Polypharmacy in Palliative Care: Two Deprescribing Tools Compared with a Clinical Review.

Lack of guidance is highlighted as a barrier to deprescribing in palliative care. Two deprescribing tools exist, but with inclusion and exclusion criteria that limit utility. The tools have not previously been compared directly or used in an unselected palliative population. To compare the OncPal and STOPPFrail deprescribing tools to an expert review in an unselected palliative population. Secondary aims included a description of medicines incorrectly classified by both tools. Fifty palliative inpatients were retrospectively reviewed by an expert panel, and both tools were independently applied to the patients. Positive and negative predictive values (PPV and NPV) were calculated per patient using the expert review as the gold standard. The median number of medicines per patient was 11, with 19% of medicines deemed inappropriate. The PPV and NPV were 75% (interquartile range 50-100) and 91% (interquartile range 84-100), respectively, for OncPal, and 100% (interquartile range 50-100) and 90% (interquartile range 78-100), respectively, for STOPPFrail. There was no statistically significant difference between the tools (PPV  = 0.42 and NPV  = 0.07). The main medicines incorrectly ceased by OncPal were antianginals for stable coronary artery disease, and haloperidol for nausea by STOPPFrail. There was no significant difference between the tools. Both tools performed well in an unselected population. Some minor amendments could improve the PPV of both tools.