Volume 6, Issue 1 (2026)

Medication appropriateness is an important component of medicines management, but multimorbidity creates therapeutic decisions in which a medication can be indication-concordant yet still be poorly aligned with the patient’s overall clinical situation. Disease-specific benefit, treatment burden, practical workability, and patient priorities may diverge, change over time, or operate at different levels of care. To develop an evidence-bounded conceptual model of therapeutic optimization that treat

Precision dosing increasingly combines pharmacokinetic models, organ-function estimates, therapeutic drug monitoring, pharmacogenomic information, and exposure–response evidence. Yet the availability of more patient-specific data does not ensure that a safer or more defensible dose decision can be made. This Current Opinion argues that precision dosing should prioritize actionability before data abundance. Actionability is defined here as the extent to which a dosing signal is sufficiently relia

Pharmacogenomic results are often represented as durable patient attributes: a genotype is translated into a predicted phenotype, linked to a prescribing recommendation, and stored for future use. This representation captures the persistence of genomic information but can obscure the conditional nature of clinical actionability. Drug exposure, enzyme inhibition, polypharmacy, competing therapeutic priorities, guideline methodology, and the practical capacity to act can change while genotype rema

Deprescribing is often described as the planned reduction or discontinuation of medicines whose harms, treatment burden, or limited expected benefit no longer justify continued use. That definition is necessary but incomplete for complex polypharmacy because a safe medication change does not end when a prescription is stopped. The clinical consequences of withdrawal depend on which medicine is changed first, what other medicines and conditions remain, whether symptoms represent withdrawal or rec

Medication-related clinical decision support is commonly judged through alert counts, override rates, acceptance rates, or rule-level technical performance. These measures describe system activity but do not directly establish whether preventable medication harm has been reduced. This Current Opinion argues that medication-safety signals should instead be assigned to response modes according to the combination of potential harm, immediacy, contextual uncertainty, persistence, interpretive requir