proposed as remedies.

Looking ahead, the authors anticipate that falling costs will shift testing from genotyping panels toward genomic or exomic sequencing performed once in a lifetime and queried repeatedly as prescribing needs arise, though they acknowledge that exome sequencing has difficulty resolving gene duplications and pseudogenes, which occur with CYP2D6. More sophisticated clinical decision support systems will be needed to integrate genetic data with drug-drug interaction checking and other clinical information while reducing alert fatigue. The bottom line of the perspective is measured: emergency department patients appear to be a population that could particularly benefit from pharmacogenomic testing, but until prospective interventional trials demonstrate improved clinical outcomes in the acute care setting, widespread clinical implementation remains premature.

Subject of Research: The potential and current limitations of pharmacogenomic testing in emergency department care

Article Title: Pharmacogenetic testing to improve care in the emergency department

Article References: Gartner, H. T., Duarte, J. D., & Sheikh, S. (2026). Pharmacogenetic testing to improve care in the emergency department. Frontiers in Pharmacology, 17, Article 1866821. https://doi.org/10.3389/fphar.2026.1866821

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1866821

Keywords: pharmacogenomics, emergency medicine, adverse drug events, CYP2C19, CYP2D6, opioids, clinical decision support, genetic testing, drug-gene interactions, older adults, precision medicine, clopidogrel