The Drug Burden Index in Home Medicines Review: the development and implementation of a strategy to improve prescribing and clinical outcomes in older adults

Abstract

Background Anticholinergic and sedative medications are commonly used in older adults (aged 65 years or older) and are associated with a range of medication-related problems. The Drug Burden Index (DBI) is a pharmacological risk assessment measure of the cumulative exposure to medications with anticholinergic and sedative properties. Increasing DBI has been associated with poorer physical function, falls, frailty, hospitalisation and mortality. In a pilot randomised clinical trial study, a DBI report was provided to GPs to assess the impact of providing information about potential harms associated with prescribing anticholinergic and sedative medications in older adults. The intervention observed a conservative reduction in anticholinergic and sedative medication exposure but the study was not powered to assess the impact on clinical outcomes. Home Medicines Review (HMR) is an Australian government-funded medication review model in the primary care setting that aims to maximise patient benefits from medications and prevent medication-related problems. The model involves a collaborative process involving the accredited pharmacist (AP), general practitioner (GP) and the patient. However, there is a lack of data on whether HMR improves prescribing and clinical outcomes in older adults using anticholinergic and sedative medications. Computerised Clinical Decision Support Systems (CCDSS) are designed to generate actions intended to improve clinical decision making and optimise health outcomes. There is evidence of CCDSS improving anticholinergic or sedative prescribing, but there is limited evidence on improving clinical outcomes in older adults. To date, a DBI CCDSS has not been developed nor implemented in the HMR setting to assess the impact on prescribing anticholinergic and sedative medication, and associations with clinical outcomes in older adults. Aims The primary aim of this thesis was to develop a CCDSS to provide information on DBI designed for general practitioners and pharmacists. Specific aims were to: i) evaluate the knowledge of pharmacists on calculating the DBI; ii) develop, validate and evaluate the CCDSS to calculate and report the DBI; and iii) explore the perspectives of healthcare practitioners (HCP) on prescribing and deprescribing anticholinergic and sedative medications in older adults, and implementation of a report on DBI into practice. The secondary aim of this thesis was to implement the DBI into practice. Specific aims were to: i) assess the DBI CCDSS as a clinically feasible and useful tool for prescribing and deprescribing anticholinergic and sedative medications in older adults; ii) establish whether use of the DBI CCDSS intervention with the HMR service is more likely to decrease the use of anticholinergic and sedative medication use than usual care (HMR); and iii) compare and assess changes of important clinical outcomes with changes in DBI over time. Methods Investigation of pharmacists’ knowledge of the DBI An intervention was implemented that included designing education surrounding the DBI and its application in practice, and assessing knowledge of pharmacists in the form of a continuing professional development (CPD) education article with four multiple-choice questions (MCQs). De-identified demographic data on participants were collected and analysed using descriptive statistics. Development and Validation of a CCDSS to calculate the DBI The DBI Calculator© application, based on the published formula and Australian registered prescribing information was built and designed using Microsoft Access 2010. Accuracy testing involved the comparison of DBI scores computed manually (gold standard) with those computed using The DBI Calculator© for 25 published medication review case studies. Agreement was measured using Cohen’s Kappa statistic. To evaluate the CCDSS application in practice, ten APs completed a performance test and usability survey. Participants were timed performing a DBI calculation with the application for a fictitious patient case. The interface, functionality, clinical applications and user satisfaction of the application were rated using 5-point Likert-type scales. Evaluation of The DBI Calculator© report in clinical practice This was a qualitative study using focus groups with purposive samples of APs, GPs, and specialist physicians. Participants were also asked to comment on a sample DBI report (produced by The DBI Calculator©) of a hypothetical patient, and its potential role in practice. The discussions were audio recorded, transcribed verbatim, and thematically analysed to derive conceptual domains. QSR NVivo Version 10 was used for data management. Implementation of the DBI in HMR: a feasibility study An interventional implementation study was conducted. APs were recruited to participate. Each AP was educated on medication use in older adults and implementation of the DBI into HMR practice, and given access to The DBI Calculator© web-based software to generate the DBI report for inclusion in HMR reports for GPs. APs recruited patients (≥65 years) who were referred to them for HMRs. Patients were sent a letter with information about their DBI exposure, and a prompt to visit their GP to discuss their medication management options. GPs, APs and patients were asked to evaluate the DBI report. A medication inventory and data on clinical outcomes (cognition, physical function and sleepiness) were collected from patients at the time of the HMR and at 3-months. Data were analysed using quantitative and qualitative methods, descriptive statistics and regression analyses. Results Investigation of pharmacists’ knowledge of the DBI Multiple choice questions were completed by 2522 pharmacist participants (approximately 9.3% of Australian registered pharmacists): 97.9% of participants successfully completed the CPD assessment (answered three or four out of four MCQs correctly), and 76.5% of participants achieved a perfect score (answered four MCQs correctly). The question that required calculation of the DBI for a fictional patient was answered correctly least often (81.8% of participants answered this question correctly). Development and Validation of a CCDSS to calculate the DBI Excellent agreement between the manual and CCDSS application were observed (Cohen’s Kappa 0.95). In the evaluation phase, the average time (min:sec ± SD) for participants to complete the task was 7:20 ± 1:45 and The DBI Calculator© was considered useful by 80% of pharmacists for recognising anticholinergic and sedative medications in practice. Evaluation of The DBI Calculator© report in clinical practice Several enablers and barriers to deprescribing anticholinergic and sedative medications in older adults were identified. The most noteworthy barrier to deprescribing related to devolving responsibility. Predominately APs expressed frustration surrounding disregard by GPs of their recommendations to deprescribe anticholinergic and sedative medications in older adults. General practitioners expressed that deprescribing should be conducted by specialist physicians, and vice versa. The DBI report addressed some of the identified barriers to deprescribing anticholinergic and sedative medications in older adults. The healthcare practitioners also identified several opportunities and considerations for implementing the DBI report in practice, mainly highlighting that anticholinergic and sedative medications are not the only high-risk medications used in older people. Implementation of the DBI in HMR: a feasibility study The majority of respondents amongst patients (80.8%), APs (88.9%) and GPs (83.3%) considered the DBI report very useful or somewhat useful. Regarding the feasibility of the DBI report as a risk assessment tool within the HMR service, 89% of APs and 67% of GPs agreed that it would be feasible. At three months, the median DBI for patients significantly decreased from 0.82 to 0.67 (p=0.014). Additionally, of patients with a DBI>0, 36.4% had their DBI score decrease, and 6.1% had their DBI score increase. There were no significant changes in clinical outcomes such as physical function, cognition and sleepiness. Discussion Overall, this thesis demonstrated that the DBI can be operationalised into a CCDSS and implemented into the HMR service to reduce the burden of anticholinergic and sedative medications in older adults. The DBI Calculator© has the potential to be integrated with other risk assessment measures, interventions and tools as a strategy to improve prescribing of high-risk medications in older adults and related clinical outcomes.