What Medication-Related Factors Are Important For Polypharmacy And Deprescribing Assessment By Physicians And Pharmacists Using A Novel Website To Rate Clinical Cases?

Abstract Background: Little is known about how clinicians assess polypharmacy in real-world practice. Objectives: To identify which medication-related factors influence clinicians’ ratings of polypharmacy, harm from medicines and deprescribing of medicines when presented with a patient’s list of medications and comorbidities.Setting: A novel website called What Is Polypharmacy Exactly (WIPE) was created with 50 de-identified real-world cases with varying numbers of comorbidities and medicines.Methods: Participants, consisting of physicians and pharmacists were asked to rate each case from zero (lowest) to 10 (highest) on three questions: i) degree of polypharmacy, ii) potential for medication-related harm and iii) potential to deprescribe medicines. Medication-related factors including medicines count, high-risk medicines, inappropriate duplication, drug-drug and drug-disease interactions were assessed. Multiple linear regression was used to determine which medication-related factors influenced median ratings for the three questions. Main outcome measures: The primary outcome measure was the median rating for the degree of polypharmacy for each case. Secondary outcome measures were the median rating for the potential for harm from medicines for each case and median rating for the potential to deprescribe medicines for each case. Results: Ninety-two clinicians were included in the study, comprising of 76.1% pharmacists (n=70) and 23.9% physicians (n=22). The comorbidity count (P=0.001), medicines count (P<0.001), inappropriate medication duplication (P=0.017), high-risk medicines (P=0.049), use of NSAIDs (P=0.032) and antihypertensives (P=0.040) were predictive of the perceived degree of polypharmacy. The comorbidity count (P<0.001), medicines count (P<0.001), inappropriate medication duplication (P=0.016) and high-risk medicines (P=0.036) were predictive of the recognised potential for medication-related harm. The medicines count (P<0.001), inappropriate medication duplication (P=0.017), benzodiazepine use (P=0.003) and antipsychotic use (P=0.039) were predictive of the recognised potential to deprescribe medicines. Conclusion: Whilst polypharmacy has traditionally been defined using the medicines count, our findings suggest that clinicians also consider other factors such as high-risk medicines and inappropriate medicines duplication to identify patients at risk of adverse outcomes. Future polypharmacy assessment tools should additionally include these factors.