Psychotropic medicine utilisation in older people: A pharmacoepidemiological approach

Abstract

Psychotropic medicines are defined by World Health Organisation Collaborating Centre for Drug statistics Methodology as psycholeptics (central nervous system calming effect) and psychoanaleptics (central nervous system arousing effect). Psychotropic medicine utilisation has been associated with unwanted outcomes and adverse events like falls, hospitalisations, increased length of hospital stay, morbid conditions and mortality in older people, aged 65 years and over. Older individuals disproportionately use psychotropic medicines in greater numbers and at higher rates compared to their younger counterparts. The overarching aim of this thesis was to examine psychotropic medicine utilisation, adherence and persistence using real world population data of older people, aged 65 years and over. Additionally, the study was intended to monitor and assess quality use of antipsychotic medicines in residential aged care. Population-based evidence of overall psychotropic medicine utilisation was explored using pharmacoepidemiological approach and quality use of medicines assessment to make inductive arguments and ascertain findings. This work was achieved using available electronic healthcare databases and national collections as sources for population data in New Zealand. The quality use of antipsychotics survey involved participants across the country who were nurse managers employed in aged care facilities. Temporal trends in overall psychotropic utilisation increased by 22.5 % (159.6 to 195.4 DDD/TOPD) between 2005 and 2013 in New Zealanders, aged 65 years and over, and was significant when compared internationally. Psychotropic medicine utilisation based on therapeutic class varied and had dissimilar utilisation across geographical areas known as health jurisdictions or mesh blocks (ranged 7.0 to 74.0 %). Between 2005 and 2013, a significant antidepressant prevalence ratio increase (OR 1.27 (95 % CI: 1.22, 1.33)) was observed. Also, temporal trends in concomitant donepezil utilisation (n=9684) with any beta-adrenoceptor blocker significantly declined OR 0.51 (95 % CI: 0.48, 0.54) in older people over a three-year study period. Time-to-first discontinuation (TTFD) rates in donepezil new users (n=1999) was associated with adherence (TTFD was 2.2 times earlier in non-adherent compared to adherent group) or persistence (decreased by 19.0 % between 1st and 6th visit) measures and proportion of donepezil dispensings decreased (81.0 to 32.5 %). Similarly, TTFD rates in low dose second generation antipsychotic (SGA) new users (n=30297) was shortest for risperidone 101.3 (95 % CI: 85.0, 117.7; p=0.03) in days, and had the highest TTFD risk (HR 0.54) among SGAs compared to clozapine. TTFD risk in SGA new users was associated with adherence (risk was 3.6 times in non-adherent compared to adherent group). In addition, adherence was measured using variable medication possession ratio (range 0.9 to 1.1) and persistence (70.2 to 91.3 %) was measured across the study duration. Glycaemic control monitoring in new users of second-generation antipsychotics (n=25,603) in older people was consistent with guidelines for those monitored at baseline (pre-post bpac 2007 proportional change -2.78 (95 % CI:-7.28, 2.25); p=0.28) but was not significant for 60 days and later at 180 days. Quality use of antipsychotic medicines in New Zealand was dissimilar across residential aged care facilities (100 respondents from 318 facilities) by dementia status (over 60 % of participants reported common use of non-pharmacological methods in managing challenging behaviours, while 45% of participants cited administering an antipsychotic medicine for dementia). The implication of findings from these investigations provide population-based evidence which informs clinical practice and influence therapeutic or formulary decisions to help optimise the choice of psychotropic medicine treatment used in geriatric care or improve health outcomes in older people. In conclusion, this work demonstrates well-designed non-randomised studies using pharmacoepidemiological approach including the new-user design. In these population-based investigations, psychotropic medicine utilisation was quantified, measures of association between treatment discontinuation and health outcomes were ascertained, and quality use of antipsychotic medicines was assessed.