I

IIG-021 - VALMER (the Economic Value of Home Medicines Reviews) Researchers: Mr. Andrew Stafford

Dr. Peter Tenni Professor Gregory Peterson Associate Professor Chris Doran Mr. William Kelly

THE RESEARCH AND DEVELOPMENT PROGRAM IS FUNDED BY THE AUSTRALIAN GOVERNMENT DEPARTMENT OF HEALTH AND AGEING AS PART OF THE FOURTH COMMUNITY PHARMACY AGREEMENT

FINAL REPORT

II

FINAL REPORT

Acknowledgements Australian Association of Consultant Pharmacy The assistance the Australian Association of Consultant Pharmacy provided with promoting and supporting this study is acknowledged with gratitude.

Project Steering Group The guidance and input from the project steering group is acknowledged with gratitude: Ms. Jane Bowden Council on the Ageing (Tasmania) Inc

Dr. Nicole Hancock Specialist Physician Royal Hobart Hospital

Dr. Richard Ralph General Practitioner Hobart

Associate Professor Christopher Doran Health Economist University of New South Wales

HMR Assessors The participation of the team of expert assessors is also acknowledged with gratitude: Ms. Camille Boland Clinical Pharmacist Royal Hobart Hospital

Dr. Roland McCallum Specialist Physician Royal Hobart Hospital

Mrs. Deirdre Criddle Consultant Clinical Pharmacist Perth

Dr. Richard Ralph General Practitioner Hobart

Dr. Andrew Dawson Pharmacologist/Clinical Toxicologist South Asian Clinical Toxicology Research Collaboration

Dr. Jay Ramanathan Conjoint Lecturer, School of Medicine University of Western Sydney

Dr. Matt Doogue Clinical Pharmacologist/Endocrinologist Flinders Medical Centre, Flinders University

Mrs. Debbie Rigby Consultant Clinical Pharmacist Queensland

Mr. Peter Fowler Senior Clinical Pharmacist Launceston General Hospital

Dr. Pravin Shetty Consultant Physician Kalgoorlie Regional Hospital

Dr. Nicole Hancock Specialist Physician Royal Hobart Hospital

Dr. Farid Taba General Practitioner Perth

Dr. Viney Joshi Medical Superintendent Blackall Hospital

Dr. Robyn Wallace Specialist Physician Calvary Hospital Lenah Valley

Dr. Winston Liauw Medical Oncologist/Clinical Pharmacologist St. George Medical Research Foundation

The assistance of Thameemul Ansari Jainullabudeen in preparing this final report is also gratefully acknowledged.

III

FINAL REPORT

Acronyms Acronym Explanation

ABDI Australian Burden of Disease and Injury studies

ACE Angiotensin Converting Enzyme

ADE Adverse Drug Event

ATC Anatomic Therapeutic Chemical

CEAC Cost Effectiveness Acceptability Curve

DALY Disability-Adjusted Life Year

DRP Drug-Related Problem

EQ-5D EuroQOL 5D quality of life instrument

GBD Global Burden of Disease study

GP General Practitioner

HMR Home Medicines Review

ICER Incremental Cost Effectiveness Ratio

ICPC2-PLUS International Classification of Primary Care Version 2 PLUS

IQR Inter-quartile range

NSAID Non-Steroidal Anti-Inflammatory Drug

PhARIA Pharmacy Rural and Remoteness Index

PTO Person Trade-Off

QALY Quality Adjusted Life Year

QOL Quality Of Life

RMMR Residential Medication Management Review

SG Standard Gamble

TTO Time Trade-Off

VALMER The Economic Value of HMRs study

Funding notes This report was produced with the financial assistance of the Australian Government Department of Health and Ageing. The financial assistance provided must not be taken as an endorsement of the contents of this report. The Pharmacy Guild of Australia manages the Fourth Community Pharmacy Agreement Research and Development Program which supports research and development in the area of pharmacy practice. The funded projects undertaken by independent researchers and therefore, the views, hypotheses and subsequent findings of the research are not necessarily those of the Pharmacy Guild. Figures in the tables presented in this report have been rounded which may cause minor variances in these tables.

IV

FINAL REPORT

Table of Contents  Acknowledgements.......................................................................................................................................................II Acronyms.....................................................................................................................................................................III Funding notes ..............................................................................................................................................................III Table of Contents ....................................................................................................................................................... IV Table of Figures........................................................................................................................................................... V Table of Tables ............................................................................................................................................................ V  Background & rationale ................................................................................................................................................1 Objectives .....................................................................................................................................................................3 Methodology .................................................................................................................................................................4 

Study design .............................................................................................................................................................4 Steering committee...................................................................................................................................................5 Data classification.....................................................................................................................................................5 Economic evaluation.................................................................................................................................................5 Analysis.....................................................................................................................................................................7 Ethical approval ........................................................................................................................................................7 

Results ..........................................................................................................................................................................8 Pharmacist demographics ........................................................................................................................................8 Patient demographics ...............................................................................................................................................8 Drug-related problems identified...............................................................................................................................9 Economic evaluation...............................................................................................................................................11 

Discussion...................................................................................................................................................................17 Limitations...................................................................................................................................................................21 Conclusion ..................................................................................................................................................................23 References..................................................................................................................................................................24 Appendix I  Print advertisements .............................................................................................................................1 Appendix II  Consequences tables ...........................................................................................................................2 Appendix III  Derivation of Consequences table values ..........................................................................................19 Appendix IV  Screen of HMR assessment system...................................................................................................34 Appendix V  Results of pharmacist survey..............................................................................................................35 Appendix VI  Patient medical conditions ..................................................................................................................39 Appendix VII  Medications taken ...........................................................................................................................40 Appendix VIII  Nature of DRPs identified ...............................................................................................................42 Appendix IX  Uptake of recommendations...............................................................................................................43 Appendix X  Data costing tables .............................................................................................................................44 Appendix XI  Examples of highly cost-effective HMRs ............................................................................................47 Appendix XII  Re-sampled data.............................................................................................................................52 

Table of Figures Figure 1 - Study flow chart............................................................................................................................................8 Figure 2 - Distribution of changes in drug costs. ........................................................................................................12 Figure 3 - Distribution of changes in QOL and health care ........................................................................................14 Figure 4 - Cumulative health resource savings resulting from HMRs ........................................................................14 Figure 5 - Cost-effectiveness acceptability curves generated using re-sampled data. ..............................................16 Figure 6 - Example of project promotion ......................................................................................................................1 Figure 7 - EQ-5D descriptive system .........................................................................................................................22 Figure 8 - Utilities study data entry screen .................................................................................................................24 Figure 9 - Results of utility study for consequences experienced at a mild level of severity......................................25 Figure 10 - Results of utility study for consequences experienced at a moderate level of severity...........................25 Figure 11 - Results of utility study for consequences experienced at a severe level of severity ...............................26 Figure 12 - Example of assigning health-resource utilisation to consequences ........................................................28 

V

FINAL REPORTFigure 13 - Total costs and duration of consequences at mild severity level .............................................................30 Figure 14 - Total costs and duration of consequences at moderate severity level ....................................................31 Figure 15 - Total costs and duration of consequences at severe severity level.........................................................32 Figure 16- Screenshot of system used by assessors to evaluate HMRs...................................................................34 Figure 17 - Percentage of pharmacists according to their main state of practice ......................................................35 Figure 18 - Frequencies of contact with GP and receipt of management plans following HMRs ..............................37 Figure 19 - Pharmacist perception of value of communication with GPs and medication management plans..........38 Figure 20 - Re-sampled incremental cost-effectiveness ratios for HMRs vs usual care: baseline analysis ..............52 

List of Tables Table 1 - General patient demographics ......................................................................................................................8 Table 2 - DRPs identified according to D.O.C.U.M.E.N.T. classification system .........................................................9 Table 3 - Recommendations made according to D.O.C.U.M.E.N.T. classification system........................................10 Table 4 - Relative changes in drug costs according to drug class. Negative numbers indicate savings ...................11 Table 5 - Comparison between sampled HMRs and the VALMER dataset ...............................................................12 Table 6 - Analysis of costs and QOL differences for baseline assumptions ..............................................................13 Table 7 - Savings resulting from HMRs (entire sample vs upper quartile). ................................................................15 Table 8 - Results of scenario analyses.......................................................................................................................15 Table 9 - Description of consequences ........................................................................................................................5 Table 10 - Consequences table values for duration, utilities and GP visits ...............................................................10 Table 11 - Consequences table values for specialist services, investigations and hospitalisations ..........................18 Table 12 - Values assigned to PROMISe II consequence of "Seizures"....................................................................19 Table 13 - Examples of utilities for various medical conditions reported in previous studies.....................................20 Table 14 - Examples of QOL instruments ..................................................................................................................22 Table 15 - Comparison of some weights derived in the VALMER study and reference values.................................26 Table 16 - DRG data for "cerebrovascular event" ......................................................................................................27 Table 17 - Calculation of ALOS and hospitalisation costs for “cerebrovascular event” from DRG data ....................27 Table 18 - Comparison between PROMISe II and VALMER consequences table values.........................................29 Table 19 - General pharmacist demographics ...........................................................................................................36 Table 20 - Pharmacist views of differences between DVA DAA HMRs and "normal" HMRs ....................................37 Table 21 - Most common medical conditions presented according to ICPC2-PLUS chapters ..................................39 Table 22 - Most commonly taken medications grouped by ATC chapter...................................................................41 Table 24 - Most common DRPs according to associated medications ......................................................................42 Table 25 - Potential resolution of types of DRPs........................................................................................................43 Table 26 - Costs and QOL data for assessed HMRs .................................................................................................46 

1

FINAL REPORT

Background & rationale The cost of medicines accounts for a large and growing proportion of healthcare expenditure in Australia. Approximately AU$87 billion was spent on healthcare in the twelve months between July 2005 and June 2006. Of this, AU$11.5 billion was on medications, an increase of 1.6% over the previous year.1 Intuitively, medications are prescribed to ultimately improve health and health outcomes. However there is substantial evidence of a high prevalence of medication-related illness in the community.2, 3 Approximately 10% of Australian patients attending general practice report experiencing at least one adverse drug event (ADE) in the previous 6 months.4 For elderly patients and those taking multiple medications, the risk of ADEs is substantially higher. Whilst many ADEs are minor, it has been estimated that there are over 80 000 medication-related hospital admissions each year in Australia. Research suggested that approximately 50% of these hospitalisations are potentially preventable.5 Consequently, there has been considerable interest in interventions that may reduce the incidence of ADEs and consequently health expenditure. One such intervention is pharmacist- conducted medication review. Medication review is a structured evaluation of a patient’s medicines, aimed at optimising medication efficacy and minimising the number of drug-related problems (DRPs).6 Programs of pharmacist-conducted medication reviews have now been implemented in various formats in New Zealand, Europe, the USA and South America.7, 8 In Australia, the Residential Medication Management Review (RMMR) program for residents of aged care facilities was introduced in 1997. Four years later, funding for the Home Medicines Review (HMR) program commenced, whereby pharmacist-conducted medication reviews became widely available to patients living at home. As at June 2009, over 225 000 HMRs have been conducted since the program’s inception. The HMR program is currently funded under the Fourth Agreement between the Commonwealth of Australia and The Pharmacy Guild of Australia. A HMR involves collaboration between a patient’s general practitioner (GP) and their preferred community pharmacy. Following the receipt of a referral from the patient’s GP via the pharmacy, a specially accredited pharmacist interviews the patient either at home or a mutually convenient site (such as the community pharmacy). The pharmacist reviews their medication regimen, and provides the general practitioner with a report. The general practitioner and consumer then agree on a medication management plan to address any issues identified in the HMR. Patients may receive one HMR per year, unless there is a compelling need for an earlier review such as a substantial change in their medication or recent hospital discharge. HMRs are designed to assist patients to maximise the benefits of their medication regimen and minimise DRPs such as ADEs. The objectives of HMR are to:

• achieve safe, effective, and appropriate use of medications by detecting and addressing DRPs that interfere with desired patient outcomes;

• improve the patient's quality of life and health outcomes using a best practice approach, that involves cooperation between the general practitioner, pharmacist, other relevant health professionals and the patient (and where appropriate, their carer);

• improve the patient's, and health professional’s knowledge and understanding about medications; and • facilitate cooperative working relationships between members of the health care team in the interests of

patient health and well being.6 Given these objectives, it is apparent that the benefits of HMRs should not be restricted to the prevention or resolution of ADEs. A high prevalence of under-prescribing in community-dwelling patients has been identified in several studies,9, 10 even amongst those taking numerous medications for other medical conditions.11 Past studies of HMRs indicate that medication reviews may also provide an opportunity to address and resolve such issues relating to under-use of medication.9 A HMR may be offered to any patient for whom the patient’s GP feels it is clinically necessary to ensure quality use of medicines or address patient's needs. There are guidelines to aid GPs (and other health professionals) to identify patients who would be expected to benefit from a HMR. These include patients who:

• are currently taking 5 or more regular medications; • are taking more than 12 doses of medication/day; • have had significant changes made to the medication regimen in the last 3 months; • are taking medication with a narrow therapeutic index or medications requiring therapeutic monitoring; • are experiencing symptoms suggestive of an adverse drug reaction; • are experiencing sub-therapeutic response to treatment with medicines; • have suspected non-compliance or inability to manage medication related therapeutic devices; • have difficulty managing their own medicines because of literacy or language difficulties, dexterity problems

or impaired sight, confusion/dementia or other cognitive difficulties;

2

FINAL REPORT• are attending a number of different doctors, both GPs and specialists; and/or • have recently been discharged from a facility/hospital (in the last 4 weeks).12

Prior to the introduction of the HMR program, several studies were conducted to establish the benefits of HMRs. In broad terms, all of these studies identified that HMRs were appreciated and valued by most participating patients, pharmacists and general practitioners.13-17 Additionally, HMRs were found to result in the identification and resolution of DRPs.13, 14, 16, 18 Research undertaken subsequent to the introduction of the program has confirmed these benefits.18-20 Investigations into the economic benefits of HMRs have been less conclusive. Two studies found that HMRs reduced medication costs14, 16, whilst two other studies did not.13, 17 Several studies have used expert opinion to assess the potential benefits of HMRs with regard to utilisation of other health resources (such as hospitalisation and GP consultations).13, 16 These studies concluded that HMRs would most likely reduce health-resource utilisation. However these predictions were not realised in a randomised control trial which found no significant differences in quality of life, number of hospital admissions, number of non-admission hospital services, cumulative number of bed days or number of GP visits between patients who received HMRs and the control group.17 Several studies of pharmacist-conducted medication reviews from the United Kingdom have also failed to identify benefits in terms of quality of life or reduced health-resource utilisation.21-24 An evaluation of the HMR program conducted in 2005 concluded that “there is a need for collection and consolidation of information on health outcomes for patients who have had an HMR, for the purposes of ongoing valuation and to strengthen the evidence base….”.18 The VALMER study (the Economic Value of Home Medicines Reviews) was conceived to clarify the benefits of HMRs by investigating the DRPs identified in them and evaluate the potential economic outcomes of resolving these issues. The study was an 18-month project conducted by the Unit for Medication Outcomes Research and Education, School of Pharmacy, University of Tasmania, with collaborators from the Australian Association of Consultant Pharmacy and the University of New South Wales. This report presents an overview of the study and its major findings.

3

FINAL REPORT

Objectives The fundamental aim of the VALMER study was to assess the economic effects of HMRs. To achieve this aim, two broad objectives for the study were formulated:

• To quantify the following aspects of HMRs: o the number and type of DRPs identified; o the drug groups often associated with DRPs; o the recommendations made by the pharmacists to resolve the DRPs; and o the rate of uptake of the recommendations.

• To evaluate the potential outcomes of HMRs in terms of: o the number of days of “poor health” saved; o days in hospital prevented; o consultations with GP and/or specialists prevented; o investigations prevented; and o total financial costs to the health system.

4

FINAL REPORT

Methodology Study design In designing the VALMER study, consideration was given to the results of prior studies which investigated the outcomes of medication reviews. The study was initially conceived as a randomised controlled trial, powered sufficiently to detect changes in multiple health resource parameters over a period of at least 12 months. The largest study of the Australian medication review model identified that the majority of cost savings were most likely to result from reductions in hospitalisation in small numbers of patients.13 In consideration of this, the number of subjects required for a study to detect a 10% reduction in hospitalisation in older adults was calculated. Burgess et. al identified that the annual hospitalisation rate due to adverse drug reactions in elderly Australians was approximately 12.9 per 1000 person-years in 2002.2 A RCT would therefore require over 1500 patients in both the intervention and control groups to detect a 10% reduction in hospitalisation at a power of 90% (P<0.05). Such a study would be of significantly greater scale than what the VALMER budget and timeframe permitted. The research team was thus faced with the decision of whether to undertake a small RCT of short duration and limit the outcomes measured, or to assess the HMRs in another way. Past studies of medication reviews in the community setting suggested that the benefits of HMRs other than reducing hospitalisations are generally subtle, and current measures lack sufficient sensitivity to detect the benefits of HMR in small scale studies.25 Given these limitations, a randomised control design for the VALMER study was discarded in favour of an observational cohort design. For the economic analysis, the decision was made to adapt a technique developed to model interventions made in community pharmacies to model the outcomes of HMRs.26 The model utilised expert opinion to compare the effect of the intervention (the HMR) to the same patient not receiving the HMR (no intervention). There were three types of data required to undertake this analysis:

• the HMR referral from the GP, • the HMR report from the accredited pharmacist who performed the review, and • data regarding the outcomes of the recommendations made in the HMR report.

All data relating to the HMRs analysed was submitted by the accredited pharmacist who performed the HMR. Accredited pharmacists were asked to submit details pertaining to HMRs that they performed between March and November of 2008. To reduce the possibility of selection bias, participating pharmacists were requested to submit details of sequentially the next five HMRs they performed after enrolling in the study. Pharmacists who participated in the study by submitting HMRs were also asked to complete a survey to compare their characteristics to those of the greater body of accredited pharmacists. The VALMER study was advertised to accredited pharmacists using a variety of media. Advertisements were published in Australian Pharmacist, Australian Journal of Pharmacy and The Accredited Pharmacist magazines (see Appendix I for an example), and a media release was distributed by the Pharmacy Guild of Australia. The project was also promoted at the three largest accredited pharmacist continuing-education events in 2008. These events were

• the accredited pharmacist forum at the Australian Professional Pharmacy Conference, Gold Coast, Queensland (27 March 2008),

• AACP Consultant Pharmacy Clinical Seminar, Adelaide (29 May - 1 June 2008), and • the accredited pharmacist forum at the Pharmacy Australia Congress, Perth (24 October 2008).

The website www.valmer.com.au was also used to promote the study, and serve as a resource for participants to download the materials required to participate in the study. It was envisaged that outcomes data would be readily available from the medication management plan formulated by the GP following the HMR. If no medication management plan was received, then the pharmacist was asked to contact the GP directly and follow up the outcome data. Previous research involving HMRs suggested that the mean number of DRPs that would be identified in the VALMER study would be between two and five.9, 17, 19 Anecdotal evidence from experienced accredited pharmacists indicated that the acceptance of recommendations made in HMRs diminishes as the number of recommendations increases. Additionally, we proposed that most pharmacists prioritise the DRPs identified in HMRs from most to least clinically relevant. It was therefore decided to collect data regarding the outcomes of the first three recommendations made by the pharmacist in the HMR report. It was planned that these data would be used to provide an indication as to the outcomes of the recommendations made in the HMR report with the greatest potential to improve patient health and reduce health resource utilisation.

5

FINAL REPORTThere were two primary outcome measures. The first was the type and frequency of the DRPs and resolving recommendations identified in HMRs. The second outcome measure was an estimate of the quality of life and economic effects of resolving the DRPs, which was extrapolated to provide an estimate of the value of HMRs. The VALMER study was conducted between March 2008 and September 2009.

Steering committee The VALMER study was a collaborative, multidisciplinary project involving pharmacists, GPs and specialist physicians. A steering committee was established for the project in early 2008 to ensure that the interests of all stakeholders in the HMR process were represented, and to provide input for the project’s promotion. As such, the committee consisted of pharmacists active in performing medication reviews, a GP, a specialist physician and a consumer representative.

Data classification All data submitted for the study was entered into an electronic database (Access 2007, Microsoft Corporation, Redmond, Washington). All patient diagnoses and medical conditions were classified using the International Classification of Primary Care Version 2 PLUS (ICPC2-PLUS, Family Medicine Research Centre, University of Sydney). The patients’ medications were recorded using Anatomic Therapeutic Chemical (ATC, World Health Organisation) coding. The economic modelling technique that was used in the VALMER study relied upon the DRPs that were identified in the HMRs being classified according to a standardised system. Whilst most hierarchal DRP classification systems would be able to be used with the model, currently there is no uniform system employed by pharmacists for classifying DRPs. The University of Tasmania’s D.O.C.U.M.E.N.T. classification system was developed in conjunction with the economic model, so was utilised in the VALMER study to classify the interventions made in the HMRs.27 This system satisfies the major requirements for a DRP classification system defined in a review of such systems,28 and has been used in the assessment of HMRs in a previous study.19

Economic evaluation Background A health sector perspective was adopted for the VALMER study, with the major focus on the impact for government as third-party funder. A time horizon for both benefits and costs of 12 months was assumed. This was a conservative estimate to reflect the intention that the patient management plan is to be reviewed by the GP and pharmacist at 12 monthly intervals. A combination of empirical and modelled data was used to perform the economic analysis. In this evaluation, the net position associated with conducting each HMR was defined according to Equation 1:

Net position = Cost of HMR + (Medication cost after HMR - Medication cost before HMR) + (Healthcare costs after HMR - Healthcare costs before HMR)

Equation 1 - Net cost of each HMR

The effects of HMRs on productivity costs were not considered in this study as the technique used to value the consequences of the HMRs is not amenable to their inclusion. The general age of the patients reviewed in HMRs is greater than 65 years,29 and hence the omission of productivity costs was considered unlikely to substantially confound the results of the study. HMR costs The cost of each HMR was $323.80 which included both payment to the accredited pharmacy ($183.60*) and GP (Medicare Benefits Schedule (MBS) item 900, value $140.20). Information was requested from Medicare Australia to account for the rural loading for pharmacies classified as remote (Pharmacy Accessibility/Remoteness Index of Australia (PhARIA) categories 2 to 6); however Medicare Australia was unavailable to provide this data.

* as at 1 November 2008 

6

FINAL REPORT Drug costs Calculating the change in medication costs in the VALMER study was performed by costing the changes to each patient’s medication regimen that occurred as a result of the HMR. The perspective of the VALMER cost-analysis was from that of the Australian government, so only Pharmaceutical Benefits Scheme (PBS)-listed items were included in the medication cost analysis. Medications subsidised by the PBS were costed using the dispensed price for maximum quantity in the November 2008 Schedule of Pharmaceutical Benefits issued by the Australian Government Department of Health and Ageing. All items were costed on the basis of quantity of drug used per month. Medications for which the quantity required for a one month supply was different to the PBS maximum quantity were costed by calculating the fraction of the PBS maximum quantity used each month and multiplied by the dispensed price. As documentation regarding the frequency of most medications taken on a when-required basis was generally poor, these medications were all costed according to one PBS supply lasting two months. In calculating the changes in drug costs resulting from the HMRs, we assumed that any changes to the patients’ medication regimens would be sustained for 12 months following the HMR. Health resource costs A sample of 180 HMRs was assessed by a panel of GPs, medical specialists and clinical pharmacists to estimate utilisation of health resources other than medications. These resources were GP and specialist consultations, medical investigations and cost and duration of hospitalisations. The 180 HMRs were selected from the HMRs submitted for the study using stratified random sampling. All of the HMRs submitted for the study were initially stratified according to the amount of detail provided in the HMR referral. One hundred and eighty HMRs were then randomly selected from the total HMRs according to the proportions in the stratification. Assessors were invited to participate in the study based on their experience in primary care medicine and workplace experience. Sixteen assessors consisting of eight consultant physicians, four GPs and four clinical pharmacists were recruited for the analysis. Each assessor reviewed ninety HMRs, sixty of which were reviewed by all assessors and a further thirty that were reviewed by three other panellists. For the assessment, each assessor was provided with a pre-formulated list of clinical consequences that were linked to values for quality of life and health resource utilisation. Each consequence was described by three levels of severity, ranging from mild to major illness. The list of consequences and the values assigned to each are shown in Appendix II. The methodology used to derive these values is discussed in Appendix III. For each recommendation made by the pharmacists to resolve the DRPs identified in the HMRs, the assessors selected the most likely (rather than the most valuable) consequences from the list and assigned probabilities for the consequence occurring at each level of severity. Assessors were asked to assign probabilities of each consequence occurring both before and after the HMR. In addition to predictions of the outcomes of the HMRs, each assessor was asked to provide their opinion as to the probability of another healthcare professional identifying the same DRP in the following 12 months. This value, termed attribution, therefore discounted the involvement of the HMR in the identification and resolution of the DRPs. The assessment of the HMRs was performed using an online system built for the study. Appendix IV shows an example of the data entry screen. Each assessor was provided with a training manual and given the opportunity to discuss the assessment process with a member of the research team at the commencement of their assessments. Each assessor evaluated their allocated HMRs independently. For each DRP that was assessed, we used the mean probabilities assigned by the panellists of the consequence occurring before and after the HMR. Assessors were also able to indicate that the resolution of the DRP would most likely result in no consequence. We then used the cost estimates assigned to each consequence (Appendix II) to generate an estimate of the health care utilisation and quality of life (utility) for each DRP with and without the HMR occurring. To account for the possibility of a recommendation not being implemented and the DRP not being resolved, we used the outcomes data to apply an average resolution rate to the value of each DRP according to its subtype (termed uptake, Table 25, Appendix IX). To obtain the total value of each HMR, the value of each of the DRPs identified within it were summed. The cost and median length of hospital admission was derived from 2006-7 AR-DRG version 5.1 values for public hospitals Australia-wide.30 The cost used for a GP consultation (MBS item 23) was $33.55. Pathology items were costed according to the appropriate MBS item number. The cost used for an initial specialist visit (MBS item 104) was $79.05. For subsequent specialist visits, the cost used was $39.70 (MBS item 105). In addition to financial costs, each consequence incorporated an estimation of changes in health-related quality of life (QOL) occurring subsequent to the mediation review. The utilities used in the VALMER study were primarily derived from literature

7

FINAL REPORTsources.31, 32 Utilities for consequences for which no literature value could be sourced were derived specifically for the study. Six specialist physicians used the EuroQOL (EQ-5D) descriptive system to assign utilities to these consequences.† The methodology for obtaining the utilities used in the study is documented in Appendix III. Assumptions A number of key assumptions were made in determining the value of the HMRs in this study using this methodology:

• that the basis for determining the value of the consequences is an accurate estimate of both the health resource utilisation and quality of life effects incurred by each consequence,

• that the experts’ selection of consequences and assignment of pre- and post-HMR probabilities to them is appropriate and accurate,

• that the assignment of attribution values by the experts are appropriate and accurate, • that the uptake values obtained from the outcomes data are appropriate and accurate, • that the extrapolation of the assessed HMRs to the whole sample is appropriate, and • that the sample of HMRs in the VALMER study is representative of HMRs nation-wide.

Sensitivity and uncertainty analyses There is inherent uncertainty in estimating the clinical outcomes resulting from HMRs using expert opinion. To test the assumptions made regarding the attribution and uptake variables, a sensitivity analysis was conducted. The value used in the baseline scenario was the mean probability for each consequence selected by the assessors, discounted by these two factors. We therefore calculated an attributed potential value and an absolute potential value by sequentially removing these two discount factors:

• The attributed potential value was calculated by assuming that every recommendation made in the HMRs was implemented (i.e. not just those acted on by the GP). This was performed by assigning an uptake value of 100% (Table 25, Appendix IX).

• To calculate the absolute potential value of the HMRs, we further assumed that all of the value identified from the panel analysis resulted only from the HMR by assigning an attribution value of 100% to all HMRs. This scenario was the least conservative estimate of the value of HMRs, and demonstrated the absolute value of addressing the DRPs identified. In terms of methodology, this scenario is the closest to previous studies of HMRs that have used expert opinion to estimate their value.

The summary statistic calculated was the incremental cost-effectiveness ratio (ICER, or cost per QALY gained). The mean ICER does not yield any information as to the degree of uncertainty of the estimate. To derive an estimate of the uncertainty, the data were re-sampled 1000 times using a nonparametric bootstrap approach to generate a mean cost and QALY gain from the HMRs and the resulting ICERs were calculated. The uncertainty around the mean is represented by a Cost Effectiveness Acceptability Curve (CEAC), which should be interpreted as the probability that the ICER is below a threshold of acceptance.

Analysis Normally distributed variables (such as patient age and number of drugs) were summarised using means and standard deviations, and comparisons were made using t-tests. Nonparametric data were summarised using proportions for categorical data, and medians and IQRs for continuous data. Comparisons were made using χ2-tests for categorical data. Wilcoxon signed-rank tests were used to compare continuous data values before and after HMRs. Data were analysed using the statistical package SPSS 15 (SPSS Inc, Chicago, USA), and re-sampled using @RISK 5.5 (Palisade Corporation, New York, USA).

Ethical approval The VALMER study was approved by the University of Tasmania Human Research Ethics Committee (HREC 9360).

† a methodology similar to that used to derive the reference utilities

8

FINAL REPORT

Results Data collection for the VALMER study commenced on 27 March 2008, and concluded on 14 November 2008. During this time, 203 pharmacists completed an enrolment form, indicating that they would be prepared to submit HMRs for the study. At the conclusion of the data collection period, 149 pharmacists had contributed 675 HMRs. Nine HMRs were excluded from the sample as they were conducted prior to the project start date. Another five HMRs contained no data regarding the patient’s medications and were also excluded, resulting in a final sample of 661 HMRs. Of these 661 HMRs, outcomes data were available for 560 (84.7%) of them. This is shown in Figure 1.

Figure 1 - Study flow chart

Pharmacist demographics The results of the pharmacist survey are presented in Appendix V. Of the 149 pharmacists who submitted HMRs, 117 (78.5%) undertook the survey. All were accredited to perform medication reviews with the Australian Association of Consultant Pharmacy. Their general characteristics were consistent with the results of two previous surveys of accredited pharmacists conducted in 2005 (858 accredited pharmacists)18 and 2008 (560 accredited pharmacists).33

Patient demographics The general characteristics of the patients reviewed in the HMRs are summarised in Table 1. Female patients contributed a majority (57.9%) of the study population. There was a broad range of patient ages, although patients aged 65 years or older accounted for over 85% of the patient group. Over half of the patients (53.1%) had between five and ten diagnosed medical conditions; seventeen HMR referrals did not contain any information regarding the patient’s medical conditions.

Characteristic (±SD, [range]) Male

(n=278, 42.1%)

Female

(n=383, 57.9%)

Total

(n=661)

Mean age – years 74.7 ± 10.6 [30 - 96] 76.9 ± 10.2 [31 - 98] 76.0 ± 10.4 [30 - 98]

Number (% of total) aged <65 years 47 (7.1%) 46 (7.0%) 93 (14.1%)

Number (%of total) aged 65 to 75 years 197 (29.8%) 269 (40.7%) 466 (70.5%)

Number (%of total) aged >75 years 34 (5.1%) 68 (10.3%) 102 (15.4%)

Mean diagnosed medical conditions 8.4 ± 4.9 [0 - 33] 9.1 ± 5.2 [0 - 29] 8.9 ± 5.1 [0 - 33]

Total (incl. when required preparations) 11.1 ± 4.4 [2 - 28] 12.2 ± 4.5 [4 -30] 11.8 ± 4.5 [2 - 28] Mean number of medications Regular 9.1 ± 3.7 [1 - 22] 9.9 ± 3.9 [2 - 22] 9.6 ± 3.8 [1 - 22]

Did not submit HMRs (n=54 pharmacists)

Did not meet inclusion criteria (n=9 HMRs) Data unusable or incomplete (n=5 HMRs)

HMR-accredited pharmacists enrolled in study (n=203 pharmacists)

HMR data submitted by 149 pharmacists (n=675 HMRs)

HMR data submitted (n=661 HMRs) Outcomes data available (n=560 HMRs) Outcomes data unavailable (n=101 HMRs)

Assessed by expert panel (n=180 HMRs)

Not assessed by expert panel (n=481 HMRs)

Survey of participating pharmacists (n=117 pharmacists)

9

FINAL REPORTTable 1 - General patient demographics

The HMR referrals documented a total of 5846 medical conditions. Appendix VI lists the most common diagnosed medical conditions in the VALMER sample, grouped by their ICPC2-PLUS chapter. The most common diagnoses were cardiovascular conditions, such as hypertension, ischaemic heart disease and atrial fibrillation. Correspondingly, the prevalence of metabolic diseases that are significant risk factors for these illnesses, such as diabetes mellitus and hyperlipidaemia, was also high. Musculoskeletal diseases were also common diagnoses, of which osteoarthritis and osteoporosis were the most prevalent conditions. The patient sample was documented as taking a total of 7790 medications, including when-required and complementary preparations. The medications most commonly taken are shown in Appendix VII. Over two thirds of the patients were taking a lipid modifying agent, an angiotensin-converting enzyme (ACE) inhibitor or angiotensin-II receptor antagonist and/or an antiplatelet agent. This is unsurprising given the high prevalence of cardiovascular disease amongst the patients. The prevalence of musculoskeletal disease in the patients was also high. Correspondingly many patients were taking analgesics, the most common of which was paracetamol. The proportion of patients taking non-steroidal anti-inflammatory agents (NSAIDs) was substantially lower than those taking paracetamol, and the two most commonly used NSAIDs were COX-II specific agents. Almost 60% of the patients were taking a medication to treat a disorder related to gastric-acid, such as proton-pump inhibitors.

Drug-related problems identified The HMR reports documented 2323 actual or potential DRPs, equating to 3.5 (±1.8, range 0-13) DRPs per HMR. No DRPs were identified in eighteen (2.7%) HMRs. The frequency of the DRPs identified according to the D.O.C.U.M.E.N.T. classification system are shown in Table 2. The most common medical conditions and medications involved in DRPs are presented in Appendix VIII.

DRP Type DRP Subtype Number (%) of total DRPs Number (%) of Patients

Drug selection 511 (22.0%) 423 (64.0%) Drug interaction 189 (8.1%) 142 (21.5%) Contraindications apparent 135 (5.8%) 115 (17.4%) Unnecessary therapy/no apparent current indication 132 (5.7%) 111 (16.8%) Duplication 36 (1.5%) 36 (5.4%) Other drug selection problem 15 (0.6%) 15 (2.3%) Wrong dosage form 3 (0.1%) 3 (0.5%) Wrong drug 1 (0.0%) 1 (0.2%) Over or underdose prescribed 198 (8.5%) 185 (28.0%) Dose too high 117 (5.0%) 107 (16.2%) Dose too low 45 (1.9%) 43 (6.5%) Other dose problem 36 (1.5%) 35 (5.3%) Compliance & concordance 321 (13.8%) 298 (45.1%) Taking too little 133 (5.7%) 119 (18.0%) Other compliance problem 119 (5.1%) 112 (16.9%) Difficulty using dosage form 40 (1.7%) 38 (5.7%) Patient using out of date medication 16 (0.7%) 16 (2.4%) Taking too much 13 (0.6%) 13 (2.0%) Untreated indications 638 (27.5%) 500 (75.6%) Condition not adequately treated 373 (16.1%) 288 (43.6%) Therapy required 262 (11.3%) 209 (31.6%) Other untreated indication problem 3 (0.1%) 3 (0.5%) Monitoring 238 (10.2%) 196 (29.7%) Laboratory monitoring 216 (9.3%) 174 (26.3%) Non-laboratory monitoring 19 (0.8%) 19 (2.9%) Other monitoring problem 3 (0.1%) 3 (0.5%) Education or information 44 (1.9%) 40 (6.1%) Disease management or advice 15 (0.6%) 11 (1.7%) Confusion about therapy 9 (0.4%) 9 (1.4%) Demonstration of device 9 (0.4%) 9 (1.4%) Patient drug information request 6 (0.3%) 6 (0.9%) Other education or information problem 5 (0.2%) 5 (0.8%) Non-clinical 55 (2.4%) 55 (8.3%) Other non-clinical problem 14 (0.6%) 14 (2.1%) Dietary problem 13 (0.6%) 13 (2.0%) Smoking problem 13 (0.6%) 13 (2.0%) Weight management problem 11 (0.5%) 11 (1.7%) Alcohol problem 4 (0.2%) 4 (0.6%) Toxicity or adverse reaction 318 (13.7%) 261 (39.5%) Toxicity evident 247 (10.6%) 194 (29.3%) Toxicity caused by drug interaction 39 (1.7%) 36 (5.4%) Toxicity caused by dose 27 (1.2%) 26 (3.9%) Other toxicity/adverse effect problem 5 (0.2%) 5 (0.8%)

10

FINAL REPORTDRP Type DRP Subtype Number (%) of total DRPs Number (%) of Patients

TOTAL 2323 (100.0%)

Table 2 - DRPs identified according to D.O.C.U.M.E.N.T. classification system

The most frequently identified type of problem involved Untreated indications. This includes medical conditions that were either sub-optimally managed or not managed at all. Of these inadequately managed conditions, pain was the most frequently identified issue, followed by risk factors for cardiovascular disease such as hypertension, hyperlipidaemia and type 2 diabetes. Drug- selection issues, such as potentially interacting drugs or patients taking medications that were potentially contraindicated, were also frequently identified. The proportion of DRPs related to Education or information appears to be low; however these figures only represent the information contained in the HMR reports. It is likely that substantially more education and counselling was provided during the HMR interview than these figures state. With regard to the drugs most frequently implicated in the DRPs, inadequate use of analgesics (predominantly paracetamol) was the most common. Other frequently identified DRPs included potential drug interactions involving anticoagulants (primarily warfarin), ACE inhibitors and angiotensin II antagonists, HMG-CoA reductase inhibitors and diuretics. The drugs most frequently implicated in adverse drug reactions were ACE inhibitors (primarily causing cough), lipid lowering therapy (muscle pain) and calcium channel blockers (pedal oedema). Medications for obstructive airways disease were also frequently implicated in DRPs, with patients prescribed inadequate therapy, not using enough of their medications or unable to use the dosage form prescribed for them. The pharmacists made 2610 recommendations to resolve the DRPs, equating to approximately one recommendation per DRP. No recommendation was made to resolve 117 (5.0%) of the DRPs. The frequencies of the types of recommendations made by the pharmacists are shown in Table 3. The most commonly made recommendation was to perform laboratory monitoring, such as electrolyte levels or therapeutic drug monitoring. The medications most frequently requiring monitoring involved lipid modifying agents (lipid profiles), ACE inhibitors and angiotensin II antagonists (potassium and creatinine levels), and proton pump inhibitors (vitamin B12 levels).

Recommendation type Subtype

Number (%) of total recommendations

Monitoring Laboratory monitoring 470 (18.0%) Non-laboratory monitoring 100 (3.8%) For follow-up Follow-up by prescriber 133 (5.1%) Follow-up by another 47 (1.8%) Provision of education or information Patient/carer education 163 (6.2%) Prescriber information 13 (0.5%) Compliance assistance 101 (3.9%) Drug or dosage modification Dose increase 226 (8.7%) Dose decrease 191 (7.3%) Drug cease 297 (11.4%) Drug start 458 (17.5%) Formulation change 62 (2.4%) Dose schedule change 72 (2.8%) Drug switch 272 (10.4%) Other therapy change 5 (0.2%) TOTAL 2610 (100%) No recommendation No recommendation necessary 11 DRPs No recommendation made 106 DRPs

Table 3 - Recommendations made according to D.O.C.U.M.E.N.T. classification system

Outcome data were available for 560 (84.7%) of the HMRs that were submitted for the study. This equated to 1769 (67.8%) of the total number of recommendations made in the HMRs, and 1565 (67.4%) of the DRPs identified. A summary of the outcomes data is presented in Table 25 (Appendix IX). The outcomes data were grouped into two classes according to whether or not any action was taken by the GP to address the DRPs identified. DRPs with the only outcome classified as “Not accepted” were considered to be “unresolved”; any DRP that resulted in some resolving action was considered to be “potentially resolved”. These data indicate that over 80% of the DRPs identified in the HMRs would potentially have been resolved in some way. Education and information, and Compliance and concordance DRPs had a high rate of potential resolution. DRPs relating to too-high doses and

11

FINAL REPORTmedications without indications had relatively low rates of potential resolution compared to the remaining DRP types.

Economic evaluation Drug cost changes The mean monthly PBS drug cost per patient at the time of the HMR was $224.63 ± $175.51 (not including when-required medications). Cardiovascular medications accounted for the greatest proportion of the total monthly PBS drug cost, costing over twice as much per month as the next most costly medication group. For the 560 HMRs for which outcomes data were available, 247 resulted in a change in monthly costs of medications. These data indicated that the HMRs resulted in an overall reduction in the monthly regular medication cost to the PBS of $936.50. On average, this equated to a monthly drug cost saving to the PBS of $1.67 per HMR. The net effect of the different drug classes on the monthly medication costs is shown in Table 4, and the distribution of drug cost changes in Figure 2. As expected from the high prevalence of undertreated pain in the patient sample, the drug class most influential in increasing the monthly drug costs was analgesics. Drugs for airways diseases also contributed substantially to increases in drug costs, owing to the relatively high price of preparations containing tiotropium and combination corticosteroid/beta-agonist preparations. The net changes in drug costs resulting from modifications to cardiovascular drugs were small, despite the frequency of DRPs identified involving inadequate therapy of cardiovascular conditions. Additional costs due to dose increases or commenced medications in some patients were balanced by dose reductions or drug cessation in others. The drug class most influential in reducing the monthly drug cost was psychoanaleptics (which includes benzodiazepines and antidepressants). DRPs involving drugs for acid-related disorders (primarily proton-pump inhibitors) were common, and contributed substantially to reductions in the monthly drug costs. Notably, despite the relatively high prevalence of patients with inadequate management of hyperlipidaemia, changes to the cost of the medications to treat these disorders resulted in a small net cost saving.

ATC Level 2 Name Total cost change

Median change (IQR) in monthly drug cost per HMR

Number of HMRs where costs changed

Number of patients taking (n=661)

Analgesics $523.58 $8.11 ($10.21) 49 457 Drugs for obstructive airway diseases $399.25 $20.75 ($70.61) 18 223 Beta blocking agents $54.65 $3.87 ($18.95) 9 217 Agents acting on the renin-angiotensin system $48.99 $5.80 ($16.21) 26 461 Drugs for acid related disorders $30.29 $8.46 ($29.45) 3 384 Antiepileptics $22.43 $4.68 ($9.97) 5 65 All other therapeutic products $22.19 $7.40 3 92 Calcium channel blockers $6.50 $7.54 ($27.11) 12 222 Vaccines $3.20 $1.6 2 1 Thyroid therapy $1.63 $1.63 1 79 Antianemic preparations $1.29 $0.64 ($4.31) 2 132 Antidiarrhoeals, intestinal anti-inflammatory/anti-infective agents

-$0.10 -$0.10 1 48

Cardiac therapy -$2.76 -$1.25 ($15.85) 8 292 Corticosteroids for systemic use -$2.86 -$2.86 1 64 Mineral supplements -$7.75 -$7.75 1 181 Antihypertensives -$8.13 -$8.13 1 50 Urologicals -$12.95 -$5.18 ($2.59) 5 51 Lipid modifying agents -$20.99 $8.16 ($58.07) 21 480 Diuretics -$22.17 -$2.79 ($2.91) 9 229 Ophthalmologicals -$35.45 -$7.79 ($10.06) 5 178 Antibacterials for systemic use -$44.17 -$22.30 ($26.91) 3 58 Sex hormones and modulators of the genital system -$54.65 -$20.10 ($28.61) 4 48 Antiemetics and anti-nauseants -$57.83 -$9.59 ($4.79) 5 53 Antithrombotic agents -$74.84 $1.93 ($3.86) 30 456 Antigout preparations -$79.35 -$2.16 ($4.26) 14 92 Psycholeptics -$140.08 -$4.83 ($5.86) 21 166 Drugs for treatment of bone diseases -$152.69 -$46.87 ($52.91) 6 116 Drugs used in diabetes -$292.97 -$3.30 ($52.65) 15 192 Anti-inflammatory and anti-rheumatic products -$323.65 -$14.43 ($20.44) 19 200 Drugs for acid related disorders -$337.20 -$16.77 ($20.04) 37 384 Psychoanaleptics -$379.92 -$4.90 ($23.13) 27 209

Table 4 - Relative changes in drug costs according to drug class. Negative numbers indicate savings

12

FINAL REPORT

Figure 2 - Distribution of changes in drug costs. Negative numbers indicate savings; bars indicate standard error of the mean

Clinical outcomes A sample of 180 reviews was selected from the 661 VALMER HMRs for analysis by medication therapy experts to determine their probable clinical outcomes. The 661 HMRs were initially stratified according to the amount of detail provided in the HMR referral. Of the 661 HMRs submitted for the study, no DRPs were documented in of eighteen of them. As the economic analysis methodology involved assessment of individual DRPs, these eighteen HMRs were excluded from the dataset, leaving 643 HMRs. One hundred and eighty HMRs were then randomly selected from the 643 HMRs according to the proportions in the stratification. A comparison between the characteristics of the panel-evaluated HMRs and the complete VALMER dataset is shown in Table 5.

Characteristic (±SD) Sample (n=180) Total (n=661) P-value Comments

Number (%) of males 72 (40.0%) 278 (42.1%) 0.568 χ2=0.326, df=1*

Mean age (years) 77.0 ± 10.6 76.0 ± 10.4 0.228 t(179)=1.210†

Mean number of medical conditions 9.0 ± 5.4 8.9 ± 5.1 0.782 t(179)=0.277†

Mean number of total medications 12.8 ± 4.9 11.8 ± 4.5 0.006 t(179)=2.774†

Mean number of DRPs identified per HMR 4.2 ± 1.8 3.5 ± 1.8 <0.001 t(179)=4.191† *χ2 test for goodness of fit †one sample t-test

Table 5 - Comparison between sampled HMRs and the VALMER dataset

From these results, it was apparent that the patients in the sampled HMRs varied little from the complete VALMER dataset with regard to gender, age and medical conditions. Although the sample was taking a statistically significant different number of medications (approximately one more medication than the greater dataset), the overall significance of this difference was considered to be minimal. The difference in the mean number of DRPs per HMR was most likely due to the exclusion of the eighteen HMRs that did not identify any DRPs. The assessment of the HMRs by the expert panel was performed between March and July 2009. At the end of the assessment period, fourteen of the sixteen assessors had completed their assigned cases. Two consultant physicians withdrew from the study in late June due to an inability to meet the study’s deadline for the HMR assessments. One of these physicians had assessed thirteen HMRs, and the other had not commenced their assigned cases. The assessors were permitted to select and assign probabilities to as many, or as few, potential consequences as they felt appropriate for each DRP. Agreement between the assessors regarding the probable outcomes of the HMRs was poor. There was agreement between at least two of the assessors in 257 (52.8%) of the 487 DRPs assessed in 139 (72%) of the 180 HMRs assessed. Baseline scenario

13

FINAL REPORTThe results of the analysis of the baseline scenario are shown in Table 6. The assessment indicated that the HMRs would result in a significant decrease in healthcare utilisation costs, and an improvement in QOL. However, in many HMRs the absolute value of these reductions in health resource utilisation was insufficient to offset the cost of the HMR, and the ICER was $64 939 (95% CI $48 407 to $80 170) per QALY gained. Fifty five HMRs (31%) were highly cost effective at a cost-effectiveness threshold of $50 000 per QALY gained. Using a threshold of $150 000 per QALY gained, 75 HMRs (42%) would be considered cost effective. Detailed costing tables are presented in Appendix X, and examples of highly cost effective HMRs are shown in Appendix XI. Figure 20 (Appendix XII) shows the ICERs for the 1000 re-samples of the complete dataset.

Median predicted value (IQR) per HMR Median change (IQR) Parameter Before HMR After HMR Wilcoxon signed rank test Total saving (180 HMRs)

Number of GP visits 1.88 (3.20) 1.42 (2.47) 0.27 (0.83) 112.6 z=-8.00 P<0.001

Cost of GP visits $63.00 ($107.20) $47.51 ($82.76) $9.16 ($28.01) $3778 z=-8.00 P<0.001 Number of specialist visits 0.46 (0.79) 0.31 (0.61) 0.07 (0.19) 29.69 z=-7.61 P<0.001

Cost of specialist visits $29.63 ($49.86) $19.61 ($37.65) $4.64 ($13.93) $1892 z=-7.76 P<0.001 Cost of investigations $35.78 ($61.25) $27.79 ($44.79) $4.02 ($15.34) $2061 z=-7.59 P<0.001 Days in hospital 0.16 (0.40) 0.13 (0.31) 0.02 (0.07) 11.7 z=-5.15 P<0.001

Cost of hospitalisation $148.68 ($382.44) $134.09 ($350.67) $15.73 ($69.06) $11747 z=-5.10 P<0.001 Total health resource costs $299.83 ($570.48) $251.23 ($496.57) $44.05 ($211.24) $23 086 z=-6.79 P<0.001 Disability 0.011 (0.025) 0.008 (0.019) 0.001 (0.0045) 0.54 z=-5.22 P<0.001 ICER (including cost of HMR and changes in drug costs) $64 939

Table 6 - Analysis of costs and QOL differences for baseline assumptions (positive values indicate costs savings or improvement in QOL)

In summary, in the 12 months following each HMR, on average, the savings per HMR were: • 0.63 GP visits, saving $20.99; • 0.16 specialist visits, saving $10.51; • $11.45 in reduced medical investigations; • 0.065 days in hospital, saving $65.26; and • $20.04 in drug costs.

Additionally, the average gain in QOL was 0.003 QALYs. The total of these savings ($128.25) was insufficient to offset the cost of the HMR ($323.80). The savings generated and QOL effects between individual HMRs varied substantially. This is illustrated in Figure 3. In the majority of the HMRs, improvements in QOL or reductions in health care utilisation were noted. In a small number of HMRs, the experts indicated that deterioration in QOL would potentially have resulted from the interventions made in them. This was primarily due to the commencement of medications where adverse effects resulted in a loss of QOL that outweighed the potential gain in QOL in the following 12 months (e.g. statins or additional anti-hypertensives). Substantial negative consequences were likely to occur in only one HMR.‡ One assessor commented,

“Interventions likely to cause net harm were rare.” -Specialist physician

‡ This was a complex HMR involving a patient with a history of cerebrovascular attacks and ischaemic heart disease taking two antiplatelet agents and atorvastatin. The pharmacist recommended ceasing one of the antiplatelet agents which the assessors indicated was inappropriate.  

14

FINAL REPORT

Figure 3 - Distribution of changes in QOL (left) and health care utilisation (right). Positive numbers indicate savings or improved QOL; bars indicate standard error of the mean

A majority of the health resource savings was driven by a small number of HMRs that potentially resulted in substantial decreases in serious adverse events. As shown in Figure 4, the upper quartile of 45 HMRs was responsible for over 80% of the total health resource savings in the 180 HMRs.

Figure 4 - Cumulative health resource savings resulting from HMRs (straight line indicates savings in upper quartile)

In this quartile of the HMRs (based on total savings), the average saving per HMR was considerable, and completely offset the cost of the HMR. In the 12 months following these HMRs (n=45), the average saving was $632.15, resulting from the following reductions:

• 1.78 GP visits, saving $59.58; • 0.53 specialist visits, saving $32.82; • $35.39 in reduced medical investigations; • 0.27 days in hospital, saving $262.97; and • $241.38 in drug costs.

Additionally, the average gain in QOL in these HMRs was 0.011 QALYs. This is shown in Table 7.

15

FINAL REPORT

Median change (IQR) Parameter Entire sample (n=180) Upper quartile* (n=45) Number of GP visits 0.27 (0.83) 1.18 (1.31)

Cost of GP visits $9.16 ($28.01) $39.52 ($43.92) Number of specialist visits 0.07 (0.19) 0.35 (0.44)

Cost of specialist visits $4.64 ($13.93) $20.90 ($27.97) Cost of investigations $4.02 ($15.34) $22.95 ($31.82) Days in hospital 0.02 (0.07) 0.03 (0.07)

Cost of hospitalisation $15.73 ($69.06) $161.01 ($212.08) Annual medication costs $0.00 ($21.00) $104.64 ($369.00) Total health resource costs $44.05 ($211.24) $418.95 ($535.22) Disability 0.001 (0.0045) 0.006 (0.015) ICER $64 939 -$22 811

*based on savings

Table 7 - Savings resulting from HMRs (entire sample vs upper quartile). Positive numbers indicate savings or improved QOL

Scenario analyses As discussed previously, the baseline scenario was a highly conservative estimate of the economic benefits of HMRs. The estimate was sequentially discounted by two separate factors (uptake and attribution, see Methods). To investigate the degree of influence of these factors, a scenario analysis was undertaken whereby both discount values were removed. The results of the sensitivity analyses are presented in Table 8, and the CEACs are shown in Figure 5.

Median change (IQR) Analysis Utility Total costs ($) ICER†

Average saving per

HMR Baseline 0.0010 (0.0045) $44.05 ($211.24) $64 939 $128.55

z=-5.22 P<0.001 z=-6.79 P<0.001 0.0012 (0.0057) $50.14 ($243.78) $47 707 $151.31 Attributed potential value (assumes implementation of

every recommendation) z=-5.18, P<0.001 z=-6.75, P<0.001 0.0028 (0.0116) $112.39 ($410.50) $7 616 $267.21 Absolute potential value (assumes implementation of

every recommendation and 100% attribution) z=-5.42, P<0.001 z=-6.91, P<0.001 *Wilcoxon signed rank tests †includes cost of HMR and drug cost changes

Table 8 - Results of scenario analyses (positive values indicate costs savings or improvement in QOL)§

An increase in the probability of HMRs being highly cost effective at a threshold of $50 000 per QALY was identified in both the attributed potential value and absolute potential value scenarios. As the outcomes data indicated a relatively high rate of resolution of most DRP types, increasing the resolution rate in the attributed potential value did not result in a substantial reduction in the ICER. However, removing the “attribution” discount factor, and hence assuming that all of the potential savings were due solely to the HMR (as opposed to another health professional, the absolute potential value) resulted in a substantial increase in health system savings and improved quality of life. Consequently, in this scenario the ICER was substantially reduced, with a 100% probability of cost effectiveness at a threshold of $50 000 per QALY.

§ It should be noted that the values for the first three scenarios are the medians of the differences in utility and health-resource utilisation between the “before” and “after” scenarios (rather than the difference in the medians). This reflects each assessor providing a probability for consequences occurring before and after the HMR, essentially an indication of relative risk decrease or increase for the consequences. 

16

FINAL REPORT

Figure 5 - Cost-effectiveness acceptability curves generated using re-sampled data. Line indicates cost effectiveness threshold of $50 000 per QALY

17

FINAL REPORT

Discussion This report presents the results of the VALMER study which investigated the DRPs identified in HMRs, and the potential value of the outcomes of recommendations made by pharmacists to resolve them. Based on the number of HMRs assessed, to our knowledge this study is the largest investigation into HMRs that has been undertaken since the implementation of the program in 2001. The major findings of this study of over 600 HMRs include an assessment of the characteristics of patients currently receiving HMRs, classification of the nature and frequency of the DRPs identified in HMRs, and an estimation of the economic consequences of addressing these DRPs.

1. Patients review and DRPs identified The demographics of the patients referred for the HMRs in our study suggest that they would be considered to be at high risk of ADEs. Risk factors for ADEs commonly identified in previous research include increasing age, multiple chronic medical problems, taking multiple medications, and taking “high risk” medications.4 These high-risk medications include drugs affecting the cardiovascular system (such as diuretics, ACE-inhibitors and digoxin), antithrombotic agents (such as warfarin), musculoskeletal drugs (commonly NSAIDs), oral hypoglycaemic agents, and psychotropic medications such as anticholinergics, benzodiazepines, antipsychotics, sedatives and hypnotics.4 Most of the patients in our study were elderly, taking more than five medications on a regular basis (many of which were “high risk”) and were diagnosed with multiple chronic medical conditions. Therefore on face value, our results indicate that patients receiving HMRs appear to be those whom guidelines indicate would benefit from the service. Supporting this assumption is that the pharmacists in our study identified a substantial number of DRPs in the HMRs. The mean number of DRPs per HMR (3.5 per HMR) was consistent with the findings of several past Australian studies of medication reviews, which ranged from 2.513 to 8.234 DRPs per HMR. In our study, nearly half of the patients (45%) were experiencing at least one DRP relating to compliance or concordance, and almost as many (39%) were identified as experiencing a suspected or actual adverse effect from their medications. Of even greater prevalence were patients with conditions treated either inadequately or not at all, which was documented in 76% of patients. It may be broadly inferred from these findings that the HMRs addressed the program’s objectives of achieving “safe, effective, and appropriate use of medications by detecting and addressing DRPs that interfere with desired patient outcomes”.6

2. Costs Savings due to HMR Based on probable clinical outcomes, we estimated that HMRs will result in statistically significant savings to the healthcare system in both drug costs and health resource utilisation. The average savings was estimated to be a positive $128.25 per HMR in the 12 months following the review, the majority of which resulted from reduced hospitalisation costs (51%) and GP visits (16%). Though there are no previous studies that have been conducted using our methodology, general comparisons between our findings and previous research can be made. The average economic benefit found in this study ($128.25 per HMR) is relatively small compared to past studies, which have identified average savings ranging between approximately $315 and $400.13, 14, 16, 18, 35: Gilbert et al. ($315), Bennett et al. ($130-$330), Krass et al. ($400), Sorensen et al. ($275). There are several potential reasons for these differences. These include

• differences in the perspective taken for drug cost measurement, • differences in the proportion of DRPs that would achieve drug costs savings, and • different proportions of medical practitioners as expert advisors

a. Perspective taken for measurement of drug costs

In their studies of HMR, Gilbert et al.13 and Sorenson et al.17 only considered drug costs to the PBS and not the costs to patients, which is the perspective taken in our study. Neither Gilbert nor Sorenson identified that HMRs were likely to result in drug cost savings. However Krass et al. costed medications using a combination of the PBS dispensed price and the costs to patients for non-PBS items (it is unclear from their report how Bennett et al. costed medications). It is possible that the major savings reported by Bennett et al. and Krass et al. involved savings to patients rather than the PBS. Had our study also included costs to patients, the average drug cost savings will be likely to be much higher and possibly comparable to the other studies. However given that Bennett et al. reported costs savings across virtually all drug classes, this explanation is unlikely to account for the entirety of the differences.

b. Identification of a low proportion of DRPs that would result in drug costs savings In the current study, the proportion of DRPs involving too-low doses, untreated indications or requiring additional therapy, was approximately 30%. These types of DRPs are less likely to result in HMRs achieving drug costs

18

FINAL REPORTsavings compared to other DRP types, as evidenced by the minimal drop in mean monthly medication costs ($1.67 per patient per month). In a randomised controlled trial, Sorenson et al. identified a similar high proportion of these types of DRPs, and they found no differences between the HMR group and control group in drug costs. In the study by Gilbert et al., proportion of these types of DRPs was approximately 25%, and they reported a slight increase in drug costs. In the study by Krass et al., the proportion of such DRPs was only 1.5% and they found mean monthly medication costs to fall by approximately $20 per patient. Bennett et al. did not report on the types of DRPs identified, but the proportion of such DRPs was likely to be low and they found savings in mean monthly medication costs to be between $10.77 and $27.52 (depending upon whether or not a clinical audit was conducted prior to the medication review). Hence if the current study and the other studies had similar proportion of types of DRPs, the variations in economic benefit of HMRs may have been smaller and possibly comparable.

c. Greater proportion of medical practitioners as expert advisors There are substantial variations in the methods of expert assessment which can partly explain the variation between the average economic benefit found in this study and other studies. The current study had a much greater proportion of medical practitioners as assessors relative to the other studies. We identified that this type of assessor tend to value HMRs less than the pharmacists**. Hence if the current study had a more comparable proportion of medical practitioners and pharmacists to other studies, then the average economic benefit arising from HMRs may be more comparable.

3. Outcomes of HMRs in relation to financial costs to the health system The previous studies of HMRs suggested the cost savings arising from HMRs can offset the cost of HMRs. Though it was not an objective of this study to evaluate this, it should be looked at as broader assessment of outcomes of HMRs in relation to financial costs to the health system. We estimated that the average costs savings due to HMRs was $128.25 per HMR in the 12 months following the HMR, which was lower than the cost of each HMR ($323.80). It would be erroneous to conclude from just this information that HMRs cannot bring about benefits that can offset cost of each HMR due to the limitations resulting from considering only the average cost savings across the sample of HMRs assessed in this study. Furthermore, extending the time horizon for the study beyond 12 months would potentially result in additional savings. Assessment of the HMRs in groups illustrates these limitations. Of the 180 HMRs assessed by the expert panel, there were 29 HMRs (16%) which potentially resulted in cost savings greater than $323.80, which completely offset the cost of HMRs. For this group of HMRs, there is no uncertainty that they generate more benefit than the costs incurred by the health system. Amongst the remaining HMRs, some proportion of them should be able to see cost savings if ambulatory and ED costs were included. These HMRs need to be identified and they too will not impose a burden on the health system. The HMRs that are related to the DRPs involving too-low doses, untreated indications or requiring additional therapy, will likely result in minimal, negligible or even negative cost savings in medication costs in the short term. Since one of the objectives of HMRs is to address the issue of such DRPs, the health system must be willing to accept any potential burden imposed by this group of HMRs. This is because they can potentially general longer term cost benefits. A high proportion of the medications that were commenced as a result of the HMR have proven cost-effectiveness, such as virtually all of the cardiovascular and diabetes treatments. If a longer time horizon were assumed, a greater proportion of the HMRs where these agents were initiated would most likely prove cost-effective. What really needs serious attention is those HMRs which are not in the earlier groups and do not generate cost savings greater than the cost of HMRs. Even amongst these HMRs, in some of the cases there could be benefits arising in the period after 12 months or longer term. In other HMRs, there can be other complexities that may also require to be rectified such as potentially service delivery of HMRs in order to make it cost offsetting. Another issue that needs to be looked at in assessing if benefits of HMRs offset cost of HMRs, is the remuneration to GPs. Currently GPs are reimbursed $140.20 per HMR (43% of cost of HMR). This is significantly higher than levels used prior to the implementation of the program (between $30 and $50 in most studies). This cost has increased at a very significant rate over the years especially relative to the reimbursement received by pharmacists ($100 in 1997 to $183.60 in 2008). Managing that cost is essential to making HMRs cost effective. ** The mean attribution values for the assessed HMRs were 42% (GPs), 49% (specialists) and 72% (pharmacists)

19

FINAL REPORT

4. Outcomes of HMRs

a. Service Delivery It is unlikely that the number of poor HMRs is significantly different between the current study and previous studies. This is because the current study had a large number of pharmacists who participated, and each of them had passed the rigorous AACP accreditation process. Additionally, our results regarding the acceptance of the recommendations made in the HMRs imply that the GPs considered the pharmacist’s findings relevant in many instances. Furthermore, the scenarios tested in the sensitivity analysis indicated that reducing the attribution and uptake discount factors substantially increased the value of the HMRs. Implicit in this finding is that the changes resulting from the pharmacist’s recommendations were appropriate and likely to result in substantial savings. However earlier studies were conducted before the widespread implementation of HMRs and they investigated the ideal service delivery of HMRs. The current study investigated the service delivery of HMRs in the real world. Since they are different, the latter is likely to be picking up the complexities of this relatively newly implemented intervention. That can partly explain the differences in outcomes between current study and previous studies. Therefore optimising the outcomes of HMRs will involve identifying any such complexity and resolving them. This really shows the importance of constantly maintaining service delivery excellence. For instance comments made by two of the assessors who reviewed the HMRs illustrate a need for pharmacists to ensure that the focus of HMRs remains centred on the patient at all times. One assessor commented,

“Too often the pharmacist did not seem to consider the potential impact on the patient, especially (with respect to) QOL issues…medication reviews must be patient centred - otherwise they are of no value to the patient nor GP. There were a lot of theoretical recommendations, which may be of some value but, largely, tell the GP to "suck eggs".”

-Pharmacist

Another assessor indicated that there was an over-reliance on guidelines which do not consider the patient’s co-morbidities or therapeutic objectives.

“Many of these patients were elderly and intensification of treatment might tick the box but might not meet the patient’s goals. Patients do not always want to be treated according to the guidelines as different people place different value on different outcomes.”

-Specialist physician

Another factor that may influence the viability of HMRs is the potential that the “risk factors for medication misadventure” listed in this report’s introduction do not readily identify patients who would be expected to receive cost-effective HMRs. From an economic perspective, a question which intuitively follows is, “Which factors determine whether or not a HMR will be cost effective?”. Whilst no studies have directly addressed this question, some insight is provided by the results of previous research. Several studies have identified that patients who receive HMRs post-hospital discharge are highly likely to benefit from them.36-39 There is also evidence that HMRs may substantially benefit patients with heart failure, as HMRs have been associated with reductions in hospitalisation in these patients.37, 40 We identified a high degree of variability in multiple aspects of the HMR process, including the patients referred, the information contained in the referrals, the accredited pharmacists performing HMRs, the DRPs identified and the recommendations made to resolve them. The factors most influential in determining the value of the HMRs were not clear, although most were likely to influence the outcomes of the HMRs to some degree. Our findings illustrate a need for further research into HMRs to investigate factors influencing their cost-effectiveness, and ways to improve the uptake of the program in patients where they are most cost effective.

5. Cost utility analysis a. Utility analysis

Based on an expert assessment process, we estimated that, on average, HMRs will result in small but statistically significant improvements in quality of life (mean improvement in utility of 0.003). However it is important to recognise that the patients who were referred for HMRs were suffering from multiple co-morbidities, and any improvement in health was likely to be appreciated by them. Our findings that HMRs minimally improve QOL are not unexpected as there is little evidence that medication reviews substantially improve QOL. The only investigation into HMRs that has reported a substantial improvement in QOL was the 2005 evaluation of the HMR program, which reported a gain of 0.119 QALYs per HMR.18 In this study, 50 patients who had recently received HMRs were administered the EQ-5D instrument in recall mode (i.e.

20

FINAL REPORTasked to recall how they felt before the HMR, and compared to how they felt afterwards), a technique prone to recall bias. Of the four studies previously discussed, only Bennett et al. and Sorenson et al. directly examined QOL. Bennett et al. investigated two models of medication review. In both models, QOL was worse three months after the review, although there was no control group in this trial so the significance of this finding is uncertain. Sorenson et al. conducted a randomised control trial, and found no differences in QOL between the intervention and control groups, and a statistically insignificant trend towards improved severity of illness. The results from international studies of medication review have produced similar results.21-24, 41-43 Consideration of the DRPs identified in the HMRs again provides a potential explanation for the lack of substantial change in QOL. The most frequently identified DRP in the HMRs in our study was under-treated pain, and recommendations to introduce or intensify analgesics were common. Intuitively, such interventions would be expected to improve QOL. However, given that the majority of patients in our study suffered from chronic pain conditions such as osteoarthritis, the most common recommendation of increasing the dose or dose frequency of paracetamol may be unlikely to improve a patient’s pain to a point that significant improvements in QOL will be apparent. Commencing more potent analgesics such as opioids may improve pain, but adversely affect QOL in other ways, such as by causing adverse effects including constipation or sedation. In regards to the potential adverse effects identified in the HMRs in our study, those most frequently identified were relatively minor. The most common was cough resulting from ACE inhibitors, followed by muscle pain associated with statins. Resolving either issue was unlikely to substantially improve QOL for most patients.

b. Cost utility analysis The average ICER for HMRs was $64,939 per QALY. Using the WHO threshold of $50,000 per QALY to be considered cost effective, this falls outside the acceptance level. More importantly, this is an average figure whereby there are some HMRs (31%) where the average ICER satisfies the threshold level and some do not. As mentioned in the earlier part of this discussion, if ED costs and ambulatory costs were included, the average cost savings is likely to increase and the average ICER may satisfy the WHO threshold level. There is another broader acceptance criterion where an intervention with an ICER less than less than GDP per capita level is highly cost effective, less than three times of GDP per capita is cost effective, and all else being cost ineffective. By these criteria, the average ICER is cost effective. Importantly, these baseline scenario figures are a highly conservative estimate of the value of HMRs. Assuming the case of attributed potential value whereby every recommendation was assumed to be implemented, the average cost savings per HMR increased to $151.31 and ICER was reduced to $47,707. This is cost effective by WHO threshold level. In the best case scenario of absolute potential value where every recommendation was assumed to be implemented and 100% of the potential beneficial effects†† were attributed to the HMR, the average cost savings was $267.21 and ICER is $7,616 which also meets WHO threshold level. What actually requires attention is not that the average ICER is above WHO threshold level in the baseline scenario. Instead, attention should be focussed on which HMRs (if any) are still not cost effective, even after inclusion of ED costs savings, ambulatory costs savings and costs savings subsequent to the resolution of any service delivery issues.

6. Other issues A final, and perhaps most important question raised by our results relates to the most appropriate outcome measures (and duration of measurement) to assess the benefits of HMRs. We have demonstrated that, by focusing on outcomes such as drug costs to the PBS or health service utilisation, in the short term, the cost savings generated by medication reviews are not substantial in most patients. However, in consideration of the most common type of DRP that was identified in the HMRs, this was somewhat understandable. The under-management of chronic medical conditions such as hypertension, hyperlipidaemia and osteoporosis is unlikely to result in short-term adverse health outcomes. Furthermore, improving the management of these conditions is unlikely to result in short-term health benefits. Nonetheless, the long-term economic benefits of improving the management of these conditions may be substantial, and it is most likely that restricting the time horizon in our study to 12 months greatly limited the potential value of addressing these DRPs. In a recent international literature review of the cost effectiveness of medication reviews, Zermansky and Silcock assert that there is a need for a †† Importantly, this scenario also increased costs resulting from potential detrimental effects resulting from the HMR. Despite this, the benefits overwhelmingly outweighed detrimental effects 

21

FINAL REPORTlarge, controlled, long-term follow-up study of patients who received regular medication reviews.25 Until such a study is conducted, the benefits and cost-effectiveness of the HMR program will remain uncertain.

Limitations The practicalities of this study result in a number of limitations which may impact upon its findings, and our results should therefore be interpreted with caution. The first limitation results from the study’s design and the use of expert opinion to model the outcomes of the HMRs. By utilising expert opinion rather than a controlled study design, the level of evidence provided by the study is comparatively low. The following factors were considered to minimise confounding due to this limitation in the study:

• each expert was an active practitioner and had extensive experience in managing patients similar to those reviewed in the HMRs. Most experts were involved in managing such patients on a regular basis in their usual practice.

• experts were asked to utilise primarily their knowledge of relevant medical evidence relating to the HMRs when they assessed them. Several experts also possessed experience as academics and educators and were very familiar with the medical literature relevant to the HMRs.

• the values assigned to each of the consequences (e.g. hospitalisation costs, utilities etc) were obtained from reference sources wherever possible.

Implicit in the expert assessment process is that assumption that each assessor was able to accurately predict the potential outcomes of the DRPs identified in the HMRs given the information made available to them. The variation between the experts’ opinions demonstrates that this was a difficult task- the results of the sensitivity analysis indicate a substantial difference in value between the baseline scenario and the absolute minimum value. We attempted to overcome this limitation by providing the assessors with an option to not assess a case if they felt that they had insufficient information to do so. This option was used in only 0.016% of the HMR assessment, implying that the assessors were relatively confident assessing the HMRs with the data available. Using a Delphi-type process involving sequential rounds of assessment may have overcome this limitation to some degree; however this was not possible due to a lack of time and resources. It is possible that the use of systematic reviews and meta-analyses may have improved the robustness of some of the results. However for many of the DRPs identified (and indeed the recommendations made to resolve them), studies in elderly patients with numerous co-morbidities such as those reviewed in the HMRs in this study, are very limited. Hence given the broad range of issues identified in such a diverse group of patients, the research team considered it unlikely that using systematic reviews as a primary source for estimating changes in risks resulting from the HMRs was either feasible or likely to substantially improve the robustness of the results. A further limitation resulting from our modelling is that we did not perform any follow up other than the data collected by the reviewing pharmacist. We assumed that any changes to a patient’s drug therapy resulting from the HMR would be sustained for the subsequent 12 months, which may not be the case. Patients may change or discontinue medications several times throughout the following year. Furthermore, our assumptions regarding the use of “when-required” medications may have resulted in an overestimation of the increase in cost of drugs such as analgesics. Nonetheless, our findings regarding changes in drug costs are not dissimilar to those of Sorensen et al. and Gilbert et al., implying that this factor is not a significant confounder. As discussed in the previous section, the 12-month time horizon is likely to result in a second major limitation to the study. Most HMRs are undertaken for patients with multiple chronic conditions. The benefits of commencing or optimising drug therapy in such patients does not cease 12 months after the HMR; indeed, in many patients the benefits of HMR interventions will not be apparent for at least 12 months when the effect of these drugs on preventing major health sequelae occur. As a consequence of this limitation, we have generated a highly conservative estimate of the benefits of HMR. The lack of complete data for each patient resulted in a third limitation to our study. The majority of information provided in HMR reports (and hence the focus of the economic analysis) related to clinical aspects of patient therapy. Advice or counselling provided by the pharmacist in the HMR interview was documented minimally if at all. Our findings therefore provide a conservative estimate of the potential value of this aspect of HMRs, which has been identified in other studies as a major component of their benefits.18, 44 Additionally, we proposed that the first three DRPs identified in each HMR would be of the greatest significance to the patient’s health, and therefore the most valuable. Subsequently, if the pharmacists who performed the HMRs did not prioritise the issues they identified, our model would substantially discount these valuable interventions.

22

FINAL REPORTA fourth limitation resulted from the pharmacists who performed the HMRs being responsible for submitting them for the study. Despite the advertised requests for participation explicitly stating that no selection of HMRs should occur, we cannot discount the possibility that pharmacists may have submitted HMRs which they believed were of greater “value” than other HMRs they undertake. As the “average” value of the HMRs was quite low, it seems unlikely that a substantial amount of selection of HMRs occurred. Another limitation relates to the pharmacist who performed the HMR contacting the GP to obtain the outcomes data- the GP may have indicated that they accepted more recommendations than they did in reality. The reported rate of GP acceptance of recommendations made by pharmacists in HMRs varies widely between studies (from 38%45 to 90%34). The largest study of HMRs to date reported that 81% of the identified DRPs were “resolved, well managed or improving”,9 and our findings are not dissimilar to this figure. Further limitations involve aspects of the economic modelling technique we used for the study. The consequence of death was not available for the experts to select and assign probabilities to as a potential outcome of the HMRs. Death would result in a total loss of utility and potentially substantial additional healthcare expenditure in affected patients. By not modelling death, it is possible that our study undervalued the benefits of HMRs that prevented lethal events. However, we believe that this limitation is unlikely to influence our results as there is no evidence that HMRs reduce mortality, as reported in an international meta-analysis of 22 studies of pharmacist-conducted medication reviews (including that performed by Sorensen et al.).41 A final limitation results from the perspective we chose to perform the economic analysis. Our study estimated the economic outcomes of HMRs from the perspective of the Australian government as a third party payer. Consequently, out-of-pocket costs to consumers were not considered, and may have limited the extent of both drug-cost savings and healthcare utilisation costs resulting from the HMRs. Conversely, a cost which was not accounted for in our model was the expense of additional monitoring which may have occurred subsequent to the HMR. As increased monitoring was the most common recommendation, our results may overestimate of the value of HMRs. However the results of Sorensen et al. suggest that this would not be the case.17 In their study, increased monitoring was also the most frequently made recommendation, yet there was evidence of a reduction, rather than increase, in overall healthcare costs resulting from the medication review. We consider it unlikely that our results would be greatly confounded by this factor.

23

FINAL REPORT

Conclusion The aims of the VALMER study were to investigate the common types of DRPs identified in HMRs, and estimate the economic effects and cost-effectiveness of addressing them. The study demonstrated that HMRs continue to provide an opportunity for pharmacists to identify and manage DRPs in patients likely to be at high risk of adverse drug events. In the majority of cases, GPs implemented strategies to manage these DRPs, which suggest that they considered the DRPs to be relevant to the patients who were reviewed in the HMRs. Despite the inherent difficulties in assessing the outcomes of an intervention as complex as HMR, the study demonstrated that HMRs would most likely result in a statistically significant reduction in health resource utilisation and improve QOL in most patients. Additionally, some patients benefitted substantially from the resolution of these DRPs, resulting in considerable savings to the healthcare system. Despite the statistical significance of these findings, in many HMRs the economic value of these improvements is limited in the 12 months following the HMR, and insufficient to offset the cost of the HMR. Given a longer duration of follow-up, it is probable that the cost savings resulting from HMRs would be substantially higher than the figures in this study indicate. When the additional intangible benefits of HMRs are considered (such as improved patient medication management, knowledge and compliance) in conjunction with our findings that many HMRs are highly cost-effective, there is a strong argument to continue funding for HMRs into the next Community Pharmacy Agreement. In the 12 months following a HMR, a large proportion of the savings generated are achieved in a relatively small number of patients. Even with an extended duration of follow-up, a proportion of patients receiving HMRs are unlikely to benefit significantly from the service. To improve the cost effectiveness of the HMR program, investigations into the determinants of cost-effective HMRs should be made a priority. Once these factors are identified, strategies to more effectively target HMRs to patients who would benefit more from the service could be implemented to improve the cost-effectiveness and sustainability of the HMR program. Additionally, further research into the effects of HMRs should focus on measuring a broad range of health outcomes over longer time periods to elucidate their full benefits.

24

FINAL REPORT

References 1. Australia’s health 2008. Canberra: Australian Institute of Health and Welfare; 2008. 2. Burgess CL, Holman CD, Satti AG. Adverse drug reactions in older Australians, 1981-2002. Med J Aust

2005;182:267-70. 3. Zhang M, Holman CD, Preen DB, Brameld K. Repeat adverse drug reactions causing hospitalization in

older Australians: a population-based longitudinal study 1980-2003. Br J Clin Pharmacol 2007;63:163-70. 4. Easton K, Morgan T, Williamson M. Medication safety in the community: A review of the literature. Sydney:

National Prescribing Service; 2009. 5. Roughead EE, Barratt JD, Gilbert AL. Medication-related problems commonly occurring in an Australian

community setting. Pharmacoepidemiol Drug Saf 2004;13:83-7. 6. Home Medicines Review (HMR) - Medicare Australia. Medicare Australia, 2008. (Accessed 7 July, 2008, at

http://www.medicare.gov.au/provider/pbs/fourth-agreement/hmr.jsp.) 7. Rao D, Gilbert A, Strand LM, Cipolle RJ. Drug therapy problems found in ambulatory patient populations in

Minnesota and South Australia. Pharm World Sci 2007;29:647-54. 8. van Mil JW, Schulz M. A review of pharmaceutical care in community pharmacy in Europe. Harvard Health

Pol Rev 2006;7:155-68. 9. Gilbert AL, Roughead EE, Beilby J, Mott K, Barratt JD. Collaborative medication management services:

improving patient care. Med J Aust 2002;177:189-92. 10. Higashi T, Shekelle PG, Solomon DH, et al. The quality of pharmacologic care for vulnerable older

patients. Ann Intern Med 2004;140:714-20. 11. Kuijpers MA, van Marum RJ, Egberts AC, Jansen PA. Relationship between polypharmacy and

underprescribing. Br J Clin Pharmacol 2008;65:130-3. 12. Domiciliary Medication Management Review (DMMR). Australian Government Department of Health and

Ageing, 2008. (Accessed 10 September, 2009, at http://www.health.gov.au/internet/main/publishing.nsf/Content/health-epc-dmmr-answers.htm.)

13. Quality use of medicines in the community implementation trial. Report to the Department of Health and Aged Care.: University of South Australia and University of Adelaide; 2000.

14. Bennett A, Johnsen S, Smith C, Hurst R, Chen TF. A comparison study of two collaborative models for the provision of domiciliary medication review: St. George Canterbury Medico/Pharmacy project. Executive report; 2000.

15. Chen TF, Crampton M, Krass I, Benrimoj SI. Medication regimen reviews: a collaborative project between pharmacists and general practitioners. Sydney: University of Sydney; 1997.

16. Krass I, Smith C. Cost analysis of medication regimen reviews (MRR) performed by community pharmacists for ambulatory patients through liaison with local general medical practitioners: Pharmacy Department, University of Sydney; 1999.

17. Sorensen L, Stokes JA, Purdie DM, Woodward M, Elliott R, Roberts MS. Medication reviews in the community: results of a randomised, controlled effectiveness trial. Br J Clin Pharmacol 2004;58:648-64.

18. Evaluation of the Home Medicines Review program- pharmacy component. Urbis Keys Young, 2005. (Accessed 12 April, 2007, at http://www.guild.org.au/uploadedfiles/Medication_Management_Reviews/Overview/Urbis%20Keys%20Young%20evaluation.pdf.)

19. Stafford AC, Tenni PC, Peterson GM, et al. Drug-related problems identified in medication reviews by Australian pharmacists. Pharm World Sci 2009;31:216-23.

20. Blennerhassett JD, Cusack BM, Smith CD, Green L, Tribe KL. A novel medicines management pathway. J Pharm Pract Res 2006;36:175-9.

21. The MEDMAN study: a randomized controlled trial of community pharmacy-led medicines management for patients with coronary heart disease. Fam Pract 2007;24:189-200.

22. Holland R, Brooksby I, Lenaghan E, et al. Effectiveness of visits from community pharmacists for patients with heart failure: HeartMed randomised controlled trial. Br Med J 2007;334:1066-7.

23. Holland R, Lenaghan E, Harvey I, et al. Does home based medication review keep older people out of hospital? The HOMER randomised controlled trial. Br Med J 2005;330:293-5.

24. Scott A, Tinelli M, Bond C. Costs of a community pharmacist-led medicines management service for patients with coronary heart disease in England: healthcare system and patient perspectives. Pharmacoeconomics 2007;25:397-411.

25. Zermansky AG, Silcock J. Is medication review by primary-care pharmacists for older people cost effective?: a narrative review of the literature, focusing on costs and benefits. Pharmacoeconomics 2009;27:11-24.

26. Peterson G, Tenni P, Kruup H, Hasan O, Pekarsky B, Reeve J. PROMISe intervention study- final report to the Pharmacy Guild of Australia; 2003.

25

FINAL REPORT27. Peterson GM, Tenni PC. Identifying, prioritising and documenting drug-related problems. Aust Pharm

2004;23:706-9. 28. van Mil JW, Westerlund LO, Hersberger KE, Schaefer MA. Drug-related problem classification systems.

Ann Pharmacother 2004;38:859-67. 29. Home Medicine Review Quarterly Statistics End June Quarter 2008. 2008. (Accessed 23 February, 2009,

at http://www.guild.org.au/uploadedfiles/Medication_Management_Reviews/Overview/Microsoft%20Word%20-%202007-08%20Q4%20HMR%20Stats%20Bulletin%20031008.pdf.)

30. Australian hospital statistics 2006–07. Health services series no. 31. Cat. no. HSE 55. Canberra: Australian Institute of Health and Welfare; 2008.

31. Lopez AD, Murray CC. The global burden of disease, 1990-2020. Nat Med 1998;4:1241-3. 32. Stouthard MEA, Essink-Bot M-L, Bonsel GJ, on behalf of the DDWG. Disability weights for diseases: A

modified protocol and results for a Western European region. Eur J Public Health 2000;10:24-30. 33. Kelly B. Survey provides a wealth of information. The Accredited Pharmacist 2008;5:3-1. 34. Bell JS, Whitehead P, Aslani P, McLachlan AJ, Chen TF. Drug-related problems in the community setting.

Clin Drug Invest 2006;26:415-25. 35. The Domiciliary Medication Review Project: The Quality of Medication Care Group; 2000. 36. Naunton M, Peterson GM. Evaluation of home-based follow up of high-risk elderly patients discharged from

hospital. J Pharm Pract Res 2003;33:176-82. 37. Ponniah A, Anderson B, Shakib S, Doecke CJ, Angley M. Pharmacists' role in the post-discharge

management of patients with heart failure: a literature review. J Clin Pharm Ther 2007;32:343-52. 38. Ponniah A, Shakib S, Doecke CJ, Boyce M, Angley M. Post-discharge medication reviews for patients with

heart failure: a pilot study. Pharm World Sci 2008;30:810-5. 39. Yu K, Nguyen A, Shakib S, et al. Enhancing continuity of care in therapeutics: development of a post-

discharge home medicines review model. J Pharm Prac Res 2007;37:22-6. 40. Roughead EE, Barratt JD, Ramsay E, et al. The effectiveness of collaborative medicine reviews in delaying

time to next hospitalisation for heart failure patients in the practice setting: results of a cohort study. Circ Heart Fail 2009;2:424-8.

41. Holland R, Desborough J, Goodyer L, Hall S, Wright D, Loke YK. Does pharmacist-led medication review help to reduce hospital admissions and deaths in older people? A systematic review and meta-analysis. Br J Clin Pharmacol 2007;65:303-16.

42. Krska J, Avery AJ. Evaluation of medication reviews conducted by community pharmacists: a quantitative analysis of documented issues and recommendations. Br J Clin Pharmacol 2007.

43. Lenaghan E, Holland R, Brooks A. Home-based medication review in a high risk elderly population in primary care--the POLYMED randomised controlled trial. Age Ageing 2007;36:292-7.

44. Home Medicines Review Program Qualitative Research Project Final Report. Clifton Hill: Campbell Research & Consulting; 2008.

45. Boyatzis M, Batty KT. Domiciliary medication reviews by fourth year pharmacy students in Western Australia. Int J Pharm Prac 2004;12:73-1.

46. Bell CM, Chapman RH, Stone PW, Sandberg EA, Neumann PJ. An off-the-shelf help list: a comprehensive catalog of preference scores from published cost-utility analyses. Med Decis Making 2001;21:288-94.

47. Tengs T, Wallace A. One Thousand Health-Related Quality-of-Life Estimates. Medical Care 2000;38:583-637.

48. Khanna D, Tsevat J. Health-related quality of life-an introduction. Am J Manag Care 2007;13 Suppl 9:S218-23.

49. Green C, Brazier J, Deverill M. Valuing health-related quality of life. A review of health state valuation techniques. Pharmacoeconomics 2000;17:151-65.

50. Mathers C, Vos T, Stevenson C. The burden of disease and injury in Australia. Canberra: Australian Institute of Health and Welfare; 1999.

51. Begg S, Vos T, Barker B, Stevenson C, Stanley L, Lopez AD. The burden of disease and injury in Australia 2003. Canberra: Australian Institute of Health and Welfare; 2007.

52. Cost-Effectiveness Analysis (CEA) Registry Center for the Evaluation of Value and Risk in Health (CEVR), 2009. (Accessed 9 July, 2009, at https://research.tufts-nemc.org/cear/default.aspx.)

53. Rascati KL. Essentials of Pharmacoeconomics. Baltimore: Lippincott Williams & Wilkins; 2009. 54. Szende A, Oppe M, Devlin N, eds. EQ-5D value sets: inventory, comparative review and user guide.

Dordrecht: Springer; 2007. 55. Stouthard M, Essink-Bot M, Bonsel G, et al. Disability weights for diseases in the Netherlands. Rotterdam:

Department of Public Health, Erasmus University; 1997. 56. The BEACH Project - Bettering the Evaluation And Care of Health. Australian General Practice Statistics

and Classification Centre, 2009. (Accessed 8 July, 2009, at http://www.fmrc.org.au/beach.htm.) 57. Kelly B. Moving on is easy – what it leaves behind is hard. The Accredited Pharmacist 2009;6:3-4.

26

FINAL REPORT58. CPG Locator Maps. Commission for Certification in Geriatric Pharmacy, 2009. (Accessed 5 January, 2010,

at http://www.ccgp.org/consumer/locate.htm#AustraliaMap.) 59. Dosage administration aid service. Australian Government Department of Veterans' Affairs, 2009.

(Accessed 22 June, 2009, at http://www.dva.gov.au/health/DAA/pharmacy.htm.)

1

FINAL REPORT

Appendix I Print advertisements

Figure 6 - Example of project promotion (Australian Journal of Pharmacy 2009;89:22)

2

FINAL REPORT

Appendix II Consequences tables Consequence Severity Description Acidosis Mild Mild signs or symptoms of acidosis which resolve without intervention Acidosis Moderate Moderate acidosis requiring medical management without hospitalisation Acidosis Severe Severe acidosis requiring hospitalisation, electrolyte replacement and/or respiratory support Alkalosis Mild Mild signs or symptoms of alkalosis which resolve without intervention Alkalosis Moderate Moderate alkalosis requiring medical management without hospitalisation Alkalosis Severe Severe alkalosis requiring hospitalisation with electrolyte replacement Allergic reaction Mild Mild allergic reaction not requiring intervention Allergic reaction Moderate Moderate allergic reaction requiring topical or oral medications Allergic reaction Severe Severe allergic reaction requiring acute medical management Anaemia Mild Mild signs or symptoms of anaemia which resolve without intervention. E.g. haemoglobin of 100-109g/L in pregnant women, 110-119g/L in children and adult

women, and 120-129g/L in adult men Anaemia Moderate Anaemia requiring medical management and/or modification of medication regimen. E.g. haemoglobin of 70-99g/L in pregnant women, 80-109g/L in children and

adult women, and 90-119g/L in adult men Anaemia Severe Anaemia requiring hospitalisation and blood product or growth factor support. E.g. haemoglobin of 40-69g/L in pregnant women, 50-79g/L in children and adult

women, and 60-89g/L in adult men. Anxiety Mild Mild signs or symptoms of anxiety which resolve without intervention Anxiety Moderate Worsening of anxiety requiring modification of existing treatment regimen Anxiety Severe Anxiety requiring specialist medical attention Arrhythmia Mild Arrhythmia that contributes to worsening of other cardiac conditions but not to such an extent as to require medical intervention Arrhythmia Moderate Arrhythmia resulting in moderate haemodynamic or myocardial consequences requiring medical attention and treatment Arrhythmia Severe Arrhythmia resulting in significant haemodynamic or myocardial complications requiring hospitalisation Asthma Mild Mild asthma which does not require additional intervention Asthma Moderate Moderate asthma requiring medical attention and/or modification of medications Asthma Severe Severe asthma requiring high level care in a hospital setting Bleeding, non-specific Mild Easy bruising, bleeding from small cuts, petechia, ecchymosis, mild elevation of INR not requiring adjustment of dosage Bleeding, non-specific Moderate Hematoma, epistaxis, blood loss from mouth, vagina, melena, eye bleed, hematuria, hematemesis, moderate elevation of INR requiring modification of dose of

anticoagulant Bleeding, non-specific Severe Severe bleeding requiring hospitalisation, blood product and/or haemodynamic support. Bone marrow suppression Mild Mild signs or symptoms of bone marrow suppression which resolve without intervention Bone marrow suppression Moderate Bone marrow suppression requiring medical management by modification of existing medication regimen Bone marrow suppression Severe Bone marrow suppression requiring hospitalisation and blood product or growth factor support Cerebrovascular event Mild Mild symptoms which resolve (e.g. transient ischemic attack) Cerebrovascular event Moderate Resulting in significant signs and symptoms requiring medical management (e.g. reversible ischaemic neurological deficit) Cerebrovascular event Severe Resulting in severe symptoms and signs requiring hospitalisation and medical management (e.g stroke) Chronic Airways Disease Mild Mild chronic airways disease which does not require medical intervention Chronic Airways Disease Moderate Chronic airways disease requiring medical intervention and/or modification of medication Chronic Airways Disease Severe Severe chronic airways disease requiring hospitalisation and medical intervention CNS Depression Mild CNS depression interfering with normal activities, but not requiring medical intervention CNS Depression Moderate CNS depression requiring medical attention and interfering significantly with normal activities CNS Depression Severe Significant CNS depression resulting in loss of consciousness or obtundation Confusion Mild Mild signs or symptoms of confusion which resolve without intervention Confusion Moderate Confusion requiring medical management and/or modification of medication regimen Confusion Severe Confusion requiring prompt medical management and investigation Constipation Mild Mild signs or symptoms of constipation likely to resolve without intervention Constipation Moderate Constipation requiring medical management and/or modification of medication regimen

3

FINAL REPORT Consequence Severity Description Constipation Severe Constipation requiring hospitalisation and medical and/or surgical management Dementia Mild Dementia resulting in some impairment of daily activities only Dementia Moderate Dementia to the extent that independent living is not possible without limited supervision Dementia Severe Dementia to the extent that permanent supervision is required Depression Mild Mild signs or symptoms of depression which resolve without intervention Depression Moderate Worsening of depression requiring modification of treatment regimen Depression Severe Destabilisation or unmasking of depression requiring specialist medical attention Diabetes Mild Reduced control of diabetes requiring increased monitoring Diabetes Moderate Requiring modification of diabetes treatment regimen or significant complications requiring medical management Diabetes Severe Complications of diabetes requiring specialist medical attention in hospital Diarrhoea Mild Mild signs or symptoms of diarrhoea likely to resolve without intervention Diarrhoea Moderate Diarrhoea requiring medical management and/or modification of medication Diarrhoea Severe Requiring hospitalisation for significant diarrhoea-related electrolyte and hydration complications Gastrointestinal bleeding Mild Occult gastrointestinal bleeding likely to require medical management only if persistent Gastrointestinal bleeding Moderate Overt gastrointestinal bleeding requiring medical management Gastrointestinal bleeding Severe Overt gastrointestinal bleeding with haemodynamic consequences requiring prompt medical management Gastrointestinal pain Mild Mild signs or symptoms of gastrointestinal pain likely to resolve without intervention Gastrointestinal pain Moderate Gastrointestinal pain requiring medical management and/or modification of medication regimen Gastrointestinal pain Severe Gastrointestinal pain requiring prompt medical management and investigation Glaucoma Mild Mild elevation of IOP not requiring intervention Glaucoma Moderate Moderate elevation of IOP requiring medical intervention Glaucoma Severe Severe glaucoma requiring acute medical or surgical intervention Headache Mild Mild signs or symptoms of headache which resolve without intervention Headache Moderate Headache requiring oral analgesics and/or modification of medication regimen Headache Severe Severe headache requiring acute medical management and hospitalisation Heart Failure Mild Mild signs or symptoms of heart failure (e.g NYHA class II) which resolve without intervention Heart Failure Moderate Resulting in significant signs and symptoms of heart failure (e.g. NYHA class III) requiring medical management by modification of medication regimen Heart Failure Severe Significant signs and symptoms of heart failure (e.g. NYHA class IV) requiring hospitalisation and medical management (e.g. acute pulmonary oedema) Hypercalcaemia Mild Mild signs or symptoms of hypercalacemia which resolve without intervention Hypercalcaemia Moderate Hypercalcaemia requiring medical management and/or modification of medication regimen Hypercalcaemia Severe Hypercalcaemia requiring prompt medical management and investigation Hyperkalaemia Mild Mild signs or symptoms of hyperkalaemia which resolve without intervention Hyperkalaemia Moderate Hyperkalaemia requiring medical management and/or modification of medication regimen Hyperkalaemia Severe Hyperkalaemia requiring prompt medical management and investigation (e.g. palpitations, bradycardia) Hypertension Mild Mild signs or symptoms of hypertension which resolve without intervention Hypertension Moderate Moderate elevation of blood pressure requiring modification of or commencement of medical management Hypertension Severe Hypertension resulting in acute injury to target organs (e.g. renal, ocular or cerebral) requiring prompt medical management Hyperthyroidism Mild Mild signs or symptoms of hyperthyroidism likely to resolve without intervention Hyperthyroidism Moderate Hyperthyroidism requiring medical management and/or modification of medication regimen Hyperthyroidism Severe Signs or symptoms of hyperthyroidism requiring significant medical management and modification of medication regimen Hypocalcaemia Mild Mild signs or symptoms of hypocalcaemia which resolve without intervention Hypocalcaemia Moderate Hypocalcaemia requiring medical management and/or modification of medication regimen Hypocalcaemia Severe Hypocalcaemia requiring prompt medical management and investigation Hypoglycaemia Mild Mild signs or symptoms of hypoglycaemia which resolve without intervention Hypoglycaemia Moderate Hypoglycaemia requiring additional oral medical management or modification of medication regimen Hypoglycaemia Severe Hypoglycaemia requiring intravenous management Hypokalaemia Mild Mild signs or symptoms of hypokalaemia which resolve without intervention Hypokalaemia Moderate Hypokalaemia requiring medical management and/or modification of medication regimen

4

FINAL REPORT Consequence Severity Description Hypokalaemia Severe Hypokalaemia requiring prompt medical management and investigation (e.g. palpitations, tacchycardia) Hypotension Mild Clinical symptoms of hypotension not requiring medical intervention Hypotension Moderate Hypotension requiring medical attention and modification of antihypertensive therapy. Hypotension Severe Significant haemodynamic consequences of hypotension requiring hospitalisation and intravenous fluid support. Hypothyroidism Mild Mild signs or symptoms of hypothyroidism likely to resolve without intervention Hypothyroidism Moderate Hypothyroidism requiring medical management and/or modification of medication regimen Hypothyroidism Severe Signs or symptoms of hypothyroidism requiring significant medical management and modification of medication regimen Infection, general Mild Mild signs or symptoms of infection which resolve without intervention Infection, general Moderate Infection with moderate complications requiring medical attention and oral antibiotics (e.g. PSI class I pneumonia) Infection, general Severe Infection requiring hospitalisation and intravenous antibiotics (eg PSI class III pneumonia) Insomnia Mild Mild signs or symptoms of insomnia which resolve without intervention Insomnia Moderate Insomnia requiring medical management Insomnia Severe Insomnia requiring significant medical intervention and/or modification of medications Liver Disease Mild Mild liver disease likely to resolve without intervention Liver Disease Moderate Moderate liver disease requiring modification of medications and medical intervention Liver Disease Severe Severe liver disease requiring hospitalisation and acute medical management Myocardial Ischaemia Mild Mild signs or symptoms of angina (e.g. NYHA class I to II) which resolve without intervention Myocardial Ischaemia Moderate Moderate myocardial ischaemia resulting in significant signs and symptoms requiring medical management by modification of medication regimen (e.g. worsened

stable angina) Myocardial Ischaemia Severe Myocardial ischaemia resulting in severe symptoms and signs requiring hospitalisation and medical management (e.g unstable angina, myocardial infarction) Myopathy Mild Mild muscle pain not requiring intervention Myopathy Moderate Moderate muscle pain requiring modification of medications without elevation in creatine kinase Myopathy Severe Severe myopathy with elevation of creatine kinase requiring cessation of medication Nausea Mild Mild signs or symptoms of nausea likely to resolve without intervention Nausea Moderate Nausea requiring medical management and/or modification of medication Nausea Severe Requiring hospitalisation for significant vomiting-related electrolyte and hydration complications Oedema Mild Mild signs or symptoms of oedema which resolve without intervention Oedema Moderate Moderate oedema resulting in significant symptoms requiring medical management by modification of medication regimen Oedema Severe Significant oedema resulting in severe symptoms and signs requiring hospitalisation and management with intravenous diuretics Osteoporosis Mild Worsening of osteoporosis requiring increased monitoring Osteoporosis Moderate Osteoporosis requiring modification of existing treatment regimen, or commencing treatment for an "at risk" person Osteoporosis Severe Osteoporosis resulting in hospitalisation due to a major complication (e.g. fracture) Pain Mild Mild pain responding to currently prescribed agents Pain Moderate Moderate pain requiring intensification of oral analgesics Pain Severe Severe pain requiring management using specialist techniques or advice (e.g. hospitalisation, intravenous or epidural opioids or anaesthetics) Parkinsonism Mild Mild Parkinsonian symptoms of tremors and rigidity; slowness, impaired swallowing and speech; disturbance of equilibrium; patient is able to function

independently Parkinsonism Moderate Swallowing and speech severely impaired; autonomic nervous system disturbances; patients are ADL-dependent, but are able to move without help Parkinsonism Severe Severe Parkinsonian symptoms- patient wheelchair and bed-bound; severely handicapped Psychosis Mild Mild signs or symptoms of psychosis which resolve without intervention Psychosis Moderate Worsening of psychotic illness requiring modification of treatment regimen Psychosis Severe Destabilisation or unmasking of psychosis requiring specialist medical attention Rash Mild Localised or mild rash symptoms which respond to "over the counter" treatment Rash Moderate Rash requiring medical attention and topical and/or oral systemic treatment Rash Severe Widespread rash with significant consequences requiring hospitalisation and intravenous systemic treatment Renal Dysfunction Mild Mild signs or symptoms of renal dysfunction which resolve without intervention Renal Dysfunction Moderate Renal dysfunction requiring medical management and/or modification of medication regimen Renal Dysfunction Severe Acute decline in renal function requiring prompt medical management and investigation

5

FINAL REPORT Consequence Severity Description Respiratory depression Mild Mild respiratory depression which will resolve without medical intervention Respiratory depression Moderate Requiring medical intervention and/or modification of medications Respiratory depression Severe Severe respiratory depression requiring hospitalisation and medical intervention Seizures Mild Mild one-off seizure unlikely to recur or require management Seizures Moderate Seizure requiring medical attention or modification of medication regimen Seizures Severe Severe seizures requiring hospitalisation and intravenous anticonvulsants Serotonin toxicity Mild Mild signs or symptoms of serotonin toxicity which resolve without intervention Serotonin toxicity Moderate Serotonin toxicity requiring medical management and/or modification of medication regimen Serotonin toxicity Severe Serotonin toxicity requiring prompt medical management and investigation (e.g euphoria, tremor, severe anxiety, palpitations) Urinary Incontinence Mild Mild signs or symptoms of urinary incontinence which resolve without intervention Urinary Incontinence Moderate Urinary incontinence requiring medical management and/or modification of medication regimen Urinary Incontinence Severe Urinary incontinence requiring hospitalisation and medical and/or surgical management Urinary retention Mild Symptomatic urinary retention not requiring medical management Urinary retention Moderate Symptomatic urinary retention requiring medical management Urinary retention Severe Symptomatic urinary retention requiring catheterisation Urinary Tract Infection Mild Mild signs or symptoms of UTI which resolve without intervention Urinary Tract Infection Moderate Moderate UTI requiring medical attention and oral antibiotics Urinary Tract Infection Severe UTI requiring hospitalisation and intravenous antibiotics, e.g. pyelonephritis

Table 9 - Description of consequences

6

FINAL REPORT Consequence Severity

description Duration of consequence (days, ±SD)

Raw utility (365 days unless otherwise stated,±SD)

Comments regarding utilities Disability weight (specified duration,±SD)

GP visits (±SD)

Cost of GP visits (±SD)

Acidosis Mild 9.0 (± 27.4) 1.000 (± 0.000) 0.000 (± 0.000) 1.9 (± 1.5) $64.42 (± $50.66) Acidosis Moderate 9.4 (± 12.6) 0.812 (±0.103) 0.005 (± 0.003) 5.0 (± 5.9) $166.41 (± $196.27) Acidosis Severe 18.0 (± 11.9) 0.130 (±0.342) 0.043 (± 0.017) 5.3 (± 4.6) $176.81 (± $155.00) Alkalosis Mild 11.0 (± 29.6) 0.929 (±0.117) 0.002 (± 0.004) 2.4 (± 3.3) $79.18 (± $109.37) Alkalosis Moderate 9.0 (± 17.1) 0.903 (±0.124) 0.002 (± 0.003) 2.9 (± 1.0) $97.63 (± $34.89) Alkalosis Severe 16.1 (± 15.1) 0.378 (±0.367) 0.027 (± 0.016) 5.1 (± 3.6) $170.77 (± $120.44) Allergic reaction Mild 1.2 (± 1.2) 0.839 (±0.082) 0.001 (± 0.000) 1.2 (± 1.2) $40.60 (± $41.27) Allergic reaction Moderate 3.0 (± 2.3) 0.655 (±0.076) 0.003 (± 0.001) 2.1 (± 1.2) $69.78 (± $41.60) Allergic reaction Severe 8.0 (± 5.6) -0.219 (±0.430) 0.027 (± 0.009) 4.6 (± 5.0) $155.00 (± $167.75) Anaemia Mild 13.3 (± 22.4) 0.995 (±0.000) GBD mild anaemia 0.000 (± 0.000) 2.3 (± 1.8) $78.17 (± $61.40) Anaemia Moderate 15.2 (± 15.5) 0.989 (±0.000) GBD moderate anaemia 0.000 (± 0.000) 3.6 (± 1.5) $120.11 (± $50.66) Anaemia Severe 27.5 (± 24.1) 0.830 (±0.000) GBD. Mean of "severe" and "very severe"

anaemia 0.013 (± 0.000) 5.6 (± 3.1) $187.21 (± $102.66)

Anxiety Mild 17.8 (± 27.5) 0.840 (±0.147) DW. Mean of "mild-moderate" agoraphobia, diffuse anxiety disorder, obsessive-compulsive disorder, panic disorder, post-traumatic stress disorder, singular phobia and social phobia

0.008 (± 0.007) 2.9 (± 3.2) $97.63 (± $107.70)

Anxiety Moderate 36.7 (± 54.5) 0.840 (±0.147) DW. Mean of "mild-moderate" agoraphobia, diffuse anxiety disorder, obsessive-compulsive disorder, panic disorder, post-traumatic stress disorder, singular phobia and social phobia

0.016 (± 0.015) 5.0 (± 1.7) $167.75 (± $56.03)

Anxiety Severe 71.8 (± 96.2) 0.440 (±0.429) DW. Mean of "severe" agoraphobia, diffuse anxiety disorder, obsessive-compulsive disorder, panic disorder, post-traumatic stress disorder, singular phobia and social phobia

0.110 (± 0.084) 7.7 (± 2.8) $259.34 (± $92.60)

Arrhythmia Mild 7.1 (± 11.1) 0.722 (±0.095) 0.005 (± 0.002) 2.6 (± 1.8) $85.55 (± $58.71) Arrhythmia Moderate 22.6 (± 33.6) 0.478 (±0.228) 0.032 (± 0.014) 4.6 (± 2.0) $152.65 (± $67.77) Arrhythmia Severe 44.1 (± 65.8) -0.173 (±0.351) 0.142 (± 0.042) 6.9 (± 2.5) $231.83 (± $84.21) Asthma Mild 6.7 (± 7.2) 0.779 (±0.197) 0.004 (± 0.004) 1.6 (± 1.0) $55.02 (± $34.56) Asthma Moderate 13.4 (± 12.0) 0.597 (±0.068) 0.015 (± 0.002) 3.4 (± 1.4) $112.73 (± $47.98) Asthma Severe 38.0 (± 50.5) -0.358 (±0.260) 0.141 (± 0.027) 5.2 (± 2.1) $173.79 (± $71.80) Bleeding, non-specific

Mild 1.7 (± 1.6) 1.000 (±0.000) 0.000 (± 0.000) 2.1 (± 1.1) $68.78 (± $35.90)

Bleeding, non-specific

Moderate 7.3 (± 8.2) 0.950 (±0.000) ABDI 2000 weight for "moderate cases of haemophilia"

0.001 (± 0.000) 3.7 (± 2.0) $124.14 (± $65.42)

Bleeding, non-specific

Severe 24.8 (± 25.9) 0.730 (±0.000) ABDI 2000 weight for severe cases of haemophilia"

0.018 (± 0.000) 5.8 (± 4.9) $194.59 (± $165.74)

Bone marrow suppression

Mild 5.4 (± 8.9) 0.873 (±0.116) 0.002 (± 0.002) 2.7 (± 1.5) $90.59 (± $49.99)

Bone marrow suppression

Moderate 15.0 (± 26.5) 0.698 (±0.197) 0.012 (± 0.008) 4.2 (± 1.1) $140.91 (± $38.25)

Bone marrow suppression

Severe 59.0 (± 74.4) 0.371 (±0.439) 0.102 (± 0.071) 6.1 (± 2.9) $204.66 (± $96.62)

Cerebrovascular event

Mild 4.1 (± 2.7) 0.697 (±0.151) DW 0.64 0.003 (± 0.002) 3.5 (± 1.4) $117.43 (± $45.29)

Cerebrovascular Moderate 10.8 (± 8.3) 0.461 (±0.320) DW 0.37 0.016 (± 0.009) 5.0 (± 2.3) $167.75 (± $75.82)

7

FINAL REPORT Consequence Severity

description Duration of consequence (days, ±SD)

Raw utility (365 days unless otherwise stated,±SD)

Comments regarding utilities Disability weight (specified duration,±SD)

GP visits (±SD)

Cost of GP visits (±SD)

event Cerebrovascular event

Severe 180.0 (± 0.0) -0.052 (±0.147) DW 0.08 Duration set at 6 months to reflect permanent impariments

0.519 (± 0.072) 6.7 (± 3.5) $224.79 (± $116.42)

Chronic Airways Disease

Mild 9.2 (± 11.9) 0.830 (±0.315) DW "mild to moderate" COPD 0.004 (± 0.008) 2.9 (± 1.8) $95.62 (± $61.40)

Chronic Airways Disease

Moderate 18.9 (± 25.6) 0.830 (±0.315) DW "mild to moderate" COPD 0.009 (± 0.016) 4.4 (± 1.9) $147.28 (± $63.75)

Chronic Airways Disease

Severe 127.7 (± 164.5) 0.470 (±0.393) DW "severe" COPD 0.185 (± 0.137) 6.2 (± 2.9) $208.68 (± $95.95)

CNS Depression

Mild 6.7 (± 9.3) 0.609 (±0.125) 0.007 (± 0.002) 2.4 (± 1.3) $78.84 (± $41.94)

CNS Depression

Moderate 28.9 (± 41.5) 0.275 (±0.276) 0.057 (± 0.022) 5.2 (± 2.9) $174.46 (± $95.95)

CNS Depression

Severe 94.6 (± 154.0) -0.320 (±0.230) 0.342 (± 0.060) 7.4 (± 4.1) $248.27 (± $138.23)

Confusion Mild 4.9 (± 5.8) 0.698 (±0.072) 0.004 (± 0.001) 2.1 (± 1.1) $70.46 (± $36.91) Confusion Moderate 18.9 (± 26.4) 0.528 (±0.185) 0.024 (± 0.010) 4.2 (± 1.6) $140.91 (± $54.35) Confusion Severe 76.7 (± 120.8) 0.061 (±0.159) 0.197 (± 0.033) 5.6 (± 2.0) $187.88 (± $67.44) Constipation Mild 5.5 (± 11.2) 0.795 (±0.107) 0.003 (± 0.002) 1.0 (± 0.8) $32.21 (± $27.18) Constipation Moderate 9.4 (± 15.7) 0.725 (±0.000) 0.007 (± 0.000) 2.3 (± 0.9) $76.16 (± $29.52) Constipation Severe 25.0 (± 31.4) 0.216 (±0.407) 0.054 (± 0.028) 3.6 (± 1.8) $121.45 (± $61.06) Dementia Mild 51.5 (± 113.5) 0.730 (±0.384) DW 0.038 (± 0.054) 3.1 (± 1.5) $104.01 (± $51.00) Dementia Moderate 108.9 (± 155.5) 0.370 (±0.475) DW 0.188 (± 0.142) 5.3 (± 2.9) $177.82 (± $97.63) Dementia Severe 235.4 (± 179.0) 0.060 (±0.117) DW 0.606 (± 0.075) 9.2 (± 5.4) $308.66 (± $181.51) Depression Mild 8.9 (± 11.4) 0.860 (±0.235) DW 0.003 (± 0.006) 2.8 (± 2.2) $92.26 (± $73.14) Depression Moderate 34.2 (± 40.2) 0.650 (±0.279) DW 0.033 (± 0.026) 4.9 (± 1.8) $164.40 (± $60.05) Depression Severe 110.1 (± 120.9) 0.205 (±0.460) DW 0.240 (± 0.139) 6.1 (± 2.5) $204.66 (± $84.21) Diabetes Mild 6.4 (± 6.5) 0.918 (±0.091) DW "Uncomplicated diabetes" 0.93 0.001 (± 0.002) 2.9 (± 1.9) $95.62 (± $62.40) Diabetes Moderate 11.6 (± 8.4) 0.826 (±0.102) 0.006 (± 0.003) 4.6 (± 1.8) $154.33 (± $61.73) Diabetes Severe 46.9 (± 76.5) 0.609 (±0.089) 0.050 (± 0.011) 5.6 (± 3.2) $188.22 (± $106.69) Diarrhoea Mild 2.6 (± 2.9) 0.990 (±0.006) DW "digestive tract infection, uncomplicated

course (duration 2 weeks)" 0.002 (± 0.001) 1.0 (± 0.8) $33.55 (± $27.51)

Diarrhoea Moderate 6.8 (± 6.1) 0.970 (±0.104) DW "digestive tract infection, complicated course (duration 2 weeks)"

0.015 (± 0.050) 2.0 (± 0.7) $67.10 (± $22.48)

Diarrhoea Severe 12.2 (± 11.2) 0.600 (±0.311) DW "acute exacerbation of inflammatory bowel disease"

0.013 (± 0.010) 3.6 (± 1.8) $120.78 (± $61.73)

Gastrointestinal bleeding

Mild 1.9 (± 2.1) 0.975 (±0.062) 0.000 (± 0.000) 2.0 (± 1.1) $67.10 (± $35.23)

Gastrointestinal bleeding

Moderate 9.0 (± 9.1) 0.592 (±0.228) 0.010 (± 0.006) 3.4 (± 1.4) $114.07 (± $45.29)

Gastrointestinal bleeding

Severe 35.6 (± 57.0) -0.197 (±0.249) 0.117 (± 0.024) 4.9 (± 2.3) $164.40 (± $78.17)

Gastrointestinal pain

Mild 10.0 (± 27.1) 0.719 (±0.042) 0.008 (± 0.001) 1.2 (± 1.3) $40.60 (± $42.61)

Gastrointestinal pain

Moderate 10.1 (± 10.2) 0.620 (±0.085) 0.011 (± 0.002) 2.8 (± 1.3) $92.93 (± $43.62)

8

FINAL REPORT Consequence Severity

description Duration of consequence (days, ±SD)

Raw utility (365 days unless otherwise stated,±SD)

Comments regarding utilities Disability weight (specified duration,±SD)

GP visits (±SD)

Cost of GP visits (±SD)

Gastrointestinal pain

Severe 25.7 (± 38.6) -0.430 (±0.222) 0.101 (± 0.016) 4.2 (± 2.8) $141.92 (± $93.94)

Glaucoma Mild 7.7 (± 11.3) 0.980 (±0.128) DW "mild vision loss" 0.000 (± 0.003) 1.5 (± 0.8) $50.33 (± $26.84) Glaucoma Moderate 45.0 (± 77.1) 0.830 (±0.323) DW "moderate vision loss" 0.021 (± 0.040) 2.8 (± 1.0) $93.94 (± $33.55) Glaucoma Severe 74.5 (± 144.0) 0.430 (±0.305) DW "severe vision loss" 0.116 (± 0.062) 4.5 (± 2.1) $150.98 (± $70.46) Headache Mild 6.8 (± 13.7) 0.806 (±0.101) 0.004 (± 0.002) 1.0 (± 2.1) $33.55 (± $69.45) Headache Moderate 10.4 (± 11.2) 0.696 (±0.068) 0.009 (± 0.002) 2.0 (± 1.1) $67.10 (± $35.90) Headache Severe 20.5 (± 22.4) -0.279 (±0.271) 0.072 (± 0.015) 3.3 (± 1.8) $109.04 (± $58.71) Heart Failure Mild 52.4 (± 126.5) 0.705 (±0.134) DW for "Mild (NYHA 1 - 2)" 0.94 0.042 (± 0.019) 3.9 (± 2.1) $130.17 (± $70.46) Heart Failure Moderate 71.9 (± 121.6) 0.436 (±0.125) DW for "Moderate (NYHA 3)" 0.65 0.111 (± 0.025) 7.3 (± 4.2) $243.24 (± $140.91) Heart Failure Severe 156.1 (± 157.2) -0.380 (±0.188) DW for "Severe (NYHA 4)" 0.35 0.590 (± 0.081) 10.1 (± 5.3) $339.86 (± $176.14) Hypercalcaemia Mild 3.5 (± 3.4) 0.966 (±0.083) 0.000 (± 0.001) 2.1 (± 1.3) $71.46 (± $41.94) Hypercalcaemia Moderate 11.9 (± 10.3) 0.858 (±0.122) 0.005 (± 0.004) 3.8 (± 1.2) $125.81 (± $38.92) Hypercalcaemia Severe 34.9 (± 39.1) 0.324 (±0.337) 0.065 (± 0.032) 5.0 (± 1.2) $167.75 (± $40.26) Hyperkalaemia Mild 1.6 (± 2.4) 1.000 (±0.000) 0.000 (± 0.000) 1.6 (± 0.9) $54.69 (± $30.87) Hyperkalaemia Moderate 7.1 (± 9.3) 1.000 (±0.000) 0.000 (± 0.000) 3.9 (± 1.6) $130.17 (± $52.00) Hyperkalaemia Severe 22.3 (± 28.3) 0.515 (±0.297) 0.030 (± 0.018) 8.4 (± 9.0) $281.15 (± $303.29) Hypertension Mild 3.0 (± 6.9) 1.000 (±0.000) 0.000 (± 0.000) 2.1 (± 1.6) $71.46 (± $55.02) Hypertension Moderate 8.5 (± 17.0) 0.975 (±0.062) 0.001 (± 0.001) 4.6 (± 2.9) $155.34 (± $96.62) Hypertension Severe 66.4 (± 76.6) 0.459 (±0.397) 0.098 (± 0.072) 6.6 (± 2.9) $222.44 (± $96.62) Hyperthyroidism Mild 4.8 (± 3.2) 0.949 (±0.078) 0.001 (± 0.001) 2.3 (± 2.5) $75.49 (± $83.54) Hyperthyroidism Moderate 14.4 (± 10.1) 0.840 (±0.241) 0.006 (± 0.010) 4.3 (± 1.7) $142.59 (± $56.03) Hyperthyroidism Severe 43.5 (± 31.6) 0.482 (±0.221) 0.062 (± 0.026) 5.3 (± 2.1) $176.14 (± $71.13) Hypocalcaemia Mild 2.8 (± 2.7) 0.954 (±0.112) 0.000 (± 0.001) 1.8 (± 0.9) $58.71 (± $29.86) Hypocalcaemia Moderate 9.4 (± 9.4) 0.826 (±0.157) 0.004 (± 0.004) 4.3 (± 1.2) $142.59 (± $38.92) Hypocalcaemia Severe 31.0 (± 49.1) 0.311 (±0.459) 0.059 (± 0.039) 4.8 (± 1.4) $159.36 (± $46.63) Hypoglycaemia Mild 3.1 (± 4.7) 0.835 (±0.092) 0.001 (± 0.001) 1.4 (± 1.1) $46.30 (± $35.56) Hypoglycaemia Moderate 7.9 (± 9.3) 0.757 (±0.156) 0.005 (± 0.003) 3.6 (± 1.3) $121.79 (± $43.62) Hypoglycaemia Severe 14.8 (± 14.1) 0.039 (±0.272) 0.039 (± 0.011) 5.9 (± 1.4) $197.27 (± $45.63) Hypokalaemia Mild 3.9 (± 5.2) 1.000 (±0.000) 0.000 (± 0.000) 1.5 (± 0.8) $50.33 (± $25.50) Hypokalaemia Moderate 12.1 (± 19.7) 0.894 (±0.195) 0.004 (± 0.006) 3.8 (± 1.4) $125.81 (± $46.63) Hypokalaemia Severe 38.0 (± 58.7) 0.274 (±0.496) 0.076 (± 0.052) 5.3 (± 1.6) $176.14 (± $53.01) Hypotension Mild 2.9 (± 4.4) 0.815 (±0.121) 0.001 (± 0.001) 1.4 (± 1.1) $46.30 (± $35.56) Hypotension Moderate 9.5 (± 10.3) 0.694 (±0.142) 0.008 (± 0.004) 3.9 (± 1.5) $130.17 (± $48.98) Hypotension Severe 24.4 (± 31.2) -0.226 (±0.215) 0.082 (± 0.014) 6.4 (± 2.1) $214.05 (± $71.46) Hypothyroidism Mild 10.6 (± 12.9) 0.975 (±0.062) 0.001 (± 0.002) 2.3 (± 2.1) $75.49 (± $68.78) Hypothyroidism Moderate 17.0 (± 19.1) 0.924 (±0.086) 0.004 (± 0.004) 3.9 (± 0.8) $130.17 (± $27.85) Hypothyroidism Severe 33.3 (± 39.7) 0.564 (±0.247) 0.040 (± 0.022) 5.1 (± 1.6) $172.11 (± $55.02) Infection, general

Mild 3.4 (± 3.3) 0.990 (±0.106) DW "Acute non-STI urethritis (duration 1 week)" 0.005 (± 0.051) 1.5 (± 1.5) $50.33 (± $50.66)

Infection, general

Moderate 9.9 (± 6.8) 0.990 (±0.141) DW "Acyte cystitis (duration 1 week)" 0.014 (± 0.199) 2.6 (± 0.9) $88.24 (± $30.87)

Infection, general

Severe 24.9 (± 19.7) 0.990 (±0.101) DW "Acute pyelonephritis" 0.001 (± 0.007) 3.6 (± 1.2) $121.79 (± $39.92)

Insomnia Mild 10.9 (± 15.4) 0.867 (±0.067) 0.004 (± 0.002) 0.9 (± 1.0) $29.52 (± $33.21) Insomnia Moderate 20.9 (± 24.2) 0.716 (±0.203) 0.016 (± 0.012) 2.6 (± 1.1) $88.24 (± $35.56)

9

FINAL REPORT Consequence Severity

description Duration of consequence (days, ±SD)

Raw utility (365 days unless otherwise stated,±SD)

Comments regarding utilities Disability weight (specified duration,±SD)

GP visits (±SD)

Cost of GP visits (±SD)

Insomnia Severe 31.9 (± 32.0) 0.516 (±0.174) 0.042 (± 0.015) 5.5 (± 2.8) $184.53 (± $94.95) Liver Disease Mild 3.1 (± 5.6) 0.892 (±0.130) 0.001 (± 0.001) 1.3 (± 1.2) $41.94 (± $38.92) Liver Disease Moderate 14.0 (± 14.9) 0.707 (±0.183) 0.011 (± 0.007) 4.3 (± 1.0) $142.59 (± $34.89) Liver Disease Severe 67.3 (± 76.0) 0.179 (±0.443) 0.151 (± 0.082) 6.3 (± 1.3) $209.69 (± $42.94) Myocardial Ischaemia

Mild 4.3 (± 4.4) 0.591 (±0.086) DW "Mild stable angina pectoris, NYHA class I to II" 0.92

0.005 (± 0.001) 2.1 (± 1.5) $71.46 (± $48.98)

Myocardial Ischaemia

Moderate 11.8 (± 14.1) 0.478 (±0.201) DW "Severe stable angina pectoris, NYHA class III" 0.43

0.017 (± 0.006) 4.3 (± 1.4) $142.59 (± $46.63)

Myocardial Ischaemia

Severe 180.0 (± 0.0) -0.275 (±0.269) ABDI 2000 0.395 Duration set at 6 months to reflect permanent impariments

0.629 (± 0.133) 6.0 (± 1.8) $201.30 (± $59.38)

Myopathy Mild 8.9 (± 16.9) 0.720 (±0.087) 0.007 (± 0.002) 1.5 (± 1.4) $50.33 (± $47.31) Myopathy Moderate 13.5 (± 14.5) 0.598 (±0.079) 0.015 (± 0.003) 3.1 (± 1.0) $105.01 (± $33.21) Myopathy Severe 64.0 (± 122.4) -0.083 (±0.434) 0.190 (± 0.076) 5.3 (± 1.2) $176.14 (± $38.92) Nausea Mild 2.4 (± 2.2) 0.731 (±0.076) 0.002 (± 0.001) 0.9 (± 0.6) $29.52 (± $21.47) Nausea Moderate 5.8 (± 6.4) 0.589 (±0.201) 0.006 (± 0.003) 2.6 (± 0.9) $88.24 (± $30.87) Nausea Severe 12.4 (± 14.6) -0.207 (±0.348) 0.041 (± 0.012) 3.5 (± 1.2) $117.43 (± $40.26) Oedema Mild 2.1 (± 2.8) 0.885 (±0.198) 0.001 (± 0.001) 0.8 (± 0.5) $25.16 (± $15.43) Oedema Moderate 8.9 (± 9.6) 0.661 (±0.081) 0.008 (± 0.002) 3.0 (± 1.1) $100.65 (± $35.90) Oedema Severe 51.4 (± 85.3) 0.225 (±0.240) 0.109 (± 0.034) 5.0 (± 1.3) $167.75 (± $43.95) Osteoporosis Mild 46.5 (± 128.7) 0.991 (±0.000) GBD 0.001 (± 0.000) 2.3 (± 1.3) $75.49 (± $42.94) Osteoporosis Moderate 68.6 (± 123.1) 0.991 (±0.000) GBD 0.002 (± 0.000) 4.5 (± 2.8) $150.98 (± $93.27) Osteoporosis Severe 109.8 (± 124.1) 0.897 (±0.218) DW mean of hip/leg, arm/shoulder, ankle fracture 0.031 (± 0.066) 4.8 (± 1.0) $159.36 (± $34.89) Pain Mild 6.0 (± 10.0) 0.690 (±0.089) 0.005 (± 0.001) 1.6 (± 1.2) $54.69 (± $39.92) Pain Moderate 13.0 (± 8.5) 0.483 (±0.369) 0.018 (± 0.013) 3.9 (± 1.6) $130.17 (± $52.00) Pain Severe 139.1 (± 165.6) -0.317 (±0.204) 0.502 (± 0.078) 7.0 (± 3.0) $234.85 (± $101.32) Parkinsonism Mild 63.3 (± 147.8) 0.520 (±0.406) DW 0.083 (± 0.070) 2.5 (± 2.2) $83.88 (± $73.81) Parkinsonism Moderate 80.2 (± 140.2) 0.210 (±0.265) DW 0.174 (± 0.058) 5.3 (± 3.5) $177.82 (± $117.43) Parkinsonism Severe 206.3 (± 176.5) 0.080 (±0.161) DW 0.520 (± 0.091) 7.5 (± 4.3) $251.63 (± $144.27) Psychosis Mild 7.2 (± 11.5) 0.790 (±0.379) DW "one psychotic epidose, no permanent

impairments" 0.004 (± 0.007) 2.5 (± 1.4) $83.88 (± $46.97)

Psychosis Moderate 71.7 (± 116.5) 0.290 (±0.279) DW "Several psychotic episodes, some permanent impairment"

0.139 (± 0.055) 6.0 (± 3.6) $201.30 (± $120.78)

Psychosis Severe 119.7 (± 136.7) 0.472 (±0.305) DW "Several psychotic episodes, obvious permanent impairments"

0.173 (± 0.100) 5.7 (± 3.3) $191.24 (± $110.72)

Rash Mild 2.8 (± 5.5) 0.765 (±0.139) 0.002 (± 0.001) 0.3 (± 0.5) $10.07 (± $16.78) Rash Moderate 16.8 (± 22.4) 0.696 (±0.103) 0.014 (± 0.005) 2.0 (± 1.4) $67.10 (± $46.97) Rash Severe 73.2 (± 143.6) 0.016 (±0.394) 0.197 (± 0.079) 3.8 (± 1.3) $127.49 (± $43.62) Renal Dysfunction

Mild 1.7 (± 2.7) 0.954 (±0.112) 0.000 (± 0.001) 1.3 (± 1.5) $43.62 (± $50.33)

Renal Dysfunction

Moderate 7.5 (± 7.1) 0.894 (±0.195) GBD "treated renal failure" 0.89 0.002 (± 0.004) 4.0 (± 1.7) $134.20 (± $57.04)

Renal Dysfunction

Severe 59.3 (± 118.4) 0.500 (±0.366) DW "Diabetic nephropathy"; used by ABDI for "end stage renal failure" 0.71 "Diabetic nephropathy"; used by ABDI for "end stage renal failure"

0.081 (± 0.059) 6.0 (± 2.6) $201.30 (± $87.23)

Respiratory Mild 1.7 (± 2.7) 0.725 (±0.115) 0.001 (± 0.001) 1.0 (± 0.9) $33.55 (± $30.20)

10

FINAL REPORT Consequence Severity

description Duration of consequence (days, ±SD)

Raw utility (365 days unless otherwise stated,±SD)

Comments regarding utilities Disability weight (specified duration,±SD)

GP visits (±SD)

Cost of GP visits (±SD)

depression Respiratory depression

Moderate 15.0 (± 13.3) 0.517 (±0.184) 0.020 (± 0.008) 4.0 (± 1.7) $134.20 (± $57.04)

Respiratory depression

Severe 79.0 (± 113.6) -0.273 (±0.122) 0.275 (± 0.026) 5.0 (± 2.2) $167.75 (± $73.81)

Seizures Mild 2.0 (± 2.0) 0.600 (±0.321) 0.002 (± 0.002) 1.8 (± 1.2) $60.39 (± $40.26) Seizures Moderate 7.0 (± 5.8) 0.392 (±0.314) 0.012 (± 0.006) 4.0 (± 2.3) $134.20 (± $77.17) Seizures Severe 29.7 (± 36.7) -0.249 (±0.224) DW "epilepsy" 0.89 0.102 (± 0.018) 4.5 (± 2.4) $150.98 (± $80.52) Serotonin toxicity

Mild 1.8 (± 2.6) 0.699 (±0.326) 0.001 (± 0.002) 1.0 (± 0.6) $33.55 (± $20.13)

Serotonin toxicity

Moderate 9.0 (± 6.4) 0.411 (±0.461) 0.015 (± 0.011) 4.3 (± 2.9) $144.27 (± $97.30)

Serotonin toxicity

Severe 14.8 (± 15.6) -0.200 (±0.264) 0.049 (± 0.011) 4.2 (± 1.6) $140.91 (± $53.68)

Urinary Incontinence

Mild 5.7 (± 11.9) 1.000 (±0.000) ABDI "Occasional urine leakage" 0.000 (± 0.000) 0.5 (± 0.5) $16.78 (± $16.78)

Urinary Incontinence

Moderate 53.5 (± 79.2) 0.975 (±0.000) ABDI "Moderate incontinence" 0.004 (± 0.000) 2.8 (± 1.0) $93.94 (± $33.55)

Urinary Incontinence

Severe 171.3 (± 175.5) 0.843 (±0.000) ABDI "Severe incontinence" 0.074 (± 0.000) 4.5 (± 1.9) $150.98 (± $63.75)

Urinary retention

Mild 1.7 (± 2.7) 0.972 (±0.000) ABDI "Symptomatic benign prostatic hypertrophy" 0.000 (± 0.000) 0.7 (± 0.8) $23.49 (± $26.84)

Urinary retention

Moderate 7.5 (± 11.1) 0.849 (±0.000) ABDI "Urethral stricture associated with benign prostatic hypertrophy"

0.003 (± 0.000) 2.5 (± 0.8) $83.88 (± $26.84)

Urinary retention

Severe 20.5 (± 34.4) 0.843 (±0.000) ABDI "Urinary incontinence associated with benign prostatic hypertrophy"

0.009 (± 0.000) 3.7 (± 2.4) $124.14 (± $80.52)

Urinary Tract Infection

Mild 1.2 (± 1.0) 0.990 (±0.106) DW "Acute non-STI urethritis (duration 1 week)" 0.000 (± 0.018) 0.7 (± 0.5) $23.49 (± $16.78)

Urinary Tract Infection

Moderate 3.2 (± 1.6) 0.990 (±0.141) DW "Acyte cystitis (duration 1 week)" 0.000 (± 0.065) 1.7 (± 0.5) $57.04 (± $16.78)

Urinary Tract Infection

Severe 9.3 (± 8.5) 0.990 (±0.101) DW "Acute pyelonephritis" 0.000 (± 0.003) 2.5 (± 0.8) $83.88 (± $26.84)

DW - Dutch weight ABDI - Australian Burden of Disease and Injury GBD - Global Burden of Disease

Table 10 - Consequences table values for duration, utilities and GP visits

11

FINAL REPORT

Consequence Severity description

Specialist visits (±SD)

Cost of specialist visits (±SD)

Investigations Cost of investigations

AR-DRG code (version 5.1, national public hospitals 2006-2007)

Comments re DRG values

Average duration of admission (days)

Cost of admission

Acidosis Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Acidosis Moderate 1.0 (± 1.0) $75.89 (± $76.68) FBP, EUC $35 Acidosis Severe 4.6 (± 4.1) $221.18 (±

$199.93) FBP, EUC, ABG $69.3 K62A, K62B, K62C Mean of three

DRGs weighted for number of occurrences

3.7 $3,437.85

Alkalosis Mild 0.1 (± 0.3) $7.11 (± $23.72) Nil $0 Alkalosis Moderate 1.4 (± 1.3) $95.33 (± $86.54) FBP, EUC $35 Alkalosis Severe 3.3 (± 1.4) $169.17 (±

$73.46) FBP, EUC, ABG $69.3 K62A, K62B, K62C Mean of three

DRGs weighted for number of occurrences

3.7 $3,437.85

Allergic reaction Mild 0.1 (± 0.3) $6.32 (± $22.92) Nil $0 Allergic reaction Moderate 0.5 (± 1.0) $39.53 (± $79.05) Nil $0 Allergic reaction Severe 2.4 (± 0.9) $133.84 (±

$50.05) RAST $27.3 X61Z 1.3 $1,293.00

Anaemia Mild 0.4 (± 0.7) $33.20 (± $52.96) FBP, FeS, B12, folate

$94.05

Anaemia Moderate 1.5 (± 1.2) $100.49 (± $80.26)

FBP, FeS, B12, folate, colonoscopy

$402.9

Anaemia Severe 3.9 (± 1.6) $194.97 (± $77.59)

FBP, FeS, B12, folate, colonoscopy, endoscopy

$566.55 Q61A, Q61B, Q61C 1.9 $5,311.11

Anxiety Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Anxiety Moderate 1.4 (± 2.4) $95.33 (±

$163.61) TFT, EUC $193.2

Anxiety Severe 5.3 (± 2.5) $248.57 (± $119.33)

TFT, EUC $38.5 U65Z 4.2 $3,766.00

Arrhythmia Mild 0.5 (± 1.2) $35.57 (± $95.65) ECG, EUC $46.65 Arrhythmia Moderate 2.5 (± 2.1) $136.62 (±

$117.66) ECG, EUC, FBE $63.85

Arrhythmia Severe 4.6 (± 2.5) $219.99 (± $122.80)

ECG, EUC, FBE $63.85 F70A, F70B, F71A, F71B Mean of four DRGs weighted for number of occurrences

3 $2,640.48

Asthma Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Asthma Moderate 0.6 (± 0.9) $43.48 (± $73.52) Spiro $18.95 Asthma Severe 3.2 (± 2.6) $165.60 (±

$135.39) Spiro, CXR, sputum MCS

$113.7 E69A, E69B Mean of two DRGs weighted for number of occurrences

3.5 $3,194.67

Bleeding, non-specific Mild 0.3 (± 0.7) $26.09 (± $56.13) FBE $17.2 Bleeding, non-specific Moderate 2.0 (± 1.8) $116.77 (±

$105.99) FBE $17.2

12

FINAL REPORT Consequence Severity

description Specialist visits (±SD)

Cost of specialist visits (±SD)

Investigations Cost of investigations

AR-DRG code (version 5.1, national public hospitals 2006-2007)

Comments re DRG values

Average duration of admission (days)

Cost of admission

Bleeding, non-specific Severe 3.8 (± 2.7) $190.21 (± $135.15)

FBE, COAG $31.25 B63Z, B70A, B70B, B70C, B70D, B78A, B78B, E67A, E67B, E73A, E73B, E73C, F42A, F75A, F75B, F75C, G42A, G42B, G45A, G45B, G61A, G61B, G70A, G70B, H63A, H63B, L65A, L65B, L67A, L67B, L67C, Q60A, Q60B, Q60C, Q61A, Q61B, Q61C, S65A, S65B, S65C

Mean of forty DRGs weighted for number of occurrences

3.6 $3,393.41

Bone marrow suppression

Mild 0.5 (± 0.7) $39.53 (± $56.13) FBE $17.2

Bone marrow suppression

Moderate 2.6 (± 1.6) $142.57 (± $86.64)

FBE $17.2

Bone marrow suppression

Severe 5.4 (± 2.8) $253.73 (± $131.56)

FBE, BMBx $118.75 Q61A, Q61B, Q61C, Q60A, Q60B, Q60C, Q62Z

Mean of seven DRGs weighted for number of occurrences

2.2 $2,114.07

Cerebrovascular event Mild 1.7 (± 1.4) $106.84 (± $89.24)

FBE, carotid doppler

$186.7

Cerebrovascular event Moderate 3.3 (± 2.5) $168.38 (± $127.45)

FBE, carotid doppler, CT brain

$381.75 B69A, B69B Mean of two DRGs weighted for number of occurrences

3.6 $3,172.63

Cerebrovascular event Severe 6.2 (± 4.7) $283.51 (± $216.66)

FBE, carotid doppler, CT brain

$381.75 B70A, B70B, B70C, B70D Mean of four DRGs weighted for number of occurrences

8.6 $7,901.59

Chronic Airways Disease

Mild 0.1 (± 0.3) $7.91 (± $25.30) Spiro $18.95

Chronic Airways Disease

Moderate 1.2 (± 1.4) $85.01 (± $102.01)

Spiro, CXR $79.7

Chronic Airways Disease

Severe 4.4 (± 2.5) $212.05 (± $122.84)

Spiro, CXR, sputum MCS

$113.7 E65A, E65B Mean of two DRGs weighted for number of occurrences

6.2 $4,969.64

CNS Depression Mild 0.3 (± 0.7) $23.72 (± $52.96) Nil $0 CNS Depression Moderate 1.7 (± 1.3) $106.84 (±

$84.22) FBE, EUC $35

CNS Depression Severe 9.1 (± 14.6)

$400.62 (± $644.51)

FBE, EUC, CT brain

$230.05 B80Z, W61Z Mean of two DRGs weighted for number of occurrences

2 $9,619.77

Confusion Mild 0.4 (± 0.7) $31.62 (± $55.34) FBE $17.2 Confusion Moderate 1.2 (± 1.3) $86.99 (± $95.69) FBE, EUC,

CXR, FolB12 $139.5

Confusion Severe 4.0 (± 2.9) $196.17 (± FBE, EUC, $334.55 B64A, B64B Mean of two 7.6 $5,662.98

13

FINAL REPORT Consequence Severity

description Specialist visits (±SD)

Cost of specialist visits (±SD)

Investigations Cost of investigations

AR-DRG code (version 5.1, national public hospitals 2006-2007)

Comments re DRG values

Average duration of admission (days)

Cost of admission

$143.52) CXR, FolB12, CT brain

DRGs weighted for number of occurrences

Constipation Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Constipation Moderate 0.2 (± 0.5) $15.02 (± $37.94) Nil $0 Constipation Severe 2.0 (± 1.3) $118.75 (±

$79.56) Colon $308.85 G65A, G65B Mean of two

DRGs weighted for number of occurrences

3.8 $3,538.17

Dementia Mild 0.6 (± 1.1) $47.43 (± $84.58) FBE, TFT, FolB12

$236.35

Dementia Moderate 1.9 (± 0.8) $112.80 (± $45.73)

FBE, TFT, FolB12

$236.35

Dementia Severe 2.5 (± 1.7) $138.60 (± $91.48)

FBE, TFT, FolB12

$236.35 B67A, B67B, B67C Mean of three DRGs weighted for number of occurrences

6.4 $4,832.71

Depression Mild 0.3 (± 1.0) $23.72 (± $75.10) FBE, TFT $192.6 Depression Moderate 0.9 (± 0.9) $67.19 (± $74.31) FBE, TFT $192.6 Depression Severe 5.2 (± 3.9) $245.79 (±

$181.98) FBE, TFT $192.6 U63A, U63B, U64Z Mean of three

DRGs weighted for number of occurrences

12.8 $7,932.04

Diabetes Mild 0.1 (± 0.3) $7.91 (± $25.30) FBG, ACR $30.25 Diabetes Moderate 0.8 (± 0.9) $59.29 (± $67.98) FBG, ACR $30.25 Diabetes Severe 4.3 (± 2.1) $208.08 (±

$101.34) FBG, ACR, Urine MCS, HbA1c

$68.05 K60A, K60B Mean of two DRGs weighted for number of occurrences

4.8 $4,481.62

Diarrhoea Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Diarrhoea Moderate 0.4 (± 0.7) $31.62 (± $55.34) Faeces MCS $53.25 Diarrhoea Severe 2.3 (± 2.1) $128.68 (±

$121.24) Faeces MCS, colon

$362.1 G43Z, G44A, G44B, G44C, G67B, G68A

Mean of six DRGs weighted for number of occurrences

2.1 $4,220.00

Gastrointestinal bleeding

Mild 0.3 (± 0.7) $23.72 (± $52.96) FOB $9.05

Gastrointestinal bleeding

Moderate 2.1 (± 1.1) $120.74 (± $65.96)

FOB, FBE $26.25 G61B 1.6 $1,501.00

Gastrointestinal bleeding

Severe 3.8 (± 2.7) $190.21 (± $133.15)

FOB, FBE, Endo

$189.9 G61A 3.3 $2,830.00

Gastrointestinal pain Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Gastrointestinal pain Moderate 0.6 (± 0.7) $45.85 (± $52.96) FBE $17.2 Gastrointestinal pain Severe 2.0 (± 1.1) $118.75 (±

$67.09) FBE, Endo $180.85 G67A 5.2 $4,597.00

Glaucoma Mild 1.0 (± 1.5) $79.05 (± $118.58)

LVA $32.85

14

FINAL REPORT Consequence Severity

description Specialist visits (±SD)

Cost of specialist visits (±SD)

Investigations Cost of investigations

AR-DRG code (version 5.1, national public hospitals 2006-2007)

Comments re DRG values

Average duration of admission (days)

Cost of admission

Glaucoma Moderate 3.2 (± 1.7) $166.39 (± $88.39)

LVA $32.85

Glaucoma Severe 4.7 (± 2.9) $225.94 (± $139.41)

LVA $32.85 C15A, C15B Mean of two DRGs weighted for number of occurrences

1.5 $2,976.15

Headache Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Headache Moderate 0.5 (± 1.1) $39.53 (± $84.58) Nil $0 Headache Severe 1.5 (± 1.2) $98.90 (± $79.12) FBE, CT head $212.25 B77Z 1.6 $1,566.00 Heart Failure Mild 1.3 (± 1.7) $88.98 (±

$118.87) FBE $17.2

Heart Failure Moderate 2.1 (± 1.1) $123.91 (± $65.74)

FBE, Echo $247.85

Heart Failure Severe 4.3 (± 2.1) $208.08 (± $103.79)

FBE, Echo, CXR

$308.6 E64Z, F62A, F62B Mean of three DRGs weighted for number of occurrences

6.6 $5,518.11

Hypercalcaemia Mild 0.5 (± 0.8) $39.53 (± $60.08) Nil $0 Hypercalcaemia Moderate 2.0 (± 1.6) $118.75 (±

$95.00) CaPhos, PTH $40.45

Hypercalcaemia Severe 4.1 (± 1.4) $203.31 (± $66.95)

CaPhos, PTH $40.45 K62A, K62B, K62C Mean of three DRGs weighted for number of occurrences

3.7 $3,437.85

Hyperkalaemia Mild 0.0 (± 0.0) $0.00 (± $0.00) EUC $17.8 Hyperkalaemia Moderate 0.8 (± 0.9) $59.29 (± $70.35) EUC, FBE $35 Hyperkalaemia Severe 3.5 (± 1.7) $178.30 (±

$86.09) EUC, FBE $35 F70A, F70B, F71A, F71B Mean of four

DRGs weighted for number of occurrences

3 $2,640.48

Hypertension Mild 0.1 (± 0.4) $10.28 (± $27.67) EUC $17.8 Hypertension Moderate 0.0 (± 0.0) $0.00 (± $0.00) EUC, Echo $248.45 Hypertension Severe 3.1 (± 1.5) $163.61 (±

$76.32) EUC, Echo $248.45 F67A, F67B Mean of two

DRGs weighted for number of occurrences

3 $2,467.71

Hyperthyroidism Mild 0.0 (± 0.0) $0.00 (± $0.00) TFT $175.4 Hyperthyroidism Moderate 1.8 (± 1.5) $108.83 (±

$92.66) TFT $175.4

Hyperthyroidism Severe 3.1 (± 1.1) $163.61 (± $59.07)

TFT $175.4 K06Z 2.4 $6,406.00

Hypocalcaemia Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Hypocalcaemia Moderate 1.4 (± 1.1) $94.14 (± $72.31) CaPhos, PTH $40.45 Hypocalcaemia Severe 3.6 (± 1.3) $183.46 (±

$65.70) CaPhos, PTH $40.45 K62A, K62B, K62C Mean of three

DRGs weighted for number of occurrences

3.7 $3,437.85

15

FINAL REPORT Consequence Severity

description Specialist visits (±SD)

Cost of specialist visits (±SD)

Investigations Cost of investigations

AR-DRG code (version 5.1, national public hospitals 2006-2007)

Comments re DRG values

Average duration of admission (days)

Cost of admission

Hypoglycaemia Mild 0.0 (± 0.0) $0.00 (± $0.00) BSL $9.75 Hypoglycaemia Moderate 0.9 (± 1.0) $69.56 (± $78.26) BSL $9.75 Hypoglycaemia Severe 3.5 (± 2.0) $178.30 (±

$101.89) BSL $9.75 K60A, K60B, K62A, K62B,

K62C Mean of five DRGs weighted for number of occurrences

5.3 $4,023.69

Hypokalaemia Mild 0.3 (± 0.7) $19.76 (± $56.13) EUC $17.8 Hypokalaemia Moderate 0.9 (± 1.1) $69.56 (± $89.33) EUC, FBE $35 Hypokalaemia Severe 3.3 (± 0.7) $168.38 (±

$36.78) EUC, FBE $35 F70A, F70B, F71A, F71B Mean of four

DRGs weighted for number of occurrences

3 $2,640.48

Hypotension Mild 0.0 (± 0.0) $0.00 (± $0.00) EUC $17.8 Hypotension Moderate 0.4 (± 0.5) $30.04 (± $41.11) EUC, Echo $248.45 Hypotension Severe 3.7 (± 2.3) $185.84 (±

$113.32) EUC, Echo $248.45 F73A, F73B, F75A, F75B,

F75C Mean of five DRGs weighted for number of occurrences

3.4 $3,197.70

Hypothyroidism Mild 0.0 (± 0.0) $0.00 (± $0.00) TFT $175.4 Hypothyroidism Moderate 1.0 (± 1.4) $79.05 (±

$111.46) TFT $175.4

Hypothyroidism Severe 2.3 (± 1.4) $128.68 (± $79.49)

TFT $175.4 K64A, K64B Mean of two DRGs weighted for number of occurrences

3.149369 $3,850.86

Infection, general Mild 0.5 (± 1.4) $39.53 (± $111.46)

Nil $0

Infection, general Moderate 0.4 (± 1.1) $30.04 (± $83.79) FBE $17.2 Infection, general Severe 1.3 (± 1.2) $88.98 (± $82.57) FBE, CXR,

Urine MCS $98.65 B72A, B72B, C60A, C60B,

C63A, C63B, D40Z, D63A, D63B, D66A, D66B, D67A, D67B, E62A, E62B, E62C, E74A, E74B, E74C, F61Z, F75A, F75B, F75C, G67A, G67B, G68A, G68B, G69Z, G70A, G70B, H63A, H63B, I02A, I02B, I12A, I12B, I12C, I20Z, I30Z, I67A, I67B, J12A, J12B, J12C, J60A, J60B, J64A, J64B, J67A, J67B, K62A, K62B, K64A, K64B, L60A, L60B, L60C, L67A, L67B, L67C, Q60A, Q60B, Q60C, Q61A, R01B, R03B, R61A, R61B, R61C, S65A, S65B, S65C, T60B, T61B, T63B, T64B, U62B, U63B

Mean of 78 DRGs weighted for number of occurrences

3.688887 $3,508.95

16

FINAL REPORT Consequence Severity

description Specialist visits (±SD)

Cost of specialist visits (±SD)

Investigations Cost of investigations

AR-DRG code (version 5.1, national public hospitals 2006-2007)

Comments re DRG values

Average duration of admission (days)

Cost of admission

Insomnia Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Insomnia Moderate 0.1 (± 0.4) $10.28 (± $27.67) FBE $17.2 Insomnia Severe 1.3 (± 1.3) $88.98 (± $91.11) FBE $17.2 Liver Disease Mild 0.0 (± 0.0) $0.00 (± $0.00) LFTs $17.8 Liver Disease Moderate 1.8 (± 0.9) $108.83 (±

$55.35) LFTs, FBE $35

Liver Disease Severe 3.8 (± 1.2) $188.23 (± $58.22)

LFTs, FBE $35 H63A, H62B Mean of two DRGs weighted for number of occurrences

4.137553 $4,058.04

Myocardial Ischaemia Mild 0.9 (± 1.5) $69.56 (± $115.41)

Chol, FBE $28.95

Myocardial Ischaemia Moderate 2.3 (± 1.5) $128.68 (± $85.21)

Chol, FBE, Echo

$259.6 F74Z 1.5 $1,464.00

Myocardial Ischaemia Severe 4.1 (± 2.1) $203.31 (± $103.38)

Chol, FBE, Echo, angio

$564.8 F41A, F41B, F42A, F42B, F60A, F60B, F60C, F72A, F72B

3.39035 $5,414.61

Myopathy Mild 0.1 (± 0.4) $10.28 (± $27.67) Nil $0 Myopathy Moderate 0.8 (± 1.2) $59.29 (± $91.70) FBE, CK $26.95 Myopathy Severe 2.1 (± 1.7) $123.91 (±

$100.64) FBE, CK $26.95

Nausea Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Nausea Moderate 0.6 (± 1.2) $49.80 (± $94.07) EUC $17.8 Nausea Severe 1.5 (± 1.1) $98.90 (± $70.55) EUC $17.8 K62A, K62B, K62C Mean of three

DRGs weighted for number of occurrences

3.7 $3,437.85

Oedema Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Oedema Moderate 0.3 (± 0.7) $19.76 (± $56.13) EUC, CXR $78.55 Oedema Severe 2.6 (± 1.3) $143.76 (±

$71.06) EUC, CXR $78.55 E64Z 5.7 $5,408.00

Osteoporosis Mild 0.1 (± 0.4) $10.28 (± $27.67) DXA, vit D $137.6 Osteoporosis Moderate 1.8 (± 1.4) $108.83 (±

$86.44) DXA, vit D $137.6

Osteoporosis Severe 3.6 (± 1.2) $183.46 (± $60.14)

DXA, vit D $137.6 I60Z, I61Z, I63Z, I74A, I74B, I74C, I75A, I75B, I75C, I77A, I77B, I78A, I78B

Mean of thirteen DRGs weighted for number of occurrences

3.724095 $3,490.57

Pain Mild 0.0 (± 0.0) $0.00 (± $0.00) Nil $0 Pain Moderate 0.5 (± 0.9) $39.53 (± $73.52) Nil $0 Pain Severe 3.4 (± 1.5) $173.54 (±

$77.53) Nil $0 B03A, B03B, B07A, B07B Mean of four

DRGs weighted for number of occurrences

3.726299 $7,193.60

Parkinsonism Mild 1.2 (± 1.2) $86.99 (± $86.99) FBE, CT head $212.25 Parkinsonism Moderate 2.8 (± 1.0) $150.51 (±

$53.75) FBE, CT head $212.25

17

FINAL REPORT Consequence Severity

description Specialist visits (±SD)

Cost of specialist visits (±SD)

Investigations Cost of investigations

AR-DRG code (version 5.1, national public hospitals 2006-2007)

Comments re DRG values

Average duration of admission (days)

Cost of admission

Parkinsonism Severe 4.5 (± 3.3) $218.00 (± $159.87)

FBE, CT head $212.25 B67A, B67B, B67C Mean of three DRGs weighted for number of occurrences

6.4 $4,832.71

Psychosis Mild 0.2 (± 0.4) $15.81 (± $31.62) FBE $17.2 Psychosis Moderate 6.2 (± 8.8) $285.49 (±

$405.21) FBE, CT head $212.25

Psychosis Severe 6.2 (± 3.4) $285.49 (± $156.56)

FBE, CT head $212.25 U61A, U61B, U62A, U62B Mean of four DRGs weighted for number of occurrences

23.78974 $10,683.33

Rash Mild 0.2 (± 0.4) $15.81 (± $31.62) Nil $0 Rash Moderate 1.5 (± 1.5) $98.90 (± $98.90) Nil $0 Rash Severe 3.5 (± 2.4) $178.30 (±

$122.26) Skin biopsy $48.25 J67A, J67B, J68A, J68B Mean of four

DRGs weighted for number of occurrences

2.475833 $2,220.69

Renal Dysfunction Mild 0.2 (± 0.4) $15.81 (± $31.62) UEC, FBE $35 Renal Dysfunction Moderate 1.8 (± 1.2) $110.81 (±

$73.87) UEC, FBE $35

Renal Dysfunction Severe 3.7 (± 1.8) $186.24 (± $90.60)

UEC, FBE, ACR $55.5 L60A, L60B, L60C Mean of three DRGs weighted for number of occurrences

6.524047 $6,395.05

Respiratory depression Mild 0.2 (± 0.4) $15.81 (± $31.62) Nil $0 Respiratory depression Moderate 0.8 (± 1.0) $63.24 (± $79.05) Spiro $18.95 Respiratory depression Severe 3.5 (± 2.1) $178.30 (±

$106.98) Spiro, ABG $53.25 E67A, E67B, E75A, E75B,

E75C Mean of five DRGs weighted for number of occurrences

3.29972 $3,568.52

Seizures Mild 1.2 (± 1.6) $86.99 (± $115.99)

FBE, CT head, EUC

$230.05

Seizures Moderate 2.7 (± 2.2) $146.54 (± $119.40)

FBE, CT head, EUC

$230.05

Seizures Severe 4.7 (± 2.9) $225.94 (± $139.41)

FBE, CT head, EUC, MRI head

$633.25 B76A, B76B Mean of five DRGs weighted for number of occurrences

2.603352 $2,780.51

Serotonin toxicity Mild 0.2 (± 0.4) $15.81 (± $31.62) Nil $0 Serotonin toxicity Moderate 3.2 (± 4.4) $166.39 (±

$228.79) FBE $17.2

Serotonin toxicity Severe 2.3 (± 1.2) $130.66 (± $68.17)

FBE $17.2

Urinary Incontinence Mild 0.2 (± 0.4) $15.81 (± $31.62) Urine MCS $20.7 Urinary Incontinence Moderate 1.5 (± 0.8) $98.90 (± $52.75) Urine MCS $20.7 Urinary Incontinence Severe 3.0 (± 1.1) $158.45 (±

$58.10) Urine MCS, FBE

$37.9 L65A, L65B Mean of two DRGs weighted

2.83597 $2,554.40

18

FINAL REPORT Consequence Severity

description Specialist visits (±SD)

Cost of specialist visits (±SD)

Investigations Cost of investigations

AR-DRG code (version 5.1, national public hospitals 2006-2007)

Comments re DRG values

Average duration of admission (days)

Cost of admission

for number of occurrences

Urinary retention Mild 0.2 (± 0.4) $15.81 (± $31.62) Urine MCS $20.7 Urinary retention Moderate 1.2 (± 1.0) $86.99 (± $72.49) Urine MCS $20.7 Urinary retention Severe 2.7 (± 1.4) $146.54 (±

$75.98) Urine MCS $20.7 L06A, L06B, L09A, L09B Mean of four

DRGs weighted for number of occurrences

2.931248 $4,159.12

Urinary Tract Infection Mild 0.2 (± 0.4) $15.81 (± $31.62) Nil $0 Urinary Tract Infection Moderate 0.3 (± 0.8) $23.72 (± $63.24) FBE, Urine

MCS $37.9

Urinary Tract Infection Severe 1.0 (± 1.1) $79.05 (± $86.96) FBE, Urine MCS, Renal US

$147 L63A, L63B, L63C Mean of three DRGs weighted for number of occurrences

4.330972 $3,644.55

'ACR'-Albumin:creatinine ratio, 'Angio'-Angiography, 'ABG'-Arterial Blood Gasses, 'FolB12'-B12 and Folate, 'BSL'-Blood glucose level, 'BMBx'-Bone marrow biopsy, 'CaPhos'-Calcium and phosphate (serum), 'Carotid doppler'-Carotid doppler, 'CXR'-Chest Xray, 'COAG'-Coagulation studies, 'Colon'-Colonoscopy, 'CT head'-Computed tomography head, 'DXA'-Dual energy X-ray absorptiometry, 'Echo'-Echocardiogram, 'Endo'-Endoscopy, 'FOB'-Faecal occult blood, 'Faeces MCS'-Faeces MCS, 'FBE'-Full Blood Examination, 'HbA1c'-Glycosylated haemoglobin, 'RAST'-IgE detection, 'FeS'-Iron Studies, 'LS'-Lipid studies, 'LFT'-Liver function tests, 'LVA'-Low vision assessment, 'MRI Head'-Magnetic Resonance Imaging head, 'PTH'-Parathyroid hormone, 'Renal US'-Renal Ultrasound, 'Skin biopsy'-Skin biopsy, 'Spiro'-Spirometry, 'Sputum MCS'-Sputum MCS, 'TFT'-Thyroid function tests, 'UEC'-Urea, Electrolytes, Creatinine, 'Urine MCS'-Urine MCS, 'Vit D'-Vitamin D, 'ECG'-Electrocardiogram, 'CK'-Creatine kinase

Table 11 - Consequences table values for specialist services, investigations and hospitalisations

19

FINAL REPORT

Appendix III Derivation of Consequences table values Introduction A critical element of the economic assessment methodology used in the PROMISe II study and subsequently the VALMER study was the consequences table. This table described common adverse medical conditions or symptoms resulting from medication misadventure, termed consequences. Additionally, each consequence was assigned an estimate of the healthcare resources that would be required to resolve the consequence, should a patient experienced that consequence. For the PROMISe II study, the healthcare resources assigned to each consequence were:

• Impact on Health Status o This was a scale of impact based on the severity of the particular consequence, with 1 being mild

impact on health and 3 being a severe impact on health. • Duration of health status impact

o This was a value in days of the duration of the health impact. • Duration and cost of hospital admission

o The duration (in days) and cost of any hospital admission associated with the consequence. • Number and cost of general practitioner consultations

o The number and cost of community based general practitioner consultations required to manage the particular consequence.

• Number and cost of specialist consultations o The number and cost of specialist consultations required to manage the particular consequence.

• Investigations and pathology costs o The costs of typically required investigation or pathology tests required in the management of the

particular consequence. An example of the values assigned to the consequence of “Seizures” in PROMISe II is shown in Table 12.

Table 12 - Values assigned to PROMISe II consequence of "Seizures"

In PROMISE II, the values assigned to each of these parameters were obtained from combination of literature sources and expert opinion. For hospitalisation data, the duration and cost was available from Australian Hospital Statistics.30 To value the remaining parameters, a two-stage process was used to generate point estimates for each parameter. Initial estimates for these parameters were made by the project team, and then reviewed and modified by a consensus group process that included a physician, a general practitioner and two experienced clinical pharmacists. For the VALMER study, several changes to the consequences table were undertaken to improve the validity of the economic analysis. These changes involved

• updating the table to include the most current hospitalisation data • attempting to address quality of life issues associated with the consequences, and

Severity Code Severity Description

Health Status Impact

Duration of Health

Status Impact

Duration of Hospital

Admission

Cost of Hospital

Admission

Number of GP

Consults

Cost of GP

Consults

Mild Mild one-off seizure unlikely to recur or require management

1 7 0.00 $0 2 $56

Moderate Requiring medical attention or modification of medication regimen

2 60 0.00 $0 4 $112

Severe Severe seizures requiring hospitalisation and intravenous anticonvulsants

3 90 2.05 $1,606 2 $56

Severity Code

Number of Specialist Consults

Cost of Specialist Consults

Investigation or Other Cost

Total Direct Costs

Admission Cost and Duration Source

Investigation or Other Cost Notes

Mild 0 $0 $104 $160 EEG Moderate 0 $0 $122 $234 EEG, Bloods Severe 2 $192 $0 $1,854 B76B

20

FINAL REPORT• exploring the uncertainty around the point estimates of the various health resource parameters included in

the table. This section details these changes and the finalisation of the VALMER consequences table. Changes made for the VALMER study

1. Quality of Life Defining Utilities

The “health-status impact” used in the PROMISe II methodology was an arbitrary scale of the effect of each consequence on quality of life (QOL). Essentially, this measure assumes that one day of ill health due to one consequence at a particular severity level has the same effect on QOL as any other consequence. This assumption is inconsistent both intuitively and in the health economic literature. To illustrate this issue, consider the QOL effects of the mild level of two consequences, bleeding and cerebrovascular ischaemia. According to the PROMISe II methodology, the QOL of a patient with minor bruising is equivalent to the QOL of a patient experiencing a temporary ischaemic attack. Clinicians familiar with both of these consequences would most likely agree that this is an unreasonable assumption. To address this issue, the VALMER research team sought to replace the “health-status impact” with a unit more frequently employed in health economics. Most commonly, the influence of a particular disease or condition on QOL is expressed in terms of utility (Q, also referred to as utility weights, quality weights, preference weights or preference values).46 These utility weights reflect the desirability of residing or existing in a particular health state. Typically, each health state is rated on a scale from 1.0 (the best health attainable) to 0 (dead)‡‡. Examples of utilities for various conditions are shown in Table 13.

Condition Utility Condition Utility Osteoarthritis of the hip Stroke resulting in cognitive deficit

• Mild 0.69 • Mild 0.54 • Moderate 0.38 • Moderate 0.37 • Severe 0.19 • Severe 0.08

Angina Chronic hepatitis 0.94 • Mild 0.88 • Moderate 0.832 Chronic renal disease 0.63 • Severe 0.533

Table 13 - Examples of utilities for various medical conditions reported in previous studies 47

A related measure is the disability adjusted life year (DALY). In a similar fashion to the utility weights of QALYs, DALYs are also calculated using measures of QOL anchored between 0 and 1, termed disability weights (D). A fundamental difference between Qs and Ds, however, is that a D of 0 indicates full health and a D of 1.0 is equivalent to death. This inversion is because the QALY (quality adjusted life year) measures equivalent healthy years lived, whereas the DALY measures years of lost health. The D for a given condition i may therefore be converted to a Q using Equation 2, and vice versa.

Qi = 1 - Di

Equation 2 - Conversion of D weights to Q weights

Use of utilities in VALMER To develop utilities for the VALMER consequences, three techniques that have been employed in previous studies were considered. These were direct utility measurement, literature values and health state measures.48 Direct utility measurement involves the estimation of utilities using techniques such as rating scale (also referred to as visual-analogue scale), standard gamble (SG), time trade-off (TTO), person trade-off (PTO) and magnitude estimation. There has been considerable debate as to which of these techniques is the most appropriate, and currently none is seen as being definitively superior to the others.49 Direct utility measurement via any of these techniques is a time consuming and complex task. Given these complexities and the limited budget allocated for incorporating utilities into the study, the project team chose to not develop utilities for this study using direct measurement techniques. ‡‡ Some systems permit health states to be rated at less than 0, or “worse than death”. 

21

FINAL REPORTLiterature sources

A literature search was therefore conducted to identify sets of utilities that had been developed in previous research. Two large studies were identified that developed generic utilities for many chronic medical conditions. The first of these was the World Health Organisation 1990 Global Burden of Disease (GBD) study.31 This study developed and utilised disability weights using a PTO technique to provide estimates of lost health due to 107 disease-related causes in terms of DALYs. A significant limitation of the GBD disability weights are that they have been developed for a global context rather than a national one. This implies that a particular disease will cause the same level of disability in an individual regardless of their living conditions, an assumption criticised by several authors.32 Additionally, the GBD study valued many diseases which have little relevance to the health of developed nations, such as leprosy, filariasis and malaria. To address these two limitations, researchers from the Netherlands modified the GBD study PTO methodology in the “Dutch weights” study.32 The Dutch weights study derived disability weights for 52 diseases of public health importance to western European nations. Not only did this study provide disability weights for these diseases, it also weighted the disability of different stages or severity levels of many of these diseases. The study ultimately provided disability weights for 175 different disease stages, sequelae or severity levels for the 52 diseases. Another major strength of this study was that the functional health status of each disease stage or severity was described using a standardised format in addition to a short vignette, as used in the GBD study. This additional description was introduced to more readily compare different diseases or states of the same disease. The utilities developed for the GBD and Dutch weights studies were utilised in the Australian Burden of Disease and Injury (ABDI) studies to estimate the health lost due to disease and injury in Australia.50, 51 The project team therefore initially followed this approach to assign utilities to the VALMER consequences. Utilities derived in the Dutch weights study were used as a primary reference. For consequences which were not described by the Dutch weights, the GBD utilities were used where available. Of the 156 levels of severity of the 52 consequences formulated for VALMER, the Dutch weights and GBD studies directly correlated with forty-one. The project team was therefore faced with the question of the most appropriate way to obtain utilities for the remaining consequences.

Developing utilities To assign utilities to the remaining consequences (i.e. those not described by the Dutch weights or GBD studies), previous studies investigating QOL for health conditions similar to the consequences were assessed. However this approach proved problematic, despite large summaries of studies that reported utilities for numerous medical conditions being available.47, 52 Determining consistent, accurate and comparable utilities from literature sources for the VALMER consequences was not a straightforward process, primarily as utilities derived from one technique (such as SG or TTO) often do not directly equate to utilities obtained using an alternate technique.49, 53 Given these limitations regarding literature values, it was decided to investigate the potential for developing utilities specifically for the study using another technique. As discussed previously, direct utility measurement (using SG, TTO, RS etc) is complex and time-consuming. An alternative that is used extensively in health economic literature is to use a pre-scored multi-attribute health status classification system. These systems use a variety of means (usually questions relating to QOL) to map responses to the system to value sets derived by one of the direct utility measurement techniques. Multi-attribute health status classification system instruments may be either generic or disease-specific. Generic heath status instruments allow the comparison of QOL between essentially any different disease or condition, whereas disease-specific instruments are limited to the diseases which they were developed to assess. Examples of these instruments are shown in Table 14.

22

FINAL REPORT

Type Examples Generic QOL instruments • Medical Outcome Study Short Form Health Surveys (eg SF-12,

SF-36) • EuroQOL-5D (EQ-5D) • Health Utilities Index (HUI-2 and HUI-3) Disease-specific QOL instruments Asthma • Living With Asthma Questionnaire • Life Activities Questionnaire for Adult Asthma Diabetes mellitus • Diabetes-Specific QOL Instrument (DQOL) HIV/AIDS • Functional Assessment of HIV Infection • HIV Patient-Reported Status and Experience Scale (HIV-PARSE) • AIDS Health Assessment Questionnaire (AIDS-HAQ) Osteoporosis • Osteoporosis QOL Questionnaire (OQOL)

Table 14 - Examples of QOL instruments

Both types of instruments assess several different aspects of the patient’s health status (termed “domains”). These domains include physical functioning, psychological functioning, social role and functioning, and general health perceptions.53 The instrument may report the patient’s health status as different scores for each domain or as a single summary index. One of the most widely used generic health status instruments is the EuroQOL (EQ-5D).54 It was developed to provide a simple, generic measure of health for clinical and economic assessments.§§ The EQ-5D consists of two pages- a visual analogue scale and a descriptive system (Figure 7).

By placing a tick in one box in each group below, please indicate which statements best describe your own health state today Mobility I have no problems in walking around I have some problems in walking around I am confined to bed Personal Care I have no problems with personal care I have some problems washing or dressing myself I am unable to wash or dress myself Usual Activities (e.g. work, study, housework, family or leisure activities)

I have no problems with performing my usual activities I have some problems with performing my usual activities I am unable to perform my usual activities Pain/Discomfort I have no pain or discomfort I have moderate pain or discomfort I have extreme pain or discomfort Anxiety/Depression I am not anxious or depressed I am moderately anxious or depressed I am extremely anxious or depressed

Figure 7 - EQ-5D descriptive system 54

§§ A slightly modified version of the EQ-5D was utilised in deriving the Dutch weights 

23

FINAL REPORTThe descriptive system assesses QOL in five domains - mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each domain may be allocated three levels of health impact– no problems (level 1), some problems (level 2) and severe problems (level 3). Respondents are asked to indicate which level best describes their health status for each domain. The five individual responses may then be combined into a five digit number that describes the respondent’s health state. For example, a patient who responds that they have no problems with mobility, self-care or usual activities, but severe pain and moderate anxiety, would be summarised as 11132. As each of the five domains has three possible responses, there are 243 possible health states that can be defined by the instrument. A response to the EQ-5D descriptive system may be converted to a utility using a value set based on the general population. Whilst there are value sets for several countries, currently a value set has not been created for the Australian population. Many Australian studies have therefore used the United Kingdom TTO value set 54 as this population’s demographics are closest to Australia’s. EQ-5D descriptive system responses may be mapped to United Kingdom TTO utilities using Equation 3.

Q= 1 - 0.081 - 0.069MO2 - 0.314MO3 - 0.104SC2 - 0.214SC3 - 0.036UA2 - 0.094UA3 - 0.123PD2 - 0.386PD3 - 0.071AD2 - 0.236AD3 - 0.269N3

Where MOX = mobility SC X = self care UAX = usual activities PDX = pain/discomfort

AD X = anxiety/depression N3=1 if one or more “severe” responses, 0 if nil

Equation 3 - Conversion of EQ-5D descriptive system responses to UK TTO utilities 54

As the Dutch weights utilised a slightly modified version of the EQ-5D in deriving them, the VALMER research team therefore chose to undertake a small study using the EQ-5D to develop utilities for the consequences not described in the Dutch weights or GBD studies.

The VALMER utilities study

Participants

The first decision that was made was who should undertake the assessment of the consequences. Both the GBD and Dutch weights investigators decided to base their weights on the assessment of each condition by health professionals. The rationale for this was that knowledge of and insight into each condition was deemed to be essential for the validity of the studies.55 For the Dutch weight study, three panels of fifteen medical experts with general medical experience and extensive research training (all of whom were medical doctors, predominantly GPs) completed the study. For the VALMER utility study, six specialist physicians were invited to participate based on their experience in primary care medicine. No formal method of recruitment was applied.

Methods

Participation in the study involved completion of an online survey whereby each participant was shown the description of each level of severity of each consequence asked to rate the QOL of each using the EQ-5D descriptive system. Prior to commencing the study, each participating medical practitioner was provided with an information sheet. An example of the online form is shown in Figure 8.

24

FINAL REPORT

Figure 8 - Utilities study data entry screen

A significant deviation from the methodology employed in the GBD and Dutch weight studies involved how the duration of the consequences were addressed. These studies primarily valued chronic diseases, and considered that each health state remained constant throughout the year. Relatively few states were comprised of diseases occurring either in an episodic pattern (e.g. an asthma attack) or with a brief duration followed by a full recovery (e.g. headache or constipation). For episodic conditions, the Dutch weights did not assess attacks; rather, such conditions were described as chronic and measures to prevent attacks, the side-effects of such measures and the fear of suffering an attack were included in the description of the consequence. Brief conditions followed by a full recovery were presented for valuation as an annual profile, e.g. one year in good health with two weeks of influenza during that year. Hence the entire year, and not simply the episode of illness was presented for assessment. For the VALMER utility study, the majority of the consequences were single incidents described by an expected duration. The participants were therefore asked to indicate the expected QOL of a person currently experiencing the consequence rather than asking them to extrapolate the QOL effects of the consequence over a 12-month period. Once the immediate QOL effects of the consequences were determined, they were then extrapolated to a 12-month period using the duration data assessed in a subsequent study.

Results

The six participating specialists completed the study over a one month period in 2008. Four specialists were Tasmanian, one was South Australian and one was practicing in Western Australia. The level of agreement between the specialists was low (Fleiss’ kappa = 0.15), although this was expected with the large number of categories assignable by the EQ-5D (243). The results of the study for the consequences at a mild level of severity are shown in Figure 9.

25

FINAL REPORT

Figure 9 - Results of utility study for consequences experienced at a mild level of severity (bars indicate standard error of the mean)

On face value, the consequences which would be expected to result in a more substantial loss of QOL were so rated by the experts. These included conditions such as cerebrovascular events, seizures and myocardial ischaemia. Conditions likely to result in minimal symptoms (if any), such as minor biochemical abnormalities and hypertension, were rated as having little effect on QOL. The results for the consequences at a moderate level of severity are shown in Figure 10. Again, most conditions which would be expected to result in a greater loss of QOL were rated as such.

Figure 10 - Results of utility study for consequences experienced at a moderate level of severity (bars indicate standard error of the mean)

The utilities derived for the consequences at a severe level of severity are shown in Figure 11. In general there appears to be substantially greater variability between the assessors compared to the mild severity level.

26

FINAL REPORT

Figure 11 - Results of utility study for consequences experienced at a severe level of severity (bars indicate standard error of the mean)

Comparison to reference values Table 15 compares the utilities obtained in the study to the Dutch weights and GBD for several consequences.

Condition Description

VALMER value (mean)

Reference value Comments

Severe cerebrovascular event -0.0517 0.64 Dutch weight “Stroke, mild permanent

impairments” 0.37 Dutch weight “Stroke, moderate

permanent impairments”

• Resulting in severe symptoms and signs requiring hospitalisation and medical management (e.g stroke)

0.08 Dutch weight “Stroke, severe permanent impairments”

Heart failure • Mild signs or symptoms of heart failure

(e.g NYHA class II) which resolve without intervention

0.7050 0.94 Dutch weight "Mild heart failure, NYHA class I to II"

• Resulting in significant signs and symptoms of heart failure (e.g. NYHA class III) requiring medical management by modification of medication regimen

0.4358 0.65 Dutch weight "Moderate heart failure, NYHA class III"

• Significant signs and symptoms of heart failure (e.g. NYHA class IV) requiring hospitalisation and medical management (e.g. acute pulmonary oedema)

-0.3800 0.35 Dutch weight "Severe heart failure, NYHA class IV"

Myocardial Ischaemia • Mild signs or symptoms of angina (e.g.

NYHA class I to II) which resolve without intervention

0.5913 0.92 Dutch weight "Mild stable angina pectoris, NYHA class I to II"

• Moderate myocardial ischaemia resulting in significant signs and symptoms requiring medical management by modification of medication regimen (e.g. worsened stable angina)

0.4783 0.43 Dutch weight "Severe stable angina pectoris, NYHA class III"

• Myocardial ischaemia resulting in severe symptoms and signs requiring hospitalisation and medical management (e.g unstable angina, myocardial infarction)

-0.2750 0.395 Australian Burden of Disease & Injury. Not valued in Dutch weights

Table 15 - Comparison of some weights derived in the VALMER study and reference values.

27

FINAL REPORTFor all but one of the examples listed, the utilities derived in the study are less than the reference values, i.e. the loss of QOL is greater. There are two interrelated reasons for this. The first is that the UK TTO values are not anchored between 0 and 1, unlike the reference values. As shown in Figure 11, many of the severe consequences were rated as being worse than death (a negative utility). The second reason relates to the assessors in our study being asked to rate the consequence as if the patient was currently experiencing it, rather than extrapolating the consequence to 12 months. Such differences are not unexpected, and it is somewhat debatable whether these factors will substantially confound the results of further work that uses these utilities. The EQ-5D is designed to capture a person’s health state at a particular moment in time, 54 and hence will capture health states where patients rate their QOL as being worse than death. It is unlikely that a patient would rate their QOL so poor for an entire year. Indeed, no condition in the Dutch weights or GBD studies was rated with a negative QOL. It is probable that conversion of the utilities derived in this study to represent the loss of QOL of each consequence throughout 12 months minimises both issues. In addition, as data regarding the distribution of the utilities obtained in the study is available, uncertainty analyses in subsequent studies may now undertaken.

2. Updating values of health resource utilisation assigned to consequences

Hospitalisation costs Extensive Australian data regarding the duration and cost of hospitalisation is available for many common conditions grouped according to Australian Refined Diagnosis Related Groups (AR-DRGs).30 In PROMISe II, National Hospital Cost Weights for AR-DRG Version 4.2, 2002-3 were used to calculate the hospitalisation costs. For the VALMER consequences table, the cost and median length of hospital admission (ALOS) was derived from updated data (2006-7 AR-DRG version 5.1 values for public hospitals Australia-wide). For consequences associated with several DRGs, the number of separations (occurrences) was used as a proxy indicator for the probability of each DRG occurring.*** The following worked example illustrates the process used to calculate the hospitalisation costs used in the study. The consequence of severe cerebrovascular event is described as “CVA resulting in severe symptoms and signs requiring hospitalisation and medical management (e.g stroke)”. There are four DRGs relating to hospital admissions due to stroke. These, the number of separations, their average length of stay (ALOS) and cost per separation are shown in Table 16.

DRG Description Number of separations

ALOS (days)

Mean cost per separation

B70A Stroke with catastrophic complication or comorbidity 6504 17.00 $15233.00 B70B Stroke with severe complication or comorbidity 7194 9.60 $8487.00 B70C Stroke without catastrophic or severe complication or

comorbidity 7155 6.30 $5533.00 B70D Stroke, died or transferred within 5 days 6230 1.50 $2292.00

Table 16 - DRG data for "cerebrovascular event"30

To calculate the probability of each DRG contributing to a hospitalisation resulting from the consequence, the ALOS and mean cost per separation were multiplied by the percentage of total number of separations linked to the consequence (Table 17).

DRG Number of separations

Fraction of separations

ALOS (days) x fraction of separations

Mean cost per separation x fraction of separations

B70A 6504 0.2 4.1 $3658.2 B70B 7194 0.3 2.6 $2254.4 B70C 7155 0.3 1.7 $1461.8 B70D 6230 0.2 0.3 $527.2

TOTAL 27083 8.60 $7901.59

Table 17 - Calculation of ALOS and hospitalisation costs for “cerebrovascular event” from DRG data

Using this methodology, each level of consequence resulting in hospitalisation was costed and the length of stay determined. *** it is unclear how the investigators addressed this issue in PROMISe II 

28

FINAL REPORTGP costs, specialist costs and cost of additional medical investigations

Introduction and methods

Quantifying the health-resource utilisation for the remaining parameters was problematic as datasets equivalent to the Australian Hospital Statistics are not readily available for patients managed in general practice. The most extensive Australian database for GP encounters is the Bettering the Evaluation And Care of Health (BEACH project56); unfortunately the VALMER budget did not permit for this database to be used. Subsequently, a small study to investigate the health-resource utilisation of each of the consequences was conducted. The study involved an online panel who were asked to provide their opinions as to the likely health-service utilisation for each severity level of each consequence. As per the utilities study, the research team decided to use health professional’s assessments of each condition as knowledge of each condition was again deemed to be essential for the validity of the study. Unlike the utilities study where the opinions of medical specialists were sought, this study involved GPs. This was because it was felt that GPs would be most familiar with the management of the consequences at the different levels of severity. No formal method of recruitment was applied. Each GP was asked to give their opinion regarding each of the following parameters for each consequence at each level of severity:

• duration of health status impact • number and cost of general practitioner consultations • number and cost of specialist consultations, and • investigations and pathology costs.

From the responses of the GPs, costs were then assigned to each level of each consequence. The cost used for a GP consultation (Medicare Benefits Schedule item 23) was $33.55. Pathology items were costed according to the appropriate Medicare Benefits Schedule item number. The cost used for an initial specialist visit (Medicare Benefits Schedule item 104) was $79.05. For subsequent specialist visits, the cost used was $39.70 (Medicare Benefits Schedule item 105). Prior to commencing the study, each participating GP was provided with an information sheet. Each GP provided their opinion using an online form, an example of which is shown in Figure 12.

Figure 12 - Example of assigning health-resource utilisation to consequences

29

FINAL REPORTResults

Eight general practitioners completed this study. The results of the study are illustrated in Figure 13, Figure 14 and Figure 15. Predictably, the costs and duration associated with most of the consequences increased according to the severity level. As with the utilities study, data regarding the distribution of the costs and duration of the consequences obtained in the study may be used to undertake uncertainty analyses. A comparison between some of the values obtained in this study and the PROMISe II consequences table are shown in Table 18. The table illustrates several similarities between certain total values (such as moderate myocardial ischaemia and mild and moderate cerebrovascular events), but substantial differences in others (e.g. severe cerebrovascular events). Most of the differences in cost appear to result from increases in the cost of hospitalisation. Substantial differences between the two studies with regard to the predicted duration of the consequences are evident. The assessors in the VALMER study indicated that the consequences would affect patient health for less time than the PROMISe II.

Consequence • Severity level Hospitalisation GP visits Specialist visits

Study Duration Duration Cost Number Cost Number Cost

Cost of investigations

Total costs

Myocardial Ischaemia • Mild PROMISe

II 360 0.00 $0.00 2 $56 0.0 $0.00 $35 $91 VALMER 4.2 0.00 $0.00 2.1 $71.46 0.88 $69.56 $28.95 $169.98 • Moderate PROMISe

II 360 1.60 $1,278 2 $56 2 $192 $0.00 $1,526 VALMER 11.7 1.5 $1464.00 4.2 $142.59 2.25 $128.68 $259.60 $1994.86 • Severe PROMISe

II 360 4.75 $4,526 4 $112 2 $192 $0.00 $4,639 VALMER 49 3.4 $5414.61 6 $201.30 4.13 $203.31 $564.80 $6384.02 Cerebrovascular event • Mild PROMISe

II 30 0.00 $0.00 4 $112 0.0 $0.00 $285 $397 VALMER 4.0 0.0 $0.00 3.5 $45.29 1.7 $106.84 $186.70 $410.97 • Moderate PROMISe

II 180 6.31 $4,032 6 $168 0.0 $0.00 $0.00 $4,200 VALMER 10.8 3.6 $3172.63 5 $75.82 3.25 $168.38 $381.75 $3890.50 • Severe PROMISe

II 360 8.14 $6,257 8 $224 0.0 $0.00 $0.00 $6,481 VALMER 158.2 8.6 $7901.59 6.7 $116.42 6.15 $283.51 $381.75 $8791.63 Heart Failure • Mild PROMISe

II 60 0.00 $0.00 2 $56 0.0 $0.00 $35 $91 VALMER 52.4 3.9 $130.17 1.2 $88.98 $17.20 $236.35 • Moderate PROMISe

II 360 0.00 $0.00 8 $224 0.0 $0.00 $35 $259 VALMER 71.9 7.2 $243.24 2.1 $123.91 $247.85 $615.00 • Severe PROMISe

II 360 7.58 $5,368 4 $112 2 $192 $35 $5,707 VALMER 156.1 10.1 $339.86 4.2 $208.08 $308.60 $6374.65

Table 18 - Comparison between PROMISe II and VALMER consequences table values

30

FINAL REPORT

Figure 13 - Total costs and duration of consequences at mild severity level

31

FINAL REPORT

Figure 14 - Total costs and duration of consequences at moderate severity level

32

FINAL REPORT

Figure 15 - Total costs and duration of consequences at severe severity level

33

FINAL REPORT Conclusion A table of consequences commonly resulting from interventions made in HMRs was formulated based on Australian hospitalisation data, literature sources and two small studies conducted for the study. An important consideration regarding the values derived in the study is that they are based predominantly on expert opinion rather than empirical data. Hence they are simply an estimation of the QOL, duration of ill health and resource utilisation resulting from each consequence. The aim of these studies was to update and potentially improve the validity of the PROMISe II consequences table. It is the opinion of the research team that these studies have achieved this aim and produced a more realistic approximation for each parameter. However the values remain an approximation, and studies utilising them must recognise this limitation and interpret results obtained from using the values with caution.

34

FINAL REPORT

Appendix IV Screen of HMR assessment system

Figure 16- Screenshot of system used by assessors to evaluate HMRs

35

FINAL REPORT

Appendix V Results of pharmacist survey The characteristics of the pharmacists who participated in the VALMER study were assessed with an online survey that was distributed in April 2009. The survey was undertaken to ensure that the VALMER participants were generally representative of the greater body of accredited pharmacists currently performing medication reviews. The VALMER pharmacist survey results were benchmarked against two comprehensive surveys of AACP-accredited pharmacists conducted in 2005 (858 accredited pharmacists) and 2008 (560 accredited pharmacists). Most recent figures indicate that 1810 pharmacists are accredited with the AACP 57 and 46 with SHPA.58 The 2005 survey reported that 91% of accredited pharmacists had conducted at least one HMR in the past 12 months, hence the 149 pharmacists who submitted HMRs for the VALMER study represent approximately 9% of accredited pharmacists who perform HMRs. All pharmacists participating in VALMER were accredited to perform medication reviews with AACP- no SHPA-accredited pharmacist submitted HMRs for the study. The main state of practice of the VALMER pharmacists is shown in Figure 17. Each Australian state or territory was represented by at least one pharmacist. When compared to the national number of AACP- accredited pharmacists, a greater proportion of Victorian and Tasmanian pharmacists participated in VALMER, potentially due to local awareness of the study.

Figure 17 - Percentage of pharmacists according to their main state of practice

General characteristics Of the 149 pharmacists who submitted HMRs, 117 (78.5%) undertook the survey. Their general characteristics are presented in Table 19. These data are consistent with the 2008 AACP survey data, which reported that female pharmacists constituted the majority of accredited pharmacists and were generally younger than males. The main areas of employment for the VALMER pharmacists also mirrored the AACP data, where over 70% of participants worked in either community pharmacies or as consultant pharmacists.

36

FINAL REPORT

Characteristic Male (n=27, 23.1%)

Female (n=90, 76.9%)

Total (n=117)

Year of completion of initial pharmacist training Median 1983.5 1989 1988 Range 1963-2006 1967-2007 1963-2007 Year of initial AACP accreditation Median 2003 2003 2003 Range 1997-2008 1996-2008 1996-2009 Area of main employment (number, % of total) Community pharmacy proprietor 4 (3.4%) 7 (6.0%) 11 (9.4%) Community pharmacy employee 5 (4.3%) 28 (23.9%) 33 (28.2%) Provider of medication reviews 10 (8.5%) 28 (23.9%) 38 (32.5%) Hospital pharmacy employee 2 (1.7%) 7 (6.0%) 9 (7.7%)

Other (e.g. NPS facilitators, academia) 6 (5.1%) 20 (17.1%) 26 (22.2%)

HMRs completed (median [range]) Ever 300 [12 - 1090] 217 [4-2500] 235 [4 - 3500] In 2008 73 [4 - 1090] 59 [2 - 500] 60 [2 - 1090]

Table 19 - General pharmacist demographics

The most comprehensive analysis of the time taken to perform HMRs was reported in the 2005 evaluation of the HMR program.18 This report identified that pharmacists spend an average of 53 minutes interviewing patients in HMRs, with the other tasks associated with HMR (such as preparation, report writing and travel) requiring a further 2.6 hours. The results of the VALMER pharmacist survey were very similar to these data. The pharmacists reported a mean interview duration of 54 (±16.1) minutes, and mean time to complete other HMR tasks of 161 (±80.6) minutes. HMRs performed for Department of Veterans’ Affairs DAA program In early 2008, the Department of Veterans’ Affairs (DVA) commenced a program of subsidising dosage administration aids (DAAs) for eligible veterans.59 The initial implementation of this program required every veteran who was to receive the service to have a HMR prior to its commencement†††. Data collection for VALMER commenced shortly after the DAA program was introduced, hence many of the VALMER HMRs may have been performed for this indication alone. This introduced a potential confounder for VALMER for the following reasons:

• Patients may have received a HMR prior to the commencement of the DAA program, and were referred for another HMR shortly afterwards simply to commence a DAA. This may limit the potential benefits of the second HMR (which was submitted as part of VALMER), and result in an underestimation of the benefits of HMR.

• Given that the patient was considered eligible of a DAA, they may be at a greater risk of medication misadventure than a “typical” HMR patient. This would overestimate the benefits of HMR.

• GPs who would not normally refer patients for HMRs may have had to do so to arrange DAAs. The effect of this may have been variable, in that the GP may be more or less receptive to any recommendations resulting from the HMR, resulting in an unpredictable effect on the VALMER results.

• To assess the influence of the DAA program on the VALMER HMRs, the VALMER survey asked pharmacists about the HMRs that they performed as part of the DVA DAA program. The results of these questions are presented in Table 20. Eighty five pharmacists (72.6%) had performed at least one DAA HMR.

††† This restriction has now been changed - a HMR is no longer a requirement for the DAA service, although DVA continue to strongly recommend that veterans be assessed via HMR as part of the DAA program 

37

FINAL REPORT

Difference to a “normal” HMR (number (%) of total respondents) Parameter Substantially more Somewhat more About the same Somewhat less Substantially less Complexity 2 (2.4%) 9 (10.6%) 67 (78.8%) 6 (7.1%) 1 (1.2%) Number of clinical issues 1 (1.2%) 10 (11.8%) 67 (78.8%) 6 (7.1%) 1 (1.2%) Number of compliance issues 3 (3.5%) 17 (20.0%) 55 (64.7%) 8 (9.4%) 2 (2.4%) Potential to improve health 10 (11.8%) 22 (25.9%) 52 (61.2%) 1 (1.2%) 0 (0.0%)

Table 20 - Pharmacist views of differences between DVA DAA HMRs and "normal" HMRs

The majority of pharmacists believed that there were few differences between the HMRs that they had performed as part of the DVA DAA program and “normal” HMRs in any of the parameters assessed. The greatest potential for confounding appears to be in relation to compliance issues. The data from this part of the survey implies that HMRs performed as part of the DVA DAA program were unlikely to be substantially different to “typical” HMRs. From the results of the pharmacist survey, it was concluded that the characteristics of the pharmacists participating in VALMER were generally consistent with the results of previous studies of these pharmacists. The sample of pharmacists may therefore be viewed as representative of the pharmacists currently performing HMRs across Australia. Pharmacists were also asked about the frequency and value of discussing HMRs with the referring GP. When asked about the frequency of discussion, the majority of pharmacists indicated that this occurred for less than half of the HMRs they performed. Despite this lack of communication, most pharmacists believed that communication between them and the referring GP was highly valuable to the outcomes of the HMR process. Most pharmacists also found written medication management plans to be valuable; less than one third of pharmacists reported receiving a plan following a HMR. The frequencies of response to these questions are shown below.

Figure 18 - Frequencies of contact with GP and receipt of management plans following HMRs

38

FINAL REPORT

Figure 19 - Pharmacist perception of value of communication with GPs and medication management plans

39

FINAL REPORT

Appendix VI Patient medical conditions

ICPC2 chapter Number (% of total) of patients diagnosed with condition Condition Males Females Total Cardiovascular 547 (82.8%) Hypertension, uncomplicated 154 (23.3%) 248 (37.5%) 402 (60.8%) Ischaemic heart disease without angina 77 (11.6%) 60 (9.1%) 137 (20.7%) Atrial fibrillation/flutter 62 (9.4%) 58 (8.8%) 120 (18.2%) Heart failure 43 (6.5%) 49 (7.4%) 92 (13.9%) Stroke/cerebrovascular accident 26 (3.9%) 43 (6.5%) 69 (10.4%) Endocrine, metabolic and nutritional 457 (69.1%) Lipid disorder 102 (15.4%) 168 (25.4%) 270 (40.8%) Diabetes, non-insulin dependent 102 (15.4%) 113 (17.1%) 215 (32.5%) Gout 49 (7.4%) 25 (3.8%) 74 (11.2%) Vitamin/nutritional deficiency 21 (3.2%) 34 (5.1%) 55 (8.3%) Hypothyroidism/myxoedema 9 (1.4%) 39 (5.9%) 48 (7.3%) Musculoskeletal 421 (63.7%) Osteoarthrosis, other 61 (9.2%) 122 (18.5%) 183 (27.7%) Osteoporosis 33 (5.0%) 112 (16.9%) 145 (21.9%) Fracture: other 16 (2.4%) 40 (6.1%) 56 (8.5%) Back syndrome without radiating pain 22 (3.3%) 31 (4.7%) 53 (8.0%) Back syndrome with radiating pain 14 (2.1%) 30 (4.5%) 44 (6.7%) Digestive 351 (53.1%) Oesophagus disease 78 (11.8%) 117 (17.7%) 195 (29.5%) Diverticular disease 15 (2.3%) 45 (6.8%) 60 (9.1%) Hiatus hernia 9 (1.4%) 20 (3.0%) 29 (4.4%) Peptic ulcer, other 14 (2.1%) 14 (2.1%) 28 (4.2%) Benign/uncertain neoplasm digestive 12 (1.8%) 15 (2.3%) 27 (4.1%) Psychological 265 (40.1%) Depressive disorder 48 (7.3%) 71 (10.7%) 119 (18.0%) Tobacco abuse 41 (6.2%) 37 (5.6%) 78 (11.8%) Sleep disturbance 27 (4.1%) 39 (5.9%) 66 (10.0%) Dementia (incl senile, Alzheimer's) 13 (2.0%) 8 (1.2%) 21 (3.2%) Chronic alcohol abuse 12 (1.8%) 4 (0.6%) 16 (2.4%) Respiratory 265 (40.1%) Chronic obstructive pulmonary disease 60 (9.1%) 44 (6.7%) 104 (15.7%) Asthma 32 (4.8%) 67 (10.1%) 99 (15.0%) Respiratory disease, other 10 (1.5%) 11 (1.7%) 21 (3.2%) Allergic rhinitis 9 (1.4%) 11 (1.7%) 20 (3.0%) General & unspecified 237 (35.9%) Allergy/allergic reaction NOS 54 (8.2%) 130 (19.7%) 184 (27.8%) Effect of prosthetic device 18 (2.7%) 13 (2.0%) 31 (4.7%) Abnormal result investigation NOS 8 (1.2%) 12 (1.8%) 20 (3.0%) Eye 159 (24.1%) Glaucoma 15 (2.3%) 30 (4.5%) 45 (6.8%) Eye/adnexa disease, other 11 (1.7%) 29 (4.4%) 40 (6.1%) Macular degeneration 8 (1.2%) 23 (3.5%) 31 (4.7%) Cataract 12 (1.8%) 18 (2.7%) 30 (4.5%) Urological 154 (23.3%) Urinary disease, other 27 (4.1%) 55 (8.3%) 82 (12.4%) Cystitis/urinary infection, other 3 (0.5%) 17 (2.6%) 20 (3.0%) Incontinence urine 5 (0.8%) 15 (2.3%) 20 (3.0%) Skin 134 (20.3%) Malignant neoplasm of skin 21 (3.2%) 27 (4.1%) 48 (7.3%) Dermatitis, contact/allergic 10 (1.5%) 21 (3.2%) 31 (4.7%) Neurological 121 (18.3%) Vertigo/dizziness 9 (1.4%) 13 (2.0%) 22 (3.3%) Peripheral neuritis/neuropathy 10 (1.5%) 8 (1.2%) 18 (2.7%) Male genital 72 (10.9%) Malignant neoplasm prostate 25 (3.8%) 0 (0.0%) 25 (3.8%) Ear 59 (8.9%) Vertiginous syndrome 6 (0.9%) 11 (1.7%) 17 (2.6%) Blood, Blood form Organs & Immune Mechanism 54 (8.2%) Iron deficiency anaemia 5 (0.8%) 16 (2.4%) 21 (3.2%) Anaemia other/unspecified 5 (0.8%) 11 (1.7%) 16 (2.4%) Female genital 34 (5.1%) Menopausal symptom/complaint 0 (0.0%) 14 (2.1%) 14 (2.1%)

Table 21 - Most common medical conditions presented according to ICPC2-PLUS chapters

40

FINAL REPORT

Appendix VII Medications taken

ATC Level 2 Medication Group Number (%) of patients taking Most frequent individual medications Male Female Total Lipid modifying agents 205 (31.0%) 275 (41.6%) 480 (72.6%) Atorvastatin 90 (13.6%) 116 (17.5%) 206 (31.2%) Simvastatin 56 (8.5%) 65 (9.8%) 121 (18.3%) Omega-3-triglycerides 30 (4.5%) 61 (9.2%) 91 (13.8%) Agents acting on the renin-angiotensin system 198 (30.0%) 263 (39.8%) 461 (69.7%) Perindopril 61 (9.2%) 35 (5.3%) 96 (14.5%) Irbesartan 33 (5.0%) 51 (7.7%) 84 (12.7%) Ramipril 42 (6.4%) 38 (5.7%) 80 (12.1%) Analgesics 167 (25.3%) 290 (43.9%) 457 (69.1%) Paracetamol 142 (21.5%) 252 (38.1%) 394 (59.6%) Codeine, combinations excl. psycholeptics 27 (4.1%) 45 (6.8%) 72 (10.9%) Oxycodone 12 (1.8%) 23 (3.5%) 35 (5.3%) Antithrombotic agents 212 (32.1%) 244 (36.9%) 456 (69.0%) Acetylsalicylic acid 129 (19.5%) 155 (23.4%) 284 (43.0%) Warfarin 59 (8.9%) 58 (8.8%) 117 (17.7%) Clopidogrel 50 (7.6%) 49 (7.4%) 99 (15.0%) Drugs for acid related disorders 155 (23.4%) 229 (34.6%) 384 (58.1%) Esomeprazole 48 (7.3%) 67 (10.1%) 115 (17.4%) Omeprazole 34 (5.1%) 64 (9.7%) 98 (14.8%) Pantoprazole 31 (4.7%) 40 (6.1%) 71 (10.7%) Cardiac therapy 142 (21.5%) 150 (22.7%) 292 (44.2%) Glyceryl trinitrate 66 (10.0%) 72 (10.9%) 138 (20.9%) Digoxin 42 (6.4%) 44 (6.7%) 86 (13.0%) Isosorbide mononitrate 32 (4.8%) 23 (3.5%) 55 (8.3%) Diuretics 87 (13.2%) 142 (21.5%) 229 (34.6%) Furosemide 69 (10.4%) 105 (15.9%) 174 (26.3%) Spironolactone 20 (3.0%) 16 (2.4%) 36 (5.4%) Indapamide 10 (1.5%) 24 (3.6%) 34 (5.1%) Drugs for obstructive airway diseases 88 (13.3%) 135 (20.4%) 223 (33.7%) Salbutamol 57 (8.6%) 90 (13.6%) 147 (22.2%) Salmeterol and other drugs for obstructive airway diseases 37 (5.6%) 55 (8.3%) 92 (13.9%) Tiotropium bromide 39 (5.9%) 40 (6.1%) 79 (12.0%) Calcium channel blockers 87 (13.2%) 135 (20.4%) 222 (33.6%) Amlodipine 26 (3.9%) 44 (6.7%) 70 (10.6%) Diltiazem 21 (3.2%) 29 (4.4%) 50 (7.6%) Lercanidipine 12 (1.8%) 27 (4.1%) 39 (5.9%) Beta blocking agents 100 (15.1%) 117 (17.7%) 217 (32.8%) Metoprolol 45 (6.8%) 53 (8.0%) 98 (14.8%) Atenolol 31 (4.7%) 49 (7.4%) 80 (12.1%) Carvedilol 10 (1.5%) 8 (1.2%) 18 (2.7%) Psychoanaleptics 80 (12.1%) 129 (19.5%) 209 (31.6%) Amitriptyline 13 (2.0%) 33 (5.0%) 46 (7.0%) Sertraline 11 (1.7%) 19 (2.9%) 30 (4.5%) Mirtazapine 11 (1.7%) 13 (2.0%) 24 (3.6%) Antiinflammatory and antirheumatic products 70 (10.6%) 130 (19.7%) 200 (30.3%) Glucosamine 28 (4.2%) 72 (10.9%) 100 (15.1%) Meloxicam 15 (2.3%) 36 (5.4%) 51 (7.7%) Celecoxib 9 (1.4%) 18 (2.7%) 27 (4.1%) Drugs used in diabetes 94 (14.2%) 98 (14.8%) 192 (29.0%) Metformin 67 (10.1%) 66 (10.0%) 133 (20.1%) Gliclazide 38 (5.7%) 47 (7.1%) 85 (12.9%) Insulin (human) 10 (1.5%) 14 (2.1%) 24 (3.6%) Mineral supplements 46 (7.0%) 135 (20.4%) 181 (27.4%) Calcium carbonate 20 (3.0%) 84 (12.7%) 104 (15.7%) Potassium chloride 14 (2.1%) 35 (5.3%) 49 (7.4%) Magnesium (different salts in combination) 2 (0.3%) 7 (1.1%) 9 (1.4%) Vitamins 48 (7.3%) 133 (20.1%) 181 (27.4%) Colecalciferol 21 (3.2%) 67 (10.1%) 88 (13.3%) Multivitamins and other minerals, incl. combinations 13 (2.0%) 34 (5.1%) 47 (7.1%) Ascorbic acid (vit C) 6 (0.9%) 14 (2.1%) 20 (3.0%) Ophthalmologicals 58 (8.8%) 120 (18.2%) 178 (26.9%) Artificial tears and other indifferent preparations 37 (5.6%) 77 (11.6%) 114 (17.2%) Latanoprost 13 (2.0%) 15 (2.3%) 28 (4.2%) Timolol 3 (0.5%) 9 (1.4%) 12 (1.8%) Laxatives 72 (10.9%) 102 (15.4%) 174 (26.3%) Senna glycosides, combinations 31 (4.7%) 44 (6.7%) 75 (11.3%) Senna glycosides 7 (1.1%) 21 (3.2%) 28 (4.2%) Lactulose 11 (1.7%) 16 (2.4%) 27 (4.1%)

41

FINAL REPORTATC Level 2 Medication Group Number (%) of patients taking Most frequent individual medications Male Female Total Psycholeptics 60 (9.1%) 106 (16.0%) 166 (25.1%) Temazepam 27 (4.1%) 55 (8.3%) 82 (12.4%) Oxazepam 6 (0.9%) 31 (4.7%) 37 (5.6%) Diazepam 16 (2.4%) 15 (2.3%) 31 (4.7%) Antianemic preparations 46 (7.0%) 86 (13.0%) 132 (20.0%) Hydroxocobalamin 19 (2.9%) 26 (3.9%) 45 (6.8%) Folic acid 13 (2.0%) 30 (4.5%) 43 (6.5%) Ferrous sulfate 5 (0.8%) 19 (2.9%) 24 (3.6%) Corticosteroids, dermatological preparations 47 (7.1%) 78 (11.8%) 125 (18.9%) Betamethasone 24 (3.6%) 31 (4.7%) 55 (8.3%) Mometasone 15 (2.3%) 27 (4.1%) 42 (6.4%) Hydrocortisone 9 (1.4%) 8 (1.2%) 17 (2.6%)

Table 22 - Most commonly taken medications grouped by ATC chapter

42

FINAL REPORT

Appendix VIII Nature of DRPs identified

DRP Subtype Medical condition Number (%) of total DRPs Condition not adequately treated General symptom/complaint, other * 74 (3.2%) Condition not adequately treated Hypertension, uncomplicated 44 (1.9%) Condition not adequately treated Lipid disorder 27 (1.2%) Therapy required Diabetes, non-insulin dependent 25 (1.1%) Condition not adequately treated Osteoporosis 24 (1.0%) Toxicity evident Constipation 24 (1.0%) Condition not adequately treated Osteoarthrosis, other 20 (0.9%) Condition not adequately treated Diabetes, non-insulin dependent 20 (0.9%) Dose too high Urinary disease, other† 19 (0.8%) Therapy required Osteoporosis 17 (0.7%) Therapy required Lipid disorder 16 (0.7%) Toxicity evident Cough 16 (0.7%) Toxicity evident Vertigo/dizziness 15 (0.6%) Toxicity evident Mouth/tongue/lip symptom/complaint‡ 15 (0.6%) Toxicity evident Sleep disturbance 14 (0.6%) Laboratory monitoring Vitamin/nutritional deficiency 14 (0.6%) Condition not adequately treated Constipation 13 (0.6%) Contraindications apparent Urinary disease, other† 13 (0.6%) Toxicity evident Nausea 11 (0.5%) Condition not adequately treated Chronic obstructive pulmonary disease 11 (0.5%) Dose too high Oesophagus disease 11 (0.5%) Condition not adequately treated Back symptom/complaint 11 (0.5%) Smoking problem Tobacco abuse 11 (0.5%) Laboratory monitoring Urinary disease, other† 11 (0.5%) *usually pain †usually renal insufficiency ‡ usually xerostomia

Table 23 - Most common DRPs according to associated medical conditions

DRP Subtype Medication group (ATC Level 2) Number (%) of total DRPs Condition not adequately treated Analgesics 149 (6.4%) Drug interaction Antithrombotic agents 73 (3.1%) Drug interaction Renin-angiotensin system agents 63 (2.7%) Condition not adequately treated Renin-angiotensin system agents 43 (1.9%) Toxicity evident Renin-angiotensin system agents 41 (1.8%) Drug interaction Antiinflammatory / antirheumatic products 40 (1.7%) Laboratory Monitoring Drugs for acid related disorders 39 (1.7%) Drug interaction Lipid modifying agents 39 (1.7%) Drug interaction Diuretics 35 (1.5%) Unnecessary therapy/no indication Drugs for acid related disorders 34 (1.5%) Toxicity evident Lipid modifying agents 32 (1.4%) Condition not adequately treated Lipid modifying agents 32 (1.4%) Toxicity evident Calcium channel blockers 31 (1.3%) Laboratory Monitoring Renin-angiotensin system agents 31 (1.3%) Condition not adequately treated Drugs for obstructive airway diseases 31 (1.3%) Taking too little Drugs for obstructive airway diseases 29 (1.2%) Toxicity evident Psychoanaleptics 29 (1.2%) Difficulty using dosage form Drugs for obstructive airway diseases 29 (1.2%) Condition not adequately treated Drugs used in diabetes 27 (1.2%) Contraindications apparent Antiinflammatory / antirheumatic products 25 (1.1%) Laboratory Monitoring Lipid modifying agents 25 (1.1%) Toxicity evident Antithrombotic agents 25 (1.1%) Toxicity evident Drugs for acid related disorders 25 (1.1%) Dose too high Drugs for acid related disorders 24 (1.0%) Drug interaction Cardiac therapy 24 (1.0%) Laboratory Monitoring Drugs used in diabetes 24 (1.0%)

Table 24 - Most common DRPs according to associated medications

43

FINAL REPORT

Appendix IX Uptake of recommendations

DRP Type Number (%) of DRPs DRP Subtype Potentially resolved Unresolved Total

Percentage potentially resolved

Drug selection Duplication 27 (1.7%) 3 (0.2%) 30 (1.9%) 90.0% Drug interaction 91 (5.8%) 23 (1.5%) 114 (7.3%) 79.8% Wrong drug 1 (0.1%) 0 (0.0%) 1 (0.1%) 100.0% Wrong dosage form 1 (0.1%) 1 (0.1%) 2 (0.1%) 50.0%

Unnecessary therapy/no apparent current indication 59 (3.8%) 25 (1.6%) 84 (5.4%) 70.2%

Contraindications apparent 81 (5.2%) 26 (1.7%) 107 (6.8%) 75.7% Other drug selection problem 6 (0.4%) 3 (0.2%) 9 (0.6%) 66.7% Over or underdose prescribed Dose too high 60 (3.8%) 26 (1.7%) 86 (5.5%) 69.8% Dose too low 27 (1.7%) 3 (0.2%) 30 (1.9%) 90.0% Other dose problem 17 (1.1%) 7 (0.4%) 24 (1.5%) 70.8% Compliance & concordance Taking too little 99 (6.3%) 5 (0.3%) 104 (6.6%) 95.2% Taking too much 10 (0.6%) 0 (0.0%) 10 (0.6%) 100.0% Difficulty using dosage form 31 (2.0%) 4 (0.3%) 35 (2.2%) 88.6% Patient using out of date medication 13 (0.8%) 1 (0.1%) 14 (0.9%) 92.9% Other compliance problem 79 (5.0%) 8 (0.5%) 87 (5.6%) 90.8% Untreated indications Condition not adequately treated 204 (13.0%) 47 (3.0%) 251 (16.0%) 81.3% Therapy required 127 (8.1%) 40 (2.6%) 167 (10.7%) 76.0% Other untreated indication problem 1 (0.1%) 0 (0.0%) 1 (0.1%) 100.0% Monitoring Laboratory monitoring 108 (6.9%) 19 (1.2%) 127 (8.1%) 85.0% Non-laboratory monitoring 11 (0.7%) 2 (0.1%) 13 (0.8%) 84.6% Other monitoring problem 3 (0.2%) 0 (0.0%) 3 (0.2%) 100.0% Education or information Patient drug information request 6 (0.4%) 0 (0.0%) 6 (0.4%) 100.0% Confusion about therapy 6 (0.4%) 1 (0.1%) 7 (0.4%) 85.7% Demonstration of device 8 (0.5%) 0 (0.0%) 8 (0.5%) 100.0% Disease management or advice 10 (0.6%) 1 (0.1%) 11 (0.7%) 90.9% Other education or information problem 4 (0.3%) 0 (0.0%) 4 (0.3%) 100.0% Non-clinical Weight management problem 7 (0.4%) 1 (0.1%) 8 (0.5%) 87.5% Dietary problem 11 (0.7%) 0 (0.0%) 11 (0.7%) 100.0% Smoking problem 7 (0.4%) 1 (0.1%) 8 (0.5%) 87.5% Alcohol problem 1 (0.1%) 0 (0.0%) 1 (0.1%) 100.0% Other non-clinical problem 9 (0.6%) 1 (0.1%) 10 (0.6%) 90.0% Toxicity or adverse reaction Toxicity caused by dose 13 (0.8%) 4 (0.3%) 17 (1.1%) 76.5% Toxicity caused by drug interaction 20 (1.3%) 4 (0.3%) 24 (1.5%) 83.3% Toxicity evident 119 (7.6%) 29 (1.9%) 148 (9.5%) 80.4% Other toxicity/adverse effect problem 3 (0.2%) 0 (0.0%) 3 (0.2%) 100.0% TOTAL 1280 (81.8%) 285 (18.2%) 1565 (100.0%)

Table 25 - Potential resolution of types of DRPs. “Potentially resolved” indicates that the HMR resulted in some action occurring that would resolve the DRP

44

FINAL REPORT

Appendix X Data costing tables

Before HMR After HMR Change resulting from HMR HMR Utility Health

resource costs Utility Health resource costs QOL Health

resource costs Annual

drug cost

Total cost or saving

3 0.0831 $1,983.72 0.0498 $1,347.20 0.0333 $636.52 -$1,262.58 -$626.06 4 0.0710 $1,029.23 0.0816 $1,195.53 -0.0106 -$166.30 $0.00 -$166.30 5 0.0039 $287.75 0.0015 $97.27 0.0024 $190.48 -$117.18 $73.30 6 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 -$241.56 -$241.56 11 0.0440 $1,680.93 0.0153 $634.21 0.0287 $1,046.72 $140.40 $1,187.12 14 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $120.96 $120.96 15 0.0034 $134.07 0.0041 $146.62 -0.0007 -$12.54 $0.00 -$12.54 16 0.0004 $40.02 0.0009 $79.54 -0.0005 -$39.52 $0.00 -$39.52 18 0.0002 $44.80 0.0009 $108.78 -0.0007 -$63.98 -$117.42 -$181.40 21 0.0182 $907.52 0.0178 $1,030.62 0.0004 -$123.10 $26.64 -$96.46 23 0.0256 $479.01 0.0351 $435.09 -0.0095 $43.92 $0.00 $43.92 24 0.0684 $1,605.20 0.0488 $1,327.03 0.0196 $278.17 $463.44 $741.61 26 0.1444 $3,187.28 0.1572 $3,183.28 -0.0127 $4.00 -$225.90 -$221.90 34 0.0582 $1,461.56 0.0429 $1,322.31 0.0153 $139.25 $1,230.00 $1,369.25 35 0.1955 $5,483.57 0.2011 $4,977.45 -0.0055 $506.13 $104.64 $610.77 36 0.0295 $347.63 0.0184 $216.97 0.0112 $130.66 $0.00 $130.66 39 0.0076 $82.84 0.0111 $399.98 -0.0035 -$317.14 $0.00 -$317.14 40 0.0044 $69.79 0.0010 $18.74 0.0034 $51.05 -$19.20 $31.85 45 0.0224 $624.81 0.0251 $643.26 -0.0026 -$18.44 $0.00 -$18.44 49 0.0044 $425.20 0.0037 $352.06 0.0007 $73.15 $0.00 $73.15 50 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $0.00 $0.00 51 0.0280 $600.36 0.0154 $329.20 0.0126 $271.16 $0.00 $271.16 53 0.1376 $3,423.21 0.1127 $3,195.60 0.0249 $227.61 $230.16 $457.77 54 0.0412 $1,088.44 0.0444 $1,117.67 -0.0032 -$29.22 -$64.84 -$94.06 56 0.0735 $1,690.29 0.0736 $1,516.59 -0.0001 $173.70 $705.06 $878.76 58 0.0344 $652.80 0.0334 $591.12 0.0010 $61.68 $382.56 $444.24 66 0.0035 $396.13 0.0035 $396.13 0.0000 $0.00 $0.00 $0.00 69 0.0900 $4,374.15 0.0851 $4,295.15 0.0049 $79.00 $0.00 $79.00 72 0.0502 $973.73 0.0293 $644.93 0.0208 $328.79 $89.52 $418.31 75 0.0275 $644.59 0.0191 $445.06 0.0085 $199.53 $0.00 $199.53 76 0.0430 $1,313.25 0.0370 $1,219.89 0.0060 $93.36 $0.00 $93.36 78 0.0023 $51.15 0.0013 $31.58 0.0010 $19.56 $0.00 $19.56 80 0.0283 $913.70 0.0193 $605.03 0.0090 $308.67 $59.64 $368.31 81 0.0132 $516.84 0.0080 $256.98 0.0052 $259.86 $0.00 $259.86 82 0.0007 $32.76 0.0019 $115.10 -0.0013 -$82.34 $0.00 -$82.34 83 0.0179 $457.86 0.0145 $380.15 0.0033 $77.71 -$194.52 -$116.81 89 0.0015 $70.17 0.0035 $169.64 -0.0021 -$99.46 -$48.48 -$147.94 92 0.0014 $23.61 0.0022 $61.62 -0.0008 -$38.01 $0.00 -$38.01 104 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 -$126.36 -$126.36 106 0.1439 $4,279.96 0.0629 $1,644.90 0.0810 $2,635.06 $0.00 $2,635.06 110 0.0001 $15.89 0.0004 $103.54 -0.0003 -$87.65 $0.00 -$87.65 113 0.0497 $1,788.44 0.0544 $1,986.27 -0.0048 -$197.84 $229.38 $31.54 114 0.0005 $26.03 0.0003 $11.88 0.0002 $14.15 $0.00 $14.15 121 0.0425 $825.31 0.0501 $789.20 -0.0076 $36.11 $0.00 $36.11 124 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $97.52 $97.52 126 0.0111 $662.65 0.0141 $691.36 -0.0030 -$28.71 $90.78 $62.07 127 0.0339 $797.49 0.0268 $599.09 0.0071 $198.40 -$71.94 $126.46 129 0.0591 $1,146.94 0.0412 $745.30 0.0179 $401.64 -$351.24 $50.40 138 0.0116 $292.58 0.0153 $423.64 -0.0037 -$131.07 $0.00 -$131.07 140 0.0472 $1,823.95 0.0925 $3,249.56 -0.0453 -$1,425.60 $0.00 -$1,425.60 143 0.0016 $74.78 0.0008 $59.50 0.0007 $15.27 -$287.76 -$272.49 144 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $0.00 $0.00 145 0.0059 $217.07 0.0036 $156.91 0.0024 $60.16 $42.24 $102.40 147 0.0291 $949.14 0.0235 $802.79 0.0056 $146.35 $0.00 $146.35 148 0.0146 $447.43 0.0089 $295.84 0.0057 $151.59 $0.00 $151.59 153 0.0021 $17.29 0.0010 $8.62 0.0012 $8.67 $90.42 $99.09 156 0.0004 $27.10 0.0008 $53.22 -0.0004 -$26.12 $0.00 -$26.12 164 0.0759 $1,853.68 0.0528 $1,425.40 0.0232 $428.28 $0.00 $428.28 165 0.0007 $38.56 0.0003 $21.78 0.0004 $16.78 $0.00 $16.78 170 0.0060 $329.88 0.0039 $222.90 0.0021 $106.98 $0.00 $106.98 172 0.0024 $60.72 0.0016 $36.87 0.0009 $23.85 $0.00 $23.85 186 0.0019 $88.87 0.0012 $51.88 0.0007 $36.99 -$71.94 -$34.95 187 0.0338 $634.31 0.0242 $523.60 0.0096 $110.71 -$23.10 $87.61 200 0.0169 $642.11 0.0146 $600.84 0.0023 $41.27 $0.00 $41.27 209 0.0108 $418.30 0.0088 $307.37 0.0020 $110.93 -$98.76 $12.17 222 0.0164 $118.12 0.0306 $121.05 -0.0143 -$2.93 $0.00 -$2.93 223 0.0247 $280.54 0.0170 $179.14 0.0077 $101.40 -$135.00 -$33.60 225 0.0366 $951.57 0.0473 $1,285.90 -0.0107 -$334.33 $0.00 -$334.33

45

FINAL REPORTBefore HMR After HMR Change resulting from HMR

HMR Utility Health resource costs Utility Health

resource costs QOL Health resource costs

Annual drug cost

Total cost or saving

226 0.0004 $31.60 0.0003 $26.09 0.0001 $5.51 $0.00 $5.51 227 0.0022 $346.87 0.0021 $348.39 0.0000 -$1.53 $455.64 $454.11 229 0.0276 $459.17 0.0204 $343.38 0.0072 $115.79 $643.32 $759.11 230 0.0005 $39.48 0.0009 $77.85 -0.0004 -$38.38 $0.00 -$38.38 236 0.0069 $503.84 0.0031 $241.95 0.0039 $261.88 $30.80 $292.68 238 0.0033 $169.41 0.0023 $115.01 0.0010 $54.40 $0.00 $54.40 249 0.0039 $621.88 0.0037 $547.81 0.0001 $74.07 $201.24 $275.31 250 0.0005 $64.87 0.0004 $57.19 0.0001 $7.68 $0.00 $7.68 251 0.0119 $589.78 0.0089 $411.79 0.0030 $177.99 -$71.94 $106.05 255 0.0084 $138.82 0.0062 $101.67 0.0022 $37.15 $292.70 $329.85 261 0.0006 $109.50 0.0003 $61.35 0.0002 $48.15 $0.00 $48.15 268 0.0170 $670.49 0.0178 $777.67 -0.0008 -$107.18 $0.00 -$107.18 270 0.0012 $16.26 0.0018 $25.76 -0.0007 -$9.50 -$65.28 -$74.78 271 0.0089 $377.82 0.0079 $402.69 0.0010 -$24.87 $0.00 -$24.87 278 0.0062 $263.39 0.0052 $182.43 0.0010 $80.96 $914.03 $994.99 281 0.0369 $675.13 0.0267 $592.65 0.0101 $82.47 $0.00 $82.47 285 0.0379 $769.48 0.0482 $871.20 -0.0103 -$101.72 $78.87 -$22.85 289 0.0008 $71.32 0.0005 $39.52 0.0003 $31.80 $0.00 $31.80 291 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $140.76 $140.76 295 0.0154 $939.99 0.0190 $559.40 -0.0037 $380.59 $0.00 $380.59 297 0.0143 $477.54 0.0177 $606.79 -0.0034 -$129.25 $193.86 $64.61 299 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $0.00 $0.00 310 0.0019 $54.67 0.0052 $155.15 -0.0033 -$100.48 $0.00 -$100.48 313 0.0912 $2,293.26 0.0682 $1,777.24 0.0230 $516.02 $0.00 $516.02 315 0.0181 $907.05 0.0210 $903.19 -0.0029 $3.86 $0.00 $3.86 326 0.0682 $1,622.56 0.0567 $1,402.14 0.0116 $220.42 $111.78 $332.20 331 0.0000 $0.00 0.0007 $10.35 -0.0007 -$10.35 $44.76 $34.41 332 0.0370 $1,839.47 0.0207 $819.76 0.0163 $1,019.71 $50.64 $1,070.35 336 0.0379 $1,278.29 0.0361 $1,226.65 0.0018 $51.64 -$59.84 -$8.20 340 0.0003 $21.24 0.0002 $18.38 0.0001 $2.86 $0.00 $2.86 343 0.0049 $252.92 0.0060 $247.97 -0.0011 $4.95 $0.00 $4.95 345 0.0010 $41.34 0.0007 $28.91 0.0004 $12.42 $0.00 $12.42 348 0.0103 $164.65 0.0087 $135.51 0.0016 $29.14 -$333.00 -$303.86 352 0.0367 $1,376.45 0.0406 $1,126.12 -0.0038 $250.33 $0.00 $250.33 355 0.0142 $211.73 0.0070 $99.19 0.0072 $112.54 $0.00 $112.54 360 0.0047 $127.73 0.0135 $264.21 -0.0089 -$136.48 $273.72 $137.24 367 0.0139 $360.17 0.0103 $257.43 0.0036 $102.74 -$277.06 -$174.31 369 0.0031 $40.35 0.0017 $21.62 0.0014 $18.73 $0.00 $18.73 370 0.0173 $392.74 0.0135 $302.36 0.0038 $90.38 $0.00 $90.38 378 0.0161 $140.74 0.0147 $147.22 0.0014 -$6.48 $4.98 -$1.50 382 0.0471 $784.34 0.0321 $703.41 0.0150 $80.94 $358.08 $439.02 383 0.0149 $535.88 0.0072 $311.82 0.0077 $224.07 $0.00 $224.07 384 0.0115 $339.34 0.0047 $178.65 0.0068 $160.69 $0.00 $160.69 385 0.0054 $176.89 0.0047 $145.51 0.0006 $31.38 -$269.04 -$237.66 392 0.0220 $545.68 0.0155 $308.36 0.0064 $237.32 $0.00 $237.32 393 0.0064 $228.21 0.0051 $98.14 0.0014 $130.07 $0.00 $130.07 394 0.0168 $771.12 0.0073 $273.62 0.0095 $497.50 $0.00 $497.50 395 0.0072 $434.16 0.0317 $614.59 -0.0244 -$180.44 $45.12 -$135.32 402 0.0297 $2,763.34 0.0038 $371.88 0.0259 $2,391.45 $0.00 $2,391.45 404 0.0040 $132.96 0.0028 $93.69 0.0012 $39.27 $0.00 $39.27 409 0.0091 $250.44 0.0113 $260.91 -0.0022 -$10.47 $0.00 -$10.47 411 0.0065 $207.48 0.0022 $161.91 0.0043 $45.58 $0.00 $45.58 414 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $0.00 $0.00 415 0.0002 $22.72 0.0002 $18.57 0.0000 $4.15 $0.00 $4.15 418 0.0062 $308.40 0.0057 $261.63 0.0004 $46.77 $0.00 $46.77 426 0.0050 $109.52 0.0042 $71.33 0.0009 $38.19 $380.76 $418.95 431 0.0017 $78.39 0.0020 $67.41 -0.0003 $10.98 $0.00 $10.98 433 0.0316 $514.54 0.0299 $466.84 0.0017 $47.70 $0.00 $47.70 434 0.0324 $551.37 0.0282 $486.36 0.0041 $65.01 $0.00 $65.01 436 0.0264 $603.82 0.0228 $509.44 0.0036 $94.38 $0.00 $94.38 440 0.0106 $183.14 0.0137 $254.49 -0.0030 -$71.35 $0.00 -$71.35 441 0.0135 $124.13 0.0067 $62.07 0.0067 $62.07 $0.00 $62.07 446 0.0124 $276.69 0.0082 $202.32 0.0042 $74.37 $0.00 $74.37 447 0.0083 $211.75 0.0068 $177.41 0.0015 $34.35 $0.00 $34.35 448 0.0034 $219.23 0.0025 $164.55 0.0009 $54.68 $25.90 $80.58 467 0.0054 $151.95 0.0058 $132.71 -0.0004 $19.24 $0.00 $19.24 473 0.0002 $42.14 0.0001 $21.07 0.0001 $21.07 $0.00 $21.07 477 0.0171 $290.31 0.0077 $153.36 0.0094 $136.95 -$362.03 -$225.08 478 0.0105 $230.86 0.0058 $128.55 0.0047 $102.31 $0.00 $102.31 479 0.0321 $748.00 0.0184 $493.35 0.0138 $254.65 -$6.66 $247.99 484 0.0022 $127.66 0.0024 $188.84 -0.0002 -$61.18 $87.84 $26.66 487 0.0187 $267.65 0.0193 $287.88 -0.0006 -$20.24 $0.00 -$20.24 488 0.0028 $141.46 0.0018 $83.57 0.0009 $57.89 -$69.30 -$11.41 496 0.0021 $135.02 0.0001 $1.22 0.0020 $133.80 $144.84 $278.64

46

FINAL REPORTBefore HMR After HMR Change resulting from HMR

HMR Utility Health resource costs Utility Health

resource costs QOL Health resource costs

Annual drug cost

Total cost or saving

499 0.0043 $307.09 0.0022 $172.65 0.0021 $134.45 $381.66 $516.11 504 0.0387 $459.96 0.0262 $302.28 0.0125 $157.69 -$351.24 -$193.55 511 0.0216 $347.41 0.0263 $348.05 -0.0047 -$0.64 $289.44 $288.80 514 0.1213 $2,591.57 0.0687 $1,446.39 0.0526 $1,145.18 $0.00 $1,145.18 517 0.0080 $255.84 0.0067 $275.56 0.0013 -$19.72 $0.00 -$19.72 519 0.0106 $203.33 0.0131 $258.83 -0.0025 -$55.49 $1,939.98 $1,884.49 524 0.0043 $231.70 0.0033 $174.07 0.0010 $57.63 $0.00 $57.63 545 0.0065 $86.79 0.0053 $66.74 0.0011 $20.05 $0.00 $20.05 555 0.0034 $101.39 0.0028 $83.96 0.0006 $17.43 $0.00 $17.43 559 0.0003 $35.39 0.0001 $18.54 0.0003 $16.85 $0.00 $16.85 563 0.0225 $441.08 0.0186 $349.78 0.0039 $91.30 $44.76 $136.06 564 0.0012 $131.81 0.0002 $33.72 0.0010 $98.09 $108.32 $206.41 569 0.0110 $393.44 0.0133 $378.37 -0.0023 $15.07 $86.16 $101.23 570 0.0262 $344.83 0.0212 $262.15 0.0050 $82.69 $0.00 $82.69 576 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $0.00 $0.00 584 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $0.00 $0.00 585 0.0191 $466.86 0.0167 $435.40 0.0024 $31.46 $57.90 $89.36 587 0.0120 $472.04 0.0110 $393.88 0.0010 $78.15 -$28.36 $49.79 596 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $0.00 $0.00 597 0.0059 $178.00 0.0062 $185.75 -0.0002 -$7.76 $0.00 -$7.76 602 0.0121 $87.70 0.0059 $41.85 0.0063 $45.85 $194.88 $240.73 603 0.0023 $62.85 0.0019 $39.68 0.0004 $23.17 $0.00 $23.17 613 0.0008 $28.78 0.0009 $26.97 -0.0001 $1.81 $0.00 $1.81 616 0.0105 $389.63 0.0085 $293.41 0.0020 $96.22 $149.40 $245.62 618 0.0048 $167.66 0.0029 $123.48 0.0019 $44.18 $0.00 $44.18 619 0.0175 $389.99 0.0162 $363.29 0.0013 $26.70 $0.00 $26.70 626 0.0097 $1,236.51 0.0088 $1,040.70 0.0008 $195.81 $817.23 $1,013.05 636 0.0116 $522.26 0.0030 $137.66 0.0087 $384.60 $0.00 $384.60 639 0.0107 $191.14 0.0146 $177.11 -0.0039 $14.02 $0.00 $14.02 660 0.0153 $754.54 0.0113 $455.56 0.0040 $298.97 $0.00 $298.97 665 0.0000 $0.00 0.0000 $0.00 0.0000 $0.00 $0.00 $0.00 666 0.0563 $1,514.19 0.0541 $1,456.22 0.0022 $57.97 $0.00 $57.97 669 0.0121 $323.10 0.0078 $237.76 0.0043 $85.34 -$106.32 -$20.98 674 0.0334 $1,229.71 0.0224 $859.08 0.0111 $370.63 $0.00 $370.63 676 0.0060 $508.08 0.0025 $211.48 0.0035 $296.60 -$1.60 $295.00 681 0.0068 $146.92 0.0047 $104.47 0.0021 $42.46 -$465.18 -$422.72 683 0.0829 $5,342.44 0.0660 $4,246.44 0.0169 $1,096.00 $0.00 $1,096.00 684 0.0152 $276.34 0.0135 $242.85 0.0017 $33.49 $0.00 $33.49

Table 26 - Costs and QOL data for assessed HMRs. Positive numbers indicate savings or improvements in QOL

47

FINAL REPORT

Appendix XI Examples of highly cost-effective HMRs The following five case summaries are examples of HMRs which potentially resulted in substantial savings to the health care system based on the results of the analysis by the expert panel. Review 11 The patient reviewed was an 87 year old male whose medical history included glaucoma, gout, high cholesterol, hypertension, osteoarthritis, renal impairment and type II diabetes. His medications at the time of the HMR were as follows:

Medication Directions Allosig 100mg Tab 100mg in the morning Avapro 300mg Tab 300mg in the morning Codalgin Forte Tab One tablet when required Coloxyl with Senna Tab One tablet when required Coversyl 5mg Tab 5mg at night Diamicron 80mg Tab 80mg in the morning Hydraderm Cream Applied when required Lipidil 145mg Tab 145mg at night MagMin 500mg Tab 500mg at night Minax 50mg Tab 50mg in the morning Panamax 500mg Tab 0.5-1.0mg every four hours when required Plavix 75mg Tab 75mg in the morning Polaramine 2mg Tab 2mg when required Simvar 40mg Tab 40mg at night Xalatan Eye Drops 0.005% One drop into the right eye at night

The patient was referred for a HMR as part of the DVA dosage administration program. During the HMR, the pharmacist identified the following DRPs:

• the patient was elderly and had renal impairment, yet he was taking the full dose of fenofibrate (Lipidil). Additionally, the combination of fenofibrate and simvastatin (Simvar) increases the risk of adverse effects on the muscles. A dose reduction of fenofibrate was recommended, with regular monitoring for muscle pain and tenderness.

• the patient was taking a magnesium supplement (Magmin), and magnesium may accumulate with renal impairment. Monitoring of magnesium levels was recommended.

Due to a reduction in the risk of liver disease and myopathy, the mean potential averted cost to the healthcare system resulting from this HMR was $1046.72. Additional savings of $140.40 per annum resulted from the reduced dose of fenofibrate.

48

FINAL REPORT Review 106 The patient reviewed was an 83 year old woman whose medical history included hypertension, depression, ischaemic heart disease with stents, emphysema and macular degeneration. Her medications at the time of the HMR were as follows:

Medication Directions Alzene 10mg Tablet 10mg at night when required Blackmores Fish Oil 1000mg Cap Three capsules in the morning Blackmores Macu-Vision Tab One tablet in the morning Cholstat 40mg Tab 40mg at night Codalgin Tab Two tablets when required Coloxyl with Senna Tab One to two tablets when required DBL Aspirin 100mg Tab 100mg in the morning Dothep 75mg Tab 150mg at night Fungilin 10mg Lozenges Taken when required Micardis 40mg Tab 40mg in the morning Nitrolingual 400mcg SL Pumpspray One to two puffs when required Panamax 500mg Tab 1.0g twice daily Seretide MDI 125/25 Inhaler Two puffs twice daily Spiriva 18mcg Powder 18mcg daily Systane (Eye Drops) Used when required Ventolin 100mcg CFC-free Inhaler Four puffs twice daily Zanidip 20mg Tablet 20mg in the morning

The patient had recently been provided with an oxygen concentrator to treat her emphysema. During the HMR, the pharmacist identified the following DRPs:

• both the patient and her daughter were unsure how to use the oxygen concentrator. The pharmacist organised for a community nurse to demonstrate the correct use of the machine.

• despite taking dothiepin (Dothep), the patient reported that her depression was poorly controlled and that she was constantly crying. The pharmacist also noted that dothiepin may impair cognitive function, and recommended changing dothiepin to an SSRI.

Due to a reduction in the risk of confusion, anxiety, COPD and depression, the mean potential averted cost to the healthcare system resulting from this HMR was $2635.06.

49

FINAL REPORT Review 332 The patient reviewed was a 77 year old woman whose medical history included hypertension, heart failure, ischaemic heart disease, osteoarthritis and type II diabetes. Her medications at the time of the HMR were as follows:

Medication Directions Aldactone 25mg Tab 12.5mg in the morning Alepam 30mg Tab 30mg at night Astrix 100mg Cap 100mg in the morning Diaformin 500mg Tab 500mg three times daily Diamicron MR 30mg Tab 120mg in the morning Dilzem CD 180mg Cap 180mg in the morning Ferro-f-tab 310mg/350mcg One tablet twice daily Flixotide Inhaler (CFC-Free) 250mcg Two puffs twice daily Lantus 100 IU/mL 10mL 6 units in the morning Lasix High Dose 500mg Tab 250mg in the morning Lovan 20mg Cap 40mg in the morning Maxolon 10mg Tab 10 mg three times daily MS Contin 15mg Tab 15mg twice daily Nexium 40mg Tab 40mg in the morning Nitrolingual 400mcg SL Pumpspray One puff when required Noten 50mg Tab 25mg in the morning Panamax 500mg Tab 500mg four times daily Pravachol 40mg Tab 40mg at night Transiderm-Nitro 25mg (Transdermal Patches) On 0730 and off at night Valpam 5mg Tab 5mg twice weekly Ventolin 100mcg CFC-free Inhaler Two puffs twice daily when required

During the HMR, the pharmacist identified the following DRPs:

• the patient was experiencing diarrhoea, nausea and vomiting. As her renal function was impaired, the pharmacist attributed the symptoms to the metformin and recommended reducing the dose.

• the patient was using her inhalers intermittently, and only applying her nitrate patches (Transiderm Nitro) when she experienced an anginal attack. The pharmacist counselled the patient regarding the correct use of these products and requested the GP to follow-up to ensure that these medications were being used appropriately.

Due to a reduction in the risk of acidosis, myocardial ischaemia and nausea, the mean potential averted cost to the healthcare system resulting from this HMR was $1019.71.

50

FINAL REPORT Review 402 The patient reviewed was an 83 year old male whose medical history included grafts of the femoral and coronary arteries, sleep apnoea and diabetes mellitus. His medications at the time of the HMR were as follows:

Medication Directions Aldactone 25mg Tab 25mg in the morning Alphapril 20mg Tab 20mg twice daily Diaformin 500mg Tab 500mg twice daily Kenacomb Otic Ear Drops Used when required Lasix 40mg Tab 40mg in the morning Lipitor 20mg Tab 20mg at night Minax 50mg Tab 50mg in the morning Mixtard 30/70 Penfill 60U twice daily Mobic 15mg Capsule 15mg in the morning Nitro-Dur 5mg Patch Applied in the morning and removed at night Nitrolingual 400mcg SL Pumpspray Used when required Panadeine Forte Tab One tablet when required Panamax 500mg Tab 1.0g twice daily Solprin 300 mg Tab 150mg in the morning Systane (Eye Drops) Instilled when required

During the HMR, the pharmacist identified the following DRPs:

• the patient was taking meloxicam (Mobic), which may cause renal failure when used in combination with enalapril (Alphapril) and frusemide (Lasix) (the pharmacist noted that the risk is very high in renally impaired diabetic patients such as this). The pharmacist recommended ceasing the meloxicam and increasing the dose of paracetamol (Panamax).

• the pharmacist was uncertain as to the indication for spironolactone (Aldactone), and recommended ceasing the drug if it was not indicated.

Due primarily to a reduction in the risk of renal dysfunction and hyperkalaemia, the mean potential averted costs to the healthcare system resulting from this HMR were $2391.45.

51

FINAL REPORT Review 514 The patient reviewed was a 77 year old woman whose medical history included atrial fibrillation, ischaemic heart disease, dementia, osteoporosis and sleep apnoea. Her medications at the time of the HMR were as follows:

Medication Directions Asmol CFC-free Inhaler Two puffs when required Betaloc 50mg Tab 25mg twice daily Buscopan 10mg Tab 20mg every four hours Caltrate 600mg + Vit D Tab Two tablets daily DBL Aspirin 100mg Tab 100mg daily Fosamax Plus Once Weekly Tab One tablet weekly Lycinate 600 mcg Tab One tablet when required Oroxine 100mcg Tab Alternating 100mcg/50mcg daily Ostelin 1000 Cap One capsule twice daily Panamax 500mg Tab 1.0g when required Stemetil 5mg Tab 5mg three times daily when required Zocor 40mg Tab 40mg at night

The patient was referred for a HMR as her GP suspected non-compliance. During the HMR, the pharmacist identified the following DRPs:

• the patient was very confused about her medications and was regularly mixing them up and not taking them as prescribed. She was also confused about when and how to use her glyceryl trinitrate (Lycinate) tablets and salbutamol (Asmol) inhaler, and the difference between them. The pharmacist recommended that the patient commence using a Webster pack and returned to the patient’s house to demonstrate how to use it following the HMR interview.

• Counselling was also provided regarding the indications for her medications. Due to a reduction primarily in the risk of myocardial ischaemia, the mean potential averted costs to the healthcare system resulting from this HMR were $1145.18.

52

FINAL REPORT

Appendix XII Re-sampled data

Figure 20 - Re-sampled incremental cost-effectiveness ratios for HMRs vs usual care: baseline analysis. Line indicates cost effectiveness threshold of $50 000 per QALY

FINAL REPORT