Medication use and deprescribing in older patients in the last 14 days of life

Abstract

Introduction: Older patients may continue to receive potential inappropriate medications (PIMs) at the end of life.Application of consensus-based tools to identify PIMs mayresult in the identification of candidate medications fordeprescribing, with the aim of overcoming the harm of inappropriate medication and improving clinical outcomes.This study aims to describe medication use and deprescribingpatterns, and to assess prescribing appropriateness for olderpeople in the last 14 days of life in the hospice setting.Methods: Longitudinal, retrospective cohort study of deceased patients (=65 years) who died between 1 January 2018and 31 December 2018 in three hospices in a region of theUnited Kingdom. We identified prescribed and deprescribedmedications and assessed medication appropriateness usingconsensus-based criteria, namely STOPPFrail[1] and criteriadeveloped by Morin et al.[2] Unexpected/sudden deaths wereexcluded. Statistical analysis was conducted using SPSS statistics 26.0.Preliminary results: Data collection is currently ongoing.To date, data from 69 deceased patients have been collected(mean age 76.1 years). Of these decedents, 62.3% were femaleand the majority (just under 90%) had cancer reported as thecause of death. During the last 14 days of life, each patientwas prescribed a mean of 17 ± 5 different medications. Themean number of medications decreased significantly betweenday 14 and the day of death from 13.2 ± 4.4 to 9.4 ± 3.7 (P <0.01). Six hundred and thirty-nine medications were discontinued, with just under 70% stopped in the last seven daysbefore death. 34.9% of those discontinued were prescribedfor chronic conditions and 22% were proton pump inhibitors. In most decedents, swallowing difficulty was the reasonfor medication discontinuation. According to the STOPPFrailcriteria [1], 42 (60.1%) of decedents received at least one PIMbetween day 14 and the day of death. There were 59 PIMs intotal for these patients; of these 20.3% were hyoscine butylbromide and 16.9% were gliclazide. Using the criteria developed by Morin et al [2], 103 medications were assessed asbeing of questionable (81.6%) or inadequate (18.4%) clinicalbenefit. Of these, 64.1% were initiated during hospice admission. There was a statistically significant association betweenmedications of questionable clinical benefit and medicationnumber during the last 14 days of life (P < 0.01). Three of thePIMs were vitamins, considered inappropriate by both sets ofcriteria. Prescribing of PIMs reduced as patients neared death.Conclusion: A substantial proportion of older patients withlife-limiting diseases receive PIMs during their last days of life.No systematic discontinuation of inappropriate medicationswas observed thus guidelines and resources are needed to facilitate rationalisation and deprescribing of drug treatmentsfor older patients in the last days of life. The small sample sizemakes the relationship between most variables insignificant;however, data extraction is still ongoing in hospices.