- Design
- Retrospective health-record analysis combined with population genome data
- Population
- 1,157,359 patients at a Singapore tertiary hospital; 9,051 sequenced Singaporeans
- Primary outcome
- Prevalence of pharmacogenomically implicated prescriptions
- Effect
- 38–43% received one annually; 18.4% estimated actionable
Using records from a Singapore tertiary hospital from 2014 to 2021 — more than 1.1 million patients — researchers counted prescriptions for medicines with strong CPIC pharmacogenomic guidance. Each year 38–43% of patients with prescriptions received at least one. The commonest were omeprazole, statins and tramadol, implicating CYP2C19, CYP2D6 and SLCO1B1.
Combining this with whole-genome data from 9,051 Singaporeans, the authors estimate that 18.4% of patients carried a genotype that could have changed one of their prescriptions. People of Indian and Malay ancestry were prescribed these drugs at younger ages than people of Chinese ancestry.
This is a modelled estimate of opportunity, not evidence that testing improves outcomes. It is relevant to India because the drugs are among the most prescribed here and CYP2C19 poor-metaboliser variants are common in South Asian populations.
- Remember that clopidogrel needs CYP2C19 activation; consider an alternative in known poor metabolisers
- Watch for poor analgesia or toxicity with tramadol and codeine, both dependent on CYP2D6
- Consider SLCO1B1 status where simvastatin causes myopathy, and prefer a different statin
- Record any known pharmacogenomic result prominently so it informs future prescribing
Why it matters
Genotype-sensitive drugs are not niche; they are among the most commonly prescribed medicines.
Don't overread it
The 18.4% is a modelled estimate; the study did not test patients or measure outcomes.
The statistics, in plain English
The estimate multiplies how often these drugs were prescribed by how common the relevant gene variants are, so it shows potential reach rather than proven benefit.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for clinical pharmacology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free