A clinical and economic evaluation of medication reviews conducted by pharmacists for community-dwelling Australians

Introduction: There is a high prevalence of drug-related problems (DRPs) in community-dwelling residents which results in increased morbidity, mortality and healthcare expenditure. The Australian government has reimbursed pharmacists to identify and prevent or resolve DRPs through Home Medicines Reviews (HMRs) in these patients since 2001. Funding for the HMR program was based on studies undertaken prior to its implementation that indicated HMRs would be cost-effective due to reductions in drug costs and health service utilisation. Since the introduction of the program, there has been limited Australian research into the economic outcomes of HMRs. Furthermore, the results of international studies have questioned the clinical and cost-effectiveness of pharmacist-led medication reviews. The overall aim of the research described in this thesis was to investigate the clinical and cost-effectiveness of HMRs. To achieve this aim, the following objectives were formulated and addressed: * to investigate novel methods for assessing the clinical and cost-effectiveness of HMRs; * to investigate the characteristics of the DRPs identified in HMRs, including the drugs and conditions involved, and the recommendations made to resolve or prevent DRPs; * to estimate the clinical effectiveness and cost-effectiveness of HMRs; and * to investigate potential avenues to optimise the clinical effectiveness and cost-effectiveness of HMRs. Method: A methodology that used expert opinion to predict the outcomes of HMRs was developed, based on previous research of interventions undertaken in community pharmacies. The development of this methodology necessitated two studies to be undertaken whereby the healthcare system costs and quality of life effects of numerous common clinical conditions resulting from medication use were estimated using a combination of expert opinion and literature values. To evaluate HMRs, an observational cohort study was conducted across all states in Australia. Pharmacists accredited to perform HMRs submitted a random sample of HMRs that they had undertaken in 2008. Information from the HMR referral, report and outcomes of the HMR were classified according to standardised systems and entered into an electronic database for analysis. A panel of experts reviewed a sample of the HMRs and estimated the clinical outcomes of the recommendations made in the HMRs. A cost-consequence and cost-utility analysis was performed to evaluate the cost-effectiveness of HMRs from an Australian government perspective. A sub-study was undertaken that investigated the relationship between cost-effective HMRs and factors that prior research suggested may influence the cost-effectiveness of HMRs. Results: A “consequences table” was developed whereby estimates of healthcare system costs and quality of life effects of 51 common clinical consequences were evaluated at three levels of severity. The estimates of these parameters appeared to be plausible and reasonable, and demonstrated moderate to high validity and reliability where testing was possible. Six hundred and sixty-one HMRs were submitted by one hundred and forty-nine pharmacists. The HMR reports documented 2323 DRPs, of which the most common were Condition not adequately treated (16.5% of DRPs), Therapy required (11.3%) and Toxicity evident (10.6%). The most common DRP was inadequate pain management which was identified in 118 (17.9%) patients. The drug groups most commonly involved in DRPs were antithrombotics, peptic ulcer and oesophageal reflux therapies, and lipid modifying agents. The pharmacists made 2727 recommendations to resolve the DRPs. The most frequently made recommendations included performing laboratory monitoring, commencing a new medication, or ceasing another. Information relating to the outcomes of the recommendations made to resolve the DRPs was available for 66% of the data (1801 recommendations). Of these recommendations, 1565 (87%) required the prescriber to act on the recommendations to implement them. Approximately three quarters of the DRPs documented in the HMRs were potentially resolved or managed. On average, each HMR was estimated to result in a saving to the health system of $85.79, which was insufficient to offset the cost of the HMR ($323.80). Savings resulted from predicted reductions in health resource utilisation (general practitioner and specialist visits, medical investigations and hospitalisations (P<0.001)). A trend in reduced drug costs was also observed (P=0.070). Significant reductions in the risk of arrhythmias, confusion and myopathy were predicted to occur as a result of the HMRs (P<0.001). Quality of life was estimated to improve minimally (0.001 QALYs per patient, P