Drug use at the end of life in older adults.

OBJECTIVE: To investigate symptom-relief and comorbid drug uses at the end of life for older people with different dying trajectories (cancer, organ failure, frailty and sudden death) in Taiwan. METHODS: In a retrospective observational study of older people aged 65 years and older who died in hospitals between 2008 and 2012, we used NHIRD to measure numbers, incremental changes and determinants of symptom-relief and comorbid drug use in the last month of outpatient care and last hospitalisation before death. RESULTS: We included 59 407 older adults (cancer 37%, organ failure 26%, frailty 35% and sudden death 2%) who died in hospitals for this study. In the last hospitalisation before death, individuals who died of cancer received greatest number of symptom-relief drugs (mean: 4.65, [SD 2.77]) and increased most the average change in the number of symptom-relief drug use (+1.60; SD 3.36). However, individuals who died of organ failure received the highest number of comorbid drugs (mean 2.88, [SD 1.95]) and also increased most the average change in the number of comorbid drug use (+0.17; SD 2.28) at last hospitalisation. Different dying trajectories were key determinants of receiving symptom-relief and comorbid drugs in our study. CONCLUSIONS: Our study suggests that the drug use of older adults at the end of life in the cancer group is different from that in the organ failure and frailty groups. Policymakers and health professionals should consider the different strategies to optimise drug use for older people with different dying trajectories near their end of life.

Medication appropriateness for patients with dementia approaching the end of their life.

AIM: To examine medication use among patients with dementia towards the end of their life and to evaluate the appropriateness of medication use by using a nationwide database. METHODS: Using Taiwan's National Health Insurance Research Database, we identified 6532 people with dementia that died between 2008 and 2012. For each person with dementia, data of medication use in the last month of outpatient setting (vs -12th month [baseline]) and last hospitalization (vs -3rd hospitalization [baseline]) before death were retrieved for study. The medications of interest were selected according to a consensus recommendation, which included five categories defining their appropriateness (i.e. always, sometimes, rarely and never appropriate, as well as no consensus). Multivariable logistic regression was carried out to analyze the determinants for use of "never appropriate" medications. RESULTS: Approximately 10% of the study participants were prescribed medications categorized as "never appropriate" in the last month of life in the outpatient settings, which was significantly lower than their baseline (-12th month: 17.5%; P < 0.0001). A similar pattern was identified in the last hospitalization before death. Older age was associated with a lower likelihood of being prescribed "never appropriate" medications (age 75-84: aOR 0.34 [0.29-0.41], P < 0.0001; age ≥85: aOR 0.34 [0.28-0.40], P < 0.0001). In contrast, patients with a history of diabetes mellitus (aOR 1.31 [1.10-1.55], P = 0.0018) were associated with a higher likelihood of being prescribed "never appropriate" medications. CONCLUSIONS: This is the first study to sophisticatedly describe medications use, particularly according to their appropriateness for palliative care, in Asian people with dementia at the end of their life. Approximately 10% of all patients were prescribed "never appropriate" medications at the end of their life, which deserves further study to evaluate the clinical impact of the quality of care. Geriatr Gerontol Int 2017: 17 (Suppl. 1): 65-74..