Polypharmacy and potentially inappropriate prescribing in community dwelling people living with Alzheimer's dementia

Abstract

Older adults are the biggest consumers of medications worldwide(1), however they are underrepresented in the clinical literature. This is true also for people living with dementia who are often excluded from clinical trials of the medications frequently prescribed(2). People living with dementia have higher levels of comorbidity than those without and significant polypharmacy associated with these co – morbidities (3,4). They are particularly vulnerable to both the cognitive and non cognitive side effects of medication and require unique consideration when prescribing(5). Chapter one of this thesis reviews the unique pharmacokinetics and pharmacodynamics associated with ageing and Dementia. There are a number of existing prescribing tools available to guide prescribing in older people. Chapter two describes my systematic review exploring the existing prescribing tools for older people and their applicability for use in those living with dementia. This analysis was guided by an expert consensus group. We found significant heterogeneity among the prescribing tools included. There are few tools that are designed with consideration for people living with dementia. We support further research and the design of a single internationally accepted prescribing tool for older people to include specific considerations for people with dementia across the spectrum of the disease. Following the extensive literature review, systematic review and expert opinion consensus I investigated prescribing practices in a cohort of people with mild to moderate Alzheimer’s disease (AD). The cohort investigated were participants in a large, investigator led, international randomised control trial ‘Nilvadapine in mild to moderate Alzheimer disease’ (NILVAD)(6,7) . This study and study population are described in detail in Chapter 3. In Chapter 4, I describe analysis of this cohort of people with mild to moderate AD for potentially inappropriate prescriptions (PIM). Using the STOPP/START (8) prescribing tool I examined the prevalence and factors associated with ongoing PIM use, in addition to the effects of this at 18 months of PIM use. Overall, over half (55.8%; 250/448) of participants were using a potentially inappropriate medication. Of those prescribed a PIM over half were prescribed more than one PIM. The likelihood of being prescribed a PIM increased with total medication burden and was associated with both adverse events and serious adverse events, although not with worsening cognitive decline or dementia severity. PIM use was associated with unscheduled GP visits and hospitalisation. In Chapter 5, I describe further analysis investigating a number medications associated with increased risk of adverse events when used in people living with dementia but frequently prescribed. Benzodiazepines (BDZR) were the most frequently prescribed PIM in our cohort. We examined benzodiazepine use for cognitive and non cognitive side effects in the group. There was no significant difference detected on cognitive scores at 18 month follow up for BDZR users compared to non users. There was however increased incidence of delirium and falls in BDZR users. Antipsychotics were also identified as a frequently prescribed PIM, they were also identified by the experts in the systematic review as requiring significant consideration when prescribing in dementia. We examined their use in our cohort. Ongoing use of antipsychotic medication was associated with greater cognitive decline at 12 and 18 months assessed by Alzheimer 10 Disease Assessment Scale - Cognitive Subsection (ADAS-Cog) score. There was also greater progression of AD on Clinical Dementia Rating – sum of boxes (CDR-sb )and Disability Assessment for Dementia (DAD) at 12 and 18 months. Cholinergic burden is an important consideration in people with AD (9). In chapter 6 we examine this in our cohort of people with mild to moderate AD. Over one quarter of participants were prescribed a medication with potential or definite anticholinergic properties at baseline (27.9%, n = 142). Higher ACB scores were significantly associated with greater Disability Assessment for Dementia (DAD) scores. The DAD score reflects functional ability, a relevant marker when considering adverse effects of medication. Finally, I examined the use of statins in our cohort. Statins are one of the most frequently prescribed medications worldwide(10) and they were frequently prescribed in our cohort. The existing literature surrounding statin use in people living with AD is conflicting (11-13 ) both in their role in prevention but also their use in those with established cognitive impairment or dementia. This is further complicated by the FDA blackbox warning on temporary cognitive impairment associated with statins (14). In our cohort the ongoing use of statins was not associated with cognitive decline or worsening of dementia. Furthermore, use of these medications was not associated with an increased risk of adverse events, serious adverse events or unscheduled healthcare utilisation. Chapter 7, the final chapter describes conclusions, limitations, implications for future research and my own reflections on how this work will impact my work as a practicing Geriatrician.