Barriers to Genetic Data Utilization
- Donica Ward-Adams
- Nov 8, 2025
- 3 min read
Early on in our adoption process, we heard from other parents that our kids are part of an ethnic group prone to severe allergies to codeine. This turned out to be a misunderstanding, but after some research, we discovered that the phenomenon they were describing is that the incidence of individuals being ultrarapid metabolizers of codeine is much higher in their part of the world than ours. This would create greater risk for them - not just for the incidence of the genetic variation, but also because physicians in the United States would have much less experience with encountering these reactions.
The genetic variation at play is most closely aligned with CYP2D6 (although CYP2C19 also factors in with related drug reactions). There are four phenotype groups that identify how drugs are metabolized. When an individual has more than two copies of CYP2D6, they are classified as an ultrarapid metabolizer. The enzyme converts codeine into morphine, and even at a normal, therapeutic dose, someone who is an ultrarapid metabolizer can experience a morphine overdose. [1]
For my two older children, their likelihood of being an ultrarapid metabolizer of codeine based on their ethnic background is in the 20-29% range, whereas for my youngest, it's 2-3%. [2] That's essentially a 1 in 4 chance of a deathly reaction with a common medication - yikes!
Their pediatrician had no idea what to do with that information. The difference in prevalence of the reaction was news to him, and he suggested that we put into their charts that they were allergic so as to at least create a conversation with the ordering physician should anyone try to prescribe it.
When we switched to coverage with an Accountable Care Organization, we received a call from a geneticist. She wanted to know if we had codeine listed on both of their charts because they'd had an anaphylactic reaction, or whether we were just concerned about the possibility they were ultrarapid metabolizers. She agreed to order genetic testing to be sure, and I was overjoyed.
We got the results back, and thankfully, they have the normal phenotype, but the letdown came when she said that the pharmacogenomic report would be in their chart as a PDF, and when physicians ordered any medications for our kids, we should remind them to check the report.
Being in healthcare IT, right from the start, I knew how unlikely it would be that the data would be referenced. Since then, I've even had spirited back and forths with doctors asking them to check the report before I tried a new medication with my child, even sending through which genetic variant to check, and being told - without them reviewing the report - that checking it was unnecessary.
So even when we have the data...we don't have a viable environment for that data to be utilized. And the barriers aren't limited to extra steps - there are knowledge deficits, inconsistencies in reporting and naming conventions, difficulty comparing reports, challenges with understanding how to make the data actionable once you understand it, and more. [3] It makes me overwhelmed with the knowledge of needless loss of life or complications due to not having the data. Or having the data, but not having it be useful.
But the flip side is also true - there is great power in understanding, and in the appropriate timing of augmented intelligence to provide pertinent, actionable information. In the field of pharmacogenomics, we can provide the appropriate dosing information to the provider up front rather than having to guess and then adjust up or down from there. We can avoid undue prolonged suffering for low metabolizers and avoid death or traumatic experiences for ultrarapid metabolizers. We could also combine symptom and genetic information to suggest potential differential diagnoses, or save those with rare disorders years without a diagnosis.
My research thus far has made me passionate about the possibilities in clinical decision support utilizing both AI and genomics. But there is so much more to learn. I invite you to explore this topic with me as I continue my research. Let's tackle these weighty but exciting topics together - our patients, our family and friends, deserve the best care possible, personalized to their bodies. "With great power comes great responsibility."
Sources:
Comments