New Zealanders' Access to Licensed and Subsidised Medicines Compared with a Nationwide Single Payer System In

Home , ATC code A, ATC code B, Drotrecogin alfa

New Zealanders’ Access To Licensed And Subsidised Medicines Compared With A Nationwide Single Payer System In Each of Australia, The United Kingdom And The United States

Yogarajan Ragupathy, Bpharm, BSc, MPS, RegPharmNZ

A thesis submitted for the degree of Doctor of Philosophy, at the University of Otago, Dunedin, New Zealand,

October 2012

ABSTRACT

Introduction: There is disagreement about whether New Zealand is falling behind in access to medicines compared to countries such as Australia (1, 2). The controversy has largely focused around the subsidy (public funding) of medicines by New Zealand’s single payer public health system, and in particular the role of the New Zealand Pharmaceutical Management Agency (PHARMAC) (3-6).

Aims: To compare New Zealanders’ access to licensed and subsidised medicines with that under a nationwide, publically funded single payer system in each of Australia, the United Kingdom and the United States.

Methods: A literature search was performed for previous international comparisons of access to medicines in the peer-reviewed literature. The search found a lack of established methods for large-scale comparisons between high-income countries. Therefore a new methodology was developed that classified all medicines licensed and subsidised under the four single payer systems using active agents and the World Health Organization’s Anatomical Therapeutic Chemical (ATC) system, and matched medicines to registration dates from the regulatory authorities in New Zealand, Australia, the United Kingdom and the United States.

Three linked investigations were performed. Investigation One compared the number, type, timeliness and innovation of medicines licensed for adults and subsidised under the four single payer systems. Investigation Two compared the number, type, timeliness, lowest age authorised and suitability of formulation of medicines licensed for children and subsidised under the four single payer systems. Investigation Three examined key informants’ perceptions of New Zealanders’ access to medicines, including in comparison to the other countries studied.

Results: Investigation One found that New Zealand had fewer licensed and subsidised entities,

licensed and subsidised fewer innovative entities that provided important health gains, and had older licensed and subsidised entities than the comparator countries.

Investigation Two found that while New Zealand licensed the second lowest number of entities in children, PHARMAC subsidised the second highest number of entities. PHARMAC also subsidised the second highest number of entities in formulations suitable for children. New Zealand also had the highest ratio of medicines licensed in children to those licensed in adults, and the highest ratio of medicines subsidised in children to those subsidised in adults.

Investigation Three found that key informants perceived New Zealand to have an effective medicines licensing system that ensured the safety and efficacy of medicines, and which ensured a good relationship with the pharmaceutical industry while remaining objective. PHARMAC was perceived to control pharmaceutical spending, be able to extract the maximum buying power from a limited budget, and to be resistant to lobbying or political influence. However, there were concerns over access to new medicines, and that medicines were assessed more stringently for cost effectiveness than other health investments.

Conclusion: New Zealand was perceived to have an effective system for licensing medicines, and contain pharmaceutical spending well. New Zealand compares well in children’s access to medicines, but adults may not have the same access to newly developed and innovative medicines as patients in Australia, the United Kingdom and the United States. Further work is needed to determine what if any impact this has on health outcomes.

ACKNOWLEDGEMENTS

I would like to thank my supervisors, Dr June Tordoff, Professor Pauline Norris and Associate Professor David Reith for their guidance and encouragement during my studies.

I would also like to thank my PhD facilitators Drs Clare Strachan and Susan Heydon for guiding me through the PhD process, and Professors Thomas Rades and Stephen Duffull for encouraging me to start the PhD journey in the first place.

I would like to thank my friend and colleague Katri Aaltonen, who shares my interest in comparing access to pharmaceuticals in high-income countries. Katri has been a pleasure to work with, and our collaboration on the New Zealand/Finland comparison got my research at Otago off to a flying start.

I would also like to thank the management and my colleagues at Waikato District Health Board for being so supportive of my combining work and studies. My line managers at Pharmacy Services, Jan Goddard and Fiona McNabb, have gone out of their way to encourage my research and given me valuable opportunities to present my research and develop my skills.

I am grateful to all the health professionals, academics, public servants, Members of Parliament, pharmaceutical industry representatives and patient advocates who gave their valuable time for interviews, and who were willing to answer my questions about pharmaceutical policy. I am also grateful to colleagues in Australia, the United Kingdom and the United States, who were always willing to answer my queries about their respective health systems.

I am grateful to the School of Pharmacy and the Division of Health Sciences for financial support during my studies, and for the funding for attending conferences both here in New Zealand and overseas.

Last but most certainly not least, I would like to thank my family for all their love and support. My wife Vithya has been a source of endless strength and love, and I will always be grateful for having her in my life.

iii

TABLE OF CONTENTS ABSTRACT ...... i

ACKNOWLEDGEMENTS ...... iii

TABLE OF CONTENTS ...... iv

LIST OF TABLES ...... viii

LIST OF FIGURES ...... xi

LIST OF ABBREVIATIONS ...... xii

GLOSSARY ...... xv

ATC CODE MASTER KEY (ATC LEVEL 1) ...... xvii

ATC CODE MASTER KEY (ATC LEVEL 2) ...... xviii

CHAPTER 1: INTRODUCTION ...... 1

1.1: What Constitutes Access? ...... 1 1.1.1: Access Across the Globe...... 1 1.1.2: Essential Medicines, and Essential Medicine Lists ...... 2 1.1.3: Characteristics of Countries and Systems That Have Good Access to Essential Medicines ...... 5 1.1.4: Access in High-income Countries ...... 8 1.1.5: Initial Scope Limitation: Choice of Countries ...... 12 1.2: The High-Income Countries Being Compared ...... 12 1.2.1: Choice of Comparators...... 12 1.2.2 Medicine Licensing Bodies in These Four Countries ...... 14 1.2.3: Medicines Funding Systems in These Countries ...... 17 1.2.4: Second Scope Limitation- Choice of Comparator Systems ...... 24 1.3: Policy Issues of Concern in These Four Countries ...... 29 1.3.1: Access Problems for Children ...... 29 1.3.2: Orphan Drugs ...... 31 1.3.3: High Cost Medicines ...... 31 1.4: Licensing Process in New Zealand, Australia the United Kingdom and United States ...... 32 1.4.1: The Drug Development Process ...... 33

1.4.2: Regulation After Licensing ...... 35 1.5: Evaluation For Public Funding ...... 37 1.5.1: Why Evaluate Pharmaceuticals For Public Funding? ...... 37 1.5.2: Economic Evaluation ...... 38 1.5.3: Social Weighting And Ethical Theories ...... 40 1.6: The Controversy In New Zealand ...... 41 1.7: Aims, Hypotheses, and Final Scope Limitation ...... 42 1.7.1: Aims ...... 42 1.7.2: Hypotheses ...... 42 1.7.3: Final Scope Limitations ...... 44 CHAPTER TWO: METHODS ...... 46

2.1: Literature Searching ...... 46 2.2: International Comparisons (Investigations One and Two) ...... 49 2.2.1: Hypotheses ...... 49 2.2.2: Developing the Methodology ...... 49 2.2.3: Data Sources...... 49 2.2.4: Selection of products (Study Population) ...... 51 2.2.5: Study Variables ...... 56 2.2.6: Data Analysis ...... 60 2.3: Key Informants’ Perceptions of New Zealanders’ Access to Medicines (Investigation Three) ...... 66 2.3.1: Objective ...... 66 2.3.2: Participant Selection ...... 66 2.3.3: Interviews ...... 68 2.3.4: Data Analysis ...... 69 CHAPTER 3: THE LICENSING AND SUBSIDY OF MEDICINES FOR ADULTS UNDER A NATIONWIDE SINGLE PAYER SYSTEM IN EACH OF NEW ZEALAND, AUTRALIA, THE UNITED KINGDOM AND THE UNITED STATES ... 70

3.1 Context ...... 70 3.2 Results ...... 71 3.2.1 Number of Licensed and Subsidised Entities ...... 71 3.2.2 Types of Licensed and Subsidised Entities ...... 72 3.2.3 Time Since First Registration of Licensed and Subsidised Entities ...... 78 v

3.2.4. Level of Innovation of Licensed and Subsidised Products ...... 89 3.2.5: Sensitivity Analysis: Removing Entities Only Available In Injectable Formulations 92 3.3: Chapter Three Discussion ...... 94 3.3.1: Summary of Results...... 94 3.3.2 Reasons For Differences Between Countries, And Their Implications ...... 95 3.3.3: Strengths And Limitations of The Study ...... 98 3.4: Conclusion ...... 100 CHAPTER 4: THE LICENSING AND SUBSIDY OF MEDICINES FOR CHILDREN UNDER NATIONWIDE SINGLE PAYER SYSTEMS IN NEW ZEALAND, AUSTRALIA, THE UNITED KINGDOM AND THE UNITED STATES ...... 101

4.1. Context ...... 101 4.2 Results ...... 102 4.2.1 Number Of Licensed And Subsidised Entities ...... 102 4.2.2: Types of Licensed and Subsidised Entities ...... 106 4.2.3: Time Since First Registration of Licensed and Subsidised Entities ...... 110 4.2.4. Lowest Age Authorised ...... 123 4.2.5: Availability of Suitable Formulations ...... 125 4.3: Chapter Discussion ...... 130 4.3.1: Results Summary ...... 130 4.3.2: Reasons For Differences Between The Countries ...... 131 4.3.3: Strengths and Weaknesses ...... 132 4. 4. Conclusion ...... 133 5.1 Context ...... 134 5.2 Results ...... 134 5.2.1 Widespread Agreement: What New Zealand Does Well ...... 135 5.2.2 Widespread Agreement: What New Zealand Could Do Better ...... 140 5.2.3 Areas of Disagreement ...... 144 5.3 Chapter Discussion ...... 146 5.3.1 Results Summary ...... 146 5.3.2 Comparison With Past Findings ...... 147 5.3.3 Limitations and Future Work ...... 148 5.4 Conclusion ...... 148 CHAPTER 6: INTERPRETATION ...... 149 vi

6.1 Implications for New Zealanders’ Access to Medicines ...... 149 6.1.1: Breadth ...... 149 6.1.2: Depth ...... 151 6.1.3: Timeliness ...... 154 6.1.4: Choice ...... 157 6.1.5: Equity...... 160 6.1.6: Continuity ...... 162 6.1.7: Cost Sharing ...... 163 6.1.8: Summary ...... 165 6.2: Comparisons with previous work ...... 165 6.2.1: Development of new methodology ...... 168 6.3: Strengths and Limitations ...... 169 6.4 Further Research ...... 173 6.5 Conclusion ...... 174 REFERENCES ...... 176

APPENDIX I: CONFERENCE PRESENTATIONS ...... 190

APPENDIX II: PUBLICATIONS ...... 191

DATA ANNEX 1: DUPLICATE ANALYSES BY ACTIVE AGENT ...... 192

DA 1.1 INVESTIGATION ONE ...... 192 DA 1.2: INVESTIGATION TWO ...... 201 DATA ANNEX TWO: SUPPLEMENTARY ANALYSES ...... 212

vii

LIST OF TABLES

TABLE 1: The key features of each system (VANF, NHS, PBS and PHARMAC) ...... 27 TABLE 2: Data sources for literature search ...... 48 TABLE 3: Variables analysed in Investigations 1 and 2 ...... 56 TABLE 4: Characteristics of key informants who took part in the study ...... 67 TABLE 5: Number of entities collapsed by ATC code that were licensed and subsidised for adults in each country. (See TABLE DA 1.1 in Data Annex 1 for duplicate active agent analysis)...... 72 TABLE 6: Numbers of licensed and subsidised entities in each country at ATC level 1 ...... 74 TABLE 7: Numbers of licensed and subsidised entities in each country at ATC level 2 ...... 75 TABLE 8: Numbers of licensed and subsidised entities used in treating diabetes, cardiovascular disease, cancer, lung disease and mental illness in each country at ATC level 2 ...... 76 Table 9: Comparisons of the comparative age (time since first registration) of licensed and subsidised entities in each country. (See TABLE DA 1.2 in Data Annex 1 for duplicate active agent analysis)...... 83 TABLE 10: Comparison of the comparative age (time since first registration) of specific sub groups. (See TABLE DA 1.4 in Data Annex 1 for duplicate active agent analysis)...... 85 TABLE 11: Differences in the comparative age (median time since first registration) of subgroups of licensed entities. (See TABLE DA 1.5 in Data Annex 1 for duplicate active agent analysis)...... 87 TABLE 12: Differences in the comparative age (median time since first registration) of subgroups of subsidised entities. (See TABLE DA 1.6 in Data Annex 1 for duplicate active agent analysis)...... 88 TABLE 13: Number of ATC level 4 groups that contain licensed and subsidised entities in each country ...... 89 TABLE 14: Summary of the 2007 licensing and subsidy status of 65 innovative medicines that provide important health gains ...... 90 TABLE 15: Percentage of entities available in non-injectable formulations (collapsed by ATC code). (See TABLE DA 1.7 in Data Annex 1 for duplicate active agent analysis)...... 92 TABLE 16: Number of entities licensed and subsidised for children, collapsed by ATC code. (See TABLE DA 1.9 in Data Annex 1 for duplicate active agent analysis)...... 103 TABLE 17: Comparison of entities licensed and subsidised for children and adults within each country. (See TABLE DA 1.10 in Data Annex 1 for duplicate active agent analysis)...... 105 TABLE 18: Number of entities licensed and subsidised for children at ATC level 1 ...... 107 TABLE 19: Number of entities licensed and subsidised for children at ATC level 2 ...... 108 TABLE 20: Comparisons of the comparative age (times since first registration) of entities licensed and subsidised for children in the four countries. (See TABLE DA 1.11 in Data Annex 1 for duplicate active agent analysis)...... 113 TABLE 21: Comparison of the comparative age of subgroups of entities licensed and subsidised for children in the four countries. (See TABLE DA 1.12 in Data Annex 1 for duplicate active agent analysis)...... 117 TABLE 22: Differences in the comparative age (median time since first registration) of subgroups of entities licensed for children. (See TABLE DA 1.13 in Data Annex 1 for duplicate active agent analysis)...... 119 TABLE 23: Differences in the comparative age (median time since first registration) of subgroups of subsidised entities, (See TABLE DA 1.14 in Data Annex 1 for duplicate active agent analysis)...... 120 TABLE 24: Comparison of median time since first registration of entities licensed and subsidised for children with those licensed and subsidised for adults in each country. (See TABLE DA 1.15 in Data Annex 1 for duplicate active agent analysis)...... 122 TABLE 25: Lowest age authorised within each country for licensed and subsidised entities collapsed by ATC Code. (See TABLE DA 1.16 in Data Annex 1 for duplicate active agent analysis)...... 124 TABLE 26: Number and proportions of suitably formulated licensed and subsidised entities in each country. (See TABLE DA 1.17 in Data Annex 1 for duplicate active agent analysis)...... 126

viii

TABLE 27: Difference in lowest age authorised of suitably formulated and non-suitably formulated entities. (See TABLE DA 1.18 in Data Annex 1 for duplicate active agent analysis)...... 129 TABLE DA 1.1: Number of entities collapsed by active agent that were licensed and subsidised for adults in each country ...... 192 TABLE DA 1.2: Comparisons of the age (time since first registration) of licensed and subsidised entities (collapsed by active agent) in each country ...... 193 TABLE DA 1.3: Differences in the age (median times since first registration) of licensed and subsidised entities in each pair of countries ...... 194 TABLE DA 1.4: Comparison of the age (time since first registration) of specific sub groups ...... 196 TABLE DA 1.5: Differences in the age (median time since first registration) of subgroups of licensed entities ...... 197 TABLE DA 1.6: Differences in the age (median time since first registration) of subgroups of subsidised entities ...... 198 TABLE DA 1.7: Percentage of entities available in non-injectable formulations (collapsed by active agent) .. 199 TABLE DA 1.8: Median time since first registration of licensed entities available in non-injectable formulation compared with all entities (collapsed by active agent) ...... 200 TABLE DA 1.9: Number of entities licensed and subsidised for children, collapsed by active agent ...... 201 TABLE DA 1.10: Comparison of entities licensed and subsidised for children and adults within each country ...... 202 TABLE DA 1.11: Comparisons of the age (times since first registration) of entities licensed and subsidised for children in the four countries ...... 203 TABLE DA 1.12: Comparison of the age of subgroups of entities licensed and subsidised for children in the four countries ...... 205 TABLE DA 1.13: Differences in the age (median time since first registration) of subgroups of entities licensed for children ...... 206 TABLE DA 1.14: Differences in the age (median time since first registration) of subgroups of entities subsidised for children ...... 207 TABLE DA 1.15: Comparison of median time since first registration of entities licensed and subsidised for children with those licensed and subsidised for adults in each country ...... 208 TABLE DA 1.16: Lowest age authorised within each country for licensed and subsidised entities collapsed by active agent ...... 209 TABLE DA 1.17: Number and proportions of suitably formulated licensed and subsidised entities in each country ...... 210 TABLE DA 1.18: Difference in lowest age authorised of suitably formulated and non-suitably formulated entities ...... 211 TABLE DA 2.1: ATC level 4 groups that were listed in only one country, or listed in three of the four countries ...... 220 TABLE DA 2.2: ATC level 4 groups that were subsidised in only one country, or subsidised in three of the four countries...... 224 TABLE DA 2.3: Median number of entities in ATC level 4 groups that contain licensed and subsidised entities in each country ...... 225 TABLE DA 2.4: Differences in the median numbers of ATC level 4 groups that contain licensed and subsidised entities in the four countries ...... 226 TABLE DA 2.5: Licensing and subsidy status of each innovative medicine by country in 2007 ...... 231 TABLE DA 2.6: Licensed entities available in non-injectable formulations compared with the total number of entities in each country at ATC level 1 ...... 234 TABLE DA 2.7: Subsidised entities available in non-injectable formulations compared with the total number of entities in each country at ATC level 1 ...... 236

TABLE DA 2.8: Median time since first registration of licensed entities available in non-injectable formulation compared with all entities (collapsed by ATC code) ...... 237

LIST OF FIGURES

FIGURE 1: Comparative age (time since first registration) of licensed entities collapsed by ATC code. ...80 FIGURE 2: Comparative age (time since first registration) of subsidised entities collapsed by ATC code.81 FIGURE 3: Comparative age (time since first registration) of licensed entities collapsed by ATC code. . 111 FIGURE 4: Comparative age (time since first registration) of subsidised entities collapsed by ATC code...... 112

LIST OF ABBREVIATIONS

ADR Adverse Drug Reaction AFP Association of Finnish Pharmacies AIDS Acquired Immune Deficiency Syndrome ATC Anatomical Therapeutic Chemical Aus Australia CBA Cost Benefit Analysis CEA Cost Effectiveness Analysis CEC Community Exceptional Circumstances (New Zealand) CMA Cost Minimisation Analysis CMDh Co-ordination Group For Mutual Recognition and Decentralised Procedures- Human (European Medicines Agency) COBRA Consolidated Omnibus Budget Reconciliation Act (United States) CUA Cost Utility Analysis DCS Discretionary Community Supply (New Zealand) DHB District Health Board EC Exceptional Circumstances (New Zealand) EMA European Medicines Agency eMC electronic Medicines Compendium EML Essential Medicines List EU European Union FDA Food and Drug Administration (United States) GMP Good Manufacturing Practice GNP Gross National Product HCM High Cost Medicine HEC Hospital Exceptional Circumstances (New Zealand) HIV Human Immunodeficiency Virus ICH International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use MAP Medical Advisory Panel (United States Department of Veterans Affairs)

xii

Medsafe New Zealand Medicines And Medical Devices Safety Authority MeSH Medical Subject Heading MHRA Medicines and Healthcare Products Regulatory Agency (United Kingdom) NGO Non-Governmental Organisation NHS National Health Service (United Kingdom) NICE National Institute for Health and Clinical Excellence (United Kingdom) NZ New Zealand OECD Organisation for Economic Co-operation and Development OFT Office of Fair Trading (United Kingdom) PBAC Pharmaceutical Benefits Advisory Committee (Australia) PBPA Pharmaceutical Benefits Pricing Authority (Australia) PBS Pharmaceutical Benefits Scheme (Australia) PCT Primary Care Trust (United Kingdom - also Pharmaceutical Cancer Treatment in New Zealand, but not used in that sense in this thesis) PDR Physicians' Desk Reference (United States) PHARMAC Pharmaceutical Management Agency of New Zealand (New Zealand) PMPRB Patented Medicines Prices Review Board (Canada) PPA Prescription Pricing Authority (United Kingdom) PPRS Pharmaceutical Price Regulation Scheme (United Kingdom) PTAC Pharmacology and Therapeutics Advisory Committee (New Zealand) QALY Quality Adjusted Life Year TGA Therapeutic Goods Administration (Australia) TNF Tumour Necrosis Factor TRIPS Trade Related Aspects of Intellectual Property Rights UK United Kingdom of Great Britain and Northern Ireland US United States of America USD PPP United States Dollar Purchasing Power Parity VA Department of Veterans Affairs (United States) VANF Department of Veterans Affairs National Formulary (United States)

xiii

VA-PBM Veterans Affairs Pharmacy Benefits Management Strategic Healthcare Group VISN Veterans Integrated Service Network (United States Department of Veterans Affairs) WHO World Health Organization WTO World Trade Organization

xiv

GLOSSARY

Active Agent The chemical substance in a medicine that achieves the medicine's intended therapeutic effect

ATC Code An identifying code for medicines developed by the WHO Collaborating Centre for Drug Statistics Methodology for use in drug utilisation research

CBA Cost Benefit Analysis, a form of pharmaceoeconomic evaluation where results are expressed in monetary units for both costs and outcomes

CEA Cost Effectiveness Analysis, a form of pharmaceoeconomic evaluation where results are expressed in costs per clinical outcome

Children People under 18 years of age

CMA Cost Minimisation Analysis, a form of pharmacoeconomic analysis where the outcomes of the comparators are assumed to be equivalent, making the comparator with the lowest costs superior

Co-payment The patient's contribution to the cost of his or her medicines

CUA Cost Utility Analysis, a form of pharmaceoeconomic evaluation where results are expressed in costs per Quality Adjusted Life Year (QALY)

Datasheet A summary of a medicine's characteristics that has been approved by a country's regulatory agency, and designed for use by health professionals. Also called a Summary of Product Characteristics (SPC) in the United Kingdom, and a Drug Label in the United States

Entity A unique active agent or ATC code

Formulation A particular dose form of a medicine, for example dermal, nasal or ocular

Licensed A medicine that has been approved for use in humans by a country's regulatory agency. Licensing is also called Marketing Approval in the United Kingdom, and Drug Registration in the United States

Medicine A product that is intended to achieve a therapeutic effect such as preventing, curing or alleviating disease, effecting contraception, or altering the normal physiological functioning of the human body

Product A particular dose form and strength of a medicine

Single Payer A health system where one organisation (usually the Government) is responsible for purchasing and paying for services

Subsidised A medicine that is entirely or partly paid for by the tax- payer, rather than by the patient

Suitable Formulation A formulation that can be used by children of all ages

Time From First The time from the earliest date of registration (licensing) of Registration the medicine in any of New Zealand, Australia, the United Kingdom or the United States

Unsuitable Formulation A formulation that can NOT be used by children of all ages

xvi

ATC CODE MASTER KEY (ATC LEVEL 1)

A: Alimentary tract and metabolism agents B: Blood and blood forming organ products C: Cardiovascular system agents D: Dermatologicals G: Genito urinary system agents and sex hormones H: Systemic hormonal preparations J: Anti-infectives for systemic use L: Antineoplastic and immunomodulating agents M: Musculo-skeletal system agents N: Nervous system agents P: Antiparasitic products R: Respiratory system agents S: Sensory organ agents V: Various

xvii

ATC CODE MASTER KEY (ATC LEVEL 2)

A10: Drugs used in diabetes B01: Antithrombotic agents C01: Cardiac therapy D07: Corticosteroids, dermal preparations G03: Sex hormones and modulators of the genital system H02: Corticosteroids for systemic use J01: Antibacterials for systemic use J05: Antivirals for systemic use J07: Vaccines L01: Antineoplastic agents L04: Immunosuppressants M01: Anti-inflammatory and antirheumatic products M03: Muscle relaxants N01: Anaesthetics N02: Analgesics N03: Antiepileptics N05: Psycholeptics N06: Psychoanaleptics R03: Drugs for obstructive airways diseases S01: Opthalmologicals

xviii

CHAPTER 1: INTRODUCTION

1.1: What Constitutes Access?

The term ‘access to medicines’ can mean different things in different countries, and in different contexts within the same country. This thesis focuses on the availability of licensed and publically funded medicines in the four countries being compared (New Zealand, Australia, the United Kingdom and the United States).

There are a number of factors that can affect access to medicines, both in high-income countries such as the four being compared, and in low or middle-income countries. These factors are examined in this section.

1.1.1: Access Across the Globe

Access to medicines can vary widely from one country to another. In 1999, high-income countries consumed over 90% of the world’s medicines by value, despite having only 15% of the world’s population (7). The proportion of the world’s medicines consumed by high- income countries rose from 89.1% to 92.9% in the period 1985 to 1999, while the proportions consumed by low and middle-income countries fell over the period from 10.9% to 7.1%. Over the same period, the proportion of the world’s population living in high-income countries declined from 18% to 15% (7). The World Health Organization (WHO) used a World Bank classification that defined high-income countries as those with a gross national product (GNP) of United States (US) $9381 per capita or more in 1999, middle-income countries as those with a GNP of US $761 to US $9380, and low-income countries as those with GNP of US $760 or less (7, 8).

This relative per capita growth in pharmaceutical consumption reversed between 2000 and 2008. High-income countries still consumed almost eight times the volume of pharmaceuticals per capita as low-income countries in 2008. However, per capital consumption increased by 29.3% in low-income countries, 22.9% in lower middle-income countries and 20.4% in upper middle income countries over the period from 2000 to 2008,

compared with 18.6% in high-income countries (9). This suggests that the previous dominance of high-income countries has decreased.

Pharmaceutical spending (which could be regarded as a proxy for access) also varies greatly across the globe. The average spending per annum in low-income countries was US $4.4 per capita in 2000, compared with an average of US $396 per capita in high-income countries (7). There were also large differences within the income groups. In the low-income group, the lowest spending per annum measured was US $0.6 per capita, and the highest was US $26. In the high-income group, spending varied from US $84 to US $549 between countries in the group (7).

Government spending per annum on medicines (which will affect individuals’ access to medicines) increased from US $110 to $167 per capita in high-income countries in the period 1990-2000 and from $5 to $8 in middle-income countries, but barely changed from $1 to $1.1 in low-income countries. Private spending per annum increased from US $130 to $229 in high-income countries and $13 to $22 in middle-income countries, but only increased from $2.6 to $3.2 in low-income countries. High-income countries had the highest proportion of publicly funded medicines, but this declined from 42.2% to 39.4% over the period (7).

By 2006, pharmaceutical spending per capita had increased to an average of US $7.61 in low income countries, of which US $1.76 (23.1%) was on publically funded medicines. In contrast, high income countries spent an average of US $431.6 per capita on pharmaceuticals, of which US $264.4 (61.3%) was on publically funded medicines (9).

1.1.2: Essential Medicines, and Essential Medicine Lists

The World Health Organization defines essential medicines as follows: “Essential medicines are those that satisfy the priority health care needs of the population. They are selected with due regard to public health relevance, evidence on efficacy and safety, and comparative cost effectiveness. Essential medicines are intended to be available within the context of functioning health systems at all times in adequate amounts, in the appropriate dosage forms, with assured quality and adequate information, and at a price the individual and the community can afford. The implementation of the concept of essential medicines is intended

to be flexible and adaptable to many different situations; exactly which medicines are regarded as essential remains a national responsibility” (10).

From this definition, the important characteristics of essential medicines are:

 They are selected according to the needs of the population they are intended for, and so will vary from one setting to another  Selection of essential medicines involves considerations of safety, efficacy, quality, and cost effectiveness  Essential medicines need to be available in appropriate dosage forms and with relevant prescribing information for the population they will be used in  The cost to both the patient and the funding system has to be considered

Essential medicines are a cost effective way of fighting serious illness, thereby improving health. Essential medicines are available for the most important infectious and non-infectious diseases worldwide (11). Good access to essential medicines saves lives and reduces suffering. Essential medicines also increase the effectiveness of the health system, as they draw the public to health professionals, which in turn give the public access to other medical services. Last but not least, essential medicines increase the cost effectiveness of both government and personal pharmaceutical spending (11).

The selection of essential medicines depends first on marketing approval in that country by the relevant regulatory authority. This is usually based on safety, quality and efficacy rather than on comparison with products that are already available. Once marketing approval is granted, the organisation responsible for meeting the costs of the drug evaluates medicines for cost effectiveness, and compares agents with other drugs available to create an essential medicines list. The World Health Organization recommends that essential medicines lists be developed for different levels of health care (such as a health centre or hospital) and taking into account the guidelines for diseases that can be treated at that level of care (12). The World Health Organization publishes Model Lists of Essential Medicines for both adults and children (13).

The World Health Organization first developed a definition of essential medicines (then called essential drugs) in 1975, and published the first WHO Model List of Essential Drugs in 1977. Since 2002, the inclusion of medicines on the WHO Model List has been evidence based, and considers public health relevance, effectiveness, safety, and cost effectiveness (14). Essential medicines lists (EMLs) are used by national public health programmes, insurance schemes, and Non-Governmental Organisations (NGOs) (12, 15).

The majority of national EMLs are in developing countries. In 1999, over 90% of low and middle-income countries had a national EML, while less than a third of high-income countries had a national EML (7). However, EMLs in high-income countries had a much larger number of medicines, an average of 903 medicines compared with an average of 276 medicines in low-income countries and 420 medicines in middle-income countries (7). Both Australia and New Zealand have national essential medicines lists. These are the pharmaceutical reimbursement lists used in their public health systems, namely the Pharmaceutical Benefits Scheme (PBS) in Australia, and the Pharmaceutical Management Agency (PHARMAC) Schedule in New Zealand (12).

In addition, high-income countries use essential medicines in hospitals and health insurance schemes (7). Greater use of essential medicines lists in high-income countries may be a means of controlling pharmaceutical spending, while providing the best value for money spent (12, 16). However, the introduction of national essential medicines lists in high-income countries has been criticised as having the potential to reduce health care and delay drug development (15).

By 2010, the WHO model list of essential medicines contained 340 medicines, and included medicines for infectious disease (such as malaria, HIV/AIDS and tuberculosis), chronic conditions such as diabetes and cancer, and reproductive health. Four out of five countries had a national medicines list based on the WHO model list (17).

1.1.3: Characteristics of Countries and Systems That Have Good Access to Essential Medicines

The World Health Organization estimates that 30% of the world’s population, 1.7 billion people, lacked access to essential medicines in 1999. In 1999, 24% of the population in medium income countries lacked access to essential medicines, as did 39% in low-income countries (7). The countries with the poorest health outcomes and life expectancies also had the lowest access to essential medicines (7). The World Health Organisation has found that lack of availability of medicines in the public sector, along with high prices in the private sector, continue to be ongoing barriers to accessing essential medicines (9).

Access to essential medicines is related to the overall performance of a country’s health system (7). The World Health Organization also lists four medicine factors that affect access to essential medicines. These are:

 Rational selection and use of medicines  Affordable prices  Sustainable financing  Reliable health and supply systems (7, 10)

1.1.3.1 Rational Selection and Use of Medicines

Rational selection means selecting medicines that suit the country’s needs. Safety, efficacy and cost effectiveness need to be taken in to account, keeping in mind that disease patterns and costs will vary from one country to another (7, 11). In addition to the use of national Essential Medicines lists, the World Health Organization gives the following examples of policies that support rational selection of medicines:

 Independent drug information  Subsidy lists  Treatment guidelines using the best available evidence  Drug information bulletins (11) 5

1.1.3.2 Affordable Prices

Even when essential medicines are available, they need to be affordable for the patient. The World Health Organization gives the following examples of policies that help achieve affordable medicine prices: (7, 10, 11)

 Using available price information (such as that paid in other countries, or for similar medicines) to negotiate the best prices  Competitive bulk purchasing  Price regulation (reducing or eliminating mark-ups, taxes, and tariffs)  Drug tendering (negotiating with multiple suppliers)  Pooled procurement (increasing purchasing power by buying for several purchasers together)  Negotiating equity in the pricing of new drugs, i.e. manufacturers setting the prices of medicines based on what the country can pay for them  Price information (what is paid for the medicine in other countries, or for therapeutic equivalents, to allow medicine purchasers to negotiate the best prices)  Price competition within the country, including using generics (costs of generic medicines are usually much less than that of patented medicines) and therapeutic equivalents  Local production of medicines (this must be economically viable, and follow standards of good manufacturing practice)  Containment of distribution costs (reducing mark-ups by making the distribution and dispensing to the patient more efficient) (7, 10, 11)

In some high-income countries, systematic pharmacoeconomic analysis is used to help negotiate the best medicine prices and/or gain the best value for money from pharmaceutical spending. The Australian Pharmaceutical Benefits Scheme (PBS) is an example of a system that uses such analyses (7). Other examples are New Zealand’s Pharmaceutical Management Agency (PHARMAC), the United Kingdom’s National Institute for Health and Clinical Excellence (NICE) and the United States Department of Veterans Affairs National Formulary

(VANF) (18-20).

1.1.3.3: Sustainable Financing

The World Health Organization gives the following examples of methods of financing that can increase access to essential drugs. These are:

 Drug benefits in social health insurance  Targeted public/NGO financing  External financing  Better use of household spending  Drug benefits in private insurance  Community and employer financing

The World Health Organization’s position is that drug benefits within social health insurance do the most to reduce disadvantage, and therefore help equity. This is the WHO’s preferred approach to drug funding. In order to have the maximum effect, this approach requires that priority is given to important diseases and poorer groups, and requires specialist knowledge. The WHO recommends increasing public funding for essential medicines without diverting money away from other health priorities by increasing the total health funding (10, 14).

External financing is funding via development loans and grants from high-income countries to low-income ones. However, this only has a limited impact on raising the per capita expenditure on medicines (7). The WHO recommends that these should be targeted at developing long term solutions to drug access problems (14).

Household spending forms the bulk of drug spending in low and middle-income countries (7). However, it creates access problems for already disadvantaged groups (14). People in low and middle-income countries are more likely to buy their medicines from poorly regulated sources than those in high-income countries, which means they are less likely to get medicines that are safe, effective and of good quality. In addition, the medicine may not be used appropriately. The WHO recommends policies that increase availability, lower prices, and

increase rational use of medicines to maximise the effectiveness of personal spending (7, 14).

1.1.3.4: Reliable Health and Supply Systems

The effective supply of medicines depends on the overall health system, including the availability of diagnostic tests and the number of trained doctors, nurses, pharmacists and other health professionals (10). The supply systems specifically for medicines consist of purchasing strategies (see Section 1.1.3.2) and distribution systems within the country (7). A good distribution system ensures that medicines are available when needed (with a low number of medicines being out of stock) and reach the patient before the medicine’s expiry date. In addition, the outlet (such as a pharmacy) needs to be physically accessible for the patient (7).

The WHO recommends that countries use a mix of public, private and NGO services that suit their needs to make distribution more efficient. Regulation is important throughout the supply chain to reduce the number of counterfeit and sub-standard medications, and reduce unregulated sales of medicines (10).

1.1.4: Access in High-income Countries

Access to medicines in high-income countries have varies depending on regulatory approval, subsidy policies, price regulation, market size, and the mix of new and older medicines available (21-24).

In order to gain marketing approval in a high-income country, a manufacturer needs to prove that the product is safe, effective, and of good quality (manufactured according to Good Manufacturing Practice standards). Safety and efficacy are usually shown in relation to placebo rather than to existing products. In addition to these three hurdles, a manufacturer may need to show the product is cost effective compared to therapeutic alternatives in order to get it subsidised by private or public insurance schemes. Cost effectiveness comparisons are sometimes called the “fourth hurdle” to access (21, 22).

Cohen et al. (22) expanded on these four hurdles when comparing patient access to pharmaceuticals in four countries (the United States, the United Kingdom, France and the Netherlands), and identified eight factors that affect access. These are:

 The number of medicines given marketing approval in each country  The time taken for market approval  The percentage of approved drugs that would be subsidised  Patient cost sharing for subsidised drugs  The percentage of subsidised drugs that had conditions of subsidy  The time difference between marketing approval and listing for subsidy  The amount of choice patients had about their subsidy coverage  The availability of drugs across the population, regardless of ability to pay (22)

The study found the United States, which has multiple insurance providers, had the shortest time to subsidy, and offered patients the greatest flexibility in choosing their benefit plans. However, the United States also had the highest cost sharing for patients (the full cost of the medicines in the case of uninsured patients) and had the least evenness of coverage (22). Coverage in the United States is heavily influenced by whether patients have health insurance, and what is covered by their insurance (25). In the case of the European countries, each government is a monopsony (sole purchaser) for pharmaceuticals in that country. These countries had longer delays to subsidy, but greater evenness of coverage and lower cost sharing (22).

Cohen et al. also found that the European countries licensed fewer of the 100 drugs in the study (86 in the United Kingdom, 85 in France, 82 in the Netherlands) than the United States. The authors attributed this in part to the fact that all 100 selected were the top 100 selling in the United States, but found the result surprising because of the potential returns from these top selling drugs (22).

One possible explanation for this result is that all three of the European countries in the study use some form of pharmaceutical price or profit control (22). Danzon et al. analyzed 85 new chemical entities that were launched between October 1994 and October 1998 in 25 of the world’s major pharmaceutical markets, including the United States, 14 European Union

Countries, Japan, Australia and New Zealand. The study found that countries with lower drug prices have fewer launches of new drugs, and longer delays before launch. The authors suggest this finding supports the hypothesis that strict price regulation delays the launch of new drugs. The authors also found that larger markets have shorter launch delays than smaller ones (23).

A study published in 2005 by Danzon and Furukawa (24) compared retail pharmaceutical spending in 12 industrialised countries, including the United States, United Kingdom, Australia and Japan. The authors analyzed the mix of formulations, strengths and therapeutic categories of medicines used in each country. The authors found that the countries differed in the mix of formulations, strengths and categories used, but did not examine the reasons for this.

The authors also found that countries differed in the mix of older and newer drugs used (24). The United States had a higher per capita use of entities that been launched within the previous ten years than the other countries, and this was especially marked for entities that had been launched within the previous five years. Australia and the United Kingdom had a higher use of entities 30 years old or older than did the United States (24).

The authors measured the ‘launch lag’ of new entities in each of the countries (24). This was defined as the average period between the ‘global launch’ of the product (market launch anywhere in the 12 markets studied) and the launch in that particular market. The US had an average launch lag of 4.4 months for entities with a global launch within the past five years, and 7.6 months for those with a global launch within the previous 6 to 10 years. The lags for the United Kingdom were 9.7 and 13.7 months respectively, and those for Australia were 17.4 and 23.4 months respectively (24).

Wonder (26) identified five criteria that could be used for assessing access to new medicines in Australia and New Zealand. These are:

 Breadth (the number and range of new medicines available)  Depth (the restrictions on access to these medicines)

 Timeliness (how early this access occurs, and the delays caused various steps such as licensing and subsidy)  Equity (Do all citizens of the country have access to subsidised medicines? Do all patients with the same condition and level of need get the same access to medicines?)  Continuity (Does the patients get subsidised medicines as long as they need them?) (26)

The author focuses on breadth and timeliness in his report, and on ‘new’ prescription-only medicines (medicines that had been listed for subsidy in Australia within 10 years of gaining marketing approval or being recommended for marketing approval) (26). The author identified 78 new medicines subsidised in Australia between 2000 and 2006. Of these, 72 were registered in New Zealand, but only 20 were subsidised. Only four were subsidised in New Zealand before they were in Australia, and on average new medicines were subsidised in Australia 14 months before they were subsidised in New Zealand (26).

Access to prescription medicines can also be affected by access to primary health care in general. Since prescription medicines, by definition, need to be prescribed by a medical practitioner or other licensed prescriber, patients who are unable to see a prescriber will have no legal way of obtaining these medicines. For example, 33% of adults in the United States reported not seeing a doctor or getting recommended care when sick because of cost, as did 22% in Australia, 14% in New Zealand, and 5% in the United Kingdom (27).

Access to medicines also depends on an effective supply chain that delivers the medicine to the patient when needed. This supply chain can be disrupted by problems in manufacturing, natural disasters, transportation problems, manufacturers withdrawing from a market, and ‘artificial’ shortages caused by wholesalers buying up available stocks in order to drive up prices. Supply chain problems can impose additional costs to the health system in addition to causing problems for individual patients (28-31).

Last but not least, even when a medicine is licensed, subsidised and available at the point of delivery to the patient, the patient needs to be able to purchase it. 21% of adults in the United States reported not filling prescriptions in the previous 12 months due to cost, compared with 12% in Australia, 7% in New Zealand, and 2% in the United Kingdom (32).

The definitions of access developed by Wonder (2006) and Cohen et al. (2007) can be combined to create a framework to define access in high-income countries.

Access can be measured in terms of:

 Breadth (The number and range of medicines licensed and subsidised)  Depth (Restriction on access, e.g. subsidy conditions)  Timeliness (How early does the access occur?)  Choice (Do patients get to choose their coverage?)  Equity (Ability to pay? Geographical? Disease? Inter-generational?)  Continuity (Do patients get medicines as long as they need them?)  Cost Sharing (Co-Payments) (22, 26)

1.1.5: Initial Scope Limitation: Choice of Countries

This thesis aims to compare New Zealand’s access to medicines with that in comparable countries. As New Zealand is a high-income country in global terms (33), it would be logical to compare New Zealand with other high-income countries rather than low-income or middle- income countries.

The thesis (and the remainder of this introduction) will therefore focus on factors that influence access to medicines in high-income countries. As the majority of essential medicine lists are in low-income and middle-income countries, essential medicine lists will not be examined further in this thesis. Similarly, access problems unique to low-income and middle- income countries will not be examined further.

1.2: The High-Income Countries Being Compared

1.2.1: Choice of Comparators

Even when the possible comparators are limited to high-income countries, the World Bank has classified 70 economies as being high-income at the time of writing (33). The possible 12

comparators were therefore further limited to high-income countries that were members of the Organisation for Economic Co-Operation and Development (OECD), a group that includes New Zealand (34).

As licensing (marketing approval) of medicines was a potential barrier to access, another criterion was that none of the countries in the study should share a way of licensing medicines, to ensure the licensing data were independent. This meant limiting the study to only one country that came under the authority of the European Medicines Agency (EMA), due to the EMA’s centralised licensing procedures (35).

As subsidy of pharmaceuticals is also a barrier to access, the subsidy systems being compared had to be similar to that in New Zealand. New Zealand uses a publically funded single system to provide pharmaceutical coverage for all New Zealand residents. This system uses a single nationwide formulary, and has a central decision making body which decides when high cost pharmaceuticals will be publically funded (18). Therefore, each comparator country had to have at least one publically funded single payer system that used a single nationwide formulary, and which had one central body that made decisions on high cost drugs.

Lastly, both the licensing process for new medicines, and the decision making process for allocating public funding for high cost medicines are governed by the laws of the country concerned. The legal frameworks of the comparator countries should therefore be similar to that in New Zealand. Therefore, the comparator countries were limited to those whose legal systems were based on English Common Law (as is New Zealand’s legal system) rather than other types of legal system (36).

With these factors in mind, the United States, United Kingdom and Australia were selected as comparators. This choice of comparators has been validated by peer-reviewed publication (37). This quantitative study compared the number, type, timeliness and level of innovation of the products licensed and publically funded in the four countries (37).

The licensing and pharmaceutical funding systems in each country are described in sections 1.2.2 and 1.2.3 respectively.

1.2.2 Medicine Licensing Bodies in These Four Countries

1.2.2.1: New Zealand

The New Zealand Medicines and Medical Devices Safety Authority (Medsafe) regulates all medicines, controlled drugs used as medicines and medical devices marketed in New Zealand (38). Medsafe both evaluates medicines before licensing them for use in New Zealand, and carries out post marketing surveillance of medicines that are on the New Zealand market. This is designed to ensure all medicines meet standards for safety, efficacy and quality (38).

Medsafe administers the Medicines Act 1981 and the Medicines Regulations 1984, which together govern the use of medicines in humans in New Zealand. This includes the manufacture, distribution, advertising, and supply of medicines for human use (38-40).

At the time of data collection, and at the time of writing, New Zealand had no shared regulatory bodies or mutual medicine licensing treaties with any other country. However, the New Zealand and Australian Governments have announced plans to develop a joint regulatory agency that will replace both Medsafe and the Australian Therapeutic Goods Administration by 2016 (41).

1.2.2.2: Australia

The Therapeutic Goods Administration (TGA) regulates all therapeutic goods (including medicines, medical devices, blood and blood products) used in humans in Australia (42, 43). The TGA both evaluates medicines before they are licensed for the Australian market and carries out post marketing surveillance of medicines that are on the Australian market. The TGA evaluates medicines for safety, efficacy and quality (42, 43).

The TGA administers the Therapeutic Goods Act 1989. The legislation governs the use of medicines in humans, including manufacture, distribution and supply of medicines for human use (44).

At the time of data collection, and at the time of writing, Australia had no shared medicine regulatory bodies or mutual medicines licensing treaties with any other country. However, as noted in Section 1.2.2.1, the New Zealand and Australian Governments have announced plans to develop a joint regulatory agency (41).

1.2.2.3: United Kingdom

The United Kingdom’s Medicines and Healthcare Products Regulatory Agency (MHRA) regulates medicines, medical devices and blood supplies in the United Kingdom. The MHRA evaluates medicines for safety, quality and efficacy (45, 46). The MHRA both evaluates medicines before they are licensed on the United Kingdom market and carries out ongoing surveillance once the product is released on the market (45, 46).

The MHRA operates under the Medicines Act 1968, and European Directive 2001/83/EC, which together govern the use of medicines in humans in the United Kingdom (47, 48). The European Directive takes precedence over the older Medicines Act, but the Medicines Act applies in areas of the law specific to the United Kingdom (47, 48).

The MHRA works in conjunction with the European Medicines Agency (EMA). The EMA evaluates medicines that are eligible for the Centralised Authorisation Procedure, which if successful will result in the medicine gaining a marketing license in all European Union countries, as well as Iceland, Lichtenstein and Norway (46, 49).

The following types of medicines for human use must be authorised through the Centralised Authorisation Procedure, rather than by national medicine regulators such as the MHRA:

 Medicines for HIV/AIDS, cancer, diabetes, neurodegenerative diseases, disorders of the immune system and diseases caused by viruses  Medicines produced via biotechnology  Advanced therapy medicines such gene therapy, somatic cell therapy and tissue engineering  Medicines that have ‘orphan medicine’ status (highly specialised medicines for rare diseases) (49). 15

Other medicines for human use may be eligible for the centralised procedure if they represent a significant therapeutic, scientific or technical advance, or if granting a centralised authorisation would benefit public health (49).

Manufacturers may also apply for the EMA’s mutual recognition procedure, through which a medicine that has been granted a marketing license in one member country is recognised by another (46, 49, 50). The EMA’s Co-ordination Group For Mutual Recognition and Decentralised Procedures- Human (CMDh) oversees this process, and if needed resolves disputes between the member countries involved (46, 49, 50).

1.2.2.4: United States

The United States Food and Drug Administration (FDA) regulates all medicines used in the United States, as well as US territories such as Guam, Puerto Rico and American Samoa (51). The FDA regulates human medicines, medical devices, blood and blood products, but also regulates veterinary products, human and food, cosmetics and radiation emitting devices. The FDA evaluates medicines for safety, quality and efficacy. The FDA both evaluates medicines before they are released on the US market and carries out post-marketing surveillance (51- 54).

The FDA operates under the Federal Food, Drug and Cosmetic Act 1938, and its amending Acts. The FDA also oversees the Best Pharmaceuticals for Children Act, and the Pediatric Research Equity Act, which aim to increase children’s access to prescription medicines (55- 59).

The United States does not currently have mutual medicines licensing treaties or joint medicines regulatory agencies with any other country.

1.2.3: Medicines Funding Systems in These Countries

1.2.3.1: New Zealand

All New Zealand residents are eligible for subsidised medicines (18). These are listed in the Pharmaceutical Schedule (18). The Schedule is a positive formulary that primarily covers pharmaceuticals used by patients living in the community, but also includes a national list of cancer medicines that may be administered in hospitals. New Zealand residents are not charged for these or other inpatient medicines, but pay a co-payment for medicines used in the community ($0 to $11 United States Dollars at 2007 exchange rates, with most patients paying $2.2). Each region manages its own budget for other inpatient hospital medicines. This means uniform nationwide access to medicines listed in the Schedule, but potential variation in access to hospital medicines (2, 18, 60-62).

The New Zealand Pharmaceutical Management Agency (PHARMAC) is a statutorily independent body that manages the Schedule, evaluates new medicines to be listed on the Schedule for clinical and cost effectiveness, makes listing decisions, negotiates with suppliers and manages a capped yearly budget for all medicines on the Schedule (62, 63). The budget is allocated to PHARMAC by District Health Boards (DHBs), and ultimately comes from central government (62, 63). Listing decisions involve advice from the Pharmacology and Therapeutics Advisory Committee (a committee of medical experts from different specialties), cost-utility analyses and the budgetary impact of the listing. PHARMAC makes listing decisions and conducts price negotiations as part of the same evaluation cycle (2, 18, 60-62).

PHARMAC uses both demand side measures (aimed at patients and physicians) and supply side measures (aimed at manufacturers) to contain costs and gain the best value for money from pharmaceutical spending (2, 18, 60-62). Demand side measures include prescriber and patient education, requirements for prior authorisation, restrictions on usage and encouraging the use of generics. Supply side measures include therapeutic and brand reference pricing, competitive bidding (including for the right to be the sole supplier of a medicine), risk sharing agreements and “bundling” agreements (where a supplier offers a lower price on one drug to get another listed). In addition to the above roles, PHARMAC also negotiates the prices of 17

many hospital medicines with manufacturers, and provides (non-binding) assessments on the cost effectiveness of some hospital medicines (2, 18, 60-62).

PHARMAC’s listing decisions are based on nine criteria set out in PHARMAC’s Operating Policies And Procedures (64), which are reproduced here:

 The health needs of all eligible people within New Zealand  The particular health needs of Māori and Pacific People  The availability and suitability of existing medicines, therapeutic medical devices and related products and related things  The clinical benefits and risks of pharmaceuticals  The cost effectiveness of meeting health needs by funding pharmaceuticals rather than using other publicly funded health and disability support services  The budgetary impact (in terms of the pharmaceutical budget and the Government’s overall health budget) of any changes to the Schedule  The direct cost to health service users  The Government’s priorities for health funding, as set out in any objectives notified by the Crown to PHARMAC, or in PHARMAC’s Funding Agreement, or elsewhere  Such other criteria as PHARMAC thinks fit. PHARMAC will carry out appropriate consultation when it intends to take any such “other criteria” into account (64)

At the time of data collection, New Zealand had three ‘exceptional circumstances’ schemes for funding medicines that are not on the Schedule or paid for in hospital (65). The Community Exceptional Circumstances Scheme pays for “rare and unusual” treatments (typically less than ten patients in the whole country). The Cancer Exceptional Circumstances scheme allows funding for rare treatments that are not in the “cancer basket” (a list of pharmaceutical cancer treatments that are publically funded for all New Zealand residents). Patients have to be individually approved for both Schemes, and the application may be subject to cost effectiveness evaluation by PHARMAC. The Hospital Exceptional Circumstances allows hospitals to fund community pharmaceuticals for discharged patients out of their own budget (65). However, having three separate Exceptional Circumstances Schemes can cause problems for patients and clinicians, and encourage shifting costs from community to hospital, or vice versa. A Government initiated review on access to high cost

and highly specialised medicines in New Zealand recommended merging the three schemes into one funding pool (66). As of March 1 2012, the three Exceptional Circumstances Schemes were closed to new applications (patients who previously had medicines funded under the Exceptional Circumstances Schemes will continue to receive them). New applicants will be assessed under a single Named Patient Pharmaceutical Assessment (Exceptional Circumstances) Scheme, which will be administered by PHARMAC (67).

1.2.3.2 Australia

All Australian residents are eligible for subsidised pharmaceuticals (including high cost medicines) under the Pharmaceutical Benefits Scheme (PBS) (18). The PBS is funded from Federal Government revenue, and uses a positive formulary that primarily subsidises community pharmaceuticals, although some hospital medicines are also subsidised under the Highly Specialised Drugs Program. Patients in private hospitals and nursing homes can also access the PBS and some public hospitals also fund outpatient and discharge medications through the PBS. The PBS subsidises approximately 75% of all prescription costs in Australia. As the PBS is centrally funded, access to PBS funded high cost medicines is uniform across Australia. (Funding for inpatient medicines in public hospitals is shared between the Federal Government and the States or Territories under separate arrangements, which can result in variations in access to medicines in hospitals). Patients pay a co-payment per prescription ($4.1 or $25.7 United States Dollars at 2007 exchange rates) (18, 68-73). New listings on the PBS (including new formulations or indications of medicines already on the PBS) are evaluated by the Pharmaceutical Benefits Advisory Committee (PBAC), an independent statutory body (18). The PBAC considers evidence from clinical trials, drug utilisation data and economic evaluation (including cost minimisation, cost effectiveness and cost utility analyses), before making a recommendation to the Minister for Health and Ageing. The PBAC can recommend that a medicine be subsidised for unrestricted use, use only under certain conditions or use only when the prescriber has obtained prior approval. Unlike PHARMAC, the PBS separates the listing recommendation from price negotiations. Another independent body, the Pharmaceutical Benefit Pricing Authority (PBPA) advises the Minister on a price for the drug and negotiates with the supplier. Methods of containing costs include brand or therapeutic reference pricing and risk sharing agreements with suppliers. Listings costing less than US$8.37 million ($10 million Australian dollars) are decided on by the

Minister, and more expensive listings by the Australian Cabinet. The Minister (or the Cabinet) cannot list a drug without a positive recommendation from the PBAC (18, 69, 70, 72-74).

Australia also has some medicine funding mechanisms outside the PBS. The Life Saving Drugs Program provides funded treatments for very rare and life threatening conditions. To be listed on the Life Saving Drugs Program, a treatment has to be shown to be necessary and effective, but have been rejected by the PBAC as not cost effective enough to list on the PBS (75). The Herceptin® Program also funds Herceptin® (trastuzumab) for late stage HER-2 positive metastatic breast cancer outside the PBS (76).

1.2.3.3 The United Kingdom

All United Kingdom residents are covered by the National Health Service (NHS), a single payer system that provides hospital and outpatient treatment along with pharmaceutical coverage. (The NHS is managed separately in England, Wales, Scotland and Northern Ireland, but all branches are part of the same health system). Manufacturers wishing to supply branded medicines to the NHS (which comprise about 80% of the NHS drug bill) enter a multi-year contract with the Department of Health under the Pharmaceutical Price Regulation Scheme (PPRS). The current PPRS is due to run until 2014. The NHS does not negotiate the prices of individual pharmaceuticals with manufacturers. Manufacturers set their own price at market launch, in return for across the board price cuts at certain times and caps on overall profit (77-79).

The NHS does not have a set positive formulary of medicines that it pays for. Instead, the NHS uses a small negative list of medicines (mainly over the counter and ‘life-style’ products) that are not to be subsidised. In theory, any medicine marketed in the UK by a manufacturer who has signed up to the PPRS must be paid for by the NHS, with non-exempt patients paying $13.7 United States Dollars at 2007 exchange rates. In practice, access to high cost medicines (HCMs) is limited by the funding available. Funding for all pharmaceuticals is controlled by regional organisations known as Primary Care Trusts (PCTs), which decide on health service funding priorities in the area. This leads to some high cost medicines being

funded in one region but not in another, a phenomenon known as ‘post-code prescribing’, and can cause legal as well as ethical problems (18, 80).

The National Institute for Health and Clinical Excellence (NICE) is charged with ensuring patients have equitable access to medicines while also making sure expensive medicines do not consume NHS funding that could be better used for other health services. NICE does this by evaluating medicines (and other health technologies) that are controversial, have a major health impact or a major budgetary impact for the NHS. (This means that unlike PHARMAC or the PBS, NICE does not consider every pharmaceutical that is eligible for subsidy. A pharmaceutical that is not cost cost-effective may still be subsidised because it has not been considered by NICE). The evaluation includes appraisal of clinical evidence and economic evaluation using cost-utility analysis. PCTs are legally obliged to fund all medicine indications that have a positive evaluation from NICE. The situation following a negative evaluation is not defined in law. (PCTs can provide funding for these indications on an individual basis. This is quite unlike the situation with PHARMAC or the PBS, where a negative evaluation will always result in the denial of public subsidy). NICE sometimes finds itself being criticised both for denying patients access to expensive medicines and for approving expensive medicines that are unaffordable for the NHS (18, 81-83).

1.2.3.4: The United States

1.2.3.4.1: The General Situation

The United States is unique among these four countries in that it has neither universal pharmaceutical coverage nor nationwide pharmaceutical price regulation. Pharmaceuticals are covered through health insurance plans. Americans can be covered by private insurance, government insurance programs, or both (25).

The level of pharmaceutical coverage is more variable in the US than in other countries. There were 50.7 million people (16.7% of all Americans) who had no health insurance coverage at any time in 2009, and would therefore have to pay for medicines out of pocket if they could afford it, or rely on charity care (25). A significant number of Americans are also underinsured (estimated at 25 million adults aged 19-64 years in 2007), and as a result would 21

have access problems and/or high costs (84). It has been estimated that over 60% of bankruptcies in the US were related to medical problems, and most of the medically bankrupt were in middle class occupations, had good education, owned their own homes, and had health insurance (25, 84, 85).

In 2009, 194.5 million Americans (63.9%) had private health insurance coverage, of which 169.7 million (55.8% of all Americans) had employment based insurance (25). Employers choose the insurance provider, and negotiate coverage for their employees. Employers typically subsidise the premiums of employees and their dependents as part of employee remuneration, and both the employer and employee get tax benefits from this (25, 86).

Many employees do not qualify for employment based insurance coverage, including part time workers, those on temporary contracts and those working for small firms (87). Employees can also find themselves ‘job locked’, as changing employers can result in pre- existing conditions being denied coverage or subject to a waiting period before coverage resumes (87). Last but not least, employment based coverage can be lost if the employee is made redundant or becomes too ill to work (87). A Federal law called COBRA (Consolidated Omnibus Budget Reconciliation Act) lets many employees continue coverage after losing employment based coverage. However, employees must pay the full premiums without employer subsidy (which averaged over $12,000 per year for a family policy), which can be unaffordable after a job loss. In addition, coverage under COBRA has to begin within 60 days of losing previous coverage, and is normally for a maximum of 18 months (87).

The public safety net in the US is limited. Government insurance programs covered 93.2 million Americans (30.6%) in 2009 (25). (The percentages of people with private insurance, government insurance and no insurance do not add up to 100 because people can have both private and government insurance). Of those with Government insurance, 43.4 million (14.3% of Americans) were covered by Medicare, a Federal Government program which covers people 65 years and above, those with permanent disabilities and end-stage renal dialysis patients (25). Medicare Part D covers pharmaceuticals, and patients can chose from a wide range of pharmaceutical plans (25).

In 2009, 47.8 million people (15.7% of Americans) were covered by Medicaid, a State run program that covers certain groups of people with low incomes and few assets (25). Medicaid 22

eligibility rules vary by State, and coverage is often limited to groups such as children, parents with dependent children, the elderly and the disabled. This can leave single adults and childless couples without coverage even if they have no income or assets. Others find their income is too high to qualify for Medicaid, but not enough to purchase individual insurance. Medicaid pharmaceutical coverage varies from State to State, as States have the power to set their own Medicaid formularies. Medicaid pharmaceutical co-payments also vary by State (25, 87).

1.2.3.4.2: The Department of Veterans Affairs National Formulary (VANF)

The Department of Veterans Affairs (VA) health system is arguably unique in the United States in that it is a nationwide, publically funded health system with a single pharmaceutical coverage plan (88). (As such, it shares many similarities with the health systems of New Zealand, Australia and the United Kingdom). It is an integrated network of primary and secondary care that eligible patients can access at any VA health facility in the US, (Non- exempt patients paid a co-payment of $8 United States Dollars in 2007). The VA health system has been praised both for quality of care and containing pharmaceutical costs (19, 20, 88-90).

The VA health system is not intended to be representative of the large and diverse United States health system. Rather, the VA health system was chosen for this study because it is a large US health system that shares many similarities with the other comparators, and thus provides a starting point for international comparison. Researchers have also in the past looked to the VA for lessons on formulary management and comparative effectiveness that can be applied elsewhere in the US health system(19, 91).

The suitability of the VANF as a comparator to the other systems examined in this thesis (especially in light of the fact that it serves a highly restricted patient population) has been defended by publication (92). However, in light of the differences between the United States and the other countries being compared in this thesis, all table entries regarding the VANF are italicized to remind the reader of these differences.

All veterans of the United States Army, Navy, Air Force, Marine Corps and Coast Guard (except those who were dishonorably discharged) are potentially eligible for VA health benefits (93). The number of veterans who can be enrolled at any one time is limited by the funding available, with priority given to veterans with service related conditions, high levels of disability, and limited incomes. Members of the Reserves and National Guard, those who provide military related services, and veterans from countries allied with the United States in some conflicts may be eligible. Spouses and dependents of some veterans may also be eligible for treatment at VA facilities (93, 94).

The VANF is a positive formulary used throughout the VA health system, and all agents listed on the VANF must be available to patients at any VA facility (88). In addition, each VA facility has a non-formulary procedure, by which patients can access non-formulary medicines if formulary medicines are not suitable. The formulary is jointly managed by the VA Pharmacy Benefits Management Strategic Healthcare Group (VA PBM), the Medical Advisory Panel (MAP, a committee of VA and Department of Defense physicians), and the formulary leaders of the regional Veterans Integrated Service Networks (VISNs). Each VISN manages its own pharmacy budget. The MAP and the VISN formulary leaders jointly decide on changes to the VANF, with VA PBM providing information on drug utilisation, cost, safety and efficacy. Cost minimisation and cost effectiveness analyses are used in decision- making. VA PBM, in conjunction with the VA’s National Acquisition Center, negotiates prices of agents on the VANF with manufacturers. Price management techniques include competitive bidding, the use of closed classes (where the VA agrees to only use certain agents in a drug class) and preferred classes (where favorable contracts exist for one or more agents in a class and physicians are encouraged to prescribe these), the use of generics, and use of Federal Supply Schedule prices (which by law are limited to 76% of non-Federal prices) (19, 20, 88, 95).

1.2.4: Second Scope Limitation- Choice of Comparator Systems

This thesis focuses on New Zealand, and compares New Zealand with other countries in order to evaluate New Zealanders’ access to licensed and subsidised medicines relative to residents in those countries. Therefore, the comparator funding systems need to be as similar to that in New Zealand as possible.

As noted in Section 1.2.3.1, New Zealand uses a publically funded single payer system that provides pharmaceutical coverage to its population (18). Therefore, insurance-based systems such as private insurance coverage in the United States (which can deny coverage to some people based on risk, and charge different amounts for coverage) would not be suitable comparators.

The New Zealand system also uses a single nationwide formulary for pharmaceutical coverage (18). Therefore, systems that offered patients the choice of multiple coverage options (such as Medicare Part D in the United States) or had regional formularies (such as Medicaid in the United States, where coverage varies from one state to another) would not be suitable comparators.

Lastly, New Zealand has a central decision making body which decides when high cost pharmaceuticals will be publically funded (18). Therefore, each comparator system had to have one (but only one) central body that makes decisions on high cost drugs. Therefore, the National Health Service in Scotland (NHS Scotland) would not be a suitable comparator as decisions on high cost medicines can be made both by NICE and the Scottish Medicines Consortium (96, 97).

With these criteria in mind, the Pharmaceutical Benefits Scheme (PBS) in Australia, the National Health Service (NHS) for England and Wales in the United Kingdom, and the Department of Veterans Affairs National Formulary (VANF) in the United States were chosen as the comparator systems to New Zealand’s PHARMAC. As with the choice of comparator countries, the choice of comparator systems has been validated by peer-reviewed publication (37).

It is worth noting that each of the systems has some unique features that are not shared by the others. For example, PHARMAC in New Zealand has a capped budget, while the PBS in Australia does not. PHARMAC and the PBS consider individually consider each agent that is eligible for subsidy, while NICE in the United Kingdom does not. Last but not least, the fact some systems (such as PHARMAC and the PBS) must conduct price negotiations while NICE does not may affect the timeline before an agent is subsidised (18). Similarly, the VANF differs from the other systems in that it serves a very specific patient population, and is 25

not representative of the overall health system in its country (88).

However, such differences are not a reason to exclude systems from international comparisons, as long as sufficient similarities exist in order to compare like with like. The choice of these particular systems for comparison has been defended by publication (92).

The key features of each system (showing both similarities and differences) are shown in Table 1 below.

System VANF NHS PBS PHARMAC

Covers all residents of No[1] Yes Yes Yes nation Type of formulary Positive list Negative list Positive list Positive list

Pharmacoeconomic analysis CMA, CEA CUA [3] CMA, CEA, CUA [2] CUA Decisions on which At national At local At national At national treatments to fund level level [4] level level Price negotiations with Yes No [5] Yes Yes suppliers Capped national No No No Yes pharmaceutical budget 2007 Prescription co- $8 $0 or $13.7 $4.1 or $25.7 $0 to $11 payments [6] [7] [8] (most paid $2.2) 2007 yearly co-payment $960 (most None [9] $229.6 or 20 co- limits patient $886.1 [8] payments per groups) family TABLE 1: The key features of each system (VANF, NHS, PBS and PHARMAC) (Table taken from Ragupathy et al., A 3-dimensional view of access to licensed and subsidized medicines under single payer systems in the United States, United Kingdom, Australia and New Zealand. Pharmacoeconomics 2012).

Table 1 Notes:

[1] Covers veterans of the US armed forces, and in some cases their dependents, members of the Reserves and National Guard, and veterans of countries allied with the US in some conflicts. Enrollment is limited, and priority is given to veterans with service related disabilities and limited incomes.

[2] Used as part of formulary management guidelines.

[3] The National Institute For Health and Clinical Excellence (NICE) only evaluates selected treatments, typically those that are controversial, high cost or have the potential for a major impact on the NHS budget.

[4] All treatments that have had a positive evaluation from NICE have to be funded across the country.

[5] Suppliers set their own prices at market launch of the drug, in exchange for caps on maximum profit and across the board price cuts on all their products at agreed times.

[6] In United States Dollars at 2007 exchange rates. VANF co-payments are per each 30-day supply of medicines, all others are per prescription. All four systems use income based co-payments or co-payment exemptions (or both) to limit the cost to patients.

[7] English patients paid $13.7 per prescription, but up to 85% of prescriptions are exempt. Wales abolished prescription co-payments on April 1st 2007.

[8] Patients with limited incomes paid $4.1 (safety net of $229.6, after which co- payments were waived) and all other patients paid US$25.7 (safety net of US$886.1, after which patients paid $4.1).

[9] Patients can purchase three monthly or twelve monthly “prescription pre-payment certificates”, which cover all co-payments in that period for a fixed price, but the onus is on the patient to do this.

1.3: Policy Issues of Concern in These Four Countries

Some groups of patients face additional barriers to access, beyond that of the population as a whole. Policy makers in these four countries have identified specific access problems for children, patients who need highly specialised medicines or medicines for extremely rare conditions (sometimes called ‘orphan drugs’) and patients who need high cost medicines, and have developed policies to address these issues (57, 59, 66, 98-101). Access problems for each group are briefly described below

1.3.1: Access Problems for Children

In many countries (including the United States, United Kingdom, Australia and New Zealand), many drugs that are licensed for adults are not licensed for children (102-106). This is because the necessary studies (such as pharmacokinetic studies, dose ranging studies and randomised controlled trials) that determine the safety, efficacy, dosage and adverse effects in children have not been conducted. This in turn results in inequitable access to medicines for children. It has previously been shown that the number and proportion of medicines listed in standard prescribing texts which had prescribing information for children decreased over the decade from 1998 to 2007 (102-106).

Medical practitioners and other prescribers are therefore often forced to prescribe for children without access to full prescribing information. This is described as ‘‘unlicensed’’ use (use of a medicine that has not been approved for use in humans by the regulatory authority in that country), or ‘‘off-label’’ use (use of a medicine in a way that differs from the conditions under which it has been licensed for use, for example by a different route or for a different indication than that for which it was licensed (107). A survey of United Kingdom paediatricians in 2003 and 2004 showed that the majority (55%) believed having to use medicines that had not been licensed in children potentially disadvantaged children. A large majority (70%) had concerns about the safety of using medicines in this manner and just under half (47%) had concerns about efficacy (108).

The use of unlicensed and off-label medicines in children may increase the risks of adverse drug reactions. A study of off-label drug use in children under 16 years of age in the community (carried out in France between 2000 and 2001) showed a statistically significant relationship between off-label use and the risk of an adverse drug reaction (ADR), with a relative risk of 3.44 (95% CI 1.26–9.38) (109).

Access can also be impaired when dosage forms suitable for children are not available, especially in the case of younger children. Suitable forms could be defined as those that enable medicine to be taken by a child, improve compliance and enable individualised dosing (110, 111). Lack of suitable dose forms may therefore lead to dosing errors and potentially prevent the child from being able to take the medicine.

In addition, the cost of the medicine may be a further barrier to access. This can depend on whether the cost is subsidised i.e. funded (partly or fully) by the Government or other third party payer. In the United States alone, 21 million children relied on public coverage for health insurance in 2005, and this increased to 35 million children in 2008 (112, 113). However, 11% of children in the US had no health insurance coverage at any time in 2007 (113). Public coverage for children in the US could expand further under the Patient Protection and Affordable Care Act, which extends Medicaid eligibility for individuals and families with an income of 133% of the Federal poverty level or below (114). Public coverage is the main means by which medicines are subsidised for children in the New Zealand, Australia and the United Kingdom (18).

Both the United States and European Union have passed legislation to increase children’s access to medicines (57, 59, 98). The legislation provides a six-month patent extension for medicines that are still on patent if the manufacturers carry out testing in children (and in the case of the European legislation, 10 years of data exclusivity for off-patent medicines) (57, 98). The legislation also obliges manufacturers to carry out testing in children if the medicine is potentially useful in children (59, 98). However, patent extensions may delay the entry of generic competitors on to the market, and therefore increase pharmaceutical purchasing costs (115). Australia and New Zealand do not have any comparable legislation.

1.3.2: Orphan Drugs

Orphan drugs are highly specialised medicines used to treat rare disorders. The FDA, EMA and TGA have all issued requirements that a medicine must meet in order to be considered an orphan drug (99-101, 116, 117). New Zealand’s Medsafe is also developing orphan drug regulations (118). While the exact requirements for an orphan drug designation vary from one jurisdiction to another, the requirements all include that the drug should only be indicated for a small patient population, and/or it should not be commercially viable for manufacturers to bring on to the market under normal circumstances (99-101, 116-118).

Gaining orphan drug designation gives manufacturers financial help in developing and marketing the product, which again varies by jurisdiction. This includes the waiving of product registration fees (Australia), and grants to help with clinical trials and product developments plus tax credits (United States) (116, 117). The European Union gives registration fee reductions, help with product development, eligibility for the centralised authorisation procedure (see Section 1.2.2.3), and ten years’ market exclusivity. Individual member states also provide additional incentives (119, 120).

1.3.3: High Cost Medicines

Many if not all high-income countries struggle with how to pay for high cost medicines (HCMs) such as biological anticancer treatments, tumour necrosis factor (TNF) inhibitors for rheumatoid arthritis and enzyme replacement therapies. For example, the annual cost of new anticancer treatments can approach or exceed the median annual family income even in relatively wealthy countries such as the United States (121, 122).

Deciding which HCMs should be funded poses unique challenges for third party payers. Many HCMs are licensed and marketed as a significant improvement over existing therapies, and in some cases may be the only treatment option available. They hold the promise of reduced mortality, extended survival times and improved quality of life. However, the cost effectiveness of HCMs can be poor, in some cases hundreds of thousands of US Dollars per Quality Adjusted Life Year (QALY) gained (122).

Assessing these HCMs purely on cost effectiveness grounds would see vulnerable and seriously ill patients left with few if any treatment options. On the other hand, funding such HCMs without adequately considering cost effectiveness would consume a disproportionate share of pharmaceutical budgets. This would create large opportunity costs, as more cost effective treatments that would benefit large numbers of patients would have to be foregone. Third party payers must therefore balance maximising the benefits of pharmaceutical spending with fairness for vulnerable patients, and take into account society’s criteria about which patient groups should be given priority. They must also make these decisions in a way that is procedurally ‘fair’. This includes having adequate transparency, considering relevant criteria in their decision-making, having mechanisms in place for decisions to be reviewed, and having procedures for making sure the first three criteria are met. Meeting all four of these criteria can be very difficult (123-125).

The process can also be complicated when there is public or political pressure to fund one HCM ahead of others. Media coverage of the controversy is much more likely to focus on the chosen medicine’s claimed effectiveness, its status as a miracle drug or it being the difference between life and death than on doubts about its effectiveness or concerns about potential harms. This can lead to other patient groups missing out, including those who need HCMs that do not receive the same publicity (126).

1.4: Licensing Process in New Zealand, Australia the United Kingdom and United States

In order to be licensed for use in humans in any of these countries, pharmaceuticals must be evaluated and approved by the regulatory agency responsible for that country (see Section 1.2.2). Manufacturers must prove to the regulatory agency that the pharmaceutical is safe, effective and has been manufactured to defined quality standards (117, 127-129).

However, the regulatory agencies do not consider cost, cost effectiveness or the impact on the country’s pharmaceutical budget when making licensing decisions, and do not make decisions on whether medicines are publically funded (38, 130-132). (See Section 1.5).

Manufacturers prove the safety, quality and efficacy of a new drug through a dossier of evidence submitted to the regulatory agency. This includes data from in-vitro studies, animal testing and clinical trials in humans. While the data required by the regulatory agencies in the four countries is similar, the precise requirements are governed by each country’s laws. Manufacturers must prepare separate dossiers that meet each regulator’s requirements (133- 137).

1.4.1: The Drug Development Process

Compounds can have severe adverse effects in humans that were not seen in animal testing. An example is the monoclonal antibody TGN1412, which provoked a severe immune reaction (a ‘cytokine storm’) and multi-organ failure in humans that was not evident in trials on monkeys. For this reason, clinical trials are carried in several phases to establish safety and efficacy in humans (138).

1.4.1.1: Conduct of Clinical Trials

There are international agreements in place to guarantee the validity of trials, and to safeguard trial participants. The International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) is a collaborative effort by regulatory agencies in the US, EU and Japan to standardise trial requirements, reduce the need for redundant trials to gain licensing in different countries, and protect trial participants (139).

The Nuremberg Code (developed as a result of the trials of medical personnel who conducted unethical and often deadly experiments on humans under the Nazis) and the World Medical Association Declaration of Helsinki set out the principles for the protection of human subjects in medical experiments, including clinical trials (140, 141).

1.4.1.2: Clinical Trial Phases

Clinical trials are carried out in phases, with each stage progressively involving larger numbers of participants. This reduces the risk to participants, but also gives the drug developer the option of halting testing if the early stages show a drug is not a promising candidate for further development.

Phase I (or Phase One) studies are intended to test whether a drug is safe in humans (the ‘cytokine storm’ with TGN1412 occurred during a Phase I study). They also provide information on the drug’s pharmacokinetics and pharmacological activity in humans, and allow investigators to establish initial doses, and determine the best route of administration. The doses used are typically much smaller than those estimated to be safe from animal and preclinical studies (129, 138, 142).

Phase I studies typically involve less than 100 participants, and these are usually healthy volunteers. (Some particularly toxic drugs undergo Phase I testing in those who have the disease the drug is intended to treat). Because of the potential risks involved in giving a drug to humans for the first time, Phase I studies are typically carried out in dedicated trial units by investigators who specialise in such studies (129, 138, 142).

Phase II (or Phase Two) trials are intended to establish efficacy as well as provide further safety information to that gained in Phase I trials. They typically involve several hundred participants who have the disease the drug is intended to treat. The trials are carried out by investigators who are specialists in treating that disease (129, 138, 142).

Phase II trials typically involve a control group, a group of participants who are as similar to the participants who receive the trial drug as possible. Assignment is typically random (i.e. participants are assigned to the trial group or control group by chance rather than by the investigators). These trials are usually blinded (i.e. the patients and/or the investigators do not know the group assignments in order to minimise bias) (129, 138, 142).

Phase III (or Phase Three) trials are typically the last stage of trial carried out before information is submitted to regulatory agencies. Phase III studies can include thousands of

participants, often in multiple trial centers and across several countries. Because of the cost of investing in such large studies, manufacturers only carry out such studies if the drug showed good evidence of safety and efficacy in Phase II studies (129, 138, 142).

Phase III studies are carried out in participants who have the disease the drug is intended to treat. Phase III studies provide further information on the clinical benefits of the drug, and allow investigators to determine if these benefits outweigh the risks. (The larger size of these trials means that rare adverse events are more likely to be detected than in Phase II trials). Phase III trials also provide information on interactions and whether use of the drug should be restricted in some groups (for example those with poor renal or liver function). This is used in producing the application submitted to the regulatory agencies (129, 138, 142).

Phase IV (or Phase Four) studies are carried out after the drug has been licensed for a particular use. This can include examining whether the risks and benefits of the drug as it is used in clinical practice differ from that in trials. (Extremely rare adverse events that were not detected in Phase III trials may also be detected in Phase IV studies). These studies can also compare the benefits of the drug against other commonly used treatments for that disease (i.e. is it better than its competitors are?). Such information can be used to make the case for drug to be publically funded (see Section 1.5) (129, 138, 142).

It is important to note that trials carried out to test whether a licensed drug has uses other than the one it is licensed for, can be used in a different population (e.g. children rather than adults) or via a different route are not Phase IV studies. These new uses must go through Phase I, II and III studies, and must be submitted to the regulatory agency as new applications (129, 138, 142).

1.4.2: Regulation After Licensing

Regulatory agencies continue overseeing drugs after they have been licensed in a given country. This oversight includes post marketing surveillance, control of drug advertising, regulation of manufacturing and distribution, and controls on sale.

1.4.2.1: Post Marketing Surveillance

Post marketing surveillance is ongoing safety monitoring of a drug once it has been launched into the market. Such surveillance can pick up rare adverse effects that were not detected in clinical trials, or adverse effects that only occurred after prolonged exposure.

Each regulator has its own requirements for post marketing surveillance. Manufacturers are generally required to report serious adverse effects or safety concerns they become aware of to the regulatory agency. This can include information from post marketing studies (Phase IV studies), reports collected from health professionals and the patients (such as the United Kingdom’s Yellow Card programme), and information from medicine safety monitoring programmes.

Information collected from post-marketing surveillance can be used to educate health professionals and the public about newly discovered risks from a drug. The regulatory agency may also require the manufacturer to add new information to the drug’s official product information, or restrict the use of the drug in some groups of people. In some cases, the regulatory agency may withdraw the drug’s license altogether (129, 143-146).

1.4.2.2: Control of Advertising

Regulatory agencies oversee drug advertising in order to ensure that drugs are only promoted for the uses for which they have been licensed, claims about the drug’s effectiveness are not being exaggerated, and the information about the drug’s risks matches that in the drug’s official licensing. This includes advertising through print, radio and television, but also face- to-face promotion of a drug to health professionals and patient groups (147-150).

1.4.2.3: Regulation of Manufacturing

Regulatory agencies oversee the manufacture of licensed medicines to ensure that manufacturers consistently manufacture licensed medicines to set standards of safely, efficacy and quality (i.e. that the product which marketed to the public is of the same standard that was

approved by the regulatory agency). This is done through codes of Good Manufacturing Practice (GMP), which are systems of practice manufacturers must follow at all stages of the manufacturing process (151-155).

1.4.2.4: Distribution

Regulatory agencies oversee the distribution and supply chain for drugs, in order to ensure that drugs which have been licensed and manufactured according to GMP reach the point of sale (such as a pharmacy) in good condition. This includes setting standards for sites involved in distribution, storage and transport of medicines, the training and qualifications of the personnel involved, and the records that must be kept (156-159).

1.4.2.5: Classification of Medicines

Regulatory agencies in each of these countries play a role in the legal classification of drugs. The regulatory agency may make the classification decision itself, or provide expert advice to those who do such as Ministers and their delegates. A drug’s classification determines the controls that will be put on its sale (for example, whether it can only be sold with a prescription, whether it needs to be sold from a pharmacy or whether it can be sold from any retail outlet) (160-163).

1.5: Evaluation For Public Funding

1.5.1: Why Evaluate Pharmaceuticals For Public Funding?

Pharmaceutical spending is a significant component of total health spending in each of these four countries, ranging from 12.1 % in the United States and 12.3% in the United Kingdom to 14.3% in Australia in 2007, with no data available in New Zealand (164). In 2009, this percentage was 9.3% in New Zealand and 12.0% in the United States, with no data for the United Kingdom and Australia (164).

Publically funded systems (such as the four being compared in this thesis) must therefore decide which pharmaceuticals provide the best ‘value for money’ and under which conditions. Such an evaluation is sometimes called the ‘fourth hurdle’ to marketing a drug, after the three hurdles (safety, efficacy and quality) that must be passed to get a drug approved by the regulatory agencies (21, 22).

As noted in Section 1.3.3, the funding agencies must also consider people who need very high cost medicines that have low cost effectiveness, incorporate society’s values in relation to the allocation of scarce resources, and must make decisions in a way that are procedurally fair (123-125).

This section describes the methodologies that funding agencies use when they make pharmaceutical funding decisions. The remit and powers of each individual agency compared in this thesis have previously been described in Sections 1.2.3.1, 1.2.3.2, 1.2.3.3 and 1.2.3.4.2.

1.5.2: Economic Evaluation

An economic evaluation is a way of calculating the costs and outcomes of two (or more) courses of action. For example, the costs and consequences of paying for drug X versus drug Y, paying for drug X versus doing nothing, and paying for drug X instead of surgery can all be calculated if information on costs and outcomes is available. Economic evaluations can be based on data derived from clinical trials, or through modeling and simulations (165).

It should be noted that while all four of the systems compared use economic evaluations in their decision making, none of them rely solely on the results of economic evaluation to make decision (see Sections 1.2.3.1, 1.2.3.2, 1.2.3.3 and 1.2.3.4.2). For example, PHARMAC has to work within the constraints of a capped yearly budget, while NICE does not (18). For this reason, a drug may be rejected on cost effectiveness grounds by one agency while being accepted by another, even when the same data set is submitted.

There are four main types of economic evaluation, which are described below.

1.5.2.1: Cost Minimisation Analysis (CMA)

A cost minimisation analysis (CMA) can be used where the two options have been shown to have little or no clinically significant difference (for example comparing a patented medicine with a licensed generic competitor, which during the licensing process would have been shown to be no worse in safety and efficacy than the originator). In this case, the superior comparator will always be the one with the lower cost (165).

1.5.2.2: Cost Effectiveness Analysis (CEA)

A cost effectiveness analysis (CEA) compares the cost per given health outcome for each of the comparators. (For example, the cost per successful renal transplant, or the cost per myocardial infarction avoided). The superior competitor will be the one with the lower cost per outcome. However, CEA cannot be used for comparing treatments that have different types of health outcomes. For example, it cannot be used for comparing a drug for preventing renal transplant rejection with a drug for preventing myocardial infarctions, as CEA has no way of comparing the value of a myocardial infarction avoided versus that of a successful renal transplant (165).

1.5.2.3: Cost Utility Analysis (CUA)

A cost utility analysis (CUA) compares the cost per quality-adjusted life year (QALY) gained for each of the comparators. QALYs are derived by multiplying the number of life years gained by the quality of life in those years. The quality of life in any given health state is measured by the preference people have for living in that health state, ranging from zero (death) to one (perfect health). The advantage of CUA is that the outcome of any treatment can be measured in QALYs. This means that CUA can be used to compare drugs for very different uses, such as preventing myocardial infarctions and preventing renal transplant rejection (165).

1.5.2.4: Cost Benefit Analysis (CBA)

A cost benefit analysis (CBA) converts the value of treatment outcomes into monetary units (for example, the monetary value of a renal transplant rejection avoided). As the outcome of any treatment can theoretically be converted into monetary outcomes, CBA can be used for comparing different types of treatment outcomes just as CUA can. However, converting health outcomes into monetary values can potentially be controversial (165).

1.5.3: Social Weighting And Ethical Theories

Decisions about allocating scarce resources (such as funding for pharmaceuticals) may also be influenced by society’s preferences about how those resources should be used. It has been shown that funding decisions which incorporate social preferences (such as reducing inequity) will produce different outcomes to those which are based on economic evaluation alone (166).

These outcomes will also be affected by which theories are used. For example, the ethical theory of utilitarianism (which seeks to maximise the benefits to society, regardless of who those benefits accrue to) will produce different results from the theory of Rawlsianism (which seeks to minimise inequities by favouring patient groups with poorer health) (167). Other treatments might be favored when decision makers apply the concept of a ‘fair innings’ (giving priority to people who would otherwise be denied their ‘natural’ lifespan) or the ‘rule of rescue’ (assisting those who are identified as being at particular risk, especially the risk of death) (167, 168). The combination of ethical values that decision makers may apply is potentially infinite, and these values may either be explicitly stated in their decision criteria or be implicit.

It should be noted that these ethical theories are different from the concept of ‘fairness’ in centralised decision making described by Mitton et al. (125). Mitton et al. describe ‘procedural fairness’ in decision making about pharmaceuticals, which includes having relevant decision making criteria, having transparency in decision making, having review mechanisms for decisions, and having the means to ensure decision makers follow the procedures they have set for themselves (125). Such a process allows a decision to made ‘in the right way’, but different ethical theories will produce different views of what the ‘right

decision’ is. For example, application of the ‘rule of rescue’ by policy makers to the funding of trastuzumab as adjuvant therapy in HER-2 positive breast cancer in Australia may have contributed to a positive funding decision (169). Conversely, PHARMAC’s decision to fund a nine-week course of trastuzumab in early stage HER-2 positive breast cancer rather than 52 weeks has been defended from a utilitarian perspective. In doing so, the author also acknowledges concerns about fairness for this vulnerable patient group (a Rawlsianism perspective) (123).

1.6: The Controversy In New Zealand

New Zealand has been held up by some authors as an example of how to contain pharmaceutical expenditure while expanding patient access to medicines (2, 61). All New Zealand residents are eligible for one hundred percent publically funded in-patient medicines and heavily subsidised outpatient pharmaceuticals. The number of subsidised medicines and the volume of dispensed prescriptions have increased over the past decade, while patient co- payments have fallen (2). Despite this, New Zealand’s per capita pharmaceutical spending was the fifth lowest in the OECD. New Zealand spent 9.4% of its health expenditure on pharmaceuticals, the fourth lowest proportion in the OECD, and well below the OECD average of 17% (2008 data). However, it should be noted that Australia, the United Kingdom and the United States also spent less than of 17% of total health expenditure on pharmaceuticals (170).

However, New Zealand’s medicines system has faced several controversies in recent years, largely centering on the role of the New Zealand Pharmaceutical Management Agency (PHARMAC). These have included concerns over access to widely used agents such as statins, lack of funding for rare but expensive treatments such trastuzumab (Herceptin®) in early stage HER-2 positive breast cancer, and the quality of medicines used in New Zealand (3, 4, 171-174).

PHARMAC manages a yearly budget for buying pharmaceuticals (allocated to it by the DHBs), and by law can never exceed this budget (2, 61, 62). PHARMAC is charged with securing the best possible health outcomes for New Zealanders from this capped budget. It does this through the Pharmaceutical Schedule, a positive formulary that lists all publically

funded outpatient medicines and all medicines in the cancer basket. It evaluates new medicines to be listed on the Schedule for clinical and cost effectiveness, negotiates the price with the manufacturer, and makes the final listing decisions. PHARMAC effectively has a monopsony position in the New Zealand market, giving it a great deal of advantage in negotiations with suppliers. Methods of cost containment include therapeutic and brand reference pricing, competitive bidding (including for the right to be the sole supplier of a medicine), risk sharing agreements, and cross-sector agreements (where a supplier offers a lower price on one drug to get another listed) (2, 61, 62).

PHARMAC’s place in the New Zealand medicines system is described in Section 1.2.3.1. In addition to these roles, PHARMAC negotiates the prices of many in-patient medicines on behalf of District Health Boards (DHBs), but does not currently play any role in managing their use (175, 176). The New Zealand Government has announced plans to gradually expand PHARMAC’s role in managing in-patient medicines. PHARMAC will eventually take over the purchasing and management of all hospital medicines. PHARMAC will also gradually take over the purchasing of medical devices such as insulin pumps (177). This potentially represents the biggest change to New Zealand’s medicines system since PHARMAC was established in 1993. As such, it is an opportune time to examine how the current system serves New Zealanders, and the potential impact of any changes.

1.7: Aims, Hypotheses, and Final Scope Limitation

1.7.1: Aims

The aim of this thesis is to compare “access” to pharmaceuticals licensed in each of the four countries and subsidised under each of the four single payer systems. As such, it aims to compare what pharmaceuticals have been licensed to be on the market of each country, and what has been approved for funding by the selected single payer system.

1.7.2: Hypotheses

A number of initial hypothesis were developed to examine the aims. These were further developed into testable hypotheses that could be rejected or confirmed.

1.7.2.1: Initial hypotheses

1 There are differences in the number and types of prescription medicines licensed for adults in New Zealand, Australia, the United Kingdom and the United States in 2007, and those subsidised for adults in New Zealand, Australia, the United Kingdom and the Department of Veterans Affairs National Formulary of the United States in 2007.

2 There are differences in the number and types of prescription medicines licensed for children in New Zealand, Australia, the United Kingdom and the United States in 2007, and those subsidised for children in New Zealand, Australia, the United Kingdom and the Department of Veterans Affairs National Formulary of the United States in 2007.

3 Children have less access than adults do to prescription medicines (in the form of medicines that are licensed for children, in formulations suitable for children, and subsidised for children) in their respective jurisdictions.

4 Any lack of access to prescription medicines for adults or children in New Zealand will be of concern to those with key knowledge of the New Zealand pharmaceutical system.

1.7.2.2: Testable Hypotheses

1.1 New Zealand licensed fewer medicines, fewer types of medicines, older medicines (greater time from earliest registration date) and less innovative medicines for adults than Australia, the United Kingdom and the United States in 2007.

1.2 New Zealand subsidised fewer medicines, fewer types of medicines, older medicines (greater time from earliest registration date) and less innovative medicines for adults than Australia, the United Kingdom and the United States Department of Veterans Affairs National Formulary in 2007.

2.1 New Zealand licensed fewer medicines, fewer types of medicines, older medicines (greater time from earliest registration date) and fewer medicines in formulations suitable for 43

children for children than Australia, the United Kingdom and the United States in 2007.

2.2 New Zealand subsidised fewer medicines, fewer types of medicines, older medicines (greater time from earliest registration date) and fewer medicines in formulations suitable for children for children than Australia, the United Kingdom and the United States Department of Veterans Affairs National Formulary in 2007.

3.1 There are fewer medicines, fewer types of medicines and older (greater time from earliest registration date) medicines licensed for children in New Zealand, Australia, the United Kingdom and the United States than those licensed for adults in their respective jurisdictions in 2007.

3.2 There are fewer medicines, fewer types of medicines and older (greater time from earliest registration date) medicines subsidised for children in New Zealand, Australia, the United Kingdom and the United States Department of Veterans Affairs National Formulary than those subsidised for adults in their respective jurisdictions in 2007.

4.1 Any deficiencies in access (number, time, type, comparative age (time from earliest registration date) and level of innovation) to prescription medicines for adults in New Zealand (compared with Australia, the United Kingdom, and the United States) will be of concern to policy makers, researchers, and others with detailed knowledge of the New Zealand pharmaceutical system.

4.2 Any deficiencies in access (number, time, type, comparative age (time from earliest registration date) to prescription medicines for children in New Zealand (compared with Australia, the United Kingdom, and the United States) will be of concern to policy makers, researchers, and others with detailed knowledge of the New Zealand pharmaceutical system.

1.7.3: Final Scope Limitations

This thesis does not compare restrictions on advertising, differences in medicine classification or distribution of medicines. It also does not compare the regulators’ effectiveness in overseeing manufacturing or monitoring the safety of medicines on the market.

It does not attempt to evaluate all aspects of access to medicines. In particular, it does not evaluate depth of access, such as the restrictions that apply to subsidised medicines. This in part because restrictions on medicines subsidised in New Zealand in 2007 have been examined in previous work (178).

It does not evaluate the quality of the decisions made by either the regulatory agencies in making licensing decisions, or the funding agencies in making funding decisions.

Rather, it focuses on the result of these decisions, in terms of the medicines available for adults and children. This will be done through two quantitative investigations and one qualitative investigation.

CHAPTER TWO: METHODS

2.1: Literature Searching

The first step in planning the search was deciding what questions needed to be answered by the search. The initial questions for the search were:

 What does “access” mean in the context of access to medicines?  What factors determine access?  How is access measured?  Do the same factors determine access in every country, or does this vary from country to country?  Have previous studies measured access in New Zealand?  If previous studies have measured access in New Zealand, are there gaps in the literature that need to be addressed?  Could testable hypotheses be developed to address these gaps?  What data sources would be needed for a study measuring access to medicines?

To answer these questions, a stepwise literature search strategy was used. Research article databases that cover medicine, pharmaceuticals and economics were searched using keywords and Medline Medical Subject Heading (MeSH) terms.

Keywords used:

 Set 1: Medicines, medication, pharmaceuticals, drugs  Set 2: Access, supply, availability, licensing, authorisation, subsidy, reimbursement  Set 3: New Zealand, NZ, Australia, United Kingdom, UK, Great Britain, England, United States, US, USA  Set 4: Medsafe, Pharmac, Pharmaceutical Management Agency, Therapeutic Goods Administration, TGA, Pharmaceutical Benefits Scheme, PBS, National Institute of Clinical Excellence, NICE, Health Insurance, Medicare, Medicaid, Department of Veterans’ Affairs, Department of Veterans’ Affairs National Formulary

Keywords in each set were searched for individually, and in combination with keywords from one, two or three other sets using the Boolean operator “and”. Keywords were mapped to Medical Subject Heading (MeSH) terms were this was possible. Wildcards, truncations, and the “explode” and “focus” commands were used where possible.

The contents of journals that frequently published relevant articles were then individually searched (all issues from January 1998 to May 2008). The websites of organisations who produced relevant articles or which were frequently cited in articles on access to medicines were examined for publications of interest.

E-mail and discussion groups linked to the websites were subscribed to, and relevant individuals were contacted by telephone and E-mail. Commissioned reports and publications cited by individuals, websites or E-mail lists were obtained directly from their authors if they were not publically available.

The data sources used in the literature search are shown in Table 2 below:

Medline (Via Ovid, 1950s to May 2008) International Pharmaceutical Abstracts (IPA) Embase (1998 to Week 20 2008) E-Con Lit (covers economic literature, including pharmacoeconomics) Articles Databases Proquest (Dissertations and theses, Pharmaceutical News Index, Health and Medical Complete, Science Journals, Social Science Journals) Web of Knowledge Web of Science Google Scholar

Pharmacoeconomics Journal

Health Affairs Contents

New Zealand Medical Journal

The World Health Organization (WHO)

Organisation For Economic Co-operation And Development (OECD)

Websites: Health Action International International International Society For Pharmacoeconomics and Outcomes Research

Organisations (ISPOR) The International Conference on Harmonisation of Technical Requirements For Registration of Pharmaceuticals For Human Use

(ICH) New Zealand Medicines And Medical Devices Safety Authority

(Medsafe) Websites: Australian Therapeutic Goods Administration (TGA) Regulatory United Kingdom Medicines and Healthcare products Regulatory Agency Agencies

(MHRA)

United States Food and Drug Administration (FDA)

The New Zealand Pharmaceutical Management Agency (PHARMAC)

Websites: The Australian Pharmaceutical Benefits Schedule (PBS)

Subsidy National Institute for Health and Clinical Excellence (NICE)

Agencies United States Department of Veterans Affairs (VA)

United States Centers for Medicare and Medicaid Services Websites: Researched Medicines Industry Association of New Zealand (now

Advocacy Medicines New Zealand)

Groups Access To Medicines Coalition

National Electronic Library of Medicines

E-mail Lists E-drug (Essential drugs in English)

FiercePharma

TABLE 2: Data sources for literature search

2.2: International Comparisons (Investigations One and Two)

2.2.1: Hypotheses

Testable hypothesis that could either confirm or reject the contention that there were differences in access to medicines for adults and children between New Zealand, Australia, the United Kingdom and the United States were developed. These are listed in Section 1.7.2.

2.2.2: Developing the Methodology

Parts of the methodology used for this thesis were initially developed in collaboration with Katri Aaltonen (University of Otago, New Zealand/University of Eastern Finland, Finland), for a cross-country comparison of the availability of medicines in New Zealand and Finland. This methodology was validated by peer-reviewed publication (178).

The methodology was then developed further for the New Zealand/Australia/UK/US comparison, and again validated by peer-reviewed publication (37).

Similarly, parts of the methodology used for the comparison of children’s access to medicines was developed for a previous study, and validated by peer-reviewed publication (106).

These prior publications are highlighted in bold text in the list of references. Publications resulting from this thesis itself are highlighted in bold italics in the list of references.

2.2.3: Data Sources

A comparison of access to medicines across different countries requires data sources that are applicable to the countries being compared, and which are widely used in these countries. The following sources were used:

2.2.3.1: New Zealand

The January to June 07 (Issue 6) of the MIMS New Ethicals was used as the primary reference source for products marketed in New Zealand (179) for both adults and children.

The online version of the New Zealand Pharmaceutical Schedule (June 2007) was used for subsidy information (180).

Information was collected on the date each product was registered in New Zealand using the Medsafe “Product /Application Search” search engine (181).

2.2.3.2: Australia

The CD-ROM version of the Australian Prescription Products guide 2007 was used as the data source for products licensed in adults and children.

The online Pharmaceutical Benefits Schedule (PBS) was used for subsidy information (182).

The Australian registration date of products was obtained from the online Australian Register of Therapeutic Goods, available from the Department of Health and Ageing Therapeutic Goods Administration (183).

2.2.3.3: United Kingdom

The electronic Medicines Compendium (eMC) 2007 was used as the primary reference source for products licensed for adults and children in the United Kingdom (184).

The United Kingdom’s Prescription Pricing Authority (PPA) website was used to find products that were not to be prescribed on the National Health Service (185). If a product was listed under “Part XVIIIA- Drugs and Other Substances Not to Be Prescribed under the NHS Pharmaceutical Services”, this product was listed as not subsidised.

The website of the National Institute For Health and Clinical Excellence (NICE) was used to find products which had been evaluated by NICE and found not be cost effective (186). These were listed as not subsidised.

The United Kingdom does not currently have a publically available database of medicine registration dates. An information request was made to the Medicines and Healthcare Products Regulatory Agency (MHRA). The MHRA responded to this request (FOI_08-386), and the information provided was used to enter the UK registration dates.

2.2.3.4: United States

The 2007 Physicians’ Desk Reference was used as the data source for products licensed for adults and children in the United States (187).

The Department of Veterans Affairs National Formulary was used as the source of subsidy information in the United States (188).

The United States registration date information for products was obtained using the Food and Drug Administration’s Drugs@FDA database (189).

2.2.4: Selection of products (Study Population)

2.2.4.1: Inclusion and exclusion criteria

A product was included in the study if it was listed in the licensing data source for that country. Medical devices, diagnostic strips, compliance aids, and non-hormonal contraceptives such as condoms, diaphragms and purely mechanical intra-uterine devices (but not intra-uterine devices that released a medicine) were excluded from the study. Products that were available without prescription were excluded from the study.

2.2.4.2: Licensing of products for children

The rules in sections 2.2.4.2.1 to 2.2.4.2.4 to determine if products were licensed for use in children, the lowest age that the products were licensed for use in, and the suitability of the products’ formulations for children. The rules were developed by the candidate in consultation with all supervisors. This classification was validated by a supervisor independently checking a sample of products to check the rules were followed accurately and consistently. The rules themselves were validated by peer reviewed publication of a previous study that used the same rules (106).

2.2.4.2.1: A medication was classified as licensed for children if the:

 reference material stated the medication was indicated for children.  reference material contained doses for children (e.g. 2-6 years, 5 ml qid).  medication was for a child specific indication (e.g. nappy rash, pulmonary surfactant in premature infants, breast milk substitute).  medication had a milligram/kilogram dosing, and nothing in (2.2.3.2.2) applied.  medication had a milligram/square meter body surface area dosing, and nothing in (2.2.3.2.2) applied.  medication had a single dose given (e.g. 250mg tid) and nothing in (2.2.3.2.2) applied.

2.2.4.2.2: A medication was classified as NOT licensed for children if the:

 only dosing given was for adults.  medication was contra-indicated in patients under 18 years.  datasheet unequivocally stated that safety has not been established in children. If the warning had caveats, an individual assessment was made of the datasheet, taking account any information on doses in children and the frequency of adverse effects in children.  datasheet unequivocally stated that effectiveness has not been proven in children. If the warning had caveats, an individual assessment was made of the datasheet, taking account any information on use in children.

 datasheet unequivocally stated that experience in children was limited. If the warning had caveats, an individual assessment was made of each datasheet, taking account any information on doses in children and the frequency of adverse effects in children.  datasheet contained the warning “Use in children is not recommended” (or words to that effect).  datasheet contained the warning “No pharmacokinetic data exists for children” (or words to that effect).  only licensed indications stated were for conditions that did not occur in children (e.g. hormone replacement therapy in menopause).

2.2.4.2.3: Lowest Age Licensed:

The lowest age the medication was licensed for was recorded. If this was not clearly stated in the text, the following methods were used in determining the lowest age:

 If the medication was contra-indicated below a certain age, then that age was taken as the lowest age licensed. (E.g. in a medication that was contra-indicated in patients under 16 years, the lowest age licensed was 16).  If doses were given by age, then the lowest age for which a dose is given was taken as the lowest age licensed.  If a medication was licensed “for children”, but no information was given as to specific ages, then it was classified as being licensed for all children down to neonates.  If dosing was given by milligrams/kilogram body weight or milligrams/metre squared body surface area, but no information was given as to specific ages, then it was classified as being licensed for all children down to neonates.

2.2.4.2.4: Suitability of formulations for children:

All medicines were classified by their route of administration and their suitability of administration for children as follows.

 Suitable Oral: Oral liquids, oral powders, oral sprays, and oral lozenges as well as soluble, dispersible and effervescent tablets 53

 Unsuitable Oral: Solid tablets and capsules, solid buccal and sublingual preparations  Inhalers: Aerosolised medications for delivery to the lungs  Nasal: Topical or systemic medications applied to the nasal mucosa  Dermal: Topical or systemic medications applied to the skin (including scalp and hemorrhoid preparations applied to the external skin of the anus)  Rectal: Topical or systemic medications applied rectally  Vaginal: Topical or systemic medication applied vaginally  Urethral: Topical or systemic medications applied to the urethra  Injectable: Intravenous, intramuscular or subcutaneous delivery by injection  Other Suitable  Other Unsuitable  Medications classified as suitable for children were: all Suitable Oral, Inhalers, Nasal, Dermal, Rectal, Vaginal, Urethral, Injectable and Other Suitable medications  Medications classified as unsuitable for children were: Unsuitable Oral and Other Unsuitable medications

2.2.4.3: ATC Classification

Before comparing the range of medicines, it was necessary to classify the medicines accurately. One option was to classify medicines by generic name. However, medicines can have different “generic names”, for example dothiepin and dosulepin for the same chemical entity. In addition, the same generic substance can have different uses, depending on formulation and indication.

To make sure that variations in the “generic” name used in different countries did not influence the results, all medicines were classified with the World Health Organization’s Anatomical Therapeutic Chemical (ATC) code system, developed by the WHO Collaborating Centre For Drug Statistics Methodology (190). Each medicine was assigned the WHO standardised name that matched its ATC Code. (Where the active ingredient in a medicine had two or more possible ATC codes, the code that matched the route and licensed indication of the medicine was assigned).

Combination products (products that had one or more active ingredients) which had an existing ATC code for the combination were assigned that code. If no code existed for the combination but one of the active ingredients in the combination product had a code for “X and combinations”, that code was used and the standardised name had the other ingredients listed in alphabetical order. If two or more active ingredients in a combination product had a code for “X and combinations”, then the code for the active ingredient that came first alphabetically was used, and all ingredients were listed alphabetically.

2.2.4.4: Collapsing products down to entities for analysis

The products were collapsed by active agent and ATC code using the data management programme STATA (191). All data management and statistical analysis was carried out using STATA. Collapsing data in this way ensured that multiple brands of therapeutically identical products would not affect the results. A unique ATC code or unique active agent was considered a unique “entity” for the purpose of analysis.

Entities were compared by both ATC Code and active agent, but the ATC code analysis is the primary analysis. The ATC system was specifically designed for drug utilisation studies, including comparisons between countries. It allows medicines to be classified in a standard manner, and makes accurate comparisons easier. However, not all medicines (or combinations of medicines) have an ATC code assigned in the ATC classification system. Also, a product that has two or more key indications may either have only one ATC Code assigned, or may have multiple ATC codes assigned based on indication and route of administration (192). As noted above, where the active ingredient in a medicine had two or more possible ATC codes, the code that matched the route and licensed indication of the medicine was assigned. A duplicate analysis by active agent acts as a further check on these decisions. (These duplicate analyses active agent are in Data Annex I).

2.2.5: Study Variables

Variable Definition

Entity A unique ATC Level 5 code (also a unique active agent in the parallel analysis by active agent)

Licensed for adults The entity had been authorised for use in adults by the regulatory authority in that country

Licensed for children The entity had been authorised for use in children by the regulatory authority in that country

Time since first registration The time (in days) since the earliest date the entity was registered in any of New Zealand, Australia, the United Kingdom or the United States.

Innovative entity An entity that has been granted “fast track” approval by the United States Food And Drug Administration and/or classified by the Canadian Patented Medicines Review Board as a as a “breakthrough or substantial improvement” (see Section 2.2.5.4).

Lowest age licensed (children) The lowest age of children for which an entity has been authorised for use in by the regulatory authority in that country.

Suitability of formulation (children) Whether or not an entity is available as formulation which can be used by all children.

TABLE 3: Variables analysed in Investigations 1 and 2

2.2.5.1: Number of Licensed and Subsidised Entities

The number of entities licensed in each country, and subsidised in each formulary was tabulated, and the proportion of subsidised entities to licensed entities in each country was calculated. As noted in Table 2, a unique level 5 ATC code was considered a unique entity.

2.2.5.2: Types of Entities

The types of entities licensed and subsidised in each country were compared by comparing the number of licensed and subsidised entities in each country at the first and second levels of the ATC system.

2.2.5.3: Time Since First Registration of Products

For the purpose of this study, the “the time since first registration” (“comparative age”) of a medicine was defined as the time from the earliest registration date of that medicine in any of New Zealand, Australia, the United Kingdom or the United States. The earliest registration date was subtracted from 31/12/2007 (the end of the data collection period) to generate the “time since first registration” of the product in days.

Pre-determined rules were used to resolve ambiguities when adding registration dates, as described below:

The starting point was always the brand name column. All the information in the brand name column was used, along with all the information given for the brand name by the regulatory agency’s database.

Where there was an exact match between the information in the brand name column and the agency, that date was used.

Where there was a close match, the “best fit” was used. For example, if the brand name had “Betaloc 190 mg BP” and the database had “Betaloc 190 mg” (without the formulary being

specified), that date was used. However, if the database had “Betaloc 190mg USP”, that date could not be used, as the information available clearly showed they were different products. The same rule was used for strengths and dose forms.

If a dose form was not specified in the brand name column, information in the route column was used, as each route corresponded to certain dose forms. These were:

 Suitable Oral: oral formulations that could be administered to children of all ages, such as oral liquids, syrups, gels, powders (including powders for reconstitution), soluble tablets, dispersible tablets, effervescent tablets  Unsuitable Oral: oral formulations that could not be administered children of all ages, such as solid tablets, capsules, chewable tablets  Ears/Eyes: eye and ear drops, creams, ointments, intra-ocular implants or devices  Dermal: topical creams, gels, ointments, lotions, powders and transdermal patches  Inhalation: all inhalation devices  Injectable: all intravenous, intramuscular and subcutaneous injections, including infusions  Nasal: devices for delivery to the nasal membrane, e.g. antihistamine sprays  Rectal: suppositories, rectal foams, enemas  Vaginal: pessaries, vaginal patches  Other Suitable/Other Unsuitable: These were very rare. Other suitable products were any other dose form that could be used by children of all ages. Other unsuitable products were any other dose form that could not be used by children of all ages.

When there were two possible products that fit all the information available, the earlier date was used. When there were more than two possible products that fit all the information available, the earliest date was used.

2.2.5.4: Level of Innovation of Products

The level of innovation of products licensed and subsidised in the four countries was compared in two ways.

First, the licensed and subsidised entities in each country were collapsed by ATC level 4 groups, and the number of level 4 groups was compared. The reason for this is that the level 4 group is the lowest grouping in the ATC system that contains more than one entity. For example, the level 4 group C10AA (HMG CoA Reductase Inhibitors, or “statins”) contains the codes C10AA01 (simvastatin), C10AA02 (lovastatin), C10AA03 (pravastatin), C10AA04 (fluvastatin), C10AA05 (atorvastatin), C10AA06 (cerivastatin), C10AA07 (rosuvastatin) and C10AA08 (pitavastatin) (190, 193).

While the entities in the same level 4 group are not therapeutically interchangeable, they have similar therapeutic, pharmacological and chemical properties. Licensing or subsidizing only a few agents within an ATC level 4 group (for example only a limited number of statins in group C10AA) may in fact be appropriate health policy, but this may not be apparent in an analysis that counts each ATC level 5 entity equally.

Comparing the number of ATC level 4 groups is a way of determining if differences in the number and time since first registration of licensed or subsidised entities are solely due to multiple entities in the same ATC level 4 groups, or if there are differences in the ATC level 4 groups between countries.

However, the World Health Organization Collaborating Centre For Drug Statistics Methodology as a rule does not create a new level 4 group until there are at least two chemical entities with marketing authorisations that fit in that group (190). This means that simply comparing level 4 groups may miss new and innovative products that have not yet been assigned a level 4 group.

For this reason, a second method of identifying new and innovative products was also used. This uses the methodology developed by Roughead et al. to identify innovative medicines that provide important health gains (194). A list was compiled of all medicines that had been granted “fast track” approval (because the medicine treated a serious illness and fulfilled an unmet need) by the United States Food And Drug Administration between 1998 and 2007, and all medicines classified by the Canadian Patented Medicines Review Board as a “breakthrough or substantial improvement” between 2000 and 2007 (195, 196). Two vaccines were excluded from the list, as vaccination schedules and pathogen prevalence differ between countries. All other medicines on the list were checked to see if they were licensed or 59

subsidised in each of the four countries in 2007.

The number of innovative products that are licensed and subsidised can arguably be used as a proxy for whether a country or system is ‘falling behind’ in access to medicines. This is an aspect of access that may be of more interest to patients and policy makers than the total number of entities, which can be influenced by ‘me-too’ products.

2.2.6: Data Analysis

2.2.6.1: Number of Entities

The number of entities licensed or subsidised in each country that were licensed or subsidised in each of the other countries were determined. The appropriate statistical tests for statistical significance were carried out (Chi squared test, 2x2 tables).

This addresses the questions of whether the number of licensed or subsidised entities differ between countries, systems, adults and children.

2.2.6.2: Types of Entities

The types of entities licensed and subsidised in each country were compared by comparing entities in each country at the first and second levels of the ATC system. The first level of the ATC system classifies medicines by the organ or system they affect (e.g. cardiovascular or nervous system) while the second level classifies medicines by their main therapeutic or pharmacological action (e.g. drugs used in diabetes or antineoplastic agents).

Data from the four countries were also examined in more detail for differences in agents used to treat diabetes, cardiovascular disease, cancer, mental illness and respiratory disease. These diseases have been identified the World Health Organization as major contributors to life years lost and burden of disease in high-income countries (197). The licensed and subsidised entities used to treat these conditions were analysed at ATC level 2 to find out which

countries had licensed or subsidised agents in level 2 groups that were not available in other countries.

This addresses the questions of whether the types of licensed or subsidised entities differ between countries, systems, adults and children.

2.2.6.3: Time Since First Registration of Products

The times from first registration of active agents and ATC Codes licensed and subsidised in each country were determined. The time from first registration of each active agent and ATC code was determined by collapsing the data on the time from first registration of each product by both active agent and ATC code.

The appropriate statistical tests for statistical significance were carried out (see Section 2.2.6.7).

The median times since first registration of licensed and subsidised entities were compared for each pair of countries to determine if there was a difference in the medians, and if this difference was statistically significant.

In addition to comparing the time since first registration of all active agents and ATC codes licensed and subsidised in the four countries, the time since first registration of specific sub groups were also compared. For example, the median time since first registration of the active agents that were licensed in New Zealand but not licensed in Australia was compared with the median time since first registration of the active agents licensed in Australia but not licensed in New Zealand.

This addresses the questions of whether the comparative age of licensed or subsidised entities differ between countries, systems, adults and children.

2.2.6.4: Level of Innovation of Products

The number of ATC level 4 groups that contained licensed and subsidised entities in each country were determined. The appropriate statistical tests for statistical significance were carried out (Chi squared test, 2x2 tables, with one degree of freedom).

The number of innovative entities (as determined in 2.2.5.4) licensed and subsidised that were licensed and subsidised in each country in 2007 were compared. Furthermore, the number of innovative entities that were newly licensed and subsidised between the end of 2007 and April 2010 in each country were compared.

This addresses the questions of whether the level of innovation of licensed or subsidised entities differ between countries, systems, adults and children.

2.2.6.5: Lowest Age Authorised (Investigation Two)

The lowest age (of children) that licensed and subsidised entities were authorised for use in were compared within each country, and between countries. The lowest age was determined using the rules in 2.2.3.2.3.

This addresses the questions of whether the lowest age authorised of licensed or subsidised entities differ between countries and systems.

2.2.6.6: Suitability of Formulations for Children (Investigation Two)

The number of licensed and subsidised entities that were in formulations suitable for children were tabulated. (The suitability of the formulations was determined using the rules in 2.2.3.2.4). The number of licensed subsidised entities in suitable formulations were compared to the total number of entities).

The lowest age (of children) that entities were authorised for use in were compared for entities in suitable formulations with those in unsuitable formulations.

This addresses the questions of whether the availability of suitable formulations of licensed or subsidised entities differ between countries and systems.

2.2.6.7: Statistical Analysis

Before testing the hypotheses (see Section 1.7.2.2: Testable Hypotheses), it was necessary to determine the level of significance. A significance level of 0.05 is commonly used (198). However, when several hypotheses are being tested, the chances of a Type I error (wrongly rejecting the hypothesis that there is no difference or no effect) increase in proportion to the number of hypotheses (198).

A significance level of 0.05 may therefore not be conservative enough. There are statistical techniques (such as the Bonferroni correction) that can adjust for the effects of testing multiple hypotheses. However, there is considerable debate among statisticians about when it is appropriate to apply such techniques. There is also a concern that such techniques are too conservative, and increase the risk of a Type II error (wrongly rejecting a difference or effect when one in fact exists) (198).

When there is uncertainty about the level of significance to set, it is recommended that the researcher consider the potential impact of both Type I and Type II errors (199, 200). Access to medicines is a potentially controversial and emotive topic (see Section 1.6: The Controversy In New Zealand). A spurious finding a difference between countries may misinform the public debate. Failing to show a difference at a conservative level of significance is unlikely to have the same effect.

Therefore, it was decided a priori that only results with a p value 0.01 or less would be accepted as being statistically significant.

The Chi Squared test (2x2 tables) was used to determine if differences in the numbers of entities licensed and subsidised in different countries were statistically significant.

The Chi Squared test (2x2 tables) was also used to determine if differences in the numbers of ATC level 4 groups that contained licensed and subsidised entities in different countries were statistically significant.

This test was chosen because licensing and subsidy status were categorical variables (i.e. an entity could either be licensed or not licensed, subsidised or not subsidised). Furthermore, the choice of study countries where licensing decisions were independent of each other (see Section 1.2.1: Choice of Comparators), and funding systems where subsidy decisions were independent of each other (see Section 1.2.4: Choice of Comparator Systems), ensures that these are independent categorical variables.

To compare times from first registration of products, the first step was to plot the distributions as histograms to determine whether data were normally distributed. As they were not normally distributed, non-parametric tests were used in this section. Medians and percentiles were used to summarise the data.

The Kruskal-Wallis test was used to test the equality of medians of times from first registration of the populations of licensed and subsidised entities in the four countries. This test is appropriate for non-parametric continuous data (the time from first registration was measured in days) where the data are independent (each country made its own registration decisions).

The Wilcoxon Rank Sum test was used to test whether differences between any two groups in times from first registration were statistically significant. The Wilcoxon Rank Sum tests the probability that the two sets of data being compared come from the same population. This test is appropriate where the data are independent (each country makes its registration decisions independently), and are capable of being ranked from least to greatest, as with the days since first registration of each entity.

2.2.6.8: Sensitivity Analysis

The primary data source for a country may not list all medicines licensed in that country. In particular, not all medicines that are solely used in hospitals may be listed in general

prescribing reference texts such as those used in this study. Furthermore, the number of such medicines listed may vary from one of the primary reference sources to another (and hence from one country being compared to another). A major difference in the listing of medicines in the primary reference sources could potentially bias the results. Similarly, some of the subsidy systems formularies may be more likely to list hospital only medicines than others.

For this reason, a sensitivity analysis was performed. None of the countries have separate legal classifications for hospital only medicines, and none of the primary reference sources list hospital only medicines separately. For the sensitivity analysis, all entities that were available only in injectable formulations in each country were removed. Being available only in injectable form was used as a proxy for hospital use, as these are difficult to self-administer in the community (with some exceptions such as insulins).

The sensitivity analysis examined the number of entities that were licensed and subsidised in non-injectable formulations in each country, and the proportions of entities licensed and subsidised in non-injectable formulations to the total number of entities licensed and subsidised in each country. The same analysis was undertaken for each ATC level 1 group, to determine if the numbers in any ATC group changed drastically when injectable only entities were removed. Lastly, the time since first registration of non-injectable entities licensed and subsidised in each country was calculated, to determine if this was different from the time since first registration of all entities.

2.3: Key Informants’ Perceptions of New Zealanders’ Access to Medicines (Investigation Three)

2.3.1: Objective

To examine key informants’ perceptions of New Zealanders’ access to medicines.

2.3.2: Participant Selection

Ethics approval was obtained from the School of Pharmacy, University of Otago. An initial list was generated of the types of organisations to be included, such as political parties, industry, government agencies, health professionals, health economists and consumer groups. People from these groups who would be suitable key informants were identified through reading literature on the controversy around New Zealand’s access to medicines, internet searches and discussions with colleagues. These were individuals who had published research or commentary on the medicines system, acted as spokespersons for stakeholder groups, or held positions that gave them important insights into the medicines system (see Section 1.6: The Controversy In New Zealand). They were purposefully selected to ensure inclusion of a range of professional groups and positions, and opinions on the medicines system. This resulted in a list of 32 people.

Respondents were initially contacted by using their work e-mail addresses or the e-mail address on their professional web pages. (This allowed the inclusion of the participant information sheets, and gave respondents an opportunity to consider the proposal before having to respond). Those who did not respond to the initial e-mail were sent a second e-mail to the same address, followed by a telephone call to a work telephone number or a number listed on a professional web page.

Twenty informants agreed to take part in the interview. The characteristics of the informants are shown in Table 4:

Professional Location Notes Affiliation Medicine (5) Metropolitan North Island Specialties: General Practice (2), (1), Metropolitan South Oncology (1), Pharmacology (2) Island (4) Pharmacy (4) Metropolitan North Island Hospital Pharmacy (3), Community (2), Provincial North Island Pharmacy Representative (1) (2) Members of Metropolitan North Island Government (1), Main Opposition Parliament (3) (3) (1), Minor Opposition Party (1)

Public Service (3) Metropolitan North Island PHARMAC (2), Medsafe (New (3) Zealand's Medicines regulator) (1)

Pharmaceutical Metropolitan North Island Represented major international Industry (2) (2) pharmaceutical companies. Patient Advocate Metropolitan North Island Umbrella group that represents (1) (1) multiple disease advocacy groups Health Economist Metropolitan South Island (1) (1) Māori Health Metropolitan North Island Māori are New Zealand's indigenous Researcher (1) (1) people, and have poorer health outcomes than other New Zealanders TABLE 4: Characteristics of key informants who took part in the study

The twelve people who declined to participate were a broadly similar group to those who did take part, and included medical practitioners, Members of Parliament (as with the Members of Parliament who did participate, those who declined came from both the Government coalition at the time and the Opposition) and a health economist. The most common reasons for declining was not having time, and not being able to agree a mutually acceptable time for an interview.

2.3.3: Interviews

All except two of the interviews were conducted face to face, with the remainder by telephone because a suitable time for a face-to-face interview could not be arranged. The interviews were semi-structured, focusing on common topics, but with the freedom to pursue additional topics of interest as they came up.

The interview data were not anonymised, but the interviewees were aware that the results were non-attributable (no interviewees would be quoted by name or have a position attributed to them by name without their express written position). All interviewees agreed to this.

The interview structure was developed using information from the controversy over access to medicines in New Zealand in the peer-reviewed literature. This was then iteratively pilot tested on health professionals and researchers (i.e. people who had similar levels of education and command of English as our expected participants, including both first and second language English speakers) to ensure reliability and validity.

The interviews focused on the following topics:

 What New Zealand does well or poorly when it comes to funding medicines  What New Zealand does well or poorly when it comes to licensing medicines  What New Zealand could learn from other countries (and vice versa)  Equality of access to medicines for all people in New Zealand  The role of the pharmaceutical industry in providing New Zealanders with access to medicines  The relationship between Government agencies and the pharmaceutical industry  The effectiveness of patient advocates and other lobby groups in influencing access to medicines  Changes that participants would like to see in New Zealand’s medicines system

2.3.4: Data Analysis

The interviews ranged from 25 minutes to an hour, with interviews around 40 minutes being typical. Interviews were digitally recorded and later transcribed verbatim by a professional transcriber. Hand written notes were taken to supplement the recordings and as a precaution against recorder failure.

The data were then coded in NVivo to find common themes and areas of disagreement using constant comparison methods, an inductive technique where data are examined critically and constantly for new meanings. The analysis method included first identifying broad categories such the demographic characteristics of the respondent and the major topic being discussed at the time (for example medicine funding or equity), before identifying sub-categories in each category to create a ‘tree’ structure.

This methodology has been validated by peer-reviewed publication (201).

CHAPTER 3: THE LICENSING AND SUBSIDY OF MEDICINES FOR ADULTS UNDER A NATIONWIDE SINGLE PAYER SYSTEM IN EACH OF NEW ZEALAND, AUTRALIA, THE UNITED KINGDOM AND THE UNITED STATES

3.1 Context

The licensing processes of medicines in New Zealand, Australia, the United Kingdom and the United States have been described in Section 1.2.2. The funding processes of medicines under New Zealand’s PHARMAC, Australia’s PBS, the United Kingdom NHS and the United States VANF have been described in Section 1.2.3. It is possible that the differences in these licensing and funding processes will lead to differences in the medicines licensed and subsidised under these single payer systems.

A number of hypotheses (see Section 1.7.2.2) were tested:

The methods of data collection and analysis for this investigation have been described in Section 2.2.

3.2 Results

3.2.1 Number of Licensed and Subsidised Entities

The number of entities that were licensed and subsidised for adults in each country are listed in Table 5.

The United Kingdom had the highest number of licensed entities, followed by the United States, Australia and New Zealand. The United Kingdom had the highest number of subsidised entities, followed by Australia, the United States VANF and New Zealand. The United Kingdom had the highest proportion of licensed entities that were subsidised, followed by Australia and New Zealand, and the United States VANF.

New Zealand had fewer licensed and subsidised entities than any of the other countries. The proportion of licensed entities that were subsidised in New Zealand is similar to that in Australia, lower than that in the United Kingdom, and higher than that in the United States VANF.

There were statistically significant differences in the patterns of licensed entities between countries (UK>US>Aus>NZ, Chi Squared test, p<0.001), whether compared by active agent or ATC Code.

There were statistically significant differences in the patterns of subsidised entities in New Zealand, Australia, and the United States VANF (Aus>US VANF>NZ, Chi Squared test, p<0.001), whether compared by active agent or ATC Code. However, no statistically significant differences were detected in the patterns of entities subsidised in the United Kingdom compared with any other country.

There were statistically significant differences in the pattern of licensed and subsidised entities within New Zealand and Australia (licensed>subsidised, Chi Squared test, p<0.001), whether compared by active agent or ATC code. However, no statistically significant differences were detected in the pattern of licensed and subsidised entities within the United

Kingdom. No statistically significant differences were detected between the pattern of entities licensed in the United States and subsidised in the United States VANF.

Country Licensed for Subsidised for Percentage of p Value (Chi2 adults adults entities licensed test, 2x2 table) for adults that were subsidised for adults

New Zealand 765a 503b 65.8% <0.001

Australia 879a 567b 64.5% <0.001

United 1020a 1016b 99.6% 0.733 Kingdom

United States 918a 505 (VANF)b 55.1% 0.061 a Chi2 test p value <0.001 for each pair of countries (2x2 tables). b Chi2 test p value <0.001 for New Zealand compared to Australia, New Zealand compared to the United States VANF and Australia compared to the United States VANF (2x2 tables). P values for the United Kingdom compared to any other country were not statistically significant.

TABLE 5: Number of entities collapsed by ATC code that were licensed and subsidised for adults in each country. (See TABLE DA 1.1 in Data Annex 1 for duplicate active agent analysis).

3.2.2 Types of Licensed and Subsidised Entities

The types of licensed and subsidised entities in the four countries were compared to determine if there were any differences between the countries. Tables 6 and 7 below compare licensing and subsidy in each country at ATC levels 1 and 2 respectively.

The types of entities licensed in each country were very similar at ATC level 1. The top four categories were the same in all four countries, but differed in order. Nervous system agents 72

(126 to 165 entities), systemic anti-infective agents (123 to 132 entities), cardiovascular system agents (81 to 122 entities), anti-neoplastic and immunomodulating agents (87 to 101 entities), and alimentary tract agents (61 to 93 entities) were the most commonly licensed in each country, while antiparasitic agents (8 to 17 entities) and systemic hormonal preparations (17 to 31 entities) were relatively uncommon.

The types of entities subsidised in each country were also very similar at ATC level 1. Nervous system agents (92 to162 entities), systemic anti-infective agents (61 to 132 entities), cardiovascular system agents (51 to 132 entities), anti-neoplastic and immunomodulating agents (59 to 96 entities), and alimentary tract agents (35 to 94 entities) were the most common types subsidised in each country, while antiparasitic agents (2 to 13 entities) and systemic hormonal preparations (8 to 31 entities) were relatively uncommon.

The types of entities licensed in each country were very similar at ATC level 2. Systemic antibacterials (49 to 59 entities), antineoplastic agents (49 to 53 entities), opthalmologicals (46 to 51 entities) and psycholeptics (34 to 45 entities) were the groups with the most entities licensed for adults in all four countries. Sex hormones (30 to 43 entities) were in the top five in New Zealand, Australia and the United Kingdom, and systemic antivirals (33 entities) in the United States VANF.

The types of entities subsidised in each country were very similar at ATC level 2. Once again systemic antibacterials (31 to 52 entities), antineoplastic agents (30 to 57 entities), opthalmologicals (24 to 51 entities), and psycholeptics (24 to 43 entities) were in the top five subsidised groups for all four countries. Antivirals (29 to 33 entities) were also in the top five in Australia and the United States VANF, while sex hormones (21 to 43 entities) were in the top five in New Zealand and the United Kingdom.

Ranking NZ NZ Aus Aus UK UK US Lic UNITED Lic Sub Lic Sub Lic Sub STATES VANF Sub

1 N:126 N:92 N:145 N:100 N:165 N:162 N:143 J:98

2 J:123 C:68 J:138 J:78 J:128 J:132 J:124 N:86

3 L:87 L:65 L: 100 L:75 C:122 C:123 C:109 L:59

4 C:81 J:61 C: 88 C:67 L: 101 L:96 L: 99 C:51

5 A:61 A:49 A:71 A:55 A:93 A:94 A:89 A:35

6 S:49 S:32 S:54 G:34 G:70 G:70 D:56 B:29

7 G:47 G:27 G:52 M:32 M:62 M:61 G:53 D:28

8 M:37 R:24 M:46 S:32 S:59 S:59 R:50 M:25

9 R:35 M:23 B:37 R:24 D:55 D:55 S:48 S:24

10 D:33 D:22 D:35 B:19 R:45 R:45 M:47 G:22

11 B:31 H:16 R:35 D:19 B:40 B:39 B:44 R:20

12 V:27 V:12 V:33 H:19 V:36 V:36 V:24 V:13

13 H:20 B:10 H:27 P:7 H:31 H:31 H:17 H:8

14 P:8 P:2 P:17 V:6 P:13 P:13 P:15 P:7

TABLE 6: Numbers of licensed and subsidised entities in each country at ATC level 1 (Please refer to ATC Code Master Key Level 1 for ATC categories)

Ranking NZ Lic NZ Sub Aus Lic Aus Sub UK Lic UK Sub US Lic UNITED STATES VANF Sub 1 J01:53 L01:41 J01:59 L01:38 L01:58 L01:57 L01:53 J01:34 2 L01:49 N05:31 L01:53 J01:37 J01:52 J01:52 J01:49 J05:33 3 S01:46 J01:31 S01:51 S01:30 S01:51 S01:51 S01:46 L01:30 4 N05:34 S01:29 N05:37 J05:29 N05:45 N05:43 J05:36 S01:24 5 G03:30 G03:21 G03:33 N05:26 G03:43 G03:43 N05:34 N05:24 6 N06:27 N06:18 J05:33 N06:25 J05:36 J05:36 N06:32 J07:17 7 J05:24 J05:18 N06:29 G03:25 A10:30 A10:30 G03:31 N06:16 8 J07:23 R03:15 B01:25 A10:18 N06:29 N06:29 N02:26 B01:16 9 B01:20 A10:14 A10:21 * R03:16* N02:29 N02:28 A10:26 R03:14 10 C01:19 N02:13 J07:21* M01:16** M01:28 M01:28 R03:24 G03:14 TABLE 7: Numbers of licensed and subsidised entities in each country at ATC level 2 (Please refer to ATC Code Master Key Level 2 for ATC categories)

*There are also 21 licensed entities in category R03: drugs for obstructive airways diseases ** There are also 16 subsidised entities in category C09: agents acting on the rennin- angiotensin system.

The numbers of licensed and subsidised entities in each country used to treat diabetes, cardiovascular disease, cancer, mental illness and respiratory disease are shown in Table 8 below.

Group NZ Aus UK US (PHARMAC) (PBS) (NHS) (VANF) A08- Antiobesity preparations 3 (0) 4 (1) 3 (3) 4 (0)

A10- Drugs used in diabetes 16 (14) 21 (18) 30 (30) 26 (6)

C01- Cardiac therapy 19 (10) 20 (9) 20 (20) 17 (9)

C02- Antihypertensives 7 (5) 7 (6) 11 (11) 10 (6)

C03- Diuretics 10 (10) 11 (9) 15 (15) 15 (9)

C04- Peripheral vasodilators 3 (3) 3 (1) 4 (4) 2 (1)

C07- Beta blocking agents 12 (11) 10 (9) 18 (18) 15 (8)

C08- Calcium channel blockers 7 (6) 8 (7) 11 (11) 9 (5)

C09- Agents acting on the renin- 12 (12) 18 (7) 24 (24) 22 (7) angiotension system C10- Lipid modifying agents 9 (9) 10 (10) 14 (14) 15 (5)

L01- Antineoplastic agents 49 (41) 58 (38) 58 (57) 53 (30)

L03- Immunostimulants 11 (7) 13 (12) 14 (11) 16 (12)

L04- Immunosuppresants 15 (9) 17 (13) 15 (14) 19 (10)

N05- Psycholeptics 34 (31) 37 (26) 45 (43) 34 (24)

N06- Psychoanaleptics 27 (18) 29 (25) 29 (29) 32 (16)

R03- Drugs for obstructive 17 (15) 21 (16) 23 (22) 24 (14) airways diseases TABLE 8: Numbers of licensed and subsidised entities used in treating diabetes, cardiovascular disease, cancer, lung disease and mental illness in each country at ATC level 2

Table 8 shows that at ATC level 2, there were only minor differences between the four countries in the agents used to treat these conditions. Neither New Zealand nor the United States VANF subsidised antiobesity agents, and the VANF had fewer agents used to treat diabetes, cardiovascular disease and mental illness than the other three systems. However, many products used to treat diabetes are classified as non-prescription medicines in the United States, and therefore excluded from this study. (See chapter discussion, Section 3.3).

The results of an ATC Level 4 analysis of agents used to treat diabetes, cardiovascular disease, cancer, mental illness and respiratory disease are included in Tables DA 2.1 and DA 2.2 in Data Annex II.

However, this analysis showed there were very few differences in the licensing of entities for these conditions between the countries. The ATC 4 Level 4 groups which had entities listed only in one country did in fact have entities licensed in that group in the other countries. Similarly, groups that did not have listed entities in any particular country did in fact have entities licensed in the group in that country. Licensed entities that were not listed in the primary sources in this study tended to either have been very recently registered (five years or less) or have been registered for over 20 years. (See chapter discussion, Section 3.3).

There were some important differences between the countries in the number of ATC Level 4 groups that contained subsidised entities for treating diabetes, cardiovascular disease, cancer, lung disease and mental illness. Some of these differences were due to the listing/non-listing of hospital only agents such as colony stimulating factors. The United Kingdom had the most ATC Level 4 groups that contained subsidised entities that were not subsidised in the other countries. However, many of these agents were subsequently withdrawn from the market (See Table DA 2.2 in Data Annex II).

New Zealand did not subsidise filgrastim, lenograstim, pegfilgrastim or sargramostim (L03AA immunostimulants, colony stimulating factors), but they could be provided for outpatients out of hospital budgets if required. Australia subsidised filgrastim and lenograstim, the UK subsidised filgrastim, lenograstim and pegfilgrastim, and the United States VANF subsidised filgrastim and sargramostim. New Zealand did not subsidise donepezil, rivastigmine or galantamine (N06DA, psychoanaleptics, anticholinesterases),

Australia and the UK subsidised all three agents, and the United States VANF subsidised donepezil and galantamine.

Australia did not subsidise acipimox or nicotinic acid (C10AD, lipid modifying agents, nicotinic acid derivatives), whereas New Zealand subsidised acipimox, the United States VANF subsidised nicotinic acid, and the United Kingdom subsidised both. Australia did not subsidise pimozide (category N05AG, psycholeptics, diphenylbutylpiperidine derivatives), which was subsidised in all three of the other countries.

The United Kingdom was the only country to have subsidised sibutramine (A08AA, antiobesity products, centrally acting antiobesity products), rimonabant (A08AX, other antiobesity drugs), inhaled insulin (A10AF, drugs used in diabetes, insulins and analogues for inhalation), and omalizumab (R03DX, other systemic drugs for obstructive airway diseases).

The United States VANF did not subsidise rosiglitazone or pioglitazone (A10BG, drugs used in diabetes, thiazolidinediones), whereas New Zealand subsidised pioglitazone, and Australia and the UK subsidised both pioglitazone and rosiglitazone. (Rosiglitazone was subsidised by the VANF at the start of 2007, but delisted in October). The VANF did not subsidise colestipol or colestyramine (C10AC, lipid lowering agents, bile acid sequestrants), while all three other countries had colestipol subsidised and New Zealand also had colestyramine subsidised. The VANF did not subsidise ezetimibe (C10AX, other lipid modifying agents) and moclobemide (N06AG, psychoanaleptics, monoamine oxidase-A inhibitors) which were both subsidised in the other three countries.

3.2.3 Time Since First Registration of Licensed and Subsidised Entities

3.2.3.1 Time Since First Registration of Licensed and Subsidised Entities for Entire Countries

This section reports the time since first registration (comparative age) of the entities that are licensed and subsidised in each country. As previously defined, the time since first registration of an entity is the time since the earliest registration of that entity in any of the

four countries (New Zealand, Australia, the United Kingdom or the United States). The greater the time since first registration, the “older” an entity is.

The distributions of times since first registration of entities were plotted as histograms. Data for both licensed or subsidised entities for each country were not normally distributed when collapsed by active agent or ATC Code.

NZ AUS

100 100

80 80

60 60

Frequency Frequency 40 40

20 20

0 0 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 25000 days days

UK US

100

80 80

Frequency

20 20

0 0 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 25000 days days

FIGURE 1: Comparative age (time since first registration) of licensed entities collapsed by ATC code. 80

NZ AUS

80 80

60 60

Frequency

0 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 25000 days days UK US

100

Frequency Frequency

40 20

0 0 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 25000 days days

FIGURE 2: Comparative age (time since first registration) of subsidised entities collapsed by ATC code.

New Zealand had the “oldest” licensed and subsidised entities (see Table 7 below), and there were significant differences in the comparative age of entities between countries.

The median time from first registration of licensed entities (when collapsed by either active agent or ATC code) is highest in New Zealand, followed by Australia, the United Kingdom and the United States. The probability that the medians were equal is <0.001 (Kruskal-Wallis test).

The median comparative age of subsidised entities (when collapsed by either active agent or ATC code) is highest in New Zealand, followed by the United States VANF, Australia and the United Kingdom. The probability that the medians were equal is <0.001 (Kruskal-Wallis test).

Summary of times since first registration of entities licensed for adults collapsed by ATC code (a) Median 25th 75th Min Max (Days) Range (Days) Percentile Percentile (Days) (Days) (Days) (Days)

New Zealand 8,936a 1,329 21,591 557 25,162 24,605 (N=765) Australia 7,795a 929 22,071 704 25,162 24,458 (N=879) United Kingdom 7,319a 551 23,978 194 25,162 24,968 (N=1020) United States 6,607a 521 23,376 195 24,927 24,732 (N=918) Summary of times since first registration of entities subsidised for adults collapsed by ATC code (b)

New Zealand 10,724b 1,826 21,591 1329 25,162 23,833 (N=503) Australia 8,065b 1,146 21,629 923 25,162 24,239 (N=567) United Kingdom 7,319b 551 23,978 194 25,162 24,968 (N=1016) United States 8,203b 557 21,591 195 24,927 24,732 (N=505) a P <0.01 for differences in median time since first registration for New Zealand compared with all other countries, and Australia compared with the United States (Wilcoxon Rank Sum Test). b P <0.01 for differences in median time since first registration for New Zealand compared with all other countries, and the USVANF compared with Australia and the United Kingdom (Wilcoxon Rank Sum Test).

Table 9: Comparisons of the comparative age (time since first registration) of licensed and subsidised entities in each country. (See TABLE DA 1.2 in Data Annex 1 for duplicate active agent analysis).

(The Kruskal-Wallis test for probability of equality of medians found p< 0.001 when comparing (a) or (b) between countries)

3.2.3.2 Time Since First Registration of Licensed And Subsidised Products for Specific Subgroups

Table 10 shows the parameters of the time since first registration of specific sub groups that were licensed (or subsidised) in one country but not in another. Tables 11 and 12 summarise the statistical comparison (Wilcoxon Rank Sum test) of these results.

Median 25th 75th Min Max Range (Days) Percentile Percentile (Days) (Days) (Days) (Days) (Days) Time from first registration of licensed entities collapsed by ATC code

Licensed in NZ but 8,740 1,477 19,085 557 21,053 20,496 not in AUS Licensed in NZ but 10,161 1,866 19,834 1,271 20,990 19,719 not in the UK Licensed in NZ but 9,954 1,719 21,031 1,271 25,162 23,891 not in the US VANF Licensed in AUS 5,464 929 21,416 704 24,927 24,223 but not in NZ Licensed in AUS 9,133 1,727 19,996 1,019 22,071 21,052 but not in the UK Licensed in AUS 8,782 1,168 21,584 923 25,162 25,162 but not in the US VANF Licensed in the UK 4,962 551 21,416 194 24,927 24,733 but not in NZ Licensed in the UK 6,478 551 19,150 194 24,248 24,054 but not in AUS Licensed in the UK 8,657 551 21,584 194 25,162 25,162 but not in US VANF Licensed in the US 4,771 521 23,011 195 24,927 24,732 VANF but not in NZ Licensed in the US 5,501 521 21,475 195 23,659 23,464 VANF but not in AUS Licensed in the US 6,940 521 21,475 195 23,659 23,464 VANF but not the UK

Median 25th 75th Min Max Range (Days) Percentile Percentile (Days) (Days) (Days) (Days) (Days) Time from first registration of subsidised entities collapsed by ATC code

Subsidised in NZ 11,587 2,881 20,424 1,607 21,591 21,591 but not in AUS Subsidised in NZ 10,879 2,881 19,731 1,607 20,275 20,275 but not in the UK Subsidised in NZ 10,663 1,893 20,667 1,329 25,162 23,833 but not in the US VANF Subsidised in AUS 4,694 1,146 20,065 923 22,071 21,148 but not in NZ Subsidised in AUS 9,880 1,866 19,834 1,019 22,071 21,052 but not in the UK Subsidised in AUS 6,871 1329 20,275 923 25,162 25,162 but not in the US VANF Subsidised in the 5,017 551 21,416 194 24,927 24,733 UK but not in NZ Subsidised in the 6,478 551 21,584 194 24,927 24,733 UK but not in AUS Subsidised in the 7,550 557 20,647 194 25,162 24,732 UK but not in US VANF Subsidised in the 6,008 557 21,416 195 24,927 24,732 US VANF but not in NZ Subsidised in the 7,550 557 20,647 195 24,927 24,732 US VANF but not in AUS Subsidised in the 8,452 1,053 19,564 195 21,475 21,280 US VANF but not the UK TABLE 10: Comparison of the comparative age (time since first registration) of specific sub groups. (See TABLE DA 1.4 in Data Annex 1 for duplicate active agent analysis).

(Subgroups are entities that were licensed (or subsidised) in one country but not in another)

Tables 11 and 12 show that for most subgroup comparisons, the differences in the median time since first registration (comparative age) were statistically significant at the 0.01 level.

Licensing of Subgroups

 The differences in comparative age were greatest for entities licensed in New Zealand but not licensed in the US compared with entities licensed in the US but not licensed in New Zealand.  The differences in comparative age were smallest for entities licensed in the UK but not licensed in the US compared with entities licensed in the US but not licensed in the UK.

Subsidy of Subgroups

 The differences in comparative age were greatest for entities subsidised in New Zealand but not subsidised in Australia compared with entities subsidised in Australia but not subsidised in New Zealand.  The smallest statistically significant differences in comparative age were between entities subsidised in the UK but not subsidised in the US VANF compared with entities subsidised in the US VANF but not in the UK.  There were no statistically significant differences in comparative age between entities subsidised in Australia but not in the US VANF and entities subsidised in the United States VANF but not in Australia.

Time since first registration of licensed entities collapsed by ATC code Licensed in NZ but 8,740 Licensed in 5,464 3,276 <0.0001 not in Australia Australia but not in NZ Licensed in NZ but 1,0161 Licensed in the UK 4,962 5199 <0.0001 not In The UK But not in NZ Licensed in NZ but 9,954 Licensed in the US 4,770.5 5,184 <0.0001 not In US but not in NZ Licensed in Australia 9,133 Licensed in the UK 6,478 2,655 <0.0001 but not in the UK but not in Australia Licensed in Australia 8,782 Licensed in the US 5,500.5 3,282 <0.0001 but not in US but not in Australia Licensed in the UK 8,657 Licensed in the US 6,940 1,717 0.0140 but not in the US But not in the UK TABLE 11: Differences in the comparative age (median time since first registration) of subgroups of licensed entities. (See TABLE DA 1.5 in Data Annex 1 for duplicate active agent analysis).

(Subgroups are entities licensed in one country but not in another)

Time since first registration of subsidised entities collapsed by ATC code Subsidised in NZ but 11,587 Subsidised in Australia 4,694 6,893 <0.0001 not in Australia but not in NZ

Subsidised in NZ but 10,879 Subsidised in the UK but 5,017 5,862 <0.0001 not in the UK not in NZ Subsidised in NZ but 10,663 Subsidised in the US 6,008 4,655 <0.0001 not in the US VANF VANF but not in NZ

Subsidised in 9,880 Subsidised in the UK but 6,478 3,402 0.0002 Australia but not in not in Australia the UK Subsidised in 6,871 Subsidised In The US 7,550 -679 0.8162 Australia but not in VANF But not in the US VANF Australia Subsidised in the UK 7,550 Subsidised in the US 8,452 -902 0.0015 but not in the US VANF but not in the UK VANF TABLE 12: Differences in the comparative age (median time since first registration) of subgroups of subsidised entities. (See TABLE DA 1.6 in Data Annex 1 for duplicate active agent analysis).

(Subgroups are entities subsidised in one country but not in another)

3.2.4. Level of Innovation of Licensed and Subsidised Products

3.2.4.1: Comparing Number of ATC Level 4 Groups And Number of Entities Per Level 4 Group

The number of ATC Level four groups that contain entities licensed and subsidised in each country are shown in Table 13.

The United Kingdom had the highest number of level 4 groups that contain entities licensed for adults, followed by the United States, Australia and then New Zealand. The United Kingdom also had the highest number of level 4 groups that contain entities subsidised for adults, followed by the United States VANF, Australia and New Zealand. The United Kingdom had the highest proportion of groups with subsidised agents to groups with licensed agents, followed by Australia, New Zealand and the United States VANF.

Country Number of level 4 Number of level 4 Percentage of level 4 groups groups that contain groups that contain that contain subsidised entities licensed for entities subsidised entities to Level 4 groups adults for adults that contain licensed entities New Zealand 386 285 73.8% Australia 392 293 74.7% United Kingdom 443 443 100% United States 420 296 (VANF) 70.5% TABLE 13: Number of ATC level 4 groups that contain licensed and subsidised entities in each country

There were statistically significant differences in the pattern of level four groups that contain licensed entities between countries (UK>US>Aus>NZ, Chi Squared test, p<0.001).

There were statistically significant differences in the pattern of level four groups that contain subsidised entities (UK>US VANF>Aus>NZ, Chi Squared test, p<0.001).

On the other hand, there were only small differences in the median number of entities per level 4 group. (These comparisons are included in Tables DA 2.3 and DA 2.4 in Data Annex II for completeness).

The only statistically significant differences in the median numbers of entities per group were the differences between the numbers of entities subsidised per group in:

 New Zealand versus United Kingdom (median difference one entity, p<0.001)  The United Kingdom versus the United States VANF (median difference one entity, p<0.001)

Therefore, the differences in the total number of entities licensed and subsidised in the four countries (Section 3.2.1) and the time since first registration of entities licensed and subsidised in the four countries (Section 3.2.3) are not solely due to multiple entities being licensed and subsidised in the same ATC Level 4 groups.

3.2.4.2: Differences Between Countries in the Number of Innovative Medicines That Provide Important Health Gains

Table 14 below summarises the licensing and subsidy status of the 65 Innovative medicines that provide important health gains in each country in 2007.

Country Number licensed Number % of licensed in 2007 subsidised in medicines that 2007 were subsidised New Zealand 39 19 48.7% Australia 49 30 61.2% United Kingdom 58 58 100% United States 64 20 (VANF) 31.3% TABLE 14: Summary of the 2007 licensing and subsidy status of 65 innovative medicines that provide important health gains

The United States had the highest number of the innovative medicines licensed, followed by Australia, the United Kingdom and New Zealand. The United Kingdom had the highest number of medicines subsidised, followed by Australia, the United States VANF and New Zealand. The United Kingdom had the highest proportion of subsidised medicines to licensed medicines, followed by Australia, New Zealand and the United States VANF.

Eleven of the medicines were newly licensed in New Zealand between the end of 2007 and April 2010, and eight were newly licensed in Australia in the same period. One of the medicines was newly subsidised in New Zealand between the end of 2007 and April 2010, ten were newly subsidised in Australia in the same period, and four were newly subsidised in the United States VANF in that period.

Eight of the medicines that were not subsidised under the Pharmaceutical Benefits Schedule in Australia (agalsidase alfa, agalsidase beta, idursulfase, alglucosidase alfa, galsulfase, miglustat, laronidase and imiglucerase) were subsidised separately under the Life Saving Drugs Program. Imiglucerase, agalsidase alfa, agalsidase beta and laronidase were funded in 2007. Galsulfase and idursulfase gained funding in 2008, miglustat in 2009 and alglucosidase alfa in 2010.

The detailed licensing and subsidy status of the 65 innovative medicines in each country is listed in Table DA 2.5 in Data Annex II.

3.2.5: Sensitivity Analysis: Removing Entities Only Available In Injectable Formulations

Tables 15 below show the number of licensed and subsidised entities in each country, when entities only available in injectable formulations have been removed.

Each country had between 21.7% to 30.8% of licensed entities, and between 18.9% to 24.8% of subsidised entities available only in injectable formulations.

Entities available in Total Percentage of entities non-injectable number of available in non- formulations entities injectable formulations Licensed New Zealand 529 765 69.2% Subsidised New Zealand 408 503 81.1% Licensed Australia 611 879 69.5% Subsidised Australia 458 567 80.8% Licensed United Kingdom 756 1020 74.1% Subsidised United Kingdom 752 1015 74.1% Licensed United States 719 918 78.3% Subsidised United States 380 505 75.2% VANF TABLE 15: Percentage of entities available in non-injectable formulations (collapsed by ATC code). (See TABLE DA 1.7 in Data Annex 1 for duplicate active agent analysis).

The numbers of licensed and subsidised entities in each ATC level 1 group in that were available in non-injectable formulations are listed in Tables DA 2.6 and DA 2.7 in Data Annex II.

The percentages of licensed and subsidised entities in each ATC level 1 group that were available in non-injectable formulations (as a proportion of all licensed and subsidised entities in that Level 1 group were similar in each country. 92

The median time since first registration of entities of licensed and subsidised available in non- injectable formulations compared to the median time since first registration of all entities are shown in Table DA 2.8 in Data Annex II. (See also Table DA 1.8 in Data Annex I for the duplicate analysis by active agent). There were no statistically significant differences in any country.

Overall, the sensitivity analysis shows that entities only available in injectable formulations (and therefore any listing differences in these entities between countries) did not influence the results.

3.3: Chapter Three Discussion

3.3.1: Summary of Results

The results show that there are differences between the countries in the number, “comparative age” (time since first registration), and level of innovation of the entities licensed in the four countries and subsidised under the four systems.

The United Kingdom had the highest number of licensed entities (Section 3.2.1) and the highest number of ATC level 4 groups that contained licensed entities (Section 3.2.4), followed by the United States, Australia and New Zealand. The United Kingdom had the highest number of subsidised entities and the highest number of ATC level 4 groups that contained subsidised entities, followed by Australia, the United States VANF and New Zealand.

The four countries were similar in the types of medicines that were licensed and subsidised at ATC levels 1 and 2 (Section 3.2.2). They were also similar in the number of licensed and subsidised entities in the classes of medicines used to treat diabetes, cardiovascular disease, cancer, lung disease and mental illness.

New Zealand had the oldest licensed entities (highest median time since first registration), followed by Australia, the United Kingdom and the United States. (Section 3.2.3). The median time since first registration of subsidised entities was highest in New Zealand (i.e. New Zealand had the oldest subsidised entities), followed by the United States VANF, Australia and the United Kingdom.

The United States has the highest number of innovative medicines that provided important health gain that were licensed, followed by Australia, the United Kingdom and New Zealand. The United Kingdom had the highest number of such medicines that were subsidised, followed by Australia, the United States VANF and New Zealand.

In summary, the United Kingdom had the most entities both licensed and subsidised, the newest subsidised entities (shortest time since first registration), and the most innovative medicines subsidised. The United States had the newest licensed entities (shortest time since first registration), and the most innovative medicines licensed. New Zealand had the least licensed and subsidised entities, the oldest licensed and subsidised entities (longest time since first registration), and the smallest number of licensed and subsidised innovative medicines.

3.3.2 Reasons For Differences Between Countries, And Their Implications

Both drug manufacturers and regulatory agencies have an impact on the number of medicines licensed in a country, and the time for a drug to be licensed in a particular country. Drug manufacturers launch fewer products and launch these products later in smaller markets, especially those markets with tighter price controls where manufacturers expect a lower price (23). The study by Danzon et al. (see Section 1.1.4) found that of 85 new chemical entities launched between 1994 and 1998, 73 were licensed in the US, 64 in the United Kingdom, 43 in Australia and 28 in New Zealand. The median delay before a drug was launched in a market was inversely correlated with the number of launches (23). As manufacturers make the decision on when to enter a market (and hence on when to submit a dossier to the regulator for licensing), they can have significant impact both on the number of medicines licensing and the time to licensing.

Rawson reported in 2000 that the time taken for a drug to be approved by regulators also differs between countries. The median time for drug approval (the difference between the date an application was made and the approval being granted) was greater in Australia (526 days) than in the UK (308 days) and the US (369 days) (30). These factors may affect the number and time since first registration of entities licensed in a particular country, as well as the number of innovative medicines that provide health gains (202).

Countries may also differ in their pharmaceutical subsidy policies. This can affect the number of subsidised entities available. New Zealand and Australia use a positive list system for subsidised pharmaceuticals, where a product has to be individually evaluated and approved for subsidy. The 95

same is true for the United States VANF (60, 182, 188). In contrast, the United Kingdom uses a negative list system, where a product has to be specifically excluded from being subsidised before subsidy will be denied (185). This means that the subsidy system in the United Kingdom is potentially much more open, and can subsidise a greater range of entities than the other three formularies. This is a possible reason why the United Kingdom had more subsidised entities in this study, including newer entities and innovative medicines that provide health gain.

However, local organisations known as Primary Care Trusts (PCTs) control the funding for pharmaceuticals in the United Kingdom and decide spending priorities for their patient populations (18). This means that variations in funding can occur, and treatments that may be funded in one region/locality may not be funded in another. The National Institute for Health and Clinical Excellence (NICE) was set up in 1999 to eliminate these regional variations by providing guidance on the clinical and cost effectiveness of treatments (203). NICE only reviews a small number of the entities potentially available, chiefly those which are expensive, controversial and/or may have a major impact on patients’ health (18). While PCTs by law have to fund all treatments that have received a positive recommendation from NICE , the situation with treatments that have not been reviewed by NICE (including alternative indications of entities that have received a positive indication) is less clear (18). The UK Department of Health advises that the lack of a positive recommendation from NICE (or the fact that a treatment has not been referred to NICE) is not grounds for refusing funding. In the absence of NICE guidance, PCTs are expected to make evidence based decisions on a case-by-case basis (204).

This means that the UK subsidy system allows patients access to a large potential pool of treatments. In theory there is nothing preventing PCTs even from funding treatments that have received a negative recommendation from NICE, though this may be difficult to justify if the PCT later has a funding shortfall. In the countries with nationwide positive list systems, the potential pool of treatments is smaller, but there is more certainty about what treatments will be available and under what criteria.

In addition, the use of a negative formulary, along with the fact that funding for pharmaceuticals is held locally rather than centrally, may make cost containment more difficult in the UK. NICE,

which does not have any responsibility for managing pharmaceutical budgets, approved all ten of the treatments that its own analysis had found to be the least cost effective between 1996 and 2005. Of these, only six were approved in Australia, and five in New Zealand. The estimated cost per quality adjusted life year of the treatments only approved in the UK ranged from £23,000 to £54,000 (205).

At the other end of the scale, New Zealand has the smallest number of entities subsidised, with the longest lag time since first registration for subsidised entities, and the smallest number of innovative medicines subsidised. The pool of potential drug treatments is therefore smaller than in the other countries. One reason for this may be the constraints of the New Zealand pharmaceutical system.

The New Zealand Pharmaceutical Management Agency (PHARMAC) was established in 1993 and operates under the New Zealand Public Health And Disability Act 2000 (206). The Act requires PHARMAC to “to secure for eligible people in need of pharmaceuticals, the best health outcomes that are reasonably achievable from pharmaceutical treatment and from within the amount of funding provided” (emphasis added). This means that PHARMAC cannot, by law, ever exceed its allocated budget.

To achieve this requirement, PHARMAC manages the New Zealand Pharmaceutical Schedule (including listing or delisting items, and setting the conditions under which items on the schedule will be subsidised), manages the budget for pharmaceuticals, and negotiates the prices of pharmaceuticals with suppliers (62). None of the other subsidy systems in this study have one agency responsible for all of these functions.

PHARMAC uses a number of criteria in making decisions about changes to the Schedule. These include the health needs of all eligible people in New Zealand, the suitability of existing medicines and medical devices, the impact on both the pharmaceutical budget and the overall health budget, and the cost effectiveness of using pharmaceuticals rather than other health and disability services (64). This means that medicines that only benefit a small number of people, newer medicines that offer an incremental advantage over existing treatments, medicines with a

high cost per quality adjusted life year and medicines that are less cost effective than using other health or disability services may not be added to the Schedule.

The fact that PHARMAC subsidised the smallest number of innovative entities (which could potentially have significant benefits to their intended patient populations) may be of concern to patients and policy makers. This may cause or reinforce concerns that New Zealand is ‘falling behind’ other countries, and needs to be examined further (see Investigation Three).

The New Zealand pharmaceutical system offers good potential for pharmaceutical cost containment. PHARMAC uses methods such as reference pricing within therapeutic subgroups, competitive tendering, exclusive supply arrangements for medicines on the Schedule and risk sharing with manufacturers to keep pharmaceutical costs down. New Zealand’s pharmaceutical spending has increased by only 2% between 1994 and 2008, whereas overall health spending rose by 7.2% (2). However, as the present study shows, the trade-off for this is fewer and older medicines being subsidised than in the other systems.

3.3.3: Strengths And Limitations of The Study

The present study was comprehensive, and examined the full range of prescription medicines listed in commonly used prescribing reference sources in each country. The analysis was not limited to particular subgroups or newly licensed or subsidised medicines. All medicines were classified using active agents and the ATC system. The latter is a system developed specifically for medicine usage comparisons. The results were compared by both systems to ensure validity, and were similar in all cases. The results were also validated with a sensitivity analysis that removed all injectable only products. No previous studies were identified that compared the full range of licensed and subsidised medicines by their time since first registration in this group of countries.

In addition to comparing the total number of entities in each country licensed and subsidised in each country (which can be potentially confounded by multiple and very similar ‘me-too’ agents that enter the market), the present study compares the number of ATC Level 4 groups that 98

contain licensed and subsidised entities. This analysis reduces the potential confounding effect of ‘me-too’ agents, as each group that may contain multiple ‘me-too’ agents is only counted once. This analysis also takes into account the fact licensing or subsidizing a few agents in an ATC 4 level 4 class, for example a limited number of HMG Co-A reductase inhibitors (statins), may be suited to a country’s health needs. PHARMAC, for example, has to work within a capped medicines budget, and limiting the number of statins funded at any one time is one way of maximizing the number of patients who are eligible for statins (172).

A limitation of the present study was that it depended on the primary data sources (which list the medicines marketed in a particular country) providing almost complete lists of the medicines licensed in that country. All four of the primary sources in turn depended on information submitted by pharmaceutical manufacturers (179, 184, 187, 207). Some products which have been recently licensed in 2007, but not been marketed, may not have been included. In addition, as listing incurs a fee, older products that are off patent and orphan drug products may not have been included in such sources.

It should also be noted that the fact that a product is licensed in one country but not in another may reflect decisions by the manufacturer rather than by the regulatory agency. The manufacturer may decide not to submit an application in a given country based on an assessment of the potential market (or lack thereof), or because experience in other markets has raised safety or efficacy concerns.

The results for subsidised products that were only recently licensed when the data were collected may have been affected by the different timelines from licensing to subsidy in the different systems. The NHS automatically subsidises products licensed in the United Kingdom unless the product is listed on the negative list or receives a negative evaluation from NICE (18). In the systems that use positive lists (such as the PBS in Australia), the manufacturer must submit a dossier to the subsidy agency, a listing decision must be made, and price negotiations must be carried out before the product is eligible for subsidy (18). This difference in the timeline between licensing and subsidy may therefore favour the NHS, but this advantage may be diminished as new licensed products in the other countries go through the subsidy process.

In addition, the positive lists systems (the VANF, PBS and PHARMAC) have, by definition, a definitive list of the products that are subsidised at any given time (18, 88). This is not the case in the United Kingdom, where any licensed product is theoretically subsidised unless banned from subsidy (18). The present study uses a negative recommendation from NICE as a proxy for not being subsidised in the UK. However, regional Primary Care Trusts make the final decisions on which products are subsidised in their region (18). A product that is rejected by NICE may therefore in fact be subsidised. The number of subsidised products in the UK may therefore be higher than shown in the present study.

In addition, the present study was limited to prescription products. As the regulator in each country makes an independent decision about the legal classification of medicines, some products that are prescription items in one country (and hence would have been included in the dataset for that country) may be non-prescription products in another country. For example, many insulin formulations are not prescription products in the United States.

3.4: Conclusion

The results of the present study show that there are differences between the countries in the number, time since first registration, and level of innovation of the entities licensed in the four countries and subsidised under the four systems.

The United Kingdom had the most licensed and subsidised entities, the newest subsidised entities, and the most innovative medicines subsidised. The United States had the newest licensed entities, and the most innovative medicines licensed.

New Zealand had the fewest licensed and subsidised entities, the oldest licensed and subsidised entities, and the smallest number of licensed and subsidised innovative medicines.

100

CHAPTER 4: THE LICENSING AND SUBSIDY OF MEDICINES FOR CHILDREN UNDER NATIONWIDE SINGLE PAYER SYSTEMS IN NEW ZEALAND, AUSTRALIA, THE UNITED KINGDOM AND THE UNITED STATES

4.1. Context

The licensing of medicines in New Zealand, Australia, the United Kingdom and the United States has been described in Section 1.2.2. The funding of medicines by New Zealand’s PHARMAC, Australia’s PBS, the United Kingdom NHS and the United States VANF has been described in Section 1.2.3. Access problems for children and the policies designed to tackle these have been described in Section 1.3.1. The differences in these licensing and funding processes may lead to differences in the medicines licensed and subsidised under a single payer system.

A number of hypotheses (see Section 1.7.2.2) were tested:

2.1 New Zealand licensed fewer medicines, fewer types of medicines, older medicines (greater time from earliest registration date) and fewer medicines in formulations suitable for children for children than Australia, the United Kingdom and the United States in 2007.

101

4.2 Results

4.2.1 Number Of Licensed And Subsidised Entities

4.2.1.1 Total Number of Entities Licensed and Subsidised For Children Within Each Country

The number of entities that were licensed and subsidised for children in each country are listed in Table 16.

102

Licensed In Subsidised In Percentage p Value (Chi2 Country Children Children Subsidised In test, 2x2 table) Children New Zealand 443a 312b 70.4% <0.001 Australia 461a 269b 58.4% <0.001 United Kingdom 570a 566b 99.3% <0.001 United States 333a 189 (VANF)b 56.8% <0.001

2 a Chi test p value <0.001 for each pair of countries (2x2 tables). 2 b Chi test p value <0.001 for each pair of countries (2x2 tables)

TABLE 16: Number of entities licensed and subsidised for children, collapsed by ATC code. (See TABLE DA 1.9 in Data Annex 1 for duplicate active agent analysis).

The United Kingdom had the highest number of entities licensed for children, followed by Australia, New Zealand and then the United States. The United Kingdom also had the highest number of entities subsidised for children, followed by New Zealand, Australia and the United States VANF. The United Kingdom had the highest proportion of subsidised entities to licensed entities, followed by New Zealand, Australia and the United States VANF.

There were statistically significant differences in the patterns of entities licensed for children between countries (UK>Aus>NZ>US, Chi Squared test, p<0.001).

There were statistically significant differences in the patterns of entities subsidised for children between countries (UK>NZ>Aus>US VANF, Chi Squared test, p<0.001).

103

4.2.1.2 Comparison of Entities Licensed and Subsidised for Children and Adults within Each Country

The numbers of entities licensed and subsidised for children and adults in each country are shown in Table 17.

New Zealand had the highest proportion of entities licensed in children compared with those licensed in adults, followed by the United Kingdom, Australia and then the United States. New Zealand had the highest proportion of entities subsidised in children compared with those subsidised in adults, followed by the United Kingdom, Australia and then the United States VANF.

There were statistically significant differences in the patterns of entities licensed for children and those licensed for adults within Australia and the US (adults>children, Chi Squared test, p <0.01). However, no differences were detected in the patterns within New Zealand and the UK.

There were statistically significant differences in the patterns of entities subsidised for children and those subsidised for adults within each country (adults>children, Chi Squared test, p<0.001).

104

Licensed by ATC Code Country Children Adults Percentage p Value (Chi2 test, 2x2 table)

New Zealand 443 765 57.9% 0.37 Australia 461 879 52.4% <0.001 United Kingdom 570 1020 55.9% 0.04 United States 333 918 36.3% <0.001 Subsidised by ATC Code New Zealand 312 503 62.0% <0.001 Australia 269 567 47.4% <0.001 United Kingdom 566 1015 55.8% <0.001 United States 189 (VANF) 505 (VANF) 37.4% <0.001 TABLE 17: Comparison of entities licensed and subsidised for children and adults within each country. (See TABLE DA 1.10 in Data Annex 1 for duplicate active agent analysis).

105

4.2.2: Types of Licensed and Subsidised Entities

4.2.2.1: Types of Entities Licensed and Subsidised for Children

Table 18 below shows the types of entities licensed and subsidised for children at ATC level 1.

The types of entities licensed for children were similar in the four countries. Systemic anti- infectives (75 to 125 entities) were the most commonly licensed in each country, and nervous system agents (37 to 87 entities) were among the three most commonly licensed groups in each country. Anti-neoplastic and immunomodulating agents (35 to 44 entities) were among the five most commonly licensed groups in New Zealand, Australia and the UK. Alimentary tract and metabolism agents (25 to 53 entities) were among the five most commonly licensed groups in Australia, the UK and the UK, and respiratory tract agents (32 to 38 entities) were among the five most commonly licensed groups in New Zealand, Australia and the US. Blood and blood forming agents (10 to 19 entities), anti-parasitic agents (8 to 13 entities), genito-urinary and sex hormone agents (5 to 6 entities in all countries except the UK, where 22 were licensed), and systemic hormonal agents (9 to 8 entities) were the least commonly licensed for children.

Systemic anti-infectives (54 to 107 entities), nervous system agents (24 to 85 entities), and antineoplastic and immunomodulating agents (14 to 42 entities) were among the five most commonly subsidised groups in all four countries, and alimentary tract and metabolism agents (21 to 53 entities) were among the five most commonly subsidised groups in New Zealand, Australia and the United Kingdom. Genito-urinary and sex hormones (3 to 8 entities in all countries except the UK, where 22 were subsidised), and anti-parasitic agents (2 to 11 entities) were relatively unlikely to be subsidised.

106

Ranking NZ Lic NZ Sub Aus Lic Aus Sub UK Lic UK Sub US Lic US VANF Sub

1 J:111 N:55 J:125 J:61 J:107 J:107 J:75 J:54 2 N:71 J:54 N:81 N:52 N:85 N:85 R:38 N:24 3 L:43 L:36 L:35 L:22 A:53 A:53 N:37 R:15 4 R:34 C:27 A:33 A:21 D:44 D:44 D:30 L:14 5 C:29 A:26 R:32 R:21 L:44 L:42 A:25 C:13 6 A:28 R:22 V:25 C:19 R:43 R:42 S:24 D:12 7 D:24 S:19 M:23 M:15 S:39 S:39 L:22 A:11 8 S:24 D:18 C:21 D:14 C:36 C:36 C:20 S:11 9 M:22 M:15 S:21 H:12 M:26 M:26 M:13 B:7 10 V:18 H:13 D:20 S:11 V:23 V:23 P:13 M:7 11 H:14 V:11 H:14 B:7 G:22 G:22 V:12 V:7 12 B:11 G:8 P:13 P:5 B:19 B:18 B:10 P:6 13 P:8 B:6 B:12 V:5 H:18 H:18 H:9 H:5 14 G:6 P:2 G:6 G:4 P:11 P:11 G:5 G:3 TABLE 18: Number of entities licensed and subsidised for children at ATC level 1 (Please refer to ATC Code Master Key ATC Level 1)

107

Table 19 below shows the types of entities licensed and subsidised for children at ATC level 2.

The most commonly licensed types of entities were very similar in all four countries. Systemic antibacterials (27 to 51 entities) and vaccines (20 to 28 entities) were among the five most commonly licensed groups in all four countries. Antineoplastic agents (19 to 26 entities) and psycholeptics (19 in each country) were among the five most commonly licensed groups in New Zealand, Australia and the UK, while systemic antivirals (20 to 27 entities) were among the five most commonly licensed groups in Australia, the UK and US.

Systemic antibacterials (18 to 45 entities) were the most commonly subsidised entities in all four countries, while systemic antivirals (16 to 23 entities) were among the five most commonly subsidised entities in all four countries. Opthalmologicals (11 to 31 entities) were among the five most commonly subsidised entities in New Zealand, the UK and the US. Antineoplastic agents (24 to 29 entities) were among the five most commonly subsidised agents in New Zealand and the UK, and drugs for obstructive airways diseases (11 to 15 entities) were among the five most commonly subsidised entities in Australia and the United States VANF.

Ranking NZ Lic NZ Sub Aus Lic Aus Sub UK Lic UK Sub US Lic US VANF Sub 1 J01:50 J01:30 J01:51 J01:33 J01:51 J01:45 J01:27 J01:18 2 L01:26 L01:24 J07:28 J05:20 J07:28 S01:31 S01:23 J05:18 3 J07:25 N05:18 J05:27 N03:15 J05:27 L01:29 J05:20 J07:14 4 S01:21 J05:16 L01:19 R03:15 L01:19 J05:23 J07:20 R03:11 5 N05:19 S01:16 N05:19 N05:13 N05:19 J07:23 R03:20 S01:11 6 J05:18 R03:14 R03:19 L01:12 R03:19 N05:21 L01:12 L01:8 7 R03:16 D07:13 S01:19 N06:12 S01:19 R03:20 N03:11 N05:7 8 D07:14 N03:11 N01:16 S01:10 N01:16 N02:17 L04:9 L04:6 9 N01:13 N02:10 N03:15 M01:8 N03:15 G03:16 N06:9 N03:6 10 N03:12 H02:8 N02:13 N02:8 N02:13 A10:15 D07:8 M03:5 TABLE 19: Number of entities licensed and subsidised for children at ATC level 2 (Please refer to ATC Code Master Key ATC Level 2)

108

There are similarities between the types of entities most commonly licensed for adults and children at ATC level 1. Systemic anti-infectives, nervous system agents and anti-neoplastic and immunomodulating agents were among the five most commonly licensed types of entities in both adults and children. However, cardiovascular system agents and alimentary tract and metabolism agents were consistently in the five most commonly licensed groups in adults, which was not the case in children. Anti-parasitic agents and systemic hormonal preparations were relatively uncommon in both adults and children.

The types of entities subsidised at ATC level 1 were also similar. Nervous system agents, systemic anti-infective agents, anti-neoplastic and immunomodulating agents, and alimentary tract agents were the most common types subsidised in both adults and children, while antiparasitic agents and systemic hormonal preparations were relatively uncommon.

There are similarities as well as differences in licensing for adults and children at ATC level 2. Systemic antibacterials, antineoplastic agents and psycholeptics were among the most commonly licensed for both adults and children in New Zealand, Australia and the UK, while systemic antibacterials and systemic antivirals were among the most commonly licensed for both adults and children in the US. Vaccines (J07) were more commonly licensed for children than for adults in all countries.

Subsidy at ATC level 2 also showed both similarities and differences. Systemic antibacterials, antineoplastic agents, opthalmologicals, psycholeptics were in the top five subsidised groups for adults for all four countries. Only systemic antibacterials were in the top five subsidised groups for children in all four countries, while opthalmologicals were in the top five in New Zealand, the UK and the US. Psycholeptics were only among the five most commonly types of entities subsidised for children in New Zealand and Australia. Conversely, systemic antivirals were among the five most commonly subsidised groups for children in all four countries, but this was only the case in Australia and the United States VANF for adults.

109

4.2.3: Time Since First Registration of Licensed and Subsidised Entities

4.2.3.1: Time Since First Registration of Licensed and Subsidised Entities for Entire Countries

This section reports the time since first registration (or comparative age) of the entities that are licensed and subsidised for children in each country. The time since first registration of an entity is the time since the earliest registration of that entity in any of the four countries (New Zealand, Australia, the United Kingdom or the United States). The greater the time since first registration, the “older” an entity is.

The distributions of the time since first registration of licensed and subsidised entities were plotted as histograms. The distribution plots showed that the time since first registration of licensed or subsidised entities for each country were not normally distributed when collapsed by active agent or ATC Code.

110

NZ AUS

60 60

Frequency Frequency

0 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 25000 (max) days (max) days

UK US

Frequency Frequency

20 10

0 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 25000 days (max) days

FIGURE 3: Comparative age (time since first registration) of licensed entities collapsed by ATC code.

111

NZ AUS

50 2.0e-04

1.5e-04

Density

Frequency

1.0e-04

5.0e-05

0 0 5000 10000 15000 20000 25000 0 5000 10000 15000 20000 25000 (max) days (max) days

UK US VANF

20 60

Frequency

20 5

0 0 0 5000 10000 15000 20000 0 5000 10000 15000 20000 25000 (max) days days

FIGURE 4: Comparative age (time since first registration) of subsidised entities collapsed by ATC code.

112

There are significant differences in the comparative age (time since first registration) of both licensed and subsidised entities. In both cases, the probability that the medians were equal is <0.001 (Kruskal-Wallis test).

Table 20 below shows the median comparative ages of licensed and subsidised entities when collapsed by ATC code.

Summary of times since first registration of entities licensed for children collapsed by ATC code (a) New Zealand 10372a 5915 13879 557 25162 24605 Australia 9426a 5212 13879 599 25162 24563 United Kingdom 9148a 5212 13879 557 25162 24605 United States 7393a 4064 12667 326 24927 24601 Summary of times since first registration of entities subsidised for children collapsed by ATC code (b) New Zealand 12049b 7779 13879 1607 25162 23555 Australia 10581b 5905 13879 1007 25162 24155 United Kingdom 9148b 5212 13879 557 25162 24605 United States 8315b 5037 13411 326 20731 20405 (VANF) a The only statistically significant difference in time since first registration of licensed entities was between entities licensed in the United States and every other country (Wilcoxon Rank Sum test, p <0.01 in each instance). b The only statistically significant difference in time since first registration of subsidised entities was between entities subsidised in New Zealand and subsidised in the United Kingdom (Wilcoxon Rank Sum test, p <0.001), and between entities subsidised in New Zealand and the United States VANF (Wilcoxon Rank Sum test, p <0.001).

TABLE 20: Comparisons of the comparative age (times since first registration) of entities licensed and subsidised for children in the four countries. (See TABLE DA 1.11 in Data Annex 1 for duplicate active agent analysis).

(Kruskal-Wallis: Probability that a or b are equal between countries is <0.001)

113

Licensing:

 The greatest significant difference was between entities licensed in the United States and those licensed in New Zealand (2979 days).  The smallest significant difference was between entities licensed in the United States and those licensed in the United Kingdom (1755 days).

Subsidy:  The difference between entities subsidised in New Zealand and the United States VANF was 3734 days.  The difference between entities subsidised in New Zealand and the UK was 2901 days.

114

4.2.3.2: Time Since First Registration of Licensed and Subsidised Products When Comparing Specific Subgroups

Table 21 shows the parameters of the time since first registration of specific subgroups, i.e. entities that were licensed (or subsidised) in one country but not in another. Tables 22 and 23 summarise the statistical comparison (Wilcoxon Rank Sum test) of these results.

115

Median 25th Percentile 75th Percentile Min Max Range (Days) (Days) (Days) (Days) (Days) (Days) Time from first registration of licensed entities collapsed by ATC code Licensed in NZ but 10,933 6,745 13,879 557 21,053 20,496 not in AUS Licensed in NZ but 10,161 5,753 13,879 1,271 20,275 19,004 not in the UK Licensed in NZ but 11,601 7,142 13,879 1,271 25,162 23,891 not in the US VANF Licensed in AUS but 8,417 4,353 13,879 599 24,927 24,328 not in NZ Licensed in AUS but 10,038 5,530 13,879 599 22,071 21,472 not in the UK Licensed in AUS but 10,271 5,832 13,879 599 25,162 24,563 not in the US VANF Licensed in the UK 7,511 4,245 12,345 650 24,927 24,277 but not in NZ Licensed in the UK 9,073 5,480 13,837 557 24,248 23,691 but not in AUS Licensed in the UK 9,783 6,027 13,879 1,146 25,162 24,016 but not in US VANF Licensed in the US 6,250 3,751 12,400 326 24,927 24,601 VANF but not in NZ Licensed in the US 6,606 3,691 12,318 326 23,659 23,333 VANF but not in AUS Licensed in the US 7,403 3,923 13,221 326 23,659 23,333 VANF but not the UK

116

Median 25th Percentile 75th Percentile Min Max Range (Days) (Days) (Days) (Days) (Days) (Days) Time from first registration of subsidised entities collapsed by ATC code Subsidised in NZ but 12,423 8,603 13,879 1,607 21,591 19,984 not in AUS Subsidised in NZ but 10,788 7,004 13,879 1,607 20,579 18,972 not in the UK Subsidised in NZ but 12,461 8,173 13,879 1,607 25,162 23,555 not in the US VANF Subsidised in AUS but 8,936 4,310 13,879 1,007 22,071 21,064 not in NZ Subsidised in AUS but 10,217 5,015 13,879 1,007 22,071 21,064 not in the UK Subsidised in AUS but 11,057 5,922 13,879 1,146 25,162 24,016 not in the US VANF Subsidised in the UK 7,138 4,071 11,085 557 24,927 24,370 but not in NZ Subsidised in the UK 8,628 4,729 13,073 557 24,927 24,370 but not in AUS Subsidised in the UK 9,164 5,373 13,879 650 25,162 24,512 but not in US VANF Subsidised in the US 7,403 4,634 11,085 326 20,065 19,739 VANF but not in NZ Subsidised in the US 7,683 5,149 12,400 326 19,085 18,759 VANF but not in AUS Subsidised in the US 7,739 5,218 13,879 326 19,834 19,508 VANF but not the UK TABLE 21: Comparison of the comparative age of subgroups of entities licensed and subsidised for children in the four countries. (See TABLE DA 1.12 in Data Annex 1 for duplicate active agent analysis).

(Subgroups are entities licensed or subsidised in one country, but not in another)

117

Licensing of Subgroups

 The differences in comparative age (median time since first registration) were greatest for entities licensed in the US and not licensed in New Zealand compared with entities licensed in New Zealand and not licensed in the US (whether collapsed by active agent or ATC code). (See Table 22).  The differences in comparative age for entities licensed in New Zealand but not licensed in Australia and those licensed in Australia but not in New Zealand was statistically significant when collapsed by ATC code, but not when collapsed by active agent.  The differences in comparative age for entities licensed in New Zealand but not licensed in the UK and those licensed in the UK but not in New Zealand was statistically significant when collapsed by ATC code, but not when collapsed by active agent.

Subsidy of Subgroups

 The difference in comparative age is greatest for entities subsidised in the United States VANF but not subsidised in New Zealand compared with entities subsidised in the United States VANF but not subsidised in New Zealand. (See Table 23).  The differences in comparative age is smallest for entities subsidised in New Zealand but not subsidised in Australia compared with entities subsidised in Australia but not subsidised in New Zealand. (Interestingly, the difference in comparative age is greatest for entities subsidised for adults in New Zealand but not subsidised in Australia compared with entities subsidised in Australia but not subsidised in New Zealand). (See Section 3.2.3.2).

118

Time since first registration of licensed entities collapsed by ATC code Licensed in NZ 10,933 Licensed in 8417 2,516 0.006 but not in Australia but Australia not in NZ Licensed in NZ 10,161 Licensed in the 7,511 2,650 <0.001 but not In The UK But not in UK NZ Licensed in NZ 11601 Licensed in the 6250 5,351 <0.001 but not In US US but not in NZ Licensed in 10,038 Licensed in the 9073 965 0.176 Australia but UK but not in not in the UK Australia Licensed in 10,271 Licensed in the 6,606 3,665 <0.001 Australia but US but not in not in US Australia Licensed in the 9783 Licensed in the 7403 2,380 0.001 UK but not in US But not in the US the UK TABLE 22: Differences in the comparative age (median time since first registration) of subgroups of entities licensed for children. (See TABLE DA 1.13 in Data Annex 1 for duplicate active agent analysis).

(Subgroups are entities licensed in one country but not in another)

119

Time since first registration of subsidised entities collapsed by ATC code Sub group 1 Median sub Sub group 2 Median sub Difference in P value group 1 group 2 median (Wilcoxon (Days) (Days) (Days) Rank Sum) Subsidised in 12,423 Subsidised in 8,936 3,487 <0.001 NZ but not in Australia but not Australia in NZ

Subsidised in 10,788 Subsidised in the 7,138 3,650 <0.001 NZ but not in the UK but not in UK NZ Subsidised in NZ 12,461 Subsidised in the 7,403 5,058 <0.001 but not in the US US VANF but VANF not in NZ

Subsidised in 10,217 Subsidised in the 8628 1,589 0.065 Australia but not UK but not in in the UK Australia

Subsidised in 11,057 Subsidised In 7,683 3,374 0.003 Australia but not The US VANF in the US VANF But not in Australia Subsidised in the 9,164 Subsidised in the 7739 1,425 0.626 UK but not in US VANF but the US VANF not in the UK

TABLE 23: Differences in the comparative age (median time since first registration) of subgroups of subsidised entities, (See TABLE DA 1.14 in Data Annex 1 for duplicate active agent analysis).

(Subgroups are entities subsidised in one country but not in another)

120

4.2.3.3: Comparing Time Since First Registration in Children with that in Adults

Table 24 below compares the comparative age (time since first registration) of entities licensed and subsidised in children with those licensed and subsidised in adults.

The United States was the only country where there was no statistically significant difference in comparative age between entities licensed for children and those licensed for adults. In every other country entities licensed for children were older (had been registered longer) than those licensed for adults.

The VANF was the only system where there was no statistically significant difference in comparative age between entities subsidised for children and those subsidised for adults. In every other system, entities subsidised for children were older (had been registered longer) than those subsidised for adults.

121

Country Median Median Difference in P value children adults medians (Wilcoxon (Days) (Days) (Days) Rank Sum test) Times since first registration of licensed entities collapsed by ATC code New Zealand 10372 8936 1436 <0.005 Australia 9426 7795 1631 <0.001 United Kingdom 9148 7319 1829 <0.001 United States 7393 6607 786 0.059 Times since first registration of subsidised entities collapsed by ATC code New Zealand 12049 10724 1325 <0.01 Australia 10581 8065 2516 <0.001 United Kingdom 9148 7319 1829 <0.001 United States (VANF) 8315 8203 112 0.544 TABLE 24: Comparison of median time since first registration of entities licensed and subsidised for children with those licensed and subsidised for adults in each country. (See TABLE DA 1.15 in Data Annex 1 for duplicate active agent analysis).

122

4.2.4. Lowest Age Authorised

The distributions of the lowest age authorised (the minimum age for which the entity is authorised for use in children) of licensed and subsidised entities in each country were plotted as histograms. The lowest age(s) authorised were not normally distributed.

Tables 25 below show the differences in the lowest age authorised of licensed and subsidised entities within each country.

There is no statistically significant difference in the lowest age authorised of licensed and subsidised entities within any country, whether collapsed by active agent or ATC code (Wilcoxon Rank Sum test).

There are statistically significant differences in the lowest age authorised of licensed entities between countries (Kruskal-Wallis test, p<0.01). Entities were authorised for use in much younger children in New Zealand, Australia and the United Kingdom than in the United States.

There are statistically significant differences in the lowest age authorised of subsidised entities between countries (Kruskal-Wallis test, p<0.01). Entities were authorised for use in much younger children in New Zealand, Australia and the United Kingdom than in the United States VANF. Entities were also authorised for use in younger children in New Zealand than in Australia, though the difference was only a month.

123

Country Median Median Difference P Value licensed for subsidised for (Months) (Wilcoxon children children Rank Sum test) (Months) (a) (Months) (b) New Zealand 0a 0b 0 0.587 Australia 1a 1b 0 0.516 United Kingdom 0a 0b 0 0.987 United States 24a 12 (VANF)a 12 0.131

a Differences between the United States and all other countries were statistically significant (p<0.001, Wilcoxon Rank Sum test). b Differences between the United States VANF and all other countries were statistically significant (p<0.001, Wilcoxon Rank Sum test). Differences between New Zealand and Australia were statistically significant (p<0.001, Wilcoxon Rank Sum test).

TABLE 25: Lowest age authorised within each country for licensed and subsidised entities collapsed by ATC Code. (See TABLE DA 1.16 in Data Annex 1 for duplicate active agent analysis).

(Kruskal-Wallis test probability (a) or (b) equal between countries <0.001)

124

4.2.5: Availability of Suitable Formulations

4.2.5.1: Number of Suitably Formulated Licensed and Subsidised Entities

Table 26 below shows the numbers and proportions of suitably formulated licensed and subsidised entities in each country.

Licensing

 The United Kingdom has the highest number of suitably formulated licensed entities, followed by New Zealand, Australia and the United States (both when collapsed by active agent and ATC Code).  The United Kingdom has the highest proportion of suitably formulated licensed entities, followed by New Zealand, the United States and Australia when collapsed by active agent.  New Zealand had the highest proportion of suitably formulated licensed entities, followed by the United Kingdom, the United States and Australia when collapsed by ATC Code.

Subsidy

 The United Kingdom had the highest number of suitably formulated subsidised entities, followed by New Zealand, Australia and the United States VANF (both when collapsed by active agent and ATC code).  The United Kingdom had the highest proportion of suitably formulated subsidised entities, followed by the United States VANF, New Zealand and Australia.

125

Country Suitably Total number of % of entities that formulated entities are suitably entities formulated Licensed by ATC code New Zealand 351 443 79.2% Australia 335 461 72.7% United Kingdom 448 570 78.6% United States 252 333 75.7% Subsidised by ATC code New Zealand 232 312 74.4% Australia 176 269 65.4% United Kingdom 455 566 80.4% United States 148 189 78.3% TABLE 26: Number and proportions of suitably formulated licensed and subsidised entities in each country. (See TABLE DA 1.17 in Data Annex 1 for duplicate active agent analysis).

126

4.2.5.2: Difference In Lowest Age Authorised of Suitably Formulated and Unsuitably Formulated Licensed and Subsidised Entities

Table 27 below shows the differences in the lowest age authorised of suitably formulated and unsuitably formulated entities.

Licensed Entities

 There is a statistically significant difference between the lowest ages authorised of licensed entities available in suitable formulations and those which are not available in suitable formulations (Wilcoxon Rank Sum test, p<0.001) in each country.  The lowest age authorised of suitably formulated entities was 0 months in New Zealand and the UK, and 12 months in the US, whether collapsed by active agent or ATC Code. The lowest age authorised in Australia was 1.75 months when collapsed by active agent and 0 months when collapsed by ATC Code.  The lowest age authorised of unsuitably formulated entities was 72 months in the US and 60 months in Australia (whether collapsed by active agent or ATC Code). In the United Kingdom, this was also 60 months collapsed by active agent, but 36 months collapsed by ATC code. In New Zealand, this was 24 months when collapsed by active agent, and only 6 months when collapsed by ATC Code.  The greatest difference in lowest age authorised of suitably formulated and unsuitably formulated entities was in the United States (60 months both when collapsed by active agent and ATC Code), closely followed by Australia (58.25 months when collapsed by active agent and 60 months when collapsed by ATC Code), and the United Kingdom (60 months collapsed by active agent and 36 months collapsed by ATC Code). The smallest difference is in New Zealand (24 months collapsed by active agent and 6 months collapsed by ATC Code).

Subsidised Entities

127

 There is a statistically significant difference between the lowest ages authorised of subsidised entities available in suitable formulations and those which are not available in suitable formulations (Wilcoxon Rank Sum test, p<0.001) in each country.  The lowest age authorised of suitably formulated entities was 0 months in New Zealand, Australia and the United Kingdom, and 6 months in the US VANF (whether collapsed by active agent or ATC Code).  The lowest age authorised of unsuitably formulated entities was 72 months in the US VANF and 36 months in Australia, whether collapsed by active agent or ATC Code. In the United Kingdom, this was 60 months when collapsed by active agent and 42 months when collapsed by ATC Code. In New Zealand, this was only 6 months when collapsed by active agent and 3.75 months when collapsed by ATC Code.  The greatest difference in lowest age authorised of suitably formulated and unsuitably formulated entities was 66 months in the United States VANF (whether collapsed by active agent or ATC code). The smallest difference was in New Zealand, 6 months when collapsed by active agent and 3.75 months when collapsed by ATC Code.

128

Country Lowest age Lowest age Difference P value authorised suitably authorised non- (Months) (Wilcoxon formulated entities suitably Rank Sum (Months) formulated test) entities (Months) Licensed by ATC code New Zealand 0 6 -6 <0.001 Australia 0 60 -60 <0.001 United Kingdom 0 36 -36 <0.001 United States 12 72 -60 <0.001 Subsidised by ATC code New Zealand 0 3.75 -3.75 <0.001 Australia 0 36 -36 <0.001 United Kingdom 0 42 -42 <0.001 United States 6 (VANF) 72 (VANF) -66 <0.001 TABLE 27: Difference in lowest age authorised of suitably formulated and non-suitably formulated entities. (See TABLE DA 1.18 in Data Annex 1 for duplicate active agent analysis).

129

4.3: Chapter Discussion

4.3.1: Results Summary

The United States has fewer licensed entities listed for children than the other countries, but licenses more newly registered entities. The United States is the only country where there is no difference in the age (time since first registration) of entities licensed in adults and those licensed in children. Entities licensed for children in the United States were newer (had been registered for a shorter period) than those licensed for children in the other countries. The later the year of first registration, the more likely entities were to be licensed in children in the United States. On the negative side, entities licensed for children in the United States were authorised for use in children who were older than in the other countries, so not licensed for younger children. The United States also has the highest proportion of licensed entities that were suitably formulated for children.

The United Kingdom had the highest number of entities licensed for children and the highest number of licensed entities that were suitably formulated for children. The United Kingdom had the most entities subsidised for children, and the most entities subsidised for children that were suitably formulated.

Entities subsidised for children in New Zealand were older than those subsidised for children in the United States VANF or the UK. The later the year of first registration, the less likely an entity is to be subsidised in New Zealand. However, New Zealand had more entities subsidised for children, had more subsidised entities that were suitably formulated, and had subsidised entities that were on average authorised for use in younger children than either Australia or the United States VANF.

130

4.3.2: Reasons For Differences Between The Countries

The reasons that the United States licenses fewer entities in total for children, but licenses more newly registered entities and licenses these in older children may be due to the incentives offered by the United States Best Pharmaceuticals For Children Act (57). The Act offers manufacturers a six-month patent extension on their products if the manufacturer carries out trials in children. The extension applies regardless of whether the trials lead to the product being licensed in children, and regardless of the age group in which the trial is carried out (57). This favours testing newer products that are still on patent in children, as the patent extension will benefit the sales of these products for adults as well as for children. It also favours carrying out trials in older children, as it is easier to test in this age group than in newborns or toddlers.

On the other hand, the US may have fewer products licensed in children because testing in children as a result of the Best Pharmaceuticals for Children Act revealed safety issues that led to the product being contra-indicated in children. Alternately, as newly registered entities are licensed in children, older products licensed in children for the same indication may have been taken off the market.

The United Kingdom’s subsidy system may be the reason the UK has more entities subsidised for children, and subsidises more entities suitably formulated for children than the other countries. The UK automatically subsidises branded medicines on market launch, subject only to a small negative list of medicines that are not to be subsidised, and the evaluation of a small number of medicines for cost effectiveness by the National Institute for Health and Clinical Excellence (NICE) (18).

New Zealand’s PHARMAC is obliged to operate from a capped budget, and must consider the impact of its listing decisions on both the pharmaceutical budget and the overall health budget, as well as the effectiveness of existing medicines and other therapies (64). This means that PHARMAC is less likely to subsidise newer products that are still on patent, and are not subject to generic competition. This is a possible reason why products subsidised for children in New Zealand had been registered longer than those subsidised in the UK and the United States VANF.

131

However, one of PHARMAC’s decision making criteria is to consider the health needs of all New Zealanders in its listing decisions (64). This means that the needs of groups such as children must be considered. This is a possible reason why New Zealand had more entities subsidised for children, had more subsidised entities that were suitably formulated, and had subsidised entities that were on average authorised for use in younger children than either Australia or the United States VANF.

4.3.3: Strengths and Weaknesses

In most respects, the strengths and weaknesses of the present study are identical to those in Investigation One. In particular, potential confounding issues with regards to manufacturers deciding to submit licensing dossiers to different regulators at different times, and differences in the timeline to subsidy in different systems apply to children as well as to adults (see Section 3.3.3).

Another potential confounding issue with subsidised medicines is unique to children. A product that is licensed for adults but not for children may still be subsidised for children when prescribed outside its official licensed indication. The reason for this is that the formularies do not specify age groups for each subsidised indication (180, 182, 185, 188). Therefore, unless subsidy for children is specifically barred by the formulary, the product may still be subsidised when prescribed for children.

Investigation Two examined two aspects of access that are specific to children, the availability of suitable formulations and the lowest age that the entities were authorised for use in. The availability of suitable formulations allows for accurate dosing, and minimizes risks such as choking. It also avoids the need to modify the dose form by methods such as crushing or extemporaneous compounding (which can alter physiochemical and dose release properties). It also avoids the need to administer the dose form by a route other than the one intended (for example a solution for injection vial the oral route), which can alter absorption. Children are not a homogenous group, and having products only authorised in an older group of children (as is the case in the United States), means that younger children are disadvantaged, as there is no prescribing information in this group. (If there is no prescribing

132

information and no suitable formulation available, younger children could be doubly disadvantaged).

4. 4. Conclusion

The countries differ in the number of entities licensed and subsidised for children, the age (time since first registration) of entities, the lowest age authorised and the availability of suitable formulations.

133

CHAPTER FIVE: KEY INFORMANTS’ PERCEPTIONS OF NEW ZEALANDERS’ ACCESS TO MEDICINES

5.1 Context

The controversy around New Zealanders’ access to medicines has been discussed in Section 1.6. The current investigation examines key informants’ perceptions of New Zealanders’ access to medicines, both in isolation and in comparison to other countries.

Two hypotheses (see Section 1.7.2.2) were tested:

4.1 Any differences in access (number, time, type and comparative age (time from first registration) and level of innovation) to prescription medicines for adults between New Zealand, Australia, the United Kingdom, and the United States will be of no concern to policy makers, researchers, and others with detailed knowledge of the New Zealand pharmaceutical system.

4.2 Any differences in access (number, time, type and comparative age (time from first registration)) to prescription medicines for children between New Zealand, Australia, the United Kingdom, and the United States will be of no concern to policy makers, researchers, and others with detailed knowledge of the New Zealand pharmaceutical system.

The methods have been discussed in Section 2.3.

5.2 Results

The results are divided into three categories and themes are discussed under those categories: areas where there was widespread agreement on what New Zealand does well, areas where there was widespread agreement on what New Zealand could do better, and areas where there was disagreement. As the respondents were not a statistically representative sample, the third section does not compare the numbers of respondents on each side of disagreement, but rather summarises the views expressed.

134

5.2.1 Widespread Agreement: What New Zealand Does Well

5.2.1.1 The overall process of licensing medicines

The majority of participants (eleven of the twenty) praised the Medsafe licensing process, with none of the remainder having a negative view of the system. The respondents who praised the Medsafe system came from across the spectrum, including pharmacists, medical practitioners, industry representatives, public servants and political figures.

Among the strengths of the Medsafe licensing system informants identified were very strong systems and processes. Informants commented that the Medsafe system was “robust”, “followed international standards”, was “better than most”, and had a “mature and rigorous” process. The system ensured the safety and efficacy of medicines, without presenting unreasonable barriers (in terms of either time or cost) in getting products on to the New Zealand market.

Another strength identified was the relationship between Medsafe and the pharmaceutical industry. The terms most commonly used to describe the relationship were “pragmatic” and “professional”. It was not described as either being hostile (thus compromising the ability to work together) or too friendly (thus compromising Medsafe’s objectivity). One informant did raise the concern that it would be better for a system designed to safeguard the public to be entirely publicly funded, rather than being funded by industry user fees. However, the informant stated this was not unique to New Zealand, and also occurred in other countries.

Other strengths identified were New Zealand’s medicine’s classification system, and post- marketing surveillance of medicines. New Zealand’s four level medicine classification system- prescription medicines, pharmacist only medicines (which can only be sold by a pharmacist), pharmacy medicines (which can only be sold in a pharmacy) and general sales medicines (which can be sold in any retail outlet)-was thought to maximise public access while maintaining appropriate controls on medicines.

135

Post marketing surveillance was seen as a particular strength, with one informant claiming New Zealand had the highest rate of adverse effect reporting from general practice and hospitals in the world. (Another described this system as “a national treasure”). A strong reporting system was seen as important because medicines were tested in small cohorts of people before launch both in New Zealand and internationally, as few as 600 patient years for long-term medicines. In addition, patterns of prescribing and use of medicines in New Zealand could be very different to the rest of the world. An example that was given was that of third generation oral contraceptives. At the time the increased risk of pulmonary embolism with these medicines was suspected, third generation contraceptives had an 80% market share in New Zealand, compared with 10% internationally.

A potential problem that was identified with Medsafe’s licensing system was the limited pool of expertise in New Zealand, especially with highly specialised areas like large biological molecules. This was identified by informants in both the pharmaceutical industry and public service. For the industry, this caused problems in getting products to market. (A particular concern that was raised was the assessment of “bio-similar” drugs, the “generic” version of patented biological medicines). For Medsafe, the lack of people specialised in this area not only limited the number of assessments, but caused problems with the peer review of assessors. A related concern raised by a medical practitioner was that the limited pool of expertise in New Zealand could lead to the same people being appointed to the same roles over and over again, possibly compromising objectivity.

5.2.1.2 PHARMAC Extracts The Maximum Buying Power From A Limited Budget

Seventeen of the twenty respondents said that the amount of funding available for medicines in New Zealand was limited. This response crossed all sectors, and came from medical practitioners, pharmacists, academics, public servants, Members of Parliament and people working in the pharmaceutical industry. Comments included “We’re not a rich country anymore”, “We don’t have an endless supply of money”, “You just have to accept there’s a fixed budget and you can’t go above that”, “There’s a high level of acceptance in the New Zealand community that we live in a financially constrained environment”, and “We’re a small country with a limited health budget”. Respondents made the point that increasing the 136

amount of money available for medicines was not simply a case of increasing the medicines budget, as this came from DHB budgets. Increasing medicine funding without increasing total DHB funding would divert money from other services.

Fifteen of the respondents also identified current and future pressures on the limited medicines budget. These responses again crossed the range of respondents. The most commonly cited pressure was the cost of new medicines. Respondents made the point that that the potential demand was unlimited, and public expectations would always exceed what could be supplied. Among the examples of expensive new medicines respondents gave were monoclonal antibodies, new oral anticoagulants, and tyrosine kinase inhibitors [such as Gleevec®/Glivec®], of which newer versions were constantly being developed. One respondent commented, “There’s now like a son of Gleevec®, a grandson of Gleevec®, which are even more expensive”. Another commented that twenty percent of products currently consume eighty percent of the medicines budget, but if costs kept rising, the budget would not be enough even to cover that twenty percent. Other pressures on the medicines budget respondents identified were the existence of budgetary silos or ring fenced budgets within the health system (which can encourage cost shifting from one area to another, and increase overall costs), lower fees for doctor visits and prescriptions (which increases the volume of prescriptions) and shifting medicines previously sold over the counter on to the Pharmaceutical Schedule.

Sixteen of the respondents said that PHARMAC had a strong focus on cost containment, and kept medicine costs under control. This response again crossed all sectors. (None of the other respondents said PHARMAC did not focus enough on cost containment or was profligate with spending). Respondents indicated that PHARMAC actively managed prices to get the best deals. PHARMAC did this through a robust tendering process. This included sole supply tendering, selecting a limited number of agents to subsidise in each class and not offering partial subsidies to the agents that were not selected (thus giving suppliers a strong incentive to offer the lowest price), a strong focus on generics, and using bundling deals to get new items on the Schedule on favourable terms. Examples respondents gave included fluoxetine initially costing almost three dollars per capsule but now costing five cents, and the upcoming change from branded atorvastatin (Lipitor®) to generic atorvastatin potentially saving tens of millions of dollars.

137

While respondents agreed that PHARMAC kept prices down and contained expenditure, there was disagreement over what this meant in practice. Some respondents felt this was a bonus for patients and the health system, freeing up money for other uses. Respondents in this group expressed views like “We probably get the biggest bang for our buck of any country in the world”, “The PHARMAC model has succeeded in what it was intended to do, which is fund within a particular budget and maintain an increased access to funded medicines”, “PHARMAC’s an outstanding organisation internationally in terms of being able to squeeze the most value out of the budget that it’s allocated” and “A lot of other countries look at our PHARMAC regime with envy but the battle with the drug companies is quite a big one for them to take on”. Other respondents believed that PHARMAC’s focus on staying within a fixed budget meant passing up opportunities for investments that could deliver greater health benefits. People in this group spoke of “win-win” opportunities that could benefit both manufacturers and the public if there was greater willingness to spend. An example that was given was patient access schemes such as those in use in the UK, where manufacturers and the health system share the cost of a drug that would otherwise be too expensive.

5.2.1.3 The Relationship Between PHARMAC and the Pharmaceutical Industry Is Improving

Roughly half (eleven) of the respondents identified a good or improving relationship between PHARMAC and the pharmaceutical industry. This view came from a broad range of respondents, including medical practitioners, pharmacists, public servants, people in the pharmaceutical industry and Members of Parliament. While other respondents identified tensions or areas of disagreement, none cited a dysfunctional or deteriorating relationship. The general view of respondents is that while there were major differences, both were able to work together in a professional manner. Respondents cited several reasons for this.

Respondents, including public servants, people working in the pharmaceutical industry and a Member of Parliament, identified a political consensus around PHARMAC. Respondents stated that PHARMAC’s decision-making processes were free from political involvement, PHARMAC’s existence was not under threat, and its structure was not likely to undergo radical change when the Government changed. (Respondents did identify very rare cases where the Government over-rode PHARMAC’s decisions, and this is discussed in more detail 138

later). A key reason given for this support was PHARMAC’s success in managing the medicines budget, and the value this had to the Government regardless of who was in power. One respondent commented, “You hate it in opposition, and you love it when you go on the Treasury bench”.

This political stability in turn has consequences for the relationship between PHARMAC and the industry. Respondents indicated that now that its existence is not under threat, PHARMAC is less focused on defending its position, and is making greater efforts to be conciliatory. Respondents also indicated that, for its part, the pharmaceutical industry has accepted that PHARMAC in its current form is here to stay, and is no longer seeking major changes to PHARMAC’s role or powers. Respondents who commented on the relationship said it was much better than it had been even a few years ago.

This does not mean there are no tensions in the relationship. Respondents working in the pharmaceutical industry have voiced frustration that PHARMAC’s strong focus on cost containment and staying within budget makes negotiations necessarily adversarial. However, respondents from both the industry and PHARMAC spoke of a commitment to working together professionally. An academic who had worked with both parties spoke of being struck by the “enormous amount of goodwill” present on all sides. While the relationship is not warm, both groups are aware of each other’s positions, and willing to work together in a pragmatic manner.

5.2.1.4 PHARMAC’s Decision Making Is Resistant to Political Influence, Lobbying or Public Pressure

As discussed above, respondents indicated a political consensus around the existence and role of PHARMAC. In addition, respondents from many sectors including public servants, members of the pharmaceutical industry and health professionals identified PHARMAC’s non-politicised and independent decision making as a strength. The key reason informants identified for this was the way PHARMAC was structured, as a statutorily independent body. This made it difficult for outside pressure to influence PHARMAC’s decisions.

139

Roughly half of the informants (eleven) identified rare cases where the Government had over- ridden PHARMAC’s decisions. All eleven cited the case of Herceptin® (trastuzumab), where the 2008 National Party led government funded 52 weeks treatment for early stage HER-2 positive breast cancer after PHARMAC had decided to fund a shorter 9-week course (3). Three of the respondents also cited the case of beta-interferons in multiple sclerosis, where the 1999 Labour Party led government directed PHARMAC to fund the treatment. There was very little support for such cases becoming routine from respondents, including health professionals, members of the pharmaceutical industry and Members of Parliament. Respondents also indicated there was likely to be little political gain from such cases in the future, as favouring one treatment risked a backlash from clinicians and patient advocates working in other areas.

Similarly, the ability of lobbying or public pressure to get one treatment funded ahead of others was limited. Respondents gave several reasons for this. Many patient groups are focused around education and support for their members rather than for campaigning. Several respondents cited diseases that had a lower profile than they should (including dementia, prostate cancer, bowel cancer and orphan diseases), which meant there was little public awareness or support for getting treatments in these areas funded. A process where lobbying or public awareness had a strong influence could see these areas falling further behind. However, an informant with insight into PHARMAC’s decision-making made it clear the committees involved expect the case for funding a particular treatment to be evidence based. This is true whether the application comes from a patient group or the pharmaceutical industry. If the evidence in the literature did not convince the committee, then “The pleas fell on deaf ears”. This was true regardless of the disease’s profile or how well organized the lobbying campaign was.

5.2.2 Widespread Agreement: What New Zealand Could Do Better

5.2.2.1 Licensing Medicines in Small Patient Groups

Eleven of the informants cited problems with licensing medicines for small patient groups. Respondents in this group included medical practitioners, pharmacists, public servants and people in the pharmaceutical industry. The two groups of medicines commonly cited were 140

those used in paediatrics, and drugs used to treat rare diseases (orphan drugs). (See Section 1.3.2: Orphan Drugs). Not having medicines licensed caused delays for patients (as supply could be unreliable), and caused additional compliance costs for doctors and hospitals in importing unlicensed medicines.

Problems licensing medicines in children were cited by nine of the respondents. Problems identified by respondents included medicines that had only been trialed in adults having to be prescribed off license in children, children being a small market and hence little incentive for manufacturers to licence products in children, and the ethical difficulties in testing medicines on children. Respondents stated these problems were not unique to New Zealand, and that the United States and European Union had successfully increased the numbers of medicines licensed for children (paediatric licensing) through giving manufacturers patent extensions as an incentive. (See Section 1.3.1: Access Problems for Children). This was not the case in New Zealand and Australia, and the drawbacks of applying for paediatric licensing outweighed any benefit for manufacturers.

Informants from both Medsafe and the Pharmaceutical industry said the drawback was not so much the cost of the initial application, which Medsafe was able to waive if necessary. Rather, it was in the detailed dossiers that had to be prepared and submitted to gain a new indication, and in the ongoing compliance costs to manufacturers. This included producing New Zealand specific packaging, updating datasheets, and updating manufacturing details with Medsafe. This was true of any low volume product, whether for paediatrics or an orphan drug. Another factor that was cited was that PHARMAC (unlike many other subsidy agencies) could subsidise unlicensed products, meaning that licensing was not a pre-requisite for subsidy.

5.2.2.2 Variation in Access to Hospital Medicines

Nine of the informants identified variations in access to hospital medicines. These respondents primary worked in health care, but also included people from the pharmaceutical industry and politics. These respondents identified several factors that affected variations between DHBs. These included the financial pressures that DHBs are under, with those that were under severe financial pressure being less likely to offer some expensive medications.

141

Another reason identified was historical use, with a DHB that had been offering a service or treatment for some time continuing it, while others decides not to offer it as a new service. Finally, DHBs evaluate the cost effectiveness of hospital medicines independently, which could lead to them making different decisions. An example that was given was the cost effectiveness of oral versus IV chemotherapy, with some DHBs opting for more expensive oral agents to avoid the costs of IV equipment and hospital visits. Respondents also pointed out that people in rural areas had less access to pharmacies and specialised hospital facilities (such as those providing chemotherapy), which could compound geographic problems in accessing medicines.

Respondents said having the Pharmaceutical Schedule facilitates equitable geographic access to medicines, and that in the past moving medicines onto the Pharmaceutical Schedule had reduced geographic variation in access. The most commonly cited example was the moving of cancer medicines on to the national cancer basket managed by PHARMAC, along with access to newer antipsychotic medicines in schizophrenia, and expensive anti-nausea drugs. Respondents believed PHARMAC taking a greater role in the management of hospital medicines would improve consistency in access to hospital medicines. It would also reduce duplication in cost effectiveness analyses, which now has to be done by each DHB individually.

However, respondents cited several concerns about PHARMAC taking over the management of hospital medicines. The first was about prescribers’ flexibility to use agents that were not listed by PHARMAC. This included concerns about delays if prescribers had to apply for authorisation, limits on treatment options for patients who did not respond to the first line agents, and flexibility to treat rare or one-off cases. A second was whether equalising access across the country would mean DHBs that currently had better than average access losing out. This was especially the case for large tertiary hospitals that were at ‘the end of the line’ for patients who could not be treated elsewhere. Another concern was whether DHBs would actually provide all the medicines PHARMAC selected.

142

5.2.2.3 Not All New Zealanders Can Access Licensed and Subsidised Medicines

Ten of the respondents pointed out that some groups of New Zealanders are less able to access licensed and subsidised medicines. The groups identified were those in lower socioeconomic groups (which included elderly people on fixed incomes), and Māori and Pacific Island peoples. (It was pointed out that Māori and Pacific Island people were disproportionately in the lower socio-economic groups). Respondents identified several barriers to people in these groups accessing medicines.

Cost was the most common barrier identified. This went beyond the prescription fee, and included the cost of seeing a general practitioner, the cost of visiting a specialist for medicines that needed specialist approval, the cost of having a telephone and having transport. Several respondents questioned whether universally lowering prescription and doctors’ fees was the right approach. While this benefited those who were already accessing primary care, the lowered fees could still be a barrier to people who were on limited incomes. Other barriers respondents identified were patient motivation to visit doctors and manage chronic diseases, cultural barriers that prevented patients from discussing some problems with doctors, and prescribers’ beliefs about an individual’s adherence that prevented these groups from being prescribed certain medicines.

Respondents believed that these barriers resulted in poorer access to medicines in these groups. Examples that were cited were lower prescribing rates of statins in Māori and Pacific Islanders, Māori being less likely to receive palliative therapy in lung and bowel cancer, and lower prescribing rates of isotretinoin for lower socioeconomic groups. (Respondents also cited poorer access to revascularisation procedures among Māori and Pacific Islanders, and undiagnosed high cholesterol and insulin resistance among Maori). The consequences were higher morbidity and mortality, and higher long-term health costs due to complications and hospitalisations.

143

5.2.3 Areas of Disagreement

5.2.3.1 Whether PHARMAC’s Decision Making Processes Need to Change

A major area of disagreement was over PHARMAC’s decision-making. Many respondents praised the process as being robust, evidence based and following set procedures. However, other respondents had concerns or criticisms about the way PHARMAC functioned. As mentioned in the methods, this section does not count how many respondents were for and against (as only small numbers commented on particular aspects, and the sample was not designed to be statistically representative), but rather summarises the views expressed.

The most common criticism of PHARMAC’s processes was that the decision making process was not transparent enough. This was cited by respondents from medicine, pharmacy, politics and the pharmaceutical industry. These respondents felt that the information on how PHARMAC came to its decisions was limited, as only the final decision and sometimes minutes of the deliberation were published. In particular, information on which of PHARMAC’s decision-making criteria were considered, and how much weight was placed on each criterion were not readily accessible. Respondents contrasted this with the United Kingdom’s National Institute for Health and Clinical Excellence (NICE), where the full cost effectiveness modeling was published online, along with lay and technical summaries. Respondents believed greater transparency would increase public confidence in the process, and help achieve democratic accountability. However, respondents also cautioned that transparency had to be balanced against the commercial secrecy of PHARMAC’s negotiations with manufacturers. Opening up the process could undermine PHARMAC’s ability to negotiate the best prices with manufacturers.

There was also disagreement on whether PHARMAC’s role in conducting cost effectiveness evaluations should have greater separation from its budget management role. The common concern cited here was that the knowledge of what could be afforded influenced the cost effectiveness analysis. (Respondents contrasted the PHARMAC process with those in Australia and Canada, where cost effectiveness analysis and price negotiation were delegated to different bodies). Alternatives that were suggested were having a set cost effectiveness threshold above which medicines would not be funded (thus giving manufacturers a strong 144

incentive to set prices accordingly), and having a set maximum time by which medicines that had been judged a high priority would be funded.

However, other respondents pointed out that separating budget management and cost effectiveness analysis would simply shift difficult decisions elsewhere, as someone would still have to decide if a medicine could be afforded or not. If a body made binding decisions about whether a medicine should be funded without having responsibility for deciding how it would be paid for, those responsible for implementing the decision would have to make cuts elsewhere. The example that was given was the conflict between the United Kingdom’s NICE and Primary Care Trusts (which manage budgets). The Primary Care Trusts faced major opportunity costs because of NICE’s binding decisions, and had to cut more cost effective services to stay within budget. Lastly, respondents said that separating cost effectiveness analysis from budget management was counter-intuitive (and very rare outside health), as those responsible for paying for goods and services typically evaluate the value of what they buy before buying it.

5.2.3.2 Whether Current Spending In Medicines Meets NZ’s Health Needs

Many respondents (including medical practitioners, pharmacists, Members of Parliament and academics) expressed the view that New Zealand’s current spending purchased the most effective medicines for New Zealanders’ health needs. At the same time, PHARMAC’s process screened out medicines that were very similar to those that were already on the market (so called “me too” drugs), those that added little value, and those that were harmful. The example cited by several respondents was COX-2 inhibitors, which were widely used worldwide before their harmful cardiovascular effects were known, but little used in New Zealand because they were not subsidised by PHARMAC.

Conversely, respondents across many sectors (including medicine, pharmacy and the pharmaceutical industry) believed New Zealand’s investment in medicines was not high enough. A common view in this group was that a capped budget by definition limited investment that would deliver greater health gains in the long term. If the budget did not grow enough to keep up with changes in technology and developments in best practice internationally, New Zealand would fall behind. Examples that were cited of areas where New

145

Zealand was not keeping up included new oncology medicines (monoclonal antibodies and targeted DNA therapy), rare cancer indications, HIV treatments and treatments for rare diseases. Respondents in this group also said that PHARMAC could be more assertive in negotiating a higher medicines budget from DHBs.

A view where there was actually very little disagreement (it was expressed by people working in medicine, pharmacy, politics, patient advocacy, public service and the pharmaceutical industry) was that medicines were assessed much more stringently for cost effectiveness than other DHB-funded investments. This was despite the fact that pharmaceuticals were only a small percentage of health spending. Examples respondents gave of investments that did not undergo the same scrutiny were dialysis, operations, transplants, radiation therapy, vaccines and devices. The argument respondents made was not for assessing medicines less stringently, but rather applying more scrutiny to other investments to make sure they were cost-effective. If pharmaceutical and non-pharmaceutical treatments were assessed on the same basis, this could potentially free up more money for medicines. However, respondents cautioned there were practical difficulties in applying this scrutiny to non-pharmaceutical treatments, such as lack of clinical trial data.

5.3 Chapter Discussion

5.3.1 Results Summary

The respondents believed New Zealand had an effective medicines licensing process, which ensured the safety and quality of medicines without causing needless delays or costs for manufacturers. PHARMAC extracted the maximum buying power from New Zealand’s medicines budget, and did this through a process that was resistant to political pressure, lobbying or public pressure. A political consensus exists around the existence and role of PHARMAC, and the relationship between PHARMAC and the pharmaceutical industry is improving.

However, there are problems with equity of access to medicines. This includes geographic inequalities in access to hospital medicines, problems in accessing medicines for lower socioeconomic groups, Māori and Pacific Islanders, and problems licensing medicines in

146

children and small patient groups. There is disagreement about the current level of investment in medicines, and over whether PHARMAC’s processes need to change.

5.3.2 Comparison With Past Findings

PHARMAC’s success in containing New Zealand’s pharmaceutical spending and negotiating favourable prices on medicines has been well documented (2). The present findings show that people working in New Zealand’s medicines system are well aware of this. The perceptions that PHARMAC is resistant to lobbying, political pressure and public opinion is consistent with PHARMAC’s history of being willing to take unpopular decisions, and defending those decision in the face of adverse publicity (3).

The finding that a political consensus exists around PHARMAC is not surprising given both PHARMAC’s history and New Zealand’s recent political history. Since its founding in 1993, PHARMAC has operated under governments led by both major parties. These governments have also included virtually all parties represented in the New Zealand Parliament (spanning the full political spectrum) as coalition partners. In that time, PHARMAC’s role has been entrenched in New Zealand law by the 2000 Public Health and Disability Act (passed under a centre-left Labour Party led government) and is now being expanded under the current centre- right National Party led government to include hospital medicines (177, 206). The finding that there are currently geographic variations in access to hospital medicines (and that that this would be reduced by centralising hospital access under PHARMAC) suggests the hospital sector would support this latest initiative, as long as concerns about prescribing flexibility are addressed.

The finding that relationships between PHARMAC and the pharmaceutical industry are improving is new, and to the best of our knowledge has not been reported by any other study. This improvement may be related to the fact that PHARMAC is not under political threat. In the past, the pharmaceutical industry has challenged PHARMAC’s role in the courts, but has not been successful in curbing PHARMAC’s powers (2). If PHARMAC’s role is not likely to be curbed either politically or legally, the rational course for the pharmaceutical industry is to work within the political reality in New Zealand. For its part, PHARMAC would gain nothing from disturbing the political consensus by engaging in conflict with the industry.

147

5.3.3 Limitations and Future Work

The finding that medicines are assessed much more stringently than other health technologies is an interesting one. Assessing other health technologies in the same way as medicines has been done in other countries, for example by the United Kingdom’s NICE. Further work would need to done to determine if this was feasible and economically viable in New Zealand, and what if any value this would add to New Zealand’s health system. As there was strong disagreement about New Zealand’s level of investment in medicines, work would also need to be done to determine the effects of current spending on patient outcomes.

An important limitation of this study is that it measured subjective perceptions, rather than objective endpoints. Informants’ perceptions may or may not fully reflect the reality of the situation. In addition, as the study was not anonymous, respondents may have censored themselves from giving responses they felt were not socially acceptable, or which would have jeopardised their own or their employers’ interests. However, many of those who took part have roles where they are expected to make public, identifiable statements on social or policy issues.

5.4 Conclusion

New Zealand has an effective and safe system for licensing medicines, though there are problems in licensing medicines for small patient groups. PHARMAC’s purchasing system contains expenditure well, and makes decisions in a way that is resistant to lobbying and public pressure. There is political consensus in New Zealand around the role of PHARMAC, and an improving relationship between PHARMAC and the pharmaceutical industry. However, there is still considerable disagreement around New Zealand’s level of investment in new medicines.

148

CHAPTER 6: INTERPRETATION

6.1 Implications for New Zealanders’ Access to Medicines

Access is a multi-dimensional concept, and several different factors need to be considered when evaluating New Zealanders’ access to medicines. As previously noted, Wonder (2006) and Cohen et al. (2007) define access in terms of:

The results of the three investigations in this thesis, combined with other work done in this area can be used together to evaluate New Zealanders’ access to medicines.

6.1.1: Breadth

Our findings (Investigations One and Two) show that the breadth of entities licensed in New Zealand for both adults and children is less than in many other countries. (It is worth noting though, that New Zealand had more entities licensed for children than the United States, and PHARMAC subsidised more entities for children than the Pharmaceutical Benefits Scheme and the Department of Veterans Affairs National Formulary). The reduced breadth of access extended to classes of entities, which suggests fewer therapeutic options are available in New Zealand.

One potential explanation for this is that New Zealand is a smaller pharmaceutical market than Australia, the United Kingdom or the United States. New Zealand spent $254 USD PPP per capita on pharmaceuticals in 2007, compared to $371 in the United Kingdom, $480 in Australia and $900 in the United States (164). This, combined with New Zealand’s smaller

149

population, means potential pharmaceutical sales are much lower in New Zealand. Danzon et al. have previously shown that manufacturers launch fewer pharmaceutical products in smaller markets (23).

A previous comparison that used a similar methodology to Investigations One and Two compared New Zealand to Finland (178). This comparison showed the breadth of access to be similar in the two countries, despite the fact that Finland had a slightly larger population and a much higher spending on pharmaceuticals than New Zealand ($430.5 USD PPP in 2007) (164, 178). However, this similarity in the breadth of access despite very different levels of pharmaceutical spending does not necessarily invalidate the above explanation. The comparison by Aaltonen et al. showed entities that were licensed and subsidised in Finland (but not in New Zealand) were predominantly those that were launched after 1990 (178). This again suggests fewer entities are being launched in New Zealand.

In addition to being a small pharmaceutical market, the New Zealand pharmaceutical market is also relatively isolated. New Zealand currently has no mutual medicines licensing treaties or joint regulatory licensing agreements with any other country. This means medicines registered in other countries are not automatically licensed in New Zealand. In contrast, some other countries have alternate routes for licensing medicines. In the European Union for example, manufacturers can apply for centralised registration of medicines through the European Medicines Agency (valid in all European Union countries), or mutual recognition of medicines that have been registered by the national regulatory agencies of other European Union countries (49).

In addition, New Zealand only has one publically funded pharmaceutical funding scheme (PHARMAC), and no widespread system of private insurance for pharmaceuticals, unlike the United States (25, 208). (See Section 6.1.4: Choice). With only PHARMAC to act as a buyer, the demand for a wide breadth of products might be less in New Zealand than in other countries.

Previous studies have only compared selected subsets of medicines in New Zealand with those in other countries (23, 26), and not examined the full breadth of the market. A paper resulting from this doctoral work is the first to examine the full breadth of products in New

150

Zealand, and compares New Zealand with Australia, the United Kingdom and the United States (37).

6.1.2: Depth

Depth of access in New Zealand (or in any other country) is a measure of the restrictions patients face in accessing licensed and subsidised medicines. Informants in Investigation Three noted that New Zealand’s four level medicine classification system (general sale, pharmacy only, pharmacist only and prescription only) aimed to maximise access while maintaining appropriate controls on the safe use of medicines (209). This suggests New Zealand patients do not face inappropriate legal restrictions on the depth of access to medicines.

Informants also noted that each District Health Board makes its own decisions on which hospital medicines are funded. While the conditions under which medicines listed in the Pharmaceutical Schedule are subsidised are published openly (60), this is not the case for hospital medicines. This makes the depth of access to funded hospital medicines in New Zealand difficult to evaluate. The depth of access to the same hospital medicine could also vary in different parts of New Zealand.

Investigations One and Two focused on the medicines listed in the Pharmaceutical Schedule in 2007, but did not evaluate depth of access. However, the depth of access to medicines listed in the Pharmaceutical Schedule in 2007 has already been evaluated in the comparison by Aaltonen et al.(178). This showed that 406 of the 471 entities subsidised in New Zealand (over 86%) were fully subsidised, though conditions applied for some subsidies. Furthermore, 326 entities (over 69% of all subsidised entities) were fully subsidised without any conditions or restrictions on their eligibility for subsidy (178).

This means that the vast majority of entities listed in the Pharmaceutical Schedule do not have restrictions on the depth of access. For the 21% of subsidised entities that do have restrictions, the conditions of access are published openly, and are the same for patients throughout New Zealand (60). This means that patients and their prescribers have certainty about which medicines patients are eligible for. This is similar to the Pharmaceutical Benefits Scheme in

151

Australia, where conditions of access are set out in the Schedule of Pharmaceutical Benefits (182). This certainty about depth of access compares well with some other countries.

As noted previously, “post-code prescribing” is a recognised problem in the United Kingdom’s NHS. As each PCT makes decisions on which medicines to pay for separately, the depth of access for any medicine can vary from one region to another. NICE evaluates some but not all medicine indications for cost effectiveness and budgetary impact. All PCTs are obliged to pay for all medicine indications which have had a positive evaluation by NICE (18, 80). (See Section 1.2.3.3: The United Kingdom).

There have been several recent initiatives to standardise and increase depth of patient access to expensive medicines in the UK. The UK government has created special funds for cancer treatments that have been rejected by NICE (210). An interim £50 million fund began in October 2010 and ran until April 2011, when a £200 million fund became available each year for the next three years (210). However, several concerns have been raised about these funds. The cost of funding these indications could be well over £200 million a year, and have been estimated as high as £600 million. This may leave the cancer funds unable to meet demand. Earmarking scarce health resources solely for cancer treatments (especially those which have been found not to be cost effective by NICE) undermines NICE’s role in providing evidence- based guidelines on the best use of NHS resources. It also disadvantages those needing expensive medicines for other diseases such as multiple sclerosis or dementia (210-212).

The 2009 PPRS also provides two new mechanisms for increasing access to expensive medicines (79). The first is through Patient Access Schemes, where drug manufacturers and the Department of Health can enter into agreements to improve the cost effectiveness of medicines. Such agreements can include the manufacturer offering discounts or rebates on the price of the medicine, or risk-sharing agreements based on how patients respond. The second mechanism is flexible pricing. The manufacturer lists the drug at a price that is cost effective to the NHS based on available evidence, but retains the option to raise the price if new evidence of effectiveness is developed, or if the drug is approved for a new indication. The price rise would typically depend on a NICE review of the new evidence and cost effectiveness at the new price (79).

152

However, all these initiatives are essentially interim measures, pending a total revamp of how the NHS pays for pharmaceuticals. The UK office of Fair Trading (OFT) reported in 2007 that PPRS neither made the best use of NHS resources nor sufficiently rewarded manufacturers for innovation (213). (This is particularly relevant for new innovative medicines that might offer significant improvements over existing medicines). The OFT recommended a shift to ‘value based pricing’, where the price of a drug would be negotiated based on the benefit it offered the NHS. In its 2010 White Paper “Equity and excellence: liberating the NHS”, the UK government signalled a shift to value based pricing when the current PPRS ends in 2014 (214). The role of NICE will also change as part of the reforms, though the details of the changes have yet to be finalised. The White Paper emphasised NICE’s roles in creating clinical and quality improvement guidelines. If the value a drug offers is decided at the time the price is negotiated, NICE’s role in deciding which treatments are cost effective, and under what conditions, could become less relevant. This in turn could worsen variations in depth of access across the UK (213, 215).

Depth of access is also hugely variable in insurance-based health systems such as those used in the United States. Access to pharmaceuticals (especially expensive pharmaceuticals) can vary depending on the insurance plan, which in the case of the employment based coverage held by the majority of Americans, is typically chosen by the employer rather than by the individual (25, 86). Employment based insurance is particularly valuable for people who need expensive medicines because the insurance risks (and hence premiums) are averaged across all employees and dependents in the insurance pool (86), but it also has risks. Many employees do not qualify for coverage, and as employers can switch coverage, employees may find themselves switched to a minimum coverage plan, including some with annual limits as low as $2000 (87). Employees can also find themselves ‘job locked’ (unable to change jobs without losing coverage) and coverage can be lost with the loss of a job (87). (See Section 1.2.3.4: The United States). Buying private health insurance individually offers greater choice and flexibility than employment based insurance, but is more expensive due to higher administrative costs, smaller tax benefits and lack of employer subsidies (86). In addition, pre-existing conditions can be excluded from coverage, or insurance can be denied altogether (87).

In summary, previous work has shown that the majority of medicines subsidised in New Zealand are subsidised without restriction (178). This suggests New Zealanders do not face 153

undue difficulties with depth of access, but this needs to be confirmed by further research. Restrictions faced by particular groups of patients may also differ from that faced by patients in general.

The differences in the depth of access between countries, variability within countries and the differences in availability of information all make depth a difficult dimension of access to measure. However, it is also an important dimension, as medicines could be licensed or subsidised in theory while being very difficult for patients to access. International comparisons of depth of access could be an area for future research.

6.1.3: Timeliness

Timeliness is a measure of how quickly medicines are licensed or subsidised. Our results showed that New Zealand was slower than the other countries to license medicines, and PHARMAC was slower than the other single payer systems to subsidise medicines. New Zealand’s lower pharmaceutical spending may make manufacturers launch medicines later in New Zealand than in other countries (23, 164). PHARMAC’s spending is also tightly controlled, and by law can never exceed its allocated yearly budget (206). As PHARMAC must also pay for all existing items on the Schedule from this budget, there may be limited funding available for newly launched medicines.

However, timeliness may not always be positive. A medicine may not turn out to be as effective as it was thought to be at the time it was first registered. An example is the use of bevacizumab (Avastin®) in breast cancer, which initially showed promising results, but was later shown to provide no improvement in mortality. This led to the United States Food and Drug Adminstration moving to withdraw its indication for breast cancer (216). Medicines may also have harmful effects that were not known to regulatory agencies at the time the medicines were launched. An example of this is the cyclo-oxygenase 2 (COX-2) inhibitor rofecoxib (Vioxx®), an anti-inflammatory agent that was withdrawn by its manufacturer after it was shown to increase the risk of myocardial infarction and stroke (217). Another is the anti-diabetic agent rosiglitazone (Avandia®), which was shown to increase the risk of myocardial infarction and heart failure (218).

154

For this reason, the methodology of Investigation One also included a measure of the degree of innovativeness of licensed and subsidised medicines. This section was developed from a method developed by Roughead et al.(194). Researchers have in the past considered the seriousness of the disease, the availability (or lack) of existing treatments, and the effectiveness of the treatment in determining how innovative a treatment is (219, 220). The methodology used in Investigation One includes these factors, but also favours medicines that provide an added benefit over those currently in use. This is especially true of medicines listed as a “breakthrough or substantial improvement” by the Canadian Patented Medicines Price Review Board (PMPRB). The criterion of added value is increasingly being used by public subsidy agencies (221, 222). “Breakthrough and substantial improvement” drugs are roughly equivalent to those in levels 7 and 8 (most innovative: treatment for a disease that had no previous treatment, or a new mode of treatment) on the scale used by the Austrian Medicines Evaluation Commission (HEK) or ASMR I and II (most innovative: major therapeutic advance, significant increase in efficacy, or significant reduction in adverse effects) by the French Transparency Commission (221, 222). This weights the methodology towards the most innovative drugs.

It’s worth noting that the UK system is undergoing major changes at the time of writing. In its 2010 white paper “Equity and excellence: liberating the NHS”, the UK government signaled a shift to value based pricing when the current PPRS ends in 2014. Value based pricing will base the price of each drug on the benefit it gives the NHS. The role of NICE will also change, though details have yet to be finalised (214). It will be interesting to see how the final definition of “value” used by the NHS compares to those used by Canadian PMPRB, French Transparency Commission and Austrian Medicines Evaluation Commission discussed above. The United Kingdom has a stated goal of supporting the pharmaceutical industry in its pharmaceutical policy, which may influence the definition of an innovative product (78, 223).

Investigation One found the US had the highest number of innovative entities licensed and the newest licensed entities. New Zealand had the lowest number of innovative entities licensed and the oldest licensed entities. (This may reflect both fewer new entities coming on to the New Zealand market, and older entities remaining on the New Zealand market while PHARMAC continues to subsidise these). This is understandable given that the US spends well over three times as much as New Zealand per capita on pharmaceuticals (224). The VA is only one of a number of public and private health systems in the US, which provide a large

155

pharmaceutical market (25). Cohen et al. previously found US patients have faster and more flexible access to high cost drugs than patients in the UK, though with higher cost sharing and more variable coverage (225). Similarly, US patients had faster access and more choices in coverage for top selling pharmaceuticals than patients in the UK, France and the Netherlands, but with the drawbacks of higher cost and less evenly spread coverage (225). According to Danzon et al., drug manufacturers launch more products and launch these products earlier in larger markets (23). Danzon et al. found the median delay before a drug was launched was inversely correlated with the number of launches (23). The US is a much more lucrative market for the launch of new products, including innovative products, prompting manufacturers to launch products first in the US.

Investigation Two found that the US licensed newer entities for children than any other country. It was also the only country where the timeliness of entities licensed for children was no different from that in adults. This may be due to policy initiatives in the US. Legislative incentives to increase the testing of medicines in children have been in place in the US since 1998, and so were in place for a decade when the data for this study were collected. The European Union enacted similar legislation in 2007. (See Section 1.3.1: Access Problems for Children). The incentives offered by the US and EU give manufacturers a six month patent extension on their products if they carry out testing in children (57, 58, 98). The patent extensions offered by these schemes have also been shown to increase pharmaceutical purchasing costs (226), and so may put additional pressure on health budgets. Investigation Two also showed that the US licensed fewer medicines for children and licensed these for older children than in the other countries. This was surprising in the light of the US patent extension initiatives. Further work may be needed to determine the overall benefits of these initiatives for children, and if the benefits outweigh the increased pharmaceutical purchasing costs in these health systems (115).

The findings that New Zealand was slower to license and subsidise medicines (and that New Zealand licensed and subsidised fewer innovative medicines) than the comparators is of concern. The health impact of this difference (both in general and for particular patient groups) should be examined further.

156

6.1.4: Choice

Choice is a measure of how much control patients have over their coverage. The four single payer systems examined in this thesis (PHARMAC, the Pharmaceutical Benefits Scheme, the National Health Service and the Department of Veterans Affairs Formulary) all greatly limit patient choice, as they do not offer patients any choice in coverage plans, degree of cost sharing, coverage limits or formularies (18, 20, 88). For this reason, choice was not one of the dimensions of access investigated in the thesis. However, there are features of the health systems in all four countries that affect choice.

New Zealand’s no-fault accident compensation scheme, administered by the Accident Compensation Corporation (ACC) provides support for people injured because of an accident. This includes the cost of medicines in some cases, including medicines that are not subsidised by PHARMAC (227). This provides an alternative means of accessing subsidised medicines for people who are injured in accidents, but not those suffering from illnesses (See Section 6.1.5: Equity).

New Zealand patients who need medicines that are not funded by either PHARMAC or ACC, and are not paid for in hospital, can apply for ‘exceptional circumstances’ public funding. New Zealand previously had three ‘exceptional circumstances’ schemes. (See Section 1.2.3.1: New Zealand). As noted previously, these schemes were closed to new patients on March 1st 2012. New patients will be assessed under a single Named Patient Pharmaceutical Assessment (Exceptional Circumstances) Policy, which is administered by PHARMAC (67).

Australia also has some pharmaceutical funding mechanisms outside the Pharmaceutical Benefits Scheme (PBS). (See Section 1.2.3.2 Australia). The Life Saving Drugs Program provides funded treatments (for very rare and life threatening conditions) which are not cost effective enough to be listed on the PBS. This is a potentially fairer approach than that of the NHS Cancer Fund, or New Zealand’s ACC, as access is not determined by having one specific disease or cause of injury. However, it potentially disadvantages patients whose treatments have not been considered by the PBAC (75). The Herceptin® Program also funds Herceptin® (trastuzumab) for late stage metastatic HER-2 positive breast cancer outside the

157

PBS. This has the same ethical problem as the NHS Cancer Fund (favouring one type of medicine over others), but to a greater degree, as it favours one treatment for one cancer, potentially leaving those who need other cancer treatments at a disadvantage (76). (See Section 6.1.5: Equity).

In the United Kingdom, both Patient Access Schemes and the NHS Cancer Fund potentially increase choice for patients in addition to increasing depth of access (See Section 6.1.2: Depth). The fact that funding decisions are made regionally rather than at the national level in the United Kingdom (18), also potentially increases choice. Individual patients or patient groups may be able to lobby the local Primary Care Trust (PCT) to pay for a particular treatment. Patients also have the option of relocating to a region whose PCT pays for the treatment they need. However, these options favour those with knowledge of which treatments are paid for by each PCT, have access to resources for lobbying (such as media contacts) and the ability to run a publicity campaign, and/or have the resources to relocate (See Section 6.1.5: Equity).

Potentially the greatest amount of choice comes from an insurance-based health care system such as that in the United States. (See Section 1.2.3.4: The United States). Danzon et al. showed that more new chemical entities are launched in the US than in the UK, Australia or New Zealand, and that entities are launched earlier in the US than in the other countries (23). The US also had a shorter time from drug licensing to subsidy than the UK or other European countries, along with much greater flexibility in choosing the type of coverage (22). The wide variety of plans in the US means that Americans as a whole potentially have access to plans that cover most pharmaceuticals, and cover these faster than nationwide single payer systems. However, as the majority of Americans have employment-based coverage, the degree of choice individual Americans have in their coverage may be more limited (see the discussion on employment based coverage in Section 6.1.2: Depth).

In addition, the limited public safety net in the United States means that choices can be limited for patients who do not have or lose employment based coverage. (In addition to loss of coverage through dismissal or redundancy, serious illness can create an impossible situation where the employee cannot work without further treatment, but cannot get further treatment without working). A Federal law called COBRA (Consolidated Omnibus Budget Reconciliation Act) allows those who lose employment based coverage to maintain health 158

insurance, but this protection is for a limited time, and potentially expensive (87). (See Section 1.2.3.4: The United States).

Government insurance programs covered less than a third of all Americans (25). People 65 years and above, those with permanent disabilities and patients with end-stage renal dialysis are eligible for pharmaceutical coverage under Medicare Part D (25). All Medicare Part D plans have to cover two drugs in each therapeutic class (unless only one drug is available), and have to cover all or almost all anticancer drugs, immunosupressants, drugs for HIV, anticonvulsants and antipsychotics (228). However, disabled patients need to have been receiving disability payments for two years before qualifying for Medicare, on top of a five month waiting period for disability payments (87). This can mean a long period without coverage for people in this group who need expensive pharmaceuticals, especially if they have lost employment based coverage after becoming too ill to work. The 29 month waiting period is much longer than the 18 months of coverage through COBRA (87).

Patients with little income and few assets may qualify for Medicaid, but the eligibility criteria vary by State. As noted previously, single adults and childless couples are often excluded (87, 228).(See Section 1.2.3.4: The United States). The Patient Protection and Affordable Care Act addresses this by expanding Medicaid eligibility to all Americans with an income of 133% of the Federal poverty level or below (114). However, the United States Supreme Court has ruled that States can opt out of this expansion (229).

American patients who have the least choice are the 16.7% of Americans (50.7 million people) who had no health insurance coverage at any time in 2009 (25). If people without insurance need expensive pharmaceuticals, they would face the unenviable choice of relying on charity care or paying the full cost of the medicines. The Patient Protection and Affordable Care Act requires most people without health insurance to buy a basic insurance plan by 2014 (114). This requirement was challenged in court, on the grounds that requiring individuals to buy a commercial product such as health insurance violates the United States Constitution, but this requirement has been upheld by the US Supreme Court (229).

As previously noted, choice was not an aspect of access examined in this thesis. However, the amount of choice available in New Zealand may affect the implications of the findings. New Zealand subsidises fewer medicines, is slower to subsidise new medicines and subsidises 159

fewer innovative medicines. Would New Zealanders therefore be willing to pay for alternatives to the PHARMAC system, such as private insurance for pharmaceuticals?

6.1.5: Equity

Equity of access is a broad concept, and could include evenness of coverage regardless of ability to pay, evenness of coverage regardless of geographic location, and equitable coverage across all diseases or injuries. (Cost sharing is covered separately in Section 6.1.7: Cost Sharing, but impacts on coverage regardless of ability to pay).

Investigation Three specifically investigated the question of whether all New Zealanders had equal access to licensed and subsidised medicines. One problem respondents indicated was that there was unequal access to hospital medicines, as each District Health Board (DHB) made its own decision on what medicines to pay for. This problem also occurs in Australia, where each State Government makes its own decision on what inpatient medicines to fund (71). Future work could look at the extent of ‘post-code prescribing’ of hospital medicines in these two countries, especially as PHARMAC’s role is being extended to the management of hospital medicines (230).

In the United Kingdom, the problem of ‘post-code’ prescribing extends beyond hospital medicines, because local Primary Care Trusts make their own decisions on which medicines to pay for and under what circumstances. This can cause inequity in access (18, 80). (See Section 1.2.3.3: The United Kingdom and Section 6.1.2: Depth). As the PHARMAC Schedule, Schedule of Pharmaceutical Benefits and Department of Veterans Affairs National Formulary all use central decision making, and have nationwide rules for access, evenness of coverage for outpatient medicines regardless of location should in theory be assured (18, 88). None of the respondents in Investigation Three indicated geographic unevenness of access to outpatient medicines was a problem in New Zealand.

Respondents in Investigation Three indicated that people in lower socio-economic groups, Māori and Pacific Islanders were more likely than other New Zealanders to have problems accessing licensed and subsidised medicines. This is consistent with work by other researchers. Jatrana and Crampton showed that 15.5% of New Zealanders had deferred seeing

160

a doctor because of cost, and 6.4% deferred collecting a prescription, with the risk increased for those with low or average incomes, having several deprivation characteristics and having more than two co-morbidities (231). Jatrana et al. also showed separately that Māori and Pacific Island people were more likely than European New Zealanders to defer collecting a prescription because of cost, even after adjusting for potential confounders (232). There are also differences in the rates of prescribing for different ethnic groups. Norris et al. showed that Māori are less likely to receive prescriptions for antibiotics than other New Zealanders and received smaller quantities of antibiotics (233). The prescription of psychotropic drugs also varies by ethnicity (234). The extent to which ethnicity and socioeconomic status impact on overall access to medicines is an area that merits further research.

Problems with access due to lack of ability to pay are not unique to New Zealand. In the United States, as noted previously, 50.7 million people had no health insurance coverage at any time in 2009 (25)..A significant number of Americans are also underinsured (estimated at 25 million adults aged 19-64 years in 2007), and as a result would have access problems and/or high costs(84).This problem also extends to children. There were 21 million children who relied on public coverage for health insurance in 2005, with a further 5.5 million children uninsured but eligible for public coverage (112). By 2009, the number of uninsured children was 7.5 million (25). These Americans would face difficulties accessing medicines due to their inability to pay.

(See Section 6.1.4: Choice for how the Patient Protection And Affordable Care Act expands insurance coverage in the US, which potentially reduces inequity due to ability to pay).

The findings of unequal access to hospital medicines in New Zealand, and poorer access for Maori, Pacific Islanders and lower socioeconomic groups, may impact on New Zealanders’ health outcomes. Since the time of data collection, prescription co-payments have been increased in New Zealand, and concerns have been raised about the impact this will have on equity and health outcomes (235). Conversely, PHARMAC’s increased role in managing hospital medicines across New Zealand may reduce the impact of geographic inequity.

161

6.1.6: Continuity

Continuity is a measure of whether patients have access to their medicines for as long as they need them. Continuity of access to individual medicines may be related to conditions of subsidy, such as only a limited quantity of medicine being funded or the treatment being only funded for a given period. This could also be viewed as an aspect of depth (See Section 6.1.2: Depth). As noted previously, less than 21% of PHARMAC subsidised medicines in New Zealand have restrictions on subsidy, and could potentially be affected by this aspect of continuity. As medicines subsidised by the PBS also have restrictions on subsidy, future work could compare this aspect of continuity in New Zealand versus Australia.

Continuity could also be affected by interruptions in medicines supply. PHARMAC uses sole supply tendering as one means of obtaining lower prices on subsidised medicines. This has in the past led to interruptions in medicines supply when the manufacturer has run out stock, or there has been a problem in the manufacturing process (236). In addition, patients may be forced to change medicines when PHARMAC makes subsidy changes. This may include changing from one brand of a medicine to another, or changing from one medicine in a chemical class to another (171, 174). It should be noted that problems with continuity of supply has also caused concern in other countries, for example the United States and Australia (28-31). It is therefore not a problem that is limited to New Zealand.

A third aspect of continuity is whether there are coverage limits on the total amount of medicines spending per patient, both per annum and over the patient’s lifetime. A different but related aspect of continuity is whether the patient can continue treatment if they have to change insurance provider. These are commonly a feature of insurance-based systems such as that in the United States (87). (See Section 1.2.3.4: The United States for details on how continuity can be disrupted when changing insurance providers, after losing a job or when an employer switches a minimum coverage plan). These are not a feature of the PHARMAC system.

Yet another aspect of continuity is continuity in regards to location (i.e. if the treatment would be interrupted if the patient had to relocate for any reason). As the PHARMAC schedule applies nationwide, there should in theory be no interruptions, as a patient who is eligible in

162

one part of the country would continue to be eligible in another. This would not be the case where decisions are made locally, such as in the UK. A patient could in theory find their treatment was not funded by the Primary Care Trust in the new location (See Section 6.1.2: Depth).

In summary, New Zealanders may face problems with continuity, especially if there are interruptions to the supply of a sole subsidy medicine. As noted above, this is not unique to New Zealand.

6.1.7: Cost Sharing

Cost sharing could include patient co-payments, part-charges (where the patient pays the portion of the drug cost that is not subsidised), and instances where the patient pays the full cost of the drug. In insurance-based systems, it can also include premiums, deductibles and coverage limits.

The PHARMAC system has the lowest co-payments of the four single payer systems, with most patients paying only $2.2 United States Dollars per three-month supply of a medicine, and a cap of 20 prescription charges per year per family (Table 1). Therefore, if all medicines were fully subsidised, a family would only pay US$44 in prescription co-payments per year, even if multiple family members each have multiple co-morbidities. Note that this charge will rise from January 2013 to $3.67 United States dollars (at 2007 exchange rates) per item, and $73.33 per family per year (237). Concerns have been raised about the negative effects of this change on access to medicines and health outcomes (235).

Aaltonen et al. showed that just under 14% of the entities that are subsidised by PHARMAC are only partly subsidised (i.e. would have a part charge) (178). The extent to which part charges contribute to difficulties patients have in getting their prescriptions filled –such as those found by Jatrana et al. (231, 232) – may be an area where further work is needed. Another possible area for further research is whether the New Zealand public sees a need for alternative ways of paying for pharmaceuticals, such as private insurance for medicines that are not subsidised by PHARMAC.

163

However, private insurance is no guarantee against high cost sharing. In the United States, where private insurance is the main means of paying for pharmaceuticals, American patients faced more cost sharing than in Europeans countries (22). Insurers are increasingly shifting towards making patients pay a percentage (typically 20 to 33%) of high cost drugs such as biological medicines rather a fixed co-payment, which can result in large out of pocket costs for vulnerable patients (238). This approach has also been adopted by public payers in the United States. Eighty-six per cent of Medicare drug plans (which cover those with the highest health needs such as people 65 years and above) used this system in 2008, along with 10% of private drug plans (238). For example, rheumatoid arthritis patients whose biological medicines were covered under Medicare faced out of pocket costs of $4000 or more per year (239).

Privately insured patients can also face high costs through co-payments and deductibles, exceeding annual or lifetime caps on their insurance, or because the recommended treatment is not covered by their plan. Cancer patients can accumulate over $100,000 in medical bills even if they are insured throughout their illness (87). Over 60% of bankruptcies in the US were related to medical problems, including among the middle class (85). In addition to the effects on the bankrupt families themselves, such bankruptcies spread costs to other patients and customers, as healthcare providers and creditors have to recover the costs of the debts that are written off.

The Patient Protection and Affordable Care Act banned lifetime limits on most types of coverage in the US from 2010, and raises the minimum annual limits gradually, banning annual limits altogether in 2014. Insurance companies were banned from denying coverage to children based on pre-existing conditions from 2010, and adults from 2014. From 2014, the law also limits the premiums insurers can charge people with pre-existing conditions when buying individual insurance (114). These changes may help limit patient cost sharing.

In summary, the New Zealand system protects patients in general from excessive cost sharing, due to relatively low co-payments and a cap on the maximum co-payment each patient needs to pay per year. However, patients who need an unfunded medicine would need to pay the full cost. This creates the possibility that some patients (or patient groups) may end up paying very high costs. It should be noted that less than 3% of New Zealanders pay over $1000

164

United States Dollars in prescription costs per year, compared with 5% of Australians and nearly 14% of Americans (240).

6.1.8: Summary

Returning to the Wonder and Cohen et al. definition of access, the three investigations of this thesis combined with previous literature show New Zealand’s access to be:

 Breadth: Poor to fair  Depth: Good (Majority fully subsidised without restrictions)  Timeliness (and innovation): Poor  Equity: Good? (More research needed)  Choice: Poor? (compared with part subsidy of newer medicines in Finland, and private insurance). However, further work is needed, as choice was not specifically investigated in this doctoral work).  Continuity: Good? (Many aspects need further research)  Cost Sharing: Good (low co-payments, most fully subsidised)

It appears therefore that New Zealand trades access to a broad range of medicines, along with immediate access to newly developed medicines, for low patient cost sharing and good depth of access to the medicines that are available.

6.2: Comparisons with previous work

This thesis builds on and extends past work that compares New Zealand’s access to medicines with that in other countries. There have been several different approaches taken in earlier research on New Zealand’s access to medicines, and these are briefly summarised.

A report by Castalia Strategic Advisors (a private advisory group specialising in policy and governance), commissioned by Pfizer New Zealand (5) looked at the strategies used by PHARMAC to obtain the lowest prices for pharmaceuticals (such as sole supply tendering 165

and reference pricing). The report raised the question of whether these policies were increasing levels of disability and ill health in New Zealand, and raising costs elsewhere in the health sector (5). The authors argued that New Zealand’s relatively low proportion of pharmaceutical spending to total health spending (compared to other OECD countries in general, and Australia in particular) was not so much due to PHARMAC’s ability to control pharmaceutical expenditure as to restrictions on access, and the substitution of more costly non-pharmacological interventions. The authors used mental health and cardiovascular health as examples, citing New Zealand’s higher rates of cardiac death, coronary bypass, suicide, intentional self-harm, and disability spending (compared to Australia) as evidence of suboptimal outcomes. They argued that these outcomes could potentially be improved by greater pharmaceutical spending, and the overall cost to the New Zealand health and welfare sectors could be reduced (5).

However, the authors themselves acknowledged that such international comparisons of health outcomes are problematic, due not just to the many factors that affect health outcomes, but also the different methods of collecting health data and sometimes incomplete datasets (5). The report’s conclusions have also been challenged by two further reports commissioned by PHARMAC (6, 241).

An independent review by the private advisory group Business and Economic Research Limited (BERL), commissioned by PHARMAC, of the Castalia report challenged Castalia’s conclusions on several grounds (241). First, New Zealand’s pharmaceutical spending as a proportion of total health spending was within one standard deviation of the OECD average, making it unlikely that the difference was statistically significant (241). Secondly, changes in the proportion of pharmaceutical spending to total health spending could also be attributed to many other factors. These include changes in the demands for health services (which in turn could be attributed to changes in ethnic composition, population ageing and lifestyle over time), to the cost and effectiveness of other health interventions, and to initiatives that improve health outcomes in different countries. The review also highlights the lack of a causal link between pharmaceutical spending and outcomes in mental health and cardiovascular health (241).

Economist Brian Easton’s review of the Castalia report (commissioned by PHARMAC) challenges Castalia’s conclusion that PHARMAC’s sole supply and reference pricing 166

methods restrict access, arguing there is little compelling evidence of important therapeutic differences within a therapeutic category, and that the reduction of some drug use may actually be beneficial (6). As with the BERL report, it highlights that Castalia have not established a causal link between pharmaceutical spending and health outcomes, and failed to consider alternative causes for these health outcomes (6). The Easton report also highlights several factors that are necessary for valid international comparisons of choice and access. These include considering: the availability of licensed but not publically subsidised medicines (which patients have the option of paying for out of pocket or through private insurance); the different arrangements for paying for pharmaceuticals in different countries (which are needed to set any comparison of access in context), and the therapeutic value (if any) added by pharmaceuticals available in one country but not in another (6). The three investigations in this thesis have examined these factors.

As seen above, it is difficult to extrapolate from pharmaceutical spending to health outcomes because both of these are affected by a broad range of factors, and establishing a causal link while accounting for confounding factors is difficult. An alternative is to focus very narrowly on individual medicines or classes of medicines. This was the approach taken by the New Zealand Medical Journal’s Special Series on New Zealanders’ access to medicines. The series consisted of reports written by clinicians or other key informants of access problems with particular medicines. The series included articles, among many others, on the availability of statins for cardiovascular disease, trastuzumab (Herceptin®) in early stage HER-2 positive breast cancer, and the quality of the generic salbutamol that replaced Ventolin® (3, 4, 171- 174). This approach is valuable for highlighting issues of concern to clinicians and pharmacists, and prompting debate on New Zealand’s access to medicines. However, such a ‘case series’ approach does not provide an overall view of New Zealand’s access to medicines, and does not compare New Zealand to other countries.

A third approach, and the one adopted in this thesis, is to measure New Zealand’s access to medicines in terms of the number and types of medicines available, and the difficulties patients face in gaining access to these medicines. These can be done both quantitatively (counts of the medicines available) and qualitatively through interviews with key informants. Both these approaches have been used in the current thesis, extending on past work done in New Zealand and overseas.

167

The Wonder Report (written by a senior health economist at Novartis Australia) examined the 78 new listings for prescription medicines on the Pharmaceutical Benefits Schedule (PBS) in Australia between December 1999 and April 2006. The report found that while 72 (92%) were licensed in New Zealand, only 20 (26%) were subsidised by PHARMAC. In addition, the medicines that were subsidised by PHARMAC were on average subsidised 14 months after they were subsidised by the PBS (26). Potential weaknesses of the report are that it focused only on newly subsidised medicines (rather than all medicines), and used medicines newly subsidised by the PBS as a starting point rather than all medicines subsidised by the PBS and PHARMAC over that period (26). However, the report considered a broad range of factors that influenced access, including breadth, depth, timeliness and innovation, which were discussed earlier in the current thesis.

A report by McCormack et al. to the Minister of Health on New Zealanders’ access to high- cost, highly specialised medicines used a different approach. The authors consulted widely with stakeholders, invited public submissions, and consulted published literature to compare New Zealand’s access with that in Australia and the United Kingdom (66). The authors concluded that while New Zealand gained very good value for pharmaceutical spending, New Zealand’s access to high cost and highly specialised medicines was less than that in other high-income countries (66) The authors made the point that any international comparison should include both positive analysis (what is available) and normative analysis (what should be available). This thesis is structured similarly, and includes both positive comparisons (breadth and timeliness in Investigations One and Two), along with normative aspects (the value added by innovative medicines in Investigation One, expectations of licensed and suitably formulated medicines for children in Investigation Two, and potential improvements that could be made to the New Zealand system in Investigation Three).

6.2.1: Development of new methodology

Before commencing the investigations in this thesis, there were no published, peer-reviewed studies that compared all licensed and subsidised medicines in New Zealand with those in other countries. In fact, our literature search found no published, peer-reviewed studies that compared all licensed and subsidised medicines in any high-income countries. Previous

168

comparisons all focused on subsets such as newly launched medicines or top selling medicines, and included less than 250 molecules (22, 23, 26).

Because of this, a new methodology needed to be developed that allowed comparison of all licensed and subsidised medicines, included multiple dimensions of access (see Section 6.1) and included both positive and normative aspects. This resulted in an international collaboration with another researcher to develop methodology for comparisons in high- income countries, and a published study that compared New Zealand with Finland, an OECD country with a similar population size and level of development (178).

The comparison of New Zealand and Finland showed that the two countries licensed and subsidised a similar number of medicines (178). Approximately 30% of entities licensed in one country were not licensed in the other, and 40% of medicines subsidised in one country were not subsidised in the other. The study also investigated timeliness and innovation, but only for subsets rather than the entire sample. Timeliness was investigated for the products licensed in one country and not in the other, and subsidised in one country but not in the other. The innovation of licensed products was also investigated, but not those of subsidised products (178).

6.3: Strengths and Limitations

Investigation One of the current thesis extends earlier work (the New Zealand/Finland comparison) by carrying out a four-way comparison of New Zealand, Australia, the United Kingdom and the United States. The timeliness of all licensed and subsidised medicines is examined, and the licensing and subsidy status of all innovative products is examined. Investigation Two extends the analysis to children, and Investigation Three adds qualitative analysis of factors such as equity. Overall, the findings are consistent with past findings that New Zealanders have access to fewer and older medicines than in other countries, such as findings in the Wonder Report and the New Zealand/Finland comparison (26, 178).

The current thesis adds a broader range of countries than have been previously compared against each other, a larger sample size that gives greater statistical power, and contributes a new methodology for international comparisons.

169

The present study is the first to use this newly developed method of international comparison (178) to compare access to medicines in these four countries. It has resulted in the first comparison of all products licensed and subsidised in New Zealand, Australia, the United Kingdom and the United States to be published in a peer-reviewed academic journal (37).

The three comparison countries (Australia, the United Kingdom and the United States) all provide important points of reference for New Zealand. All four countries are members of the Organisation for Economic Co-operation and Development (OECD) (34), and all have highly developed regulatory systems for ensuring the safety and efficacy of medicines (38, 42, 195, 242). All have significant public spending on medicines, and all have at least one single payer health system that uses a national formulary and a national body for making decisions on high cost medicines (62, 164, 185, 188, 203).

This new method examines the entire range of products listed in prescribing references, and so gives a better picture of overall access than comparing a selected subset such as newly launched products or best-selling medicines (22, 23). The large dataset (over 9000 products) creates good statistical power for detecting differences between countries. Products were compared by ATC code as well as by active agent. This is more accurate than comparing active agents alone because it captures different formulations and indications of the same active agent. This is especially important for subsidy comparisons, as subsidy listings are usually by indication and route. Linking the time since first registration to ATC code allows a comparison of how quickly countries license and subsidise new formulations and indications.

Lastly, this method includes a way of calculating the number of innovative entities licensed and subsidised in each country. (See Section 6.1.3: Timeliness for how innovative medicines add extra value over those already in use).

This thesis also compares access for children across the four countries using a modified version of the methodology described above. Children are a group who have historically been disadvantaged when it comes to access to medicines (105, 106). Both the United States and the European Union have enacted legislative incentives to improve children’s access to medicines (57, 58, 98). When the data for Investigation Two were collected, these incentives had been in place in the United States for a decade, but were only coming into effect in the United Kingdom, and had no equivalents in New Zealand or Australia. There was therefore an 170

opportunity therefore to examine the effects of a decade of incentives in the United States to improve children’s access to medicines, and to compare access with the other countries.

Investigation Three included the views of key informants across multiple sectors, including clinicians, people working in both PHARMAC and the pharmaceutical industry, patient representatives and Members of Parliament. The views of people in multiple sectors, when combined, give a much more comprehensive view of access than focusing on the views of one group. Last but not least, this thesis incorporated multiple ways of examining access to medicines. It combined quantitative and qualitative methods, examined multiple dimensions of access, and included both positive and normative views of access.

However, there are some important limitations. Any investigation depends on the accuracy of the data sources used. The texts used for licensing and subsidy information may not have included every licensed and product available in that country. As listing a product incurs a cost, manufacturers may have opted not to list orphan products, generics or products that were due to be taken off the market in the near future. Conversely, products that had been newly registered at the time the texts went to print may not have been included. As the primary data sources were not separately validated, the potential for incompleteness remains. The use of listings of registered products (obtained from the regulatory authorities) was considered in place of prescribing compendia. However, such listings may contain products that registered but never launched on the market of the country concerned, or products that were registered and marketed in the past but were subsequently withdrawn from that market. As such, listings of registered products could potentially give an inaccurate picture of the medicines available in a country at a given time.

In addition, the key informants in Investigation Three may not have been entirely candid. These people may have had agendas they did not wish to discuss, either for tactical reasons or due to restrictions imposed by their employment. Informants may also have modified their reasons to those that would have been considered socially acceptable or desirable. However, it is important to keep in mind that the aim of this investigation was to accurately capture and summarise the views the informants or their organisations expressed on pharmaceutical policy, rather than any privately held thoughts. While the interviews included people working in a wide range of areas (academia, the public sector, healthcare et cetera), the small number

171

of people working on pharmaceutical policy in New Zealand may also have meant that a wide enough range of people were not included.

In addition, as noted above, there are multiple dimensions of access to medicines. The present thesis focused more on some dimensions than on others (particularly breadth, timeliness and innovation), and may therefore have neglected others that may be as important or more important.

Also, while this thesis examined the perceptions of key informants had of New Zealanders’ access to medicines, it did not directly examine the perceptions of patients themselves. The dimensions of access that patients consider important, and the dimensions of access that patients feel are lacking in New Zealand, may be different. It would be useful to explore this in future work.

This thesis does not attempt to link any differences in access to medicines to clinical outcomes, which are the ultimate outcomes of concern to patients, the public and policy makers. Such a study would need data on the clinical outcomes themselves, as well as being able to correct for confounding factors such as differences in demographics, lifestyle factors and access to primary care and screening. It is worth noting that others have raised concerns about the availability of particular medicines, and the impact these may have on disease outcomes. Examples that have been cited include the impact of changes in the funding of HMG CoA reductase inhibitors (statins) on cardiovascular mortality, and the impact of funding of 52 weeks treatment with trastuzumab on the survival of patients with early stage HER-2 positive breast cancer (3, 171). It is equally worth noting that these concerns have been vigorously disputed (172, 243). This may be a useful avenue for future research.

Last but not least, these results are a snapshot of New Zealand’s access to medicines at the time the data were collected. Access to medicines can change over time due to many factors, such as changes in a country’s financial circumstances, changes in medicines policy or health system reform, and changes in health needs. Therefore, conclusions for today may not be valid conclusions for tomorrow.

172

6.4 Further Research

There are aspects of New Zealand’s access to medicines that merit further investigation:

Depth: A four-way comparison of the restrictions patients face when trying to get access to subsidised medicines would be very useful, but potentially difficult. A comparison of restrictions in the PHARMAC Schedule and Pharmaceutical Benefits Schedule is relatively straightforward, as both schedules set their restrictions at the national level, and publish the restrictions openly (60, 182). However, in the United Kingdom, restrictions on use are set at the local level, and this information could be much more difficult to collect (18). Similarly, while a comparison of restrictions in the Department of Veterans Affairs National Formulary with that in the PHARMAC schedule is relatively simple, a full comparison between patients in New Zealand and the United States would also need to compare depth of access via private insurance.

Cost Sharing: while co-payments in New Zealand are lower than in the other countries studied (and co-payments are capped at 20 per family per year), total cost sharing would include part charges and the cost of unsubsidised medicines. Research on what patients in New Zealand paid for their prescriptions per year compared with those in Australia, the United Kingdom and United States would be useful. The effects of switching medicines from prescription to non-prescription status have on cost sharing also need to be examined. If these medicines are bought privately after the switch rather than being publically funded, it could increase patient costs.

Choice: Are patients in New Zealand happy with the choices offered through the PHARMAC system, or are they eager for alternatives such as that offered by private insurance? This could be explored through surveys of patient attitudes in New Zealand.

Topics which were not explored in this thesis, such as the patient perspective on access to medicines, and whether the differences between countries in access to medicines shown in this thesis can be linked to differences in clinical outcomes may be useful avenues for future research.

173

This thesis also takes a ‘whole country’ or ‘whole system’ approach to access to medicines. Within the broad parameters of access to medicines examined in this thesis, some patient groups may be affected much more than other groups. (Investigation Three did show agreement among key informants that not all New Zealanders have equal access to medicines). Identifying such patient groups, and the issues of access that are of concern to these groups may be a valuable area for future research.

In addition, as noted above, the results of this thesis are a snapshot in time. Given that several years have elapsed since the first data for this study were collected, a follow up study that collected the same data for 2012 would shed light on the rate of change in the four countries. This would be especially relevant given the current economic climate, and the pressures on public finances in many countries. It would be interesting to compare how a system like PHARMAC, which has been praised for its financial responsibility (2), fared both before and after the onset of the global financial crisis. It is also worth noting that the systems examined in this thesis have undergone (and continue to undergo) changes since the data were collected. It would be interesting to evaluate whether the approaches taken by these systems to medicine evaluation and selection have either converged or diverged since the data were collected.

6.5 Conclusion

The aim of this thesis was to compare New Zealanders’ access to licensed and subsidised medicines with that in a nationwide single payer system in each of Australia, the United Kingdom and the United States. The research contained in this thesis shows New Zealand does well in some aspects of access to medicines, while needing improvement in others.

New Zealand has an effective system for licensing medicines. Medsafe is perceived to be a professional organisation that ensures the safety, efficacy and quality of medicines licensed in New Zealand. Medsafe is also perceived to have a relationship with the pharmaceutical industry that allows both parties to work effectively together, without compromising Medsafe’s objectivity. However, New Zealand licensed fewer entities for adults, and licensed fewer innovative entities that provided important health gains than the comparator countries. Entities licensed in New Zealand were older than entities in the comparator countries. This

174

means that New Zealanders have access to fewer newly developed and innovative medicines than patients in Australia, the United Kingdom and the United States.

New Zealand compares well in children’s access to medicines. New Zealand had the highest ratio of medicines licensed in children to those licensed in adults, and the highest ratio of medicines subsidised in children to those subsidised in adults. PHARMAC also subsidised the second highest number of entities in formulations suitable for children. The United States licenses more newly developed medicines for children than New Zealand (and the other two countries studied in this thesis), but only in older children. This may be due to the incentives contained in the Best Pharmaceuticals for Children Act. Further incentives may be needed to stimulate the development of medicines for younger children, and for the development of suitable formulations.

PHARMAC is perceived to be an organisation that controls pharmaceutical spending effectively, extracts the maximum buying power from a limited budget, and is resistant to lobbying and public pressure. It also perceived to be a politically independent organisation whose decisions are rarely over-ridden by the Government of the day. However, there are concerns that medicines are assessed more stringently for cost effectiveness than other health investments, and that New Zealand is falling behind in access to new medicines. The research contained in this thesis shows that New Zealand’s PHARMAC subsidises fewer medicines for adults, fewer newly developed medicines and fewer innovative medicines than the Australian PBS, United Kingdom NHS and United States VANF.

Further research is needed to determine what if any impact these differences may have on New Zealanders’ health status. This was not an outcome that was examined in this thesis, but is one that would arguably be of great concern to patients and policy makers. Further research is also needed to determine whether changes to the systems studied since the data were collected in 2007 would impact on the findings in this thesis.

175

REFERENCES

1. Wonder M, Milne R. Access to new medicines in New Zealand compared to Australia. N Z Med J. 2011;124(1346):U12-28. 2. Cumming J, Mays N, Daubé J. Analysis: how New Zealand has contained expenditure on drugs. BMJ. 2010;340:c2441. 3. Isaacs RJ, Frampton CM, Kuper-Hommel MJ. PHARMAC’s funding of 9 weeks Herceptin: many assumptions in a high-risk decision. N Z Med J. 2007;120(1259):U2676. 4. Rosevear M. PHARMAC and Herceptin for early stage breast cancer in New Zealand: Herceptin or deception? N Z Med J. 2006;119(1235):U2014. 5. Castalia Strategic Advisors. New Zealand pharmaceutical policies: time to take a fresh look [online] 2005 [cited 2012 06 Sept]. Available from: http://www.castalia- advisors.com/files/14634.pdf. 6. Easton B. New Zealand's pharmaceutical policies: a fresh look [online] 2005 [cited 2012 06 Sept]. Available from: http://www.pharmac.govt.nz/2006/05/04/040506a.pdf. 7. The World Health Organization. The world medicines situation [online] 2004 [cited 2012 26 Feb]. Available from: http://apps.who.int/medicinedocs/pdf/s6160e/s6160e.pdf. 8. The United Nations Development Programme. Human development report 2000 2000 [cited 2012 26 Feb]. Available from: http://hdr.undp.org/en/media/hdr_2000_ch0.pdf. 9. The World Health Organization. The world medicines situation [online] 2011 [cited 2013 19 May]. 10. The World Health Organization. WHO policy perspectives on medicines 8. Equitable access to essential medicines: a framework for collective action [online] 2004 [cited 2012 26 Feb]. Available from: http://apps.who.int/medicinedocs/en/d/Js4962e/1.html. 11. The World Health Organization. WHO medicines strategy: framework for action in essential medicines and medicines policy 2000-2003 [online] 2000 [cited 2012 26 Feb]. Available from: http://apps.who.int/medicinedocs/en/d/Jwhozip16e/. 12. The World Health Organization. WHO policy perspectives on medicines 4: the selection of essential medicines. [online] 2002 [cited 2012 26 Feb]. Available from: http://archives.who.int/tbs/sel/s2296e.pdf. 13. The World Health Organization. WHO model lists of essential medicines [online] [cited 2012 June 10]. Available from: http://www.who.int/medicines/publications/essentialmedicines/en/. 14. The World Health Organization. WHO medicines strategy 2004-2007: countries at the core. [online] 2004 [cited 2012 26 Feb]. Available from: http://whqlibdoc.who.int/hq/2004/WHO_EDM_2004.5.pdf. 15. Laing R, Wanning, B, Gray A, Ford N, 't Hoen E. 25 years of the WHO essential medicines lists: progress and challenges. Lancet. 2003;361(9370):1723-9. 16. Hogerzeil HV. The concept of essential medicines: lessons for rich countries. BMJ. 2004;329(7475):1169-71. 17. The World Health Organization. 10 facts on essential medicines [online] [cited 2013 25 May]. Available from: http://www.who.int/features/factfiles/essential_medicines/en/index.html. 18. Morgan SG, McMahon M, Mitton C, Roughead E, Kirk R, Kanavos P, Menon D. Centralized drug review processes in Australia, Canada, New Zealand and the United Kingdom. Health Aff (Millwood). 2006;25(2):337-47. 19. Huskamp HA, Epstein AM, Blumenthal D. The impact of a national prescribing drug formulary on prices, market share, and spending: lessons for Medicare? Health Aff (Millwood). 2003;22(3):149-57. 20. Sales MM, Cunningham FE, Glassman PA, Valentino MA, Good CB. Pharmacy benefits management in the Veterans Health Administration: 1995 to 2003. Am J Manag Care. 2005;11(2):104-12. 176

21. Maynard A, Bloor K. Dilemmas in regulation of the market for pharmaceuticals. Health Aff (Millwood). 2003;22(3):31-41. 22. Cohen J, Faden L, Predaris S, Young B. Patient access to pharmaceuticals: an international comparison. Eur J Health Econ. 2007;8(3):253-66. 23. Danzon PM, Wang YR, Wang L. The impact of price regulation on the launch delay of new drugs- evidence from 25 major markets in the 1990s. Health Econ. 2005;14(3):269-92. 24. Danzon PM, Furukawa MF. International prices and availability of pharmaceuticals in 2005. Health Aff (Millwood). 2008;27(1):221-33. 25. DeNavas-Walt C, Proctor BD, Smith JC. Income, Poverty and Health Insurance Coverage in the United States: 2009. Washington DC: United States Census Bureau 2010. 26. Wonder M. Access by patients in New Zealand to innovative new prescription only medicines; how have they been faring in recent times in relation to their Trans-Tasman counterparts? 2006 [cited 2012 06 Sept]. Available from: http://www.nzdoctor.co.nz/media/34764/aus%20nz%20wonder%20report.pdf. 27. Schoen C, Osborn R,Squires D, Doty M M, Pierson R, Applebaum S. How health insurance design affects access to care and costs, by income, in eleven countries. Health Aff (Millwood). 2010;29(12):2323-34. 28. Chabner B. Drug Shortages- a critical challenge for the generic-drug market. N Engl J Med. 2011;365(23):2147-9. 29. Kaakeh R, Sweet BV, Reilly C, Bush C, DeLoach S, Higgins B, Clark AM, Stevenson J. Impact of drug shortages on U.S. health systems. Am J Health-Syst Pharm. 2011;68(19):1811-9. 30. de Somer E. Securing the supply chain. Aust Prescr. 2011;34:105-7. 31. Gehrett BK. A prescription for drug shortages. JAMA. 2012;307(2):153-4. 32. The Commonwealth Fund. Issues in international health policy- prescription drug accessibility and affordability in the United States and abroad 2010 [cited 2013 01 Feb]. Available from: http://www.commonwealthfund.org/~/media/Files/Publications/Issue%20Brief/2010/Jun/1 408_Morgan_Prescription_drug_accessibility_US_intl_ib.pdf. 33. The World Bank. Country and lending groups [online] [cited 2012 26 Feb]. Available from: http://data.worldbank.org/about/country-classifications/country-and-lending- groups#High_income. 34. The Organisation for Economic Co-operation and Development. Members and partners [online] [cited 2012 06 Sept]. Available from: http://www.oecd.org/pages/0,3417,en_36734052_36761800_1_1_1_1_1,00.html. 35. The European Medicines Agency. European Medicines Agency [online] [cited 2012 06 Sept]. Available from: http://www.ema.europa.eu/ema/index.jsp?curl=/pages/home/Home_Page.jsp. 36. The United States Department of State. Background note- New Zealand [online] [cited 2012 06 Sept]. Available from: http://www.state.gov/r/pa/ei/bgn/35852.htm. 37. Ragupathy R, Aaltonen K, Tordoff J, Norris P, Reith D. A 3-dimensional view of access to licensed and subsidized medicines under single payer systems in the United States, United Kingdom, Australia and New Zealand. Pharmacoeconomics. 2012;30(11):1051-65. 38. The New Zealand Medicines and Medical Devices Safety Authority (Medsafe). About Medsafe [online] [cited 2012 06 Sept]. Available from: http://www.medsafe.govt.nz/other/about.asp. 39. The New Zealand Parliament. Medicines Act 1981 [online] 1981 [cited 2012 26 Feb]. Available from: http://www.legislation.govt.nz/act/public/1981/0118/latest/DLM53790.html?search=ts_act %40bill%40regulation%40deemedreg_Medicines+Act+1981_resel_25_h&p=1. 40. The New Zealand Parliament. Medicines Regulations 1984 [online] 1984 [cited 2012 26 Feb ]. Available from: http://www.legislation.govt.nz/regulation/public/1984/0143/16.0/DLM95668.html.

177

41. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Australia New Zealand Therapeutic Products Agency (ANZTPA) fact sheet [online] 2011 [cited 2012 26 Feb]. Available from: http://www.tga.gov.au/about/international-anztpa- factsheet.htm. 42. The Australian Department Of Health And Ageing Therapeutic Goods Administration. What the TGA regulates [online] 2012 [cited 2012 26 Feb]. Available from: http://www.tga.gov.au/about/tga-regulates-what.htm. 43. The Australian Department Of Health And Ageing Therapeutic Goods Administration. How the TGA regulates [online] 2012 [cited 2012 26 Feb]. Available from: http://www.tga.gov.au/about/tga-regulates-how.htm. 44. The Australian Parliament. Therapeutic Goods Act 1989 [online] 1989 [cited 2012 26 Feb]. Available from: http://www.comlaw.gov.au/Details/C2011C00589. 45. The United Kingdom Medicines And Health Care Products Regulatory Agency. What we regulate [online] 2012 [cited 2012 26 Feb]. Available from: http://www.mhra.gov.uk/Aboutus/Whatweregulate/index.htm. 46. The United Kingdom Medicines And Health Care Products Regulatory Agency. Medicines & medical devices regulation: what you need to know [online] 2012 [cited 2012 26 Feb]. Available from: http://www.mhra.gov.uk/home/groups/comms- ic/documents/websiteresources/con2031677.pdf. 47. The United Kingdom Medicines And Health Care Products Regulatory Agency. How we regulate medicines [online] [cited 2012 26 Feb]. Available from: http://www.mhra.gov.uk/Howweregulate/Medicines/index.htm. 48. The United Kingdom Medicines And Health Care Products Regulatory Agency. Overview of medicines legislation and guidance [online] [cited 2012 26 Feb]. Available from: http://www.mhra.gov.uk/Howweregulate/Medicines/Overviewofmedicineslegislationandgui dance/index.htm. 49. The European Medicines Agency. Central authorisation of medicines [online] [cited 2012 26 Feb]. Available from: http://www.ema.europa.eu/ema/index.jsp?curl=pages/about_us/general/general_content_ 000109.jsp&murl=menus/about_us/about_us.jsp&jsenabled=true. 50. The European Medicines Agency. Mutual recognition & decentralised procedures - human (CMDh) [online] [cited 2012 26 Feb]. Available from: http://www.ema.europa.eu/ema/index.jsp?curl=pages/about_us/general/general_content_ 000310.jsp&murl=menus/about_us/about_us.jsp&mid=WC0b01ac05800986a3. 51. The United States Food And Drug Administration. What does FDA do? [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/AboutFDA/Transparency/Basics/ucm194877.htm. 52. The United States Food And Drug Administration. About FDA [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/AboutFDA/Transparency/Basics/ucm192695.htm. 53. The United States Food And Drug Administration. What does FDA regulate? 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/AboutFDA/Transparency/Basics/ucm194879.htm. 54. The United States Food And Drug Administration. FDA 101: Product recalls - From first alert to effectiveness checks [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/ForConsumers/ConsumerUpdates/ucm049070.htm. 55. The United States Food And Drug Administration. Federal Food, Drug, and Cosmetic Act (FD&C Act) [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/RegulatoryInformation/Legislation/FederalFoodDrugandCosmeticActFD CAct/default.htm.

178

56. The United States Food And Drug Administration. Significant amendments to the FD&C Act [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/RegulatoryInformation/Legislation/FederalFoodDrugandCosmeticActFD CAct/SignificantAmendmentstotheFDCAct/default.htm. 57. The United States Food And Drug Adminstration. Regulatory information: Best Pharmaceuticals For Children Act. [online] [cited 2012 27 Sept]. Available from: http://www.fda.gov/RegulatoryInformation/Legislation/FederalFoodDrugandCosmeticActFD CAct/SignificantAmendmentstotheFDCAct/ucm148011.htm. 58. The United States Food And Drug Administration. Pediatric drug development: FDA Amendments Act of 2007 (FDAAA). [online] [cited 2012 27 Sept]. Available from: http://www.fda.gov/downloads/Drugs/DevelopmentApprovalProcess/DevelopmentResourc es/UCM049870.pdf. 59. The United States Food And Drug Administration. Pediatric Research Equity Act of 2003 [online] 2003 [cited 2012 03 March]. Available from: http://www.fda.gov/downloads/Drugs/DevelopmentApprovalProcess/DevelopmentResourc es/UCM077853.pdf. 60. The New Zealand Pharmaceutical Management Agency. Pharmaceutical Schedule: about the Schedule [online] [cited 2013 24 Jan]. Available from: http://www.pharmac.govt.nz/patients/Schedule. 61. Braae R, McNee W, Moore D. Managing pharmaceutical expenditure while increasing access- the Pharmaceutical Management Agency (PHARMAC) experience. Pharmacoeconomics. 1999;16(6):649-60. 62. The New Zealand Pharmaceutical Management Agency. About PHARMAC [online] [cited 2012 20 May]. Available from: http://www.pharmac.govt.nz/patients/AboutPHARMAC. 63. The New Zealand Pharmaceutical Management Agency. Introduction to PHARMAC [online] [cited 2012 22 May]. Available from: http://www.pharmac.govt.nz/2011/09/16/01INTRO.pdf. 64. The New Zealand Pharmaceutical Management Agency. Operating policies and procedures of the Pharmaceutical Management Agency ("PHARMAC") [online] [cited 2012 22 May]. Third Edition:[Available from: http://www.pharmac.govt.nz/2005/12/22/231205.pdf. 65. The New Zealand Pharmaceutical Management Agency. Information sheet for Community, Hospital and Cancer Exceptional Circumstances [online] [cited 2012 06 July]. Available from: http://www.pharmac.govt.nz/2012/03/01/EC%20info%20sheet.pdf. 66. McCormack P, Quigley J, Hansen P. Report to Minister of Health, Hon Tony Ryall- Review of access to high cost, highly specialised medicines in New Zealand [online] [cited 2012 26 Sept]. Available from: http://img.scoop.co.nz/media/pdfs/1005/Review_of_Access_to_High_Cost_Highly_Specialis ed_Medicines_31_April_2010.pdf. 67. The New Zealand Pharmaceutical Management Agency. Named Patient Pharmaceutical Assessment (Exceptional Circumstances) policy – June 2011 [online] [cited 2012 09 Sept]. Available from: http://www.pharmac.govt.nz/2011/06/27/NPPA%20policy.pdf. 68. Australian Government, Medicare Australia. Pharmaceutical Benefits Scheme (PBS) [online] [cited 2012 25 Sept]. Available from: http://www.medicareaustralia.gov.au/provider/pbs/index.jsp. 69. Graham D. The Australian Pharmaceutical Benefits Scheme. Aust Prescr. 1995;18:42-4. 70. Birkett DJ, Mitchell AS, McManus P. A cost effectiveness approach to drug subsidy and pricing in Australia. Health Aff (Millwood). 2001;20(3):104-14. 71. Gallego G, Taylor SJ, Brien JA. Provision of pharmaceuticals in Australian hospitals: equity of access? . Pharm World Sci. 2007;29(2):47-50 72. Doran E, Alexander Henry D. Australian pharmaceutical policy: price control, equity and drug innovation in Australia. J Public Health Pol. 2008;29(1):106-17.

179

73. Sweeny K. Key aspects of the Australian Pharmaceutical Benefits Scheme [online] [cited 2012 03 March]. Available from: http://www.cfses.com/documents/pharma/35- Key_aspects_of_PBS_Sweeny.pdf. 74. Sansom L. The subsidy of pharmaceuticals in Australia: processes and challenges. Aust Health Rev. 2004;28(2):194-205. 75. Australian Department of Health and Ageing. Alternative arrangements for medicines: other supply arrangements outside the Pharmaceutical Benefits Schedule [online] [cited 2012 03 March]. Available from: http://www.health.gov.au/internet/main/publishing.nsf/Content/lsdp-info. 76. Medicare Australia. Late stage metastatic breast cancer [online] [cited 2012 03 March]. Available from: http://www.medicareaustralia.gov.au/provider/patients/late-breast- cancer.jsp. 77. The United Kingdom National Health Service. About the NHS [online] 2012 [cited 2012 03 March]. Available from: http://www.nhs.uk/NHSEngland/thenhs/about/Pages/overview.aspx. 78. The United Kingdom Department of Health. Introduction to pharmaceutical price regulation [online] [cited 2012 03 March]. Available from: http://webarchive.nationalarchives.gov.uk/+/www.dh.gov.uk/en/Healthcare/Medicinesphar macyandindustry/Pharmaceuticalpriceregulationscheme/DH_4071841. 79. The United Kingdom Department of Health. The Pharmaceutical Price Regulation Scheme 2009 [online] [cited 2012 03 Mar]. Available from: http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/documents/digitalasset/d h_098498.pdf. 80. Parsons A, Johnstone A. Postcode prescribing and the Human Rights Act 1998. J R Soc Med 2001;94(4):159-60. 81. National Institute for Health and Clinical Excellence. Guide to the methods of technology appraisal [online] [cited 2012 26 Sept]. Available from: http://www.nice.org.uk/niceMedia/pdf/TAP_Methods.pdf. 82. Rawlins M, Barnett D, Stevens A. Pharmacoeconomics: NICE’s approach to decision-making. Br J Clin Pharmacol. 2010;70(3):346–9. 83. Pearson SD, Rawlins MD. Quality, innovation and value for money: NICE and the British National Health Service. JAMA. 2005;294(20):2618-22. 84. Schoen C, Collins SR, Kriss JL, Doty MM. How many are underinsured? Trends among U.S adults 2003 and 2007. Health Aff (Millwood). 2008;27(4):w298-309. Epub 2008 Jun 10. 85. Himmelstein D U, Thorne D, Warren E, Woolhandler S. Medical bankruptcy in the United States 2007: results of a national study. Am J Med. 2009;122(8):741-6. 86. Luft HS. What works and what doesn’t work well in the US healthcare system. Pharmacoeconomics. 2006;24(Supplement 2):15-28. 87. Kaiser Family Foundation and American Cancer Society. Spending to survive: cancer patient confront holes in the health insurance system. [online] [cited 2012 22 Sept]. Available from: http://www.kff.org/insurance/upload/7851.pdf. 88. Blumenthal D, Herdman R, editor. Description and analysis of the VA National Formulary. Washington DC: National Academy Press; 2000. 89. Jha AK, Perlin JB, Kizer KW, Dudley RA. Effect of the transformation of the Veterans Affairs health care system on the quality of care. N Engl J Med. 2003;348(22):2218-27. 90. Kuehn BM. Veterans health system cited by experts as a model for patient-centered care. JAMA. 2012;307(5):442-3. 91. Atkins D, Kupersmith J, Eisen S. The Veterans Affairs experience: comparative effectiveness research in a large health system. Health Aff. 2010;29(10):1906-12.

180

92. Ragupathy R, Aaltonen K, Tordoff J, Norris P, Reith D. The Authors’ Reply to Wonder and Milne: ‘‘Comparing Subsidized Access to Medicines Across Payer Systems’’. Pharmacoeconomics. 2013;31(2):175-6. 93. The United States Department of Veterans Affairs. Health benefits- veterans [online] 2012 [cited 2012 03 March]. Available from: http://www.va.gov/healthbenefits/apply/veterans.asp. 94. The United States Department of Veterans Affairs. Health benefits- family members of veterans [online] 2012 [cited 2012 03 March]. Available from: http://www.va.gov/healthbenefits/apply/family_members.asp. 95. Aspinall SL, Good CB, Glassman PA, Valentino MA. The evolving use of cost-effectiveness in formulary management within the Department of Veterans Affairs. Med Care. 2005;43(Supplement 7):20-6. 96. The United Kingdom National Institute for Health And Clinical Excellence. NICE and the NHS [online] 2012 [cited 2012 03 March]. Available from: http://www.nice.org.uk/aboutnice/whatwedo/niceandthenhs/nice_and_the_nhs.jsp. 97. The Scottish Medicines Consortium. Remit [online] 2012 [cited 2012 03 March]. Available from: http://www.scottishmedicines.org.uk/About_SMC/What_we_do/Remit. 98. The European Medicines Agency. Medicines for children: the EU Paediatric Regulation [online] [cited 2012 28 Aug]. Available from: http://www.emea.europa.eu/htms/human/paediatrics/regulation.htm. 99. The United States Food And Drug Administration. Developing products for rare diseases and conditions [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/forindustry/developingproductsforrarediseasesconditions/default.htm. 100. The European Medicines Agency. Orphan designation [online] 2012 [cited 2012 03 March]. Available from: http://www.ema.europa.eu/ema/index.jsp?curl=pages/regulation/general/general_content _000029.jsp&mid=WC0b01ac05800240ce&jsenabled=true. 101. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Orphan drugs [online] 2012 [cited 2012 03 March]. Available from: http://www.tga.gov.au/industry/pm-orphan-drugs.htm. 102. Chui J, Tordoff J, Kennedy J, Reith D. Trends in accessibility to medicines for children in New Zealand: 1998-2002. Br J Clin Pharmacol. 2004;57(3):322-7. 103. Balakrishnan K, Tordoff J, Norris P, Reith D. Establishing a baseline for the monitoring of medicines availability for children in the UK: 1998-2002. Br J Clin Pharmacol. 2007;63(1):85- 91. 104. Chui J, Tordoff J, Reith D. Changes in availability of paediatric medicines in Australia between 1998 and 2002. Br J Clin Pharmacol. 2005;59(6):736-42. 105. Young L, Lawes F, Tordoff J, Norris P, Reith D. Access to prescribing information for paediatric medicines in the USA: post-modernization. Br J Clin Pharmacol. 2009;67(3):341-6. 106. Ragupathy R, Tordoff J, Norris P, Reith D. Access to children’s medicines in the United Kingdom, Australia and New Zealand in 1998, 2002 and 2007. Pharm World Sci. 2010;32(3):386-93. 107. Turner S, Gill A, Nunn T, Hewitt B, Choonara I. Use of "off label" and unlicensed drugs in paediatric intensive care unit. Lancet. 1996;347(9000):549-50. 108. McLay JS, Tanaka M, Ekins-Daukes S, Helms PJ. A prospective questionnaire assessment of attitudes and experiences of off label prescribing among hospital based paediatricians. Arch Dis Child. 2006;91(7):584-7. 109. Horen B, Montastruc JL, Lapeyre-Mestre M. Adverse drug reactions and off-label drug use in paediatric outpatients. Br J Clin Pharmacol. 2002;54(6):665-70. 110. Schirm E, Tobi H, De Vries TW, Choonara I, De Jong-van den Berg LT. Lack of appropriate formulations of medicines for children in the community. Acta Paediatr. 2003;92(12):1486-9.

181

111. Nunn T, Williams J. Formulation of medicines for children. Br J Clin Pharmacol. 2005;59(6):674-6. 112. Hudson JL, Selden TM. Children’s eligibility and coverage: recent trends and a look ahead. Health Aff (Millwood). 2007;26(5):w618-29. Epub 2007 Aug 16. 113. National Center for Children in Poverty. Public health insurance for children [online] 2012 [cited 2012 07 July]. Available from: http://www.nccp.org/profiles/index_32.html#national. 114. The United States Congress. Patient Protection and Affordable Care Act 2010 [online] [cited 2012 24 Sept]. Available from: http://docs.house.gov/energycommerce/ppacacon.pdf. 115. Nelson RE, McAdam-Marx C, Evans ML, Ward R, Campbell B, Brixner D, Lafleur J. Patent extension policy for paediatric indications: an evaluation of the impact within three drug classes in a state Medicaid programme. Appl Health Econ Health Policy 2011;9(3):171-81. 116. The United States Food And Drug Administration. Designating an orphan product: drugs and biologics [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/ForIndustry/DevelopingProductsforRareDiseasesConditions/Howtoappl yforOrphanProductDesignation/default.htm. 117. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Australian regulatory guidelines for prescription medicines (ARGPM). 2012 [cited 2012 03 March]. Available from: http://www.tga.gov.au/pdf/pm-argpm.pdf. 118. Sharma A, Jacob A, Tandon M, Kumar D. Orphan drug: development trends and strategies. J Pharm Bioall Sci. 2010;2(4):290-9. 119. The European Medicines Agency. Orphan Incentives [online] 2012 [cited 2012 03 March]. Available from: http://www.ema.europa.eu/ema/index.jsp?curl=pages/regulation/general/general_content _000393.jsp&mid=WC0b01ac0580024c5a&jsenabled=true. 120. The European Union. Inventory of Community and Member States' incentive measures to aid the research, marketing, development and availability of orphan medicinal products. Revision 2005 [online] 2005 [cited 2012 03 March]. Available from: http://ec.europa.eu/health/files/orphanmp/doc/inventory_2006_08_en.pdf. 121. Canadian Cancer Society. Cancer drug access for Canadians [online] Canadian Cancer Society 2009 [cited 2012 26 Sept]. Available from: http://www.cancer.ca/canada- wide/about%20us/media%20centre/cw-media%20releases/cw- 2009/~/media/CCS/Canada%20wide/Files%20List/English%20files%20heading/pdf%20not%2 0in%20publications%20section/CANCER%20DRUG%20ACCESS%20FINAL%20- %20English.ashx. 122. Fojo T, Grady C. How much is life worth: cetuximab, non-small cell lung cancer, and the $440 billion question. J Natl Cancer Inst. 2009;101(15):1044-8. 123. Fenton E. Making fair funding decisions for high-cost cancer care: the case of Herceptin in New Zealand. Public Health Ethics. 2010;3(2):137-46. 124. Brock DW. Ethical and value issues in insurance coverage for cancer treatment. Oncologist.15(Supplement 1):36-42. 125. Mitton CR, McMahon M, Morgan S, Gibson J. Centralized drug review processes: are they fair? Soc Sci Med. 2006;63(1):200-11. 126. Hind D, Wailoo AJ, Sutcliffe P. Demands for 'off-licence' access to trastuzumab (Herceptin): content analysis of UK newspaper articles. Health Expect. 2011;14(1):38-47. 127. The New Zealand Medicines and Medical Devices Safety Authority (Medsafe). Quality and safety of medicines- Medsafe's evaluation & approval process [online] 2012 [cited 2012 03 March]. Available from: http://www.medsafe.govt.nz/consumers/Safety-of- Medicines/Medsafe-Evaluation-Process.asp. 128. The United Kingdom Medicines And Health Care Products Regulatory Agency. Licensing of medicines [online] 2012 [cited 2012 03 March]. Available from: http://www.mhra.gov.uk/Howweregulate/Medicines/Licensingofmedicines/index.htm.

182

129. Blanchard R. Congressional Research Service report for Congress- The U.S. drug approval process: a primer [online] 2001 [cited 2012 03 March]. Available from: http://www.law.umaryland.edu/marshall/crsreports/crsdocuments/RL30989.pdf. 130. The Australian Department Of Health And Ageing Therapeutic Goods Administration. What the TGA doesn't do [online] 2012 [cited 2012 03 March]. Available from: http://www.tga.gov.au/about/tga-doesnt-do.htm. 131. The United Kingdom Medicines And Health Care Products Regulatory Agency. Frequently asked questions about the MHRA [online] 2012 [cited 2012 03 March]. Available from: http://www.mhra.gov.uk/SearchHelp/Frequentlyaskedquestions/index.htm. 132. The United States Food And Drug Administration. What doesn’t FDA regulate? How do I contact the agencies that do? [online] 2012 [cited 2012 26 Sept]. Available from: http://www.fda.gov/AboutFDA/Transparency/Basics/ucm203499.htm. 133. The United States Food And Drug Administration. New drug application (NDA) [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandAppro ved/ApprovalApplications/NewDrugApplicationNDA/default.htm. 134. The United Kingdom Medicines And Health Care Products Regulatory Agency. Marketing authorisations 2012 [cited 2012 03 March]. Available from: http://www.mhra.gov.uk/Howweregulate/Medicines/Licensingofmedicines/Marketingautho risations/index.htm. 135. The European Medicines Agency. Regulatory and procedural guidance index [online] 2012 [cited 2012 03 March]. Available from: http://www.ema.europa.eu/ema/index.jsp?curl=pages/regulation/general/general_content _000316.jsp&murl=menus/regulations/regulations.jsp&mid=WC0b01ac05800a4902&jsenabl ed=true. 136. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Streamlined submission process for prescription medicines [online] 2012 [cited 2012 03 March]. Available from: http://www.tga.gov.au/industry/pm-ssp.htm. 137. The New Zealand Medicines and Medical Devices Safety Authority (Medsafe). Guidelines & codes [online] 2012 [cited 2012 03 March]. Available from: http://www.medsafe.govt.nz/regulatory/Guideline/NZRGM%20Volume%201.asp. 138. Barnes D. How prescription drugs are developed. Aust Prescr. 2006;29:159-61. 139. The International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) Welcome to the ICH official website 2012 [cited 2012 03 March]. Available from: http://www.ich.org/. 140. Shuster E. Fifty years later: the significance of the Nuremberg Code. N Engl J Med. 1997;337(20):1436-40. 141. The United States National Institutes of Health. World Medical Association Declaration of Helsinki [online] 2004 [cited 2012 26 Sept]. Available from: http://history.nih.gov/research/downloads/helsinki.pdf. 142. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Access to unapproved therapeutic goods- clinical trials in Australia [online] 2004 [cited 2012 03 March]. Available from: http://www.tga.gov.au/pdf/clinical-trials-guidelines.pdf. 143. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Australian guideline for pharmacovigilance responsibilities of sponsors of registered medicines by Drug Safety And Evaluation Branch [online] 2003 [cited 2012 03 March]. Available from: http://www.tga.gov.au/pdf/australian-pharmacovigilance-guideline- 050531.pdf. 144. The United Kingdom Medicines And Health Care Products Regulatory Agency. Pharmacovigilance - how we monitor the safety of medicines [online] 2012 [cited 2012 03 March]. Available from:

183

http://www.mhra.gov.uk/Safetyinformation/Howwemonitorthesafetyofproducts/Medicines/ Pharmacovigilance/index.htm. 145. The United Kingdom Medicines And Health Care Products Regulatory Agency. New drugs and vaccines under intensive surveillance [online] 2012 [cited 2012 03 March]. Available from: http://www.mhra.gov.uk/Safetyinformation/Howwemonitorthesafetyofproducts/Medicines/ BlackTriangleproducts/index.htm. 146. The New Zealand Medicines and Medical Devices Safety Authority (Medsafe). Pharmacovigilance [online] 2012 [cited 2012 03 March]. Available from: http://www.medsafe.govt.nz/regulatory/pharmacovigilance.asp. 147. The New Zealand Medicines and Medical Devices Safety Authority (Medsafe). Guideline on the regulation of therapeutic products in New Zealand, part 7: advertising of therapeutic products [online] 2011 [cited 2012 03 March]. Available from: http://medsafe.govt.nz/regulatory/Guideline/GRTPNZ/Part%207Advertising%20of%20therap eutic%20products.doc. 148. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Therapeutic Goods Advertising Code 2007 [online] 2007 [cited 2012 03 March]. Available from: http://www.comlaw.gov.au/Details/F2007L00576. 149. The United Kingdom Medicines And Health Care Products Regulatory Agency. Advertising of medicines [online] 2012 [cited 2012 03 March]. Available from: http://www.mhra.gov.uk/Howweregulate/Medicines/Advertisingofmedicines/index.htm. 150. The United States Food And Drug Administration. Prescription drug advertising: questions and answers [online] 2012 [cited 2012 03 March]. Available from: http://www.fda.gov/Drugs/ResourcesForYou/Consumers/PrescriptionDrugAdvertising/UCM 076768.htm#control_advertisements. 151. The New Zealand Medicines and Medical Devices Safety Authority (Medsafe). Guidelines and codes- New Zealand code of good manufacturing practice for manufacture and distribution of therapeutic goods [online] 2012 [cited 2012 03 March]. Available from: http://www.medsafe.govt.nz/regulatory/Guideline/NZGMPCodePart1Intro.asp. 152. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Good manufacturing practice for therapeutic goods [online] 2009 [cited 2012 03 March]. Available from: http://www.tga.gov.au/industry/manuf-gmp-tg.htm. 153. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Australian code of good manufacturing practice for medicinal products [online] 2010 [cited 2012 03 March]. Available from: http://www.tga.gov.au/industry/manuf-medicines- cgmp.htm. 154. The United Kingdom Medicines And Health Care Products Regulatory Agency. Good manufacturing practice: background [online] 2012 [cited 2012 03 March]. Available from: http://www.mhra.gov.uk/Howweregulate/Medicines/Inspectionandstandards/GoodManufa cturingPractice/Background/index.htm. 155. The United States Food And Drug Administration. Good manufacturing practice (GMP) guidelines/inspection checklist [online] 2008 [cited 2012 03 March]. Available from: http://www.fda.gov/cosmetics/guidancecomplianceregulatoryinformation/goodmanufacturi ngpracticegmpguidelinesinspectionchecklist/default.htm. 156. The New Zealand Medicines and Medical Devices Safety Authority (Medsafe). New Zealand code of good manufacturing practice for manufacture and distribution of therapeutic goods [online] 1995 [cited 2012 03 March]. Available from: http://www.medsafe.govt.nz/downloads/Part4.pdf. 157. The United Kingdom Medicines And Health Care Products Regulatory Agency. Manufacturer's and wholesale dealer's licences [online] 2012 [cited 2012 03 March]. Available from: http://www.mhra.gov.uk/Howweregulate/Medicines/Licensingofmedicines/Manufacturersa ndwholesaledealerslicences/index.htm.

184

158. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Code of good wholesaling practice for therapeutic goods for human use [online] 1991 [cited 2012 03 March]. Available from: http://www.tga.gov.au/pdf/manuf-cgwp-tg.pdf. 159. The United States Food And Drug Administration. CFR - Code of Federal Regulations Title 21 [online] 2011 [cited 2012 03 March]. Available from: http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?fr=205.50. 160. The New Zealand Medicines and Medical Devices Safety Authority (Medsafe). Classification of medicines [online] 2012 [cited 2012 03 March]. Available from: http://www.medsafe.govt.nz/profs/class/classificationCategoriesAndCriteria.asp. 161. The United Kingdom Medicines And Health Care Products Regulatory Agency. Legal status and reclassification [online] 2012 [cited 2012 03 March]. Available from: http://www.mhra.gov.uk/Howweregulate/Medicines/Licensingofmedicines/Legalstatusandr eclassification/index.htm. 162. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Pathways to a scheduling decision [online] 2011 [cited 2012 03 March]. Available from: http://www.tga.gov.au/industry/scheduling-pathway.htm#medicines. 163. The United States Food And Drug Administration. Now available without a prescription [online] 2011 [cited 2012 03 March]. Available from: http://www.fda.gov/Drugs/ResourcesForYou/Consumers/ucm143547.htm. 164. The Organisation for Economic Co-operation and Development. OECD health data 2011- frequently requested data [online] [cited 2012 03 March]. Available from: http://www.oecd.org/document/16/0,3746,en_2649_37407_2085200_1_1_1_37407,00.htm l. 165. Simoens S. Health economic assessment: a methodological primer. Int J Environ Res Public Health. 2009;6(12):2950-66. 166. Plans-Rubio P. Frameworks to set priorities for treatments based on cost-effectiveness and equity. The Open Pharmacoeconomics & Health Economics Journal. 2012;4:1-7. 167. Hansen P. A theoretical review of PHARMAC’s over-arching approach to deciding which pharmaceuticals to fund, including high cost ones. Department of Economics University of Otago; May 2006. Report No. 168. Freemantle N, Bloor K, Eastaugh J. A fair innings for NICE? Pharmacoeconomics. 2002;20(6):389-91. 169. Mackenzie R, Chapman S, Salkeld G, Holding S. Media influence on Herceptin subsidization in Australia: application of the rule of rescue? Journal of the Royal Society of Medicine. 2008;101(6):305-12. 170. The Organisation for Economic Co-operation and Development. Drawing all the benefits from pharmaceutical spending. Value for money in health spending: OECD Publishing; 2010. 171. Begg E, Sidwell A, Gardiner S, Nicholls G, Scott R. The sorry saga of statins in New Zealand- pharmacopolitics versus patient care. N Z Med J 2003;116 (1170):U360. 172. Moodie P, Metcalfe S ,McNee W. Response from PHARMAC: difficult choices. N Z Med J. 2003;116(1170):U361. 173. Metcalfe S, Moodie P, Davies A, McNee W, Dougherty S. PHARMAC responds on salbutamol. N Z Med J. 2005;118(1221):U1644. 174. Giles J, Brown J, Byrnes C, Farrell A,Graham D. PHARMAC and Ventolin in New Zealand. N Z Med J. 2005;118(1220):U1616. 175. Tordoff JM, Norris PT,Reith DM. Managing prices for hospital pharmaceuticals: a successful strategy for New Zealand? Value Health. 2005;8(3):201-8. 176. Tordoff JM, Norris PT, Reith DM. "Price Management” and its impact on hospital pharmaceutical expenditure and the availability of medicines in New Zealand hospitals. Value Health. 2008;11(17):1214-26.

185

177. Ryall T. Government extends Pharmac role [online] [cited 2012 28 Sept]. Available from: http://www.beehive.govt.nz/release/government-extends-pharmac-role. 178. Aaltonen K, Ragupathy R, Tordoff J, Reith D, Norris P. The impact of pharmaceutical cost containment policies on the range of medicines available and subsidized in Finland and New Zealand. Value Health. 2010;13(1):148-56. 179. Donohoo E, editor. MIMS New Ethicals Jan–July 2007. Auckland: CMP Media; 2007. 180. The New Zealand Pharmaceutical Management Agency. Interactive Pharmaceutical Schedule [online] [cited 2012 25 Sept]. Available from: http://www.pharmac.govt.nz/interactive/. 181. The New Zealand Medicines and Medical Devices Safety Authority. Product/application search [online] [cited 2012 24 Sept]. Available from: http://www.medsafe.govt.nz/regulatory/DbSearch.asp. 182. The Australian Department of Health And Ageing. The Pharmaceutical Benefits Schedule [online] [cited 2012 26 Sept]. Available from: http://www.pbs.gov.au/html/home. 183. The Australian Department Of Health And Ageing Therapeutic Goods Administration. Australian Register of Therapeutic Goods (ARTG) [online] 2012 [cited 2012 22 May]. Available from: https://www.ebs.tga.gov.au/. 184. Datapharm Communications Ltd. electronic Medicines Compendium (eMC) [online] [cited 2012 24 Sept]. Available from: http://emc.medicines.org.uk/. 185. The United Kingdom National Health Service Business Service Authority. NHS Electronic Drug Tariff [online] [cited 2012 24 Sept]. Available from: http://www.nhsbsa.nhs.uk/924.aspx. 186. The United Kingdom National Institute For Health And Clinical Excellence (NICE). Welcome To The National Institute For Health And Clinical Excellence [online] [cited 2010 25 Oct]. Available from: http://www.nice.org.uk/. 187. Thomson PDR. Physicians’ Desk Reference Electronic Library (2007) [CD-ROM]2007. 188. The United States Department of Veterans Affairs. VA National Formulary [online] [cited 2012 24 Sept]. Available from: http://www.pbm.va.gov/NationalFormulary.aspx. 189. The United States Food And Drug Administration. Drugs@FDA [online] [cited 2012 24 Sept]. Available from: http://www.accessdata.fda.gov/scripts/cder/drugsatfda/index.cfm. 190. The World Health Organization Collaborating Centre For Drug Statistics Methodology. ATC structure and pinciples [online] [cited 2012 26 Sept]. Available from: http://www.whocc.no/atc/structure_and_principles/. 191. STATA Corp. STATA. 10.0 ed. College Station, Texas: STATA Corp. 192. The World Health Organization Collaborating Centre For Drug Statistics Methodology. Structure and principles [online] [cited 2011 24 Oct]. Available from: http://www.whocc.no/atc/structure_and_principles/. 193. The World Health Organization. WHO Collaborating Centre For Drug Statistics Methodology [online] [cited 2012 08 May]. Available from: http://www.whocc.no/atcddd/. 194. Roughead EE, Lopert R, Sansom LN. Prices for innovative products that provide health gain: a comparison between Australia and the United States. Value Health. 2007;10(6):514-20. 195. The United States Food And Drug Administration. How drugs are developed and approved- fast track approvals [online] [cited 2012 25 May]. Available from: http://www.fda.gov/Drugs/DevelopmentApprovalProcess/HowDrugsareDevelopedandAppro ved/DrugandBiologicApprovalReports/FastTrackApprovalReports/ucm082380.htm. 196. The Canadian Patented Medicines Price Review Board. Annual reports 2000 to 2008 [online] [cited 2012 25 May]. Available from: http://www.pmprb- cepmb.gc.ca/English/view.asp?x=91. 197. The World Health Organization. The global burden of disease 2004 update [online] [cited 2012 12 Sept]. Available from: http://www.who.int/healthinfo/global_burden_disease/GBD_report_2004update_full.pdf. 198. Walker J, Almond P. Interpreting statistical findings : a guide for health professionals and students. Berkshire, England: Open University Press; 2010.

186

199. Smith MK. Type I and II errors and signifcance levels [online] [cited 2013 23 May]. Available from: http://www.ma.utexas.edu/users/mks/statmistakes/errortypes.html. 200. Case CL. Hypothesis testing [online] [cited 2013 23 May]. Available from: http://www.smccd.edu/accounts/case/biol690/hypothesis.html. 201. Ragupathy R, Tordoff J, Norris P, Reith D. Key informants’ perceptions of how PHARMAC operates in New Zealand. Int J Technol Assess Health Care. 2012;28(4):367-73. 202. Rawson N. Time required for approval of new drugs in Canada, Australia, Sweden, the United Kingdom And the United States in 1996-1998. CMAJ. 2000;162(4):501-4. 203. The United Kingdom National Institute For Health And Clinical Excellence. Who we are [online] [cited 2012 24 May]. Available from: http://www.nice.org.uk/aboutnice/whoweare/who_we_are.jsp. 204. The United Kingdom Department of Health. DH good practice guidance on managing the introduction of new healthcare interventions and links to NICE technology appraisal guidance [online] [cited 2012 27 May]. Available from: http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuid ance/DH_064983. 205. Raftery JP. Paying for costly pharmaceuticals: regulation of new drugs in Australia, England and New Zealand. Med J Aust. 2008;188(1):26-8. 206. The New Zealand Parliament. New Zealand Public Health And Disability Act 2000 [online] [cited 2010 25 Oct]. Available from: http://www.legislation.govt.nz/. 207. Innova Software. electronic Australian Prescription Products Guide (April 2007) [CD- ROM]2007. 208. Paris V, Docteur E. OECD Health Working Papers 24: Pharmaceutical pricing and reimbursement policies in Canada [online] [cited 2011 04 Jan]. Available from: http://www.oecd.org/dataoecd/21/40/37868186.pdf. 209. The New Zealand Medicines And Medical Devices Safety Authority. Classification of Medicines- Classification Process [online] [cited 2011 22 Feb]. Available from: http://www.medsafe.govt.nz/profs/class/classificationprocess.asp#General. 210. The United Kingdom Department of Health. The cancer drugs fund- a consultation [online] [cited 2012 24 Sept]. Available from: http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/documents/di gitalasset/dh_120931.pdf. 211. Lancet Editorial. New £50 million cancer fund already intellectually bankrupt. Lancet. 2010;376(9739):389. 212. Maynard A, Bloor K. The economics of the NHS cancer drugs fund. Appl Health Econ Health Policy. 2011;9(3):137-8. 213. The United Kingdom Office of Fair Trading. The Pharmaceutical Price Regulation Scheme- an OFT market study [online] [cited 2012 24 Sept]. Available from: http://www.oft.gov.uk/shared_oft/reports/comp_policy/oft885.pdf. 214. The United Kingdom Department of Health. Equity and excellence: liberating the NHS [online] [cited 2012 24 Sept]. Available from: http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuid ance/DH_117353. 215. The United Kingdom Department of Health. Equity and excellence: liberating the NHS [Online]. [cited 2010 24 Dec]. Available from: http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH _117353. 216. The United States Food And Drug Administration. FDA news release: FDA begins process to remove breast cancer indication from Avastin label [Online]. [cited 2012 09 Sept]. Available from: http://www.fda.gov/newsevents/newsroom/pressannouncements/ucm237172.htm.

187

217. The United Kingdom Medicines and Healthcare products Regulatory Agency. Immediate withdrawal of rofecoxib (Vioxx/Vioxxacutate) [Online]. [cited 2012 09 Sept]. Available from: http://www.mhra.gov.uk/Safetyinformation/Safetywarningsalertsandrecalls/Safetywarnings andmessagesformedicines/CON1004263. 218. The United Kingdom Medicines And Healthcare Products Regulatory Authority. Rosiglitazone (Avandia, Avandamet): recommended withdrawal from clinical use [cited 2012 09 Sept]. Available from: http://www.mhra.gov.uk/Safetyinformation/Safetywarningsalertsandrecalls/Safetywarnings andmessagesformedicines/CON094121. 219. Motala D, De Ponti F, Rossi P, Martini N, Montanaro M. Therapeutic innovation in the European Union: analysis of the drugs approved by the EMEA between 1995 and 2003. Br J Clin Pharmacol. 2005;59(4):475-78. 220. Motala D, De Ponti F, Poluzzi E, Martini N, Rossi P, Silvani MC, et al. An update on the first decade of the European centralized procedure: how many innovative drugs? Br J Clin Pharmacol. 2006;62(5):610-16. 221. Sermet C, Andrieu V, Godman B, Van Ganse E, Haycox A, Reynier JP. Ongoing pharmaceutical reforms in France: implications for key stakeholder groups. Appl Health Econ Health Policy. 2010;8(1):7-24. 222. Godman B, Bucsics A, Burkhardt T, Haycox A, Seyfried H, Wieninger P. Insight into recent reforms and initiatives in Austria: implications for key stakeholders. Expert Rev Pharmacoecon Outcomes Res. 2008;8(4):357-71. 223. The United Kingdom Department of Health. The Pharmaceutical Price Regulation Scheme 2009 [Online]. [cited 2012 Sept 26]. Available from: http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuid ance/DH_091825. 224. The Organisation for Economic Co-operation and Development. OECD Health Data 2009 – comparing health statistics across OECD countries [Online]. [cited 2012 25 Sept]. Available from: http://www.oecd.org/health/oecdhealthdata2009comparinghealthstatisticsacrossoecdcount ries.htm. 225. Cohen J, Cairns C, Paquette C, Faden L. Comparing patient access to pharmaceuticals in the UK and US. Appl Health Econ Health Policy. 2006;5(3):177-87. 226. Nelson RE, McAdam-Marx C, Evans ML, Ward R, Campbell B, Brixner D, Lafleur J. Patent extension policy for paediatric indications: an evaluation of the impact within three drug classes in a state Medicaid programme. Appl Health Econ Health Policy. 2011;9(3):171-81. 227. New Zealand Accident Compensation Corporation (ACC). What support can I get? Prescription medicine [Online]. [cited 2012 09 Sept]. Available from: http://www.acc.co.nz/making-a-claim/what-support-can-i-get/ECI0021. 228. Jacobson GA, Panangala SV, Hearn J. CRS report for Congress- pharmaceutical costs: a comparison of Department of Veterans Affairs (VA), Medicaid and Medicare policies [Online]. [cited 2012 24 Sept]. Available from: http://lieberman.senate.gov/assets/pdf/crs/vapharma.pdf. 229. National Federation of Independent Business et al. v. Sebelius, Secretary of Health and Human Services, et al.: Supreme Court of the United States; 2012. 230. Ryall T. Government extends Pharmac role [Online]. [cited 2012 09 Sept]. Available from: http://www.beehive.govt.nz/release/government-extends-pharmac-role. 231. Jatrana S, Crampton P. Primary health care in New Zealand: who has access? Health Policy. 2009;93(1):1-10. 232. Jatrana S, Crampton P, Norris P. Ethnic differences in access to prescription medication because of cost in New Zealand. J Epidemiol Community Health. 2011;65(5):454-60.

188

233. Norris P, Horsburgh S, Keown S, Arroll B, Lovelock K, Cumming J, Herbison P, Crampton P, Becket G. Too much and too little? Prevalence and extent of antibiotic use in a New Zealand region. J Antimicrob Chemother. 2011;66(8):1921-6. 234. Norris P, Horsburgh S, Keown S, Arroll B, Lovelock K, Cumming J, Herbison P, Crampton P, Becket G. Medicalisation or under-treatment?: Psychotropic medicines use by elderly people and children in New Zealand. Health Soc Rev. 2011;20(2):202-18. 235. Jatrana S, Crampton P, Richardson K, Norris P. Increasing prescription part charges will increase health inequalities in New Zealand. N Z Med J. 2012;125(1355):78-80. 236. MacKay P. Is PHARMAC’s sole-supply tendering policy harming the health of New Zealanders? N Z Med J. 2005;118(1214):U1433. 237. Ryall T. Budget 2012: Prescription charges help fund Health [Online]. 2012 [cited 2012 15 July]. Available from: http://www.national.org.nz/Article.aspx?articleId=38485. 238. Lee TH, Emanuel EJ. Tier 4 drugs and the fraying of the social compact. N Engl J Med. 2008;359(4):333-5. 239. Polinski JM, Mohr PE, Johnson L. Impact of Medicare Part D on access to and cost sharing for specialty biologic medications for beneficiaries with rheumatoid arthritis. Arthritis Rheum. 2009;61(6):745-54. 240. The Commonwealth Fund. Prescription Drug Accessibility and Affordability in the United States and Abroad [Online]. 2010 [cited 2012 09 July]. Available from: http://www.commonwealthfund.org/~/media/Files/Publications/Issue%20Brief/2010/Jun/1 408_Morgan_Prescription_drug_accessibility_US_intl_ib.pdf. 241. Business and Economic Research Limited (BERL). Independent review of the Castalia report on New Zealand pharmaceutical policies [Online]. 2005 [cited 2012 09 Sept]. Available from: http://www.pharmac.govt.nz/2006/05/04/040506b.pdf. 242. The United Kingdom Medicines And Health Care Products Regulatory Agency. MRHA: About us [Online]. [cited 2012 08 Sep]. Available from: http://www.mhra.gov.uk/Aboutus/index.htm. 243. Rosevear M. PHARMAC and Herceptin for early-stage breast cancer in New Zealand: Herceptin or deception? New Zealand Medical Journal. 2006;119(1235).

189

APPENDIX I: CONFERENCE PRESENTATIONS

Ragupathy R, Tordoff J, Norris P and Reith D. Life in a nine-dimensional universe: New Zealanders’ access to pharmaceuticals. Presented at the New Zealand Hospital Pharmacists Association Conference, Auckland, 2012 (Oral).

Ragupathy R, Tordoff J, Norris P and Reith D. Found in New Zealand and no-where else: kiwi, rimu, tutatara…PHARMAC? Presented at the New Zealand Hospital Pharmacists Association Conference, Tauranga, 2011 (Oral).

Ragupathy R, Tordoff J, Norris P and Reith D. What changes in New Zealand’s medicines regulation system mean for us. Presented at the University of Auckland Waikato Clinical School Research Forum, Hamilton 2011 (Oral).

Ragupathy R, Aaltonen K, Tordoff J, Norris P and Reith D. The effect of pharmaceutical cost containment on access to licensed and subsidized medicines under single payer systems in the United States, United Kingdom, Australia and New Zealand. Presented at the 16th Annual International Society For Pharmacoeconomics and Outcomes Research (ISPOR) Conference, Baltimore, United States, 2011 (Poster).

Ragupathy R, Aaltonen K, Tordoff J, Norris P and Reith D. Medicines licensed for children and subsidised under single payer systems in New Zealand, Australia, the United Kingdom and the United States. Presented at the 16th International Social Pharmacy Workshop, Lisbon, Portugal, 2010 (Oral).

190

APPENDIX II: PUBLICATIONS

Ragupathy R, Aaltonen K, Tordoff J, Norris P, Reith D. A 3-dimensional view of access to licensed and subsidized medicines under single payer systems in the United States, United Kingdom, Australia and New Zealand. Pharmacoeconomics. 2012; 30(11): 1051-65

Ragupathy R, Tordoff J, Norris P, Reith D. Key Informants’ Perceptions Of How PHARMAC Operates In New Zealand. Int J Technol Assess Health Care. 2012;28(4): 367-73.

191

DATA ANNEX 1: DUPLICATE ANALYSES BY ACTIVE AGENT

DA 1.1 INVESTIGATION ONE

Country Licensed Subsidised Percentage of p Value for adults for adults entities licensed (Chi2 test, for adults that 2x2 table) were subsidised for adults

a New Zealand 759 486b 64.0% <0.001

Australia 862a 554b 64.3% <0.001

United Kingdom 1005a 1001b 99.6% 0.746

United States 979a 503 51.4% 0.051

(VANF)b

2 a Chi test p value <0.001 for each pair of countries (2x2 tables).

2 b Chi test p value <0.001 for New Zealand compared to Australia, New Zealand compared to the United States VANF and Australia compared to the United States VANF (2x2 tables). P values for the United Kingdom compared to any other country were not statistically significant.

TABLE DA 1.1: Number of entities collapsed by active agent that were licensed and subsidised for adults in each country

192

Summary of times since first registration of entities licensed for adults collapsed by active agent (a) Median 25th 75th Min Max (Days) Range (Days) Percentile Percentile (Days) (Days) (Days) (Days) New Zealand 8,657a 1,271-4,694 13,598- 557 25,162 24,605 (N=759) 21,591 Australia 7,620a 929-4,210 13,082- 704 25,162 24,458 (N=862) 21,629 United Kingdom 7,138a 551- 3,822 12,310- 194 25,162 24,968 (N=1005) 21,629 United States 6,576a 521- 3,532 11,898- 195 24,927 24,732 (N=979) 23,659 Summary of times since first registration of entities subsidised for adults collapsed by active agent (b) New Zealand 10,606b 1,607-6,521 13,879- 970 25,162 24,192 (N=486) 21,591 Australia 7,795b 1,088-4,331 13,108- 923 25,162 24,239 (N=554) 21,629 United Kingdom 7,092b 551-3,794 12,304- 194 25,162 24,968 (N=1001) 21,629 United States 8,103b 557-46,010 13,098- 195 23,978 23,783 (N=503 (VANF)) 21,475 a P <0.01 for differences in median time since first registration for New Zealand compared with the United Kingdom, New Zealand compared with the United States , and Australia compared with the United States (Wilcoxon Rank Sum Test). P = 0.0163 for New Zealand compared with Australia. b P <0.01 for differences in median time since first registration for New Zealand compared with all other countries, and the United States VANF compared with Australia and the United Kingdom (Wilcoxon Rank Sum Test).

TABLE DA 1.2: Comparisons of the age (time since first registration) of licensed and subsidised entities (collapsed by active agent) in each country (The Kruskal-Wallis test for probability of equality of medians found p< 0.001 when comparing (a) or (b) between countries

193

Licensed by active agent Country 1 Median age Country 2 Median age Difference in P value country 1 country 2 median ages (Wilcoxon (Days) (Days) (Days) Rank Sum test)

NZ 8,657 AUS 7,620 1,037 0.0163 NZ 8,657 UK 7,138 1,519 <0.0001 NZ 8,657 US 6,576 2,081 <0.0001 AUS 7,620 UK 7,138 482 0.0285 AUS 7,620 US 6,576 1,044 <0.0001 UK 7,138 US 6,576 562 0.0831 Subsidised by active agent NZ 10,606 AUS 7,795 2,811 <0.0001 NZ 10,606 UK 7,091.5 3,514.5 <0.0001 NZ 10,606 US (VANF) 8,103 2,503 <0.0001 AUS 7,795 UK 7,091.5 704 0.0039 AUS 7,795 US (VANF) 8,103 -308 0.3666 UK 7,091.5 US (VANF) 8,103 -1,011.5 0.0002

TABLE DA 1.3: Differences in the age (median times since first registration) of licensed and subsidised entities in each pair of countries

194

Median 25th 75th Min Max Range (Days) Percentile Percentile (Days (Days) (Days) (Days) (Days) ) Time from first registration of licensed entities collapsed by active agent

Licensed in NZ but 7,810 1,279-4,053 12,527-19,085 557 21,053 20,946 not in AUS Licensed in NZ but 9,263 1,607-4,494 13,879-19,880 970 23,978 23,978 not in the UK Licensed in NZ but 9,594 970- 1,271 13,879-21,031 970 25,162 25,162 not in the US Licensed in AUS 5,460 929-3127 9,553-20,647 704 22,071 21,367 but not in NZ Licensed in AUS 8,417 1,431-5,108 13,879-20,275 1019 23,978 23,978 but not in the UK Licensed in AUS 8,379 11,46-5,177 13,879-21,584 923 25,162 25,162 but not in the US Licensed in the 4,962 551- 2,727 9,148-21,416 194 24,248 24,054 UK but not in NZ Licensed in the 6,275 551-3126 10,602-20,282 194 24,248 24,054 UK but not in AUS Licensed in the 8,065 551-4,948 13,393-21,584 194 25,162 25,162 UK but not in US Licensed in the US 5,081 521-2,600 9,553-23,376 195 24,927 24,732 but not in NZ Licensed in the US 5,901 521- 2,659 10,206-23,659 195 24,927 24,732 but not in AUS Licensed in the US 6,591 521-3,248 12,260-23,376 195 24,927 24.732 but not in UK

195

Median 25th 75th Min Max Range (Days) Percentile Percentile (Days) (Days) (Days) (Days) (Days) Time from first registration of subsidised entities collapsed by active agent

Subsidised in NZ 11,587 2,653-7,865 13,879-20,424 970 21,591 21,591 but not in AUS Subsidised in NZ 10,879 1,928- 13,879-19,834 970 23,978 23,978 but not in the UK 6,773 Subsidised in NZ 10, 610 1,607- 13,879-21,584 970 25,162 25,162 but not in the US 6,781 VANF Subsidised in 4,617 1,146-3,217 8,755-20,065 923 2,2071 21,148 AUS but not in NZ Subsidised in 8,020 1,727-4,428 1,3879-19,880 1,019 23,978 22,959 AUS but not in the UK Subsidised in 6,641 1,146-3,794 12,387-20,667 923 25,162 24,239 AUS but not in the US VANF Subsidised in the 4,962 551-2,851 8,889-21,416 194 24,248 24,054 UK but not in NZ Subsidised in the 6,234 551-3,176 10,896-21,584 194 24,248 24,054 UK but not in AUS Subsidised in the 6,244 551- 3,338 11,064-21,584 194 25,162 24,968 UK but not the US VANF Subsidised in the 6,251 557-3,749 10,427-21,416 195 23,978 23,783 US VANF but not in NZ Subsidised in the 7,550 557- 4,171 12,373-20,647 195 23,978 23,783 US VANF but not in AUS Subsidised in the 8,110 1,053-4,666 13,087-19,564 195 21,475 21,280 US VANF but not the UK

TABLE DA 1.4: Comparison of the age (time since first registration) of specific sub groups (Subgroups are entities that were licensed (or subsidised) in one country but not in another)

196

Time since first registration of licensed entities collapsed by active agent Sub group 1 Median sub Sub group 2 Median Difference P value group 1 sub group in (Wilcoxon (Days) 2 (Days) medians Rank (Days) Sum) Licensed in NZ but 7,810 Licensed in 5,460 2,170 0.0002 not in Australia Australia but not in NZ Licensed in NZ But 9,263 Licensed in the UK 4,962 4,301 <0.0001 not in the UK but not in NZ Licensed in NZ but 9,594 Licensed in the US 5,081 4,513 <0.0001 not in the US But not in NZ Licensed in Australia 8,417 Licensed in the UK 6,275 2,142 0.0001 but not in the UK but not in Australia

Licensed in Australia 8,379 Licensed in the US 5,901 2,478 <0.0001 but not in the US but not in Australia

Licensed in the UK 8,065 Licensed in the US 6,591 1,474 0.0083 but not in the US but not in the UK

TABLE DA 1.5: Differences in the age (median time since first registration) of subgroups of licensed entities (Subgroups are entities licensed in one country but not in another)

197

Time since first registration of subsidised entities collapsed by active agent Sub group 1 Median Sub group 2 Median Difference P value sub group sub in median (Wilcoxon 1 (Days) group 2 Rank Sum) Subsidised in NZ but 11,587 Subsidised in Australia 4,617 6,970 <0.0001 not in Australia but not In NZ

Subsidised in NZ but 10,879 Subsidised in the UK but 4,962 5,917 <0.0001 not in the UK not in NZ Subsidised in NZ but 10,610 Subsidised in the US 6,251 4,359 <0.0001 not in the US VANF VANF but not in NZ

Subsidised in 8,020 Subsidised in the UK but 6,641 1,379 0.0031 Australia but not in not in Australia the UK Subsidised in 6,642 Subsidised in the US 7,550 -908 0.3957 Australia but not in VANF but not in the US VANF Australia Subsidised in the UK 6,244 Subsidised in the US 8,110 -1,866 0.0005 but not in the US VANF but not in the UK VANF

TABLE DA 1.6: Differences in the age (median time since first registration) of subgroups of subsidised entities (Subgroups are entities subsidised in one country but not in another)

198

Entities available in Total Percentage of entities non-injectable number of available in non- formulations entities injectable formulations Licensed New Zealand 527 759 69.40% Subsidised New Zealand 395 486 81.30% Licensed Australia 596 862 69.1% Subsidised Australia 453 554 81.8% Licensed United Kingdom 747 1005 74.3% Subsidised United Kingdom 745 1001 74.4% Licensed United States 776 979 79.3% Subsidised United States 377 503 75.0% VANF

TABLE DA 1.7: Percentage of entities available in non-injectable formulations (collapsed by active agent)

199

Median time since first Median time Difference In P value registration of entities since first median time (Wilcoxon available in non- registration of since first Rank injectable formulations all entities registration Sum test) (Days) (Days) (Days) Licensed New 9572 8657 915 0.13 Zealand Subsidised New 10931 10606 325 0.79 Zealand Licensed Australia 8366 7620 746 0.14 Subsidised 8378 7795 583 0.37 Australia Licensed United 7511 7138 373 0.16 Kingdom Subsidised United 7425 7092 333 0.16 Kingdom Licensed United 7245 6576 669 0.06 States Subsidised United 9591 8103 1488 0.04 States VANF TABLE DA 1.8: Median time since first registration of licensed entities available in non- injectable formulation compared with all entities (collapsed by active agent)

200

DA 1.2: INVESTIGATION TWO

Country Licensed In Subsidised In Percentage p Value (Chi2 Children Children Subsidised In test, 2x2 table) Children

New Zealand 442a 299b 67.6% p<0.001

Australia 454a 259b 57.0% p<0.001

United 551a 547b 99.3% p<0.001 Kingdom

United States 330a 184 (VANF)b 55.8% p<0.001

2 a Chi test p value <0.001 for each pair of countries (2x2 tables).

2 b Chi test p value <0.001 for each pair of countries (2x2 tables).

TABLE DA 1.9: Number of entities licensed and subsidised for children, collapsed by active agent

201

Country Children Adults Percentage p Value (Chi2 test, 2x2 table) New Zealand 442 759 58.2% 0.370 Australia 454 862 52.7% p<0.01 United Kingdom 551 1005 54.8% 0.078 United States 330 979 33.7% p<0.01 Subsidised by active agent New Zealand 299 486 61.5% p<0.001 Australia 259 554 46.8% p<0.001 United Kingdom 547 1001 54.7% p<0.001 United States 184 503 (VANF) 36.6% p<0.001 (VANF)

TABLE DA 1.10: Comparison of entities licensed and subsidised for children and adults within each country

202

Summary of times since first registration of entities licensed for children collapsed by active agent (a) Median 25th 75th Min Max (Days) Range (Days) Percentile Percentile (Days) (Days) (Days) (Days)

New Zealand 9783a 5504 13879 557 25162 24605 Australia 9369a 5114 13879 599 25162 24563 United Kingdom 9034a 5114 13879 557 25162 24605 United States 7043a 4046 12667 326 23659 23333 Summary of times since first registration of entities subsidised for children collapsed by active agent (b)

New Zealand 12135b 7823 13879 1607 25162 23555 Australia 10408b 5891 13879 1007 25162 24155 United Kingdom 9083b 5131 13879 557 25162 24605 United States 8070b 5149 13411 326 20731 20405 (VANF) a The only statistically significant difference in time since first registration of licensed entities was between entities licensed in the United States and every other country (Wilcoxon Rank Sum test, p <0.01 in each instance). b The only statistically significant difference in time since first registration of subsidised entities was between entities subsidised in New Zealand and subsidised in the United Kingdom (Wilcoxon Rank Sum test, p <0.001), and between entities subsidised in New Zealand and the United States VANF (Wilcoxon Rank Sum test, p <0.001).

TABLE DA 1.11: Comparisons of the age (times since first registration) of entities licensed and subsidised for children in the four countries (Kruskal-Wallis: Probability that a or b equal between countries is <0.001)

203

Median 25th Percentile 75th Percentile Min Max Range (Days) (Days) (Days) (Days) (Days) (Days) Time from first registration of licensed entities collapsed by active agent Licensed in NZ but 9,517 5,422 13,879 557 21,053 20,496 not in AUS Licensed in NZ but 9,406 5,223 13,879 1,271 20,275 19,004 not in the UK Licensed in NZ but 10,932 5,963 13,879 1,271 25,162 23,891 not in the US Licensed in AUS 7,404 4,205 13,879 599 22,071 21,472 but not in NZ Licensed in AUS 9,406 5,223 13,879 1,271 20,275 19,004 but not in the UK Licensed in AUS 10,240 5582 13,879 599 25,162 24,563 but not in the US Licensed in the UK 7,307 4,402 12,373 650 24,248 23,598 but not in NZ Licensed in the UK 7,307 4402 12,373 650 24,248 23,598 but not in AUS Licensed in the UK 9,767 5,954 13,879 1,146 25,162 24,016 but not in US Licensed in the US 6,347 3,797 23,659 326 23,659 23,333 but not in NZ Licensed in the US 6,510 3,771 11,552 326 23,659 23,333 but not in AUS Licensed in the US 6,955 3,730 12,400 326 23,659 23,333 but not in UK

204

Median 25th Percentile 75th Percentile Min Max Range (Days) (Days) (Days) (Days) (Days) (Days) Time from first registration of subsidised entities collapsed by active agent Subsidised in NZ but 12,318 8,603 13,879 1,607 21,591 19,984 not in AUS Subsidised in NZ but 10,910 6,955 13,879 1,607 20,579 18,972 not in the UK Subsidised in NZ but 12,395 8,301 13,879 1,607 25,162 23,555 not in the US VANF Subsidised in AUS 7,984 4,245 13,879 1,007 22,071 21,064 but not in NZ Subsidised in AUS 9,594 4,634 13,879 1,007 22,071 21,064 but not in the UK Subsidised in AUS 10,372 5,919 13,879 1,088 25,162 24,074 but not in the US VANF Subsidised in the UK 6,921 4,078 11,583 557 20,647 20,090 but not in NZ Subsidised in the UK 8,182 4,728 13,073 557 24,248 23,691 but not in AUS Subsidised in the UK 9,154 5,131 13,879 650 25,162 24,512 but not the US VANF Subsidised in the US 7,307 4,590 10,750 326 20,065 19,739 VANF but not in NZ Subsidised in the US 7,578 5,212 12,239 326 19,085 18,759 VANF but not in AUS

Subsidised in the US 7,795 5,387 13,817 326 19,834 19,508 VANF but not the UK

TABLE DA 1.12: Comparison of the age of subgroups of entities licensed and subsidised for children in the four countries (Subgroups are entities licensed or subsidised in one country, but not in another)

205

Time since first registration of licensed entities collapsed by active agent Sub group 1 Median Sub group 2 Median sub Difference in P value sub group group 2 medians (Wilcoxon 1 (Days) (Days) (Days) Rank Sum) Licensed in NZ 9517 Licensed in 7404 2,113 0.032 but not in Australia but not Australia in NZ Licensed in NZ 9,406 Licensed in the 7307 2,099 0.026 But not in the UK but not in UK NZ Licensed in NZ 10932 Licensed in the 6,347 4,586 <0.001 but not in the US But not in US NZ Licensed in 9,406 Licensed in the 7,307 2,099 0.026 Australia but UK but not in not in the UK Australia Licensed in 10,240 Licensed in the 6510 3,730 <0.001 Australia but US but not in not in the US Australia Licensed in the 9767 Licensed in the 6955 2,812 <0.001 UK but not in US but not in the the US UK

TABLE DA 1.13: Differences in the age (median time since first registration) of subgroups of entities licensed for children (Subgroups are entities licensed in one country but not in another)

206

Time since first registration of subsidised entities collapsed by active agent Sub group 1 Median Sub group 2 Median Difference P value sub group sub group in median (Wilcoxon 1 (Days) 2 (Days) (Days) Rank Sum) Subsidised in NZ 12,318 Subsidised in 7984 4,334 <0.001 but not in Australia but not Australia In NZ Subsidised in NZ 10,910 Subsidised in the 6921 3,989 <0.001 but not in the UK UK but not in NZ

Subsidised in NZ 12395 Subsidised in the 7,307 5,088 <0.001 but not in the US US VANF but not VANF in NZ Subsidised in 9,594 Subsidised in the 8182 1,412 0.161 Australia but not UK but not in in the UK Australia Subsidised in 10372 Subsidised in the 7578 2,794 0.002 Australia but not US VANF but not in the US VANF in Australia Subsidised in the 9,154 Subsidised in the 7795 1,359 0.687 UK but not in the US VANF but not US VANF in the UK

TABLE DA 1.14: Differences in the age (median time since first registration) of subgroups of entities subsidised for children (Subgroups are entities licensed in one country but not in another)

207

Times since first registration of licensed entities collapsed by active agent Country Median Median Difference in P value children adults medians (Wilcoxon (Days) (Days) (Days) Rank Sum test) New Zealand 9783 8657 1126 <0.005 Australia 9369 7620 1749 <0.001 United Kingdom 9034 7138 1896 <0.001 United States 7043 6576 467 0.059 Times since first registration of subsidised entities collapsed by active agent New Zealand 12135 10606 1529 0.005 Australia 10408 7795 2613 <0.001 United Kingdom 9083 7092 1991 <0.001 United States (VANF) 8070 8103 Minus 33 0.593

TABLE DA 1.15: Comparison of median time since first registration of entities licensed and subsidised for children with those licensed and subsidised for adults in each country

208

Country Median Median Difference P value licensed for subsidised for (Months) (Wilcoxon children children Rank Sum test) (Months) (a) (Months) (b)

New Zealand 0a 0b 0 0.170

Australia 1a 1.75b -0.75 0.455

United Kingdom 0a 0b 0 0.989

United States 24a 12 (VANF)b 12 0.114 a Differences between the United States and all other countries were statistically significant (p<0.001, Wilcoxon Rank Sum test). b Differences between the United States VANF and all other countries were statistically significant (p<0.001, Wilcoxon Rank Sum test). Differences between New Zealand and Australia were statistically significant (p<0.001, Wilcoxon Rank Sum test).

TABLE DA 1.16: Lowest age authorised within each country for licensed and subsidised entities collapsed by active agent (Kruskal-Wallis test probability (a) or (b) equal between countries <0.001)

209

Licensed by Active Agent

Country Suitably Total number of % of entities that formulated entities are suitably entities formulated New Zealand 340 442 76.9% Australia 329 454 72.5% United Kingdom 431 551 78.2% United States 244 330 73.9% Subsidised by active agent New Zealand 221 229 73.9% Australia 169 259 65.3% United Kingdom 428 547 78.2% United States 142 184 77.2%

TABLE DA 1.17: Number and proportions of suitably formulated licensed and subsidised entities in each country

210

Licensed by Active Agent Country Lowest age Lowest age Difference P value authorised suitably authorised non- (Months) (Wilcoxon formulated entities suitably Rank Sum (Months) formulated test) entities (Months) New Zealand 0 24 -24 <0.001 Australia 1.75 60 -58.25 <0.001 United Kingdom 0 60 -60 <0.001 United States 12 72 -60 <0.001 Subsidised by active agent New Zealand 0 6 -6 <0.001 Australia 0 36 -36 <0.001 United Kingdom 0 60 -60 <0.001 United States 6 (VANF) 72 (VANF) -66 <0.001

TABLE DA 1.18: Difference in lowest age authorised of suitably formulated and non- suitably formulated entities

211

DATA ANNEX TWO: SUPPLEMENTARY ANALYSES

Table DA 2.1 below shows the ATC level 4 groups that were listed in only one country, or listed in three of the four countries.

Table DA 2.2 below shows the ATC level 4 groups that were subsidised in only one country, or listed in three of the four countries.

New Zealand had no insulin secretagogues listed (A10BX) while the other countries had repaglinide (Australia), repaglinide and nateglinide (UK), and repaglinide, nateglinide and exenatide (United States). Omalizumab (R03DX, other systemic drugs for obstructive airways disease) was listed in the other three countries, but not in New Zealand.

Several entities were listed only in the United Kingdom: rimonabant (A08AX, other antiobesity drugs), sitagliptin (A10BH, drugs used in diabetes, dipeptidyl peptidase 4 (DPP-4) inhibitors), guanethidine (C02CC, antihypertensives, guanidine derivatives), and piracetam (N06BX, psychoanalpeptics, other psychostimulants and nootropics). There were no other entities in any of these groups (A08AX, A10BH, C02CC or N06BX) listed in the other three countries.

The United States did not have milrinone (C01CE, cardiac therapy, phosphodiesterase inhibitors), nitroprusside (C02DD, antihypertensives, nitroferricyanide derivatives) or phentolamine (C04AB, peripheral vasodilators, imidazoline derivatives) listed in the PDR, but these were listed in the other three countries. The United States had no other entities listed in any of these groups (C01CE, C02DD and C04AB).

The United States did not have carmustine, lomustine or fotemustine (L01AD, antineoplastic agents, nitrosoureas) listed in the PDR, but carmustine and lomustine were listed in all three of the other countries, and Australia had fotemustine listed. The US did not have bleomycin or mitomycin (L01DC, antineoplastic agents, other cytotoxic antibiotics), but these were listed in the other three countries. The United States had no other entities listed in either of the L01AD or L01DC groups.

212

The US did not have chlorpromazine, levomepromazine and promazine (N05AA, psycholeptics, phenothiazines with aliphatic side-chain) listed in the PDR, whereas New Zealand had chlorpromazine and levomepromazine listed, Australia had chlorpromazine listed and the UK had chlorpromazine, levomepromazine and promazine listed. The US did not have clomethiazole, dexmedetomidine or triclofos (N05CM, psycholeptics, other hypnotics and sedatives) listed in the PDR, while New Zealand had dexmedetomidine, Australia had clomethiazole, dexmedetomidine, and the UK had clomethiazole and triclofos.

The US did not have moclobemide (N06AG, psychoanaleptics, monoamine oxidase A inhibitors) or memantine (N06DX, other anti-dementia drugs) listed, whereas these were listed in all of the other countries. The United States had no other entities listed in the N05AA, N05CM, N06AG or N06DX groups.

The United States is the only country that had mecamylamine (C02BB, antihypertensives, secondary and tertiary amines), metirosine (C02KB, antihypertensives, tyrosine hydroxylase inhibitors) or ramelteon (N05CH, psycholeptics, melatonin receptor agonists) listed. None of the other countries had any other entities listed in the C02BB, C02KB or N05CH groups.

213

Australia and the UK subsidised all three agents, and the United States VANF subsidised donepezil and galantamine.

214

ATC level 4 group Entities Entities Entities listed in Entities Notes listed in New listed in the UK listed In Zealand Australia the US

A10BX (drugs used in None A10BX02 A10BX02 A10BX02 NZ: repaglinide (NovoNorm®) was diabetes, other blood repaglinide repaglinide; repaglinid licensed in NZ, but discontinued in glucose lowering agents, A10BX03 e; 2006. Nateglinide (Starlix®) is licensed, excluding insulins) nateglinide A10BX03 but not marketed. Exenatide (Byetta®) nateglinid received marketing approval on e; 31/05/2007 A10BX04 exenatide

R03DX (drugs for None R03DX05 R03DX05 R03DX05 NZ: omalizumab (Xolair®) received obstructive airways omalizumab omalizumab omalizum marketing approval on 23/09/2004 diseases, other systemic ab drugs for obstructive airways disease)

215

ATC level 4 group Entities listed in Entities Entities listed in Entities listed Notes New Zealand listed in the UK In the US Australia A08AX (other antiobesity None None A08AX01 None NZ: New Medicine Applications were made for drugs) rimonabant rimonabant (Acomplia®, Rimonabant Winthorp®) to Medsafe in 2006, but withdrawn in 2008. Australia: Not listed on the Australian Register of Therapeutic Goods. US: FDA Briefing document (NDA 21-888, June 2007) states that rimonabant was available in several other countries, but only being considered by the FDA at that stage due to safety concerns. Not listed in Drugs@FDA (as of March 2010) A10BH (Drugs used in None None A10BH01 None NZ: sitagliptin (Januvia®) was licensed by diabetes, dipeptidyl sitagliptin Medsafe on 28/02/2008. peptidase 4 (DPP-4) Australia: Januvia® added to the Australian inhibitors) Register of Therapeutic Goods on 14/01/2008. US: Licensed by the FDA on 16/10/2006.

216

ATC level 4 group Entities listed in Entities Entities listed in Entities listed Notes New Zealand listed in the UK In the US Australia C02CC None None C02CC02 None NZ: guanethidine (Ismelin®) first licensed in (Antihypertensives, guanethidine 1969, but was discontinued in New Zealand in Guanidine derivatives) 2001. Australia: Not listed in the Australian Register of Therapeutic Goods. US: First listed in the US in 1960, discontinued N06BX None None N06BX03 None NZ: Not listed by Medsafe (psychoanalpeptics, other piracetam Australia: Not listed on the Australian Register psychostimulants and of Therapeutic Goods nootropics) US: Not listed on Drugs@FDA. Designated as an orphan drug in the US in 1987 for treatment of myoclonus, but not approved for this indication. C01CE (cardiac therapy, C01CE02 C01CE02 C01CE02 None US: milirinone is listed in Drugs@FDA, first phosphodiesterase milrinone milrinone milrinone listed 31/12/1987. inhibitors)

217

ATC level 4 group Entities listed in Entities Entities listed in Entities listed Notes New Zealand listed in the UK In the US Australia C02DD (antihypertensives, C02DD01 C02DD01 C02DD01 None US: nitroprusside is listed in Drugs@FDA, first nitroferricyanide nitroprusside nitroprusside nitroprusside listed 08/09/1981 derivatives) C04AB (peripheral C04AB01 C04AB01 C04AB01 None US: phentolamine is listed in Drugs@FDA, first vasodilators, Imidazoline phentolamine phentolamin phentolamine listed 30/01/1952 derivatives) e L01AD (antineoplastic L01AD01 L01AD01 L01AD01 None US: carmustine is listed in Drugs@FDA (first agents, nitrosoureas) carmustine; carmustine; carmustine; listed 7/03/1977), as is lomustine (4/08/ 1976) L01AD02 L01AD02 L01AD02 lomustine lomustine; lomustine; L01AD05 fotemustine

218

ATC level 4 group Entities listed in Entities listed Entities listed in Entities listed Notes New Zealand in Australia the UK In the US

L01DC (antineoplastic L01DC01 L01DC01 L01DC01 None US: bleomycin is listed in Drugs@FDA (first agents, other cytotoxic bleomycin; bleomycin; bleomycin; listed 31/07/1973), as is mitomycin antibiotics) L01DC03 L01DC03 L01DC03 (24/08/1981) mitomycin mitomycin mitomycin N05AA (psycholeptics, N05AA01 N05AA01 N05AA01 None US: chlorpromazine is listed in Drugs@FDA phenothiazines with chlorpromazine; chlorpromazin chlorpromazine; (first listed 18/09/1957), as is levopromazine aliphatic side-chain) N05AA02 e N05AA02 (no listing date) and promazine (13/04/1956) levomepromazine levomepromazine , ; N05AA03 promazine N05CM (psycholeptics, N05CM18 N05CM02 N05CM02 None US: dexmedetomidine is listed in Drugs@FDA other hypnotics and dexmedetomidine clomethiazole; clomethiazole; ( 17/12/1999), as is triclofos (no listing date) sedatives) N05CM18 N05CM07 dexmedetomid triclofos ine

219

ATC level 4 group Entities listed in Entities listed Entities listed in Entities listed Notes New Zealand in Australia the UK In the US N06AG (Psychoanaleptics, N06AG02 N06AG02 N06AG02 None US: not listed in Drugs@FDA. monoamine oxidase A moclobemide moclobemide moclobemide inhibitors) N06DX (other anti- N06DX01 N06DX01 N06DX01 None US: memantine is listed in Drugs@FDA, first dementia drugs) memantine memantine memantine listed 16/10/1953

C02BB (antihypertensives, None None None C02BB01 NZ: Not Listed by Medsafe secondary and tertiary mecamylamin Australia: Not listed on the Australian Register amines) e of Therapeutic Goods US: First listed in Drugs@FDA on 01/03/1956 C02KB (antihypertensives, None None None C02KB01 NZ: Not Listed by Medsafe tyrosine hydroxylase metirosine Australia: Not listed on the Australian Register inhibitors) of Therapeutic Goods US: First listed in Drugs@FDA on 03/10/1979. N05CH (psycholeptics, None None None N05CH02 NZ: Not Listed by Medsafe melatonin receptor ramelteon Australia: Not listed on the Australian Register agonists) of Therapeutic Goods US: Listed in Drugs@FDA on 22/07/2005. TABLE DA 2.1: ATC level 4 groups that were listed in only one country, or listed in three of the four countries

220

ATC level 4 group Subsidised Subsidised Subsidised Subsidised Notes entities listed entities listed in entities listed entities listed in in New Australia in the UK the United Zealand States VANF L03AA None L03AA02 L03AA02 L03AA02 NZ: filgrastim and lenograstim can be (Immunostimulants, filgrastim; filgrastim; filgrastim; supplied as a Discretionary Community Colony stimulating factors) L03AA10 L03AA10 L03AA09 Supply Pharmaceuticals. (Hospitals can lenograstim; lenograstim; sargramostim provide these for outpatients out of their own L03AA13 budgets, at the hospital's discretion). pegfilgrastim N06DA (psychoanaleptics, None N06DA02 N06DA02 N06DA02 anticholinesterases) donepezil; donepezil; donepezil; N06DA03 N06DA03 N06DA04 rivastigmine; rivastigmine; galantamine N06DA04 N06DA04 galantamine galantamine C10AD (lipid modifying C10AD06 None C10AD02 C10AD02 agents, nicotinic acid acipimox nicotinic acid; nicotinic acid derivatives) C10AD06 acipimox

221

ATC level 4 group Subsidised Subsidised Subsidised Subsidised Notes entities listed entities listed in entities listed entities listed in in New Australia in the UK the United Zealand States VANF N05AG (psycholeptics, N05AG02 None N05AG02 N05AG02 diphenylbutylpiperidine pimozide pimozide pimozide derivatives)

A08AA (antiobesity None None A08AA10 None UK: NICE Guidance CG43- Obesity products, centrally acting sibutramine (December 2006) recommended Sibutramine antiobesity products) as option in obesity if co-morbidities were present. This guidance was withdrawn after the MHRA revoked the marketing authorisation of Sibutramine in January 2010 due to the increased risk of heart attacks and strokes. A08AX (other antiobesity None None A08AX01 None UK: The European Medicines Agency drugs) rimonabant withdrew the marketing authorisation for Acomplia® (rimonabant) across the EU in October 2008 because its benefits did not outweigh its risks.

222

ATC level 4 group Subsidised Subsidised Subsidised Subsidised Notes entities listed entities listed in entities listed entities listed in in New Australia in the UK the United Zealand States VANF A10AF (drugs used in None None A10AF01 None UK: NICE Guidance TA113 (December diabetes, insulins and Insulin (human) 2006) did not recommend inhaled insulin for analogues for inhalation) routine use, but suggested it as a possible option in patients who had problems with injected insulin. However, the manufacturer withdrew this product from the UK market in January 2008.

R03DX (other systemic None None R03DX05 None UK: NICE Guidance TA133 Asthma- drugs for obstructive omalizumab Uncontrolled (Omalizumab), Nov 2007 airway diseases) recommended omalizumab as an option in patients over 12 years with severe persistent allergic asthma A10BG (drugs used in A10BG03 A10BG03 A10BG03 None* *Rosiglitazone was removed from the VANF diabetes, pioglitazone pioglitazone; pioglitazone; in October 2007 thiazolidinediones) A10BG02 A10BG02 rosiglitazone rosiglitazone

223

ATC level 4 group Subsidised Subsidised Subsidised Subsidised Notes entities listed entities listed in entities listed entities listed in in New Australia in the UK the United Zealand States VANF C10AX (other lipid C10AX09 C10AX09 C10AX09 None modifying agents) ezetimibe ezetimibe ezetimibe N06AG(psychoanaleptics, N06AG02 N06AG02 N06AG02 None monoamine oxidase-A moclobemide moclobemide moclobemide inhibitors) TABLE DA 2.2: ATC level 4 groups that were subsidised in only one country, or subsidised in three of the four countries

224

Table DA 2.3 below shows the median numbers of entities in the Level 4 groups that contain licensed and subsidised entities in each country. T

Table DA 2.4 shows the differences in the numbers of entities, along with the statistical significance of the differences using the Wilcoxon Rank Sum test.

The only statistically significant differences in the median numbers of entities per group were the differences between the numbers of entities subsidised per group in:

 New Zealand versus United Kingdom (median difference one entity, p<0.001)  The United Kingdom versus the United States VANF (median difference one entity, p<0.001)

Country Median number of licensed Median number of subsidised entities in level 4 groups that entities in Level 4 groups that have entities licensed for have entities subsidised for adults adults New Zealand 1 1 Australia 2 1 United Kingdom 2 2 United States 1 1 TABLE DA 2.3: Median number of entities in ATC level 4 groups that contain licensed and subsidised entities in each country

225

Comparison of Median P value Comparison of Median P value number of ATC difference ATC level 4 groups difference Level 4 groups that that contain contain licensed subsidised entities entities in: in: NZ vs. Aus 1 0.023 NZ vs. Aus 0 0.099 NZ vs. UK 1 0.037 NZ vs. UK 1 <0.001 NZ vs. US 0 0.547 NZ vs. US VANF 0 0.350 Aus vs. UK 0 0.853 Aus vs. UK 1 0.026 Aus vs. US 1 0.331 Aus vs. US VANF 0 0.001 UK vs. US 1 0.145 UK vs. US VANF 1 <0.001 TABLE DA 2.4: Differences in the median numbers of ATC level 4 groups that contain licensed and subsidised entities in the four countries

226

Generic Name US Subsidised UK Subsidised Licensed Subsidised Licensed Subsidised Licensed VANF 2007 Licensed NHS 2007 AUS PBS 2007 NZ 2007 PHARMAC 2007 2007 (6) 2007 2007 Abacavir Yes Yes Yes Yes Yes Yes Yes Yes Abatacept Yes Yes Yes Yes # * Agalsidase alfa Yes Yes Yes [7] ** Agalsidase beta Yes Yes Yes Yes [7] Yes Alemtuzumab Yes Yes Yes Yes ** Alglucosidase alfa Yes Yes Yes * [7] ** Amprenavir Yes Yes Yes Yes Yes Anastrozole Yes Yes [5] Yes Yes Yes Yes Yes Yes Apomorphine Yes Yes Yes Yes Yes Yes Yes Yes Arsenic trioxide Yes Yes Yes ** ## Yes Azacitidine Yes Yes * ** Bevacizumab Yes ## Yes Yes Yes ## Yes Bortezomib Yes ## Yes Yes Yes Yes Yes Caspofungin Yes Yes Yes Yes Yes Cetuximab Yes Yes Yes Yes Yes Yes Cinacalcet Yes Yes Yes Yes # Yes Clofarabine Yes Yes Yes **

227

Generic Name US Subsidised UK Subsidised Licensed Subsidised Licensed Subsidised Licensed VANF 2007 Licensed NHS 2007 AUS PBS 2007 NZ 2007 PHARMAC 2007 2007 (6) 2007 2007 Darunavir Yes Yes Yes Yes Yes Yes Yes Dasatinib Yes Yes Yes Yes Yes ** ## Decitabine Yes ~ Deferasirox Yes Yes Yes Yes Yes Yes Docetaxel Yes Yes Yes Yes Yes Yes Yes Yes Drotrecogin alfa (activated) Yes Yes Yes Yes Yes Yes [8]

Efavirenz Yes Yes Yes Yes Yes Yes Yes Yes Emtricitabine Yes Yes Yes Yes Yes Yes Yes Yes Emtricitabine, tenofovir Yes Yes Yes Yes ** ### ** disoproxil and efavirenz Enfuvirtide Yes Yes Yes Yes Yes Yes Yes Yes Erlotinib Yes Yes Yes Yes # Yes Etanercept Yes Yes Yes Yes Yes Yes Yes [9] Fluocinolone acetonide (1) Yes [4] ~ ~ [4]

Fosamprenavir Yes Yes Yes Yes Yes Yes

228

Generic Name US Subsidised UK Subsidised Licensed Subsidised Licensed Subsidised Licensed VANF 2007 Licensed NHS 2007 AUS PBS 2007 NZ 2007 PHARMAC 2007 2007 (6) 2007 2007 Galsulfase Yes Yes Yes Yes [7] Gefitinib Yes ** Yes Yes Yes Hydroxocobalamin (2) Yes Yes Yes Yes Yes Yes Yes Ibritumomab tiuxetan [90Y] Yes Yes Yes **

Ibuprofen(3) Yes Yes Yes Yes Yes Yes Yes Yes Idursulfase Yes Yes Yes * [7] * Imatinib Yes Yes Yes Yes Yes Yes Yes Yes Imiglucerase Yes Yes Yes Yes [7] Yes Yes Infliximab Yes Yes Yes Yes Yes Yes [10] Interferon gamma Yes Yes Yes Yes Yes Yes Yes Lapatinib Yes * Yes # Yes Laronidase Yes Yes Yes Yes [7] Yes Lenalidomide Yes ## Yes Yes Yes ## * Levofloxacin Yes Yes Yes Yes Lopinavir and ritonavir Yes Yes Yes Yes Yes Yes Yes Yes Miglustat Yes Yes Yes Yes [7] **

229

Generic Name US Subsidised UK Subsidised Licensed Subsidised Licensed Subsidised Licensed VANF 2007 Licensed NHS 2007 AUS PBS 2007 NZ 2007 PHARMAC 2007 2007 (6) 2007 2007 Nelarabine Yes Yes Yes Nitisinone Yes Yes Yes ~ Oxaliplatin Yes Yes Yes Yes Yes Yes Yes Yes Paclitaxel Yes Yes Yes Yes Yes Yes Yes Yes Palifermin Yes Yes Yes Yes Panitumumab Yes Yes Yes * Pegaptanib Yes Yes Yes Yes Peginterferon alfa-2a Yes Yes Yes Yes Yes Yes Yes Yes Pegvisomant Yes Yes Yes Yes Pemetrexed Yes Yes Yes Yes Yes Yes Posaconazole Yes Yes Yes Yes ## * Riluzole Yes ## Yes Yes Yes Yes Yes Sorafenib Yes Yes Yes Yes ## Yes Tenofovir disoproxil Yes Yes Yes Yes Yes Yes Yes Yes Tositumomab, Iodine I-131 Yes ~ Tositumomab Trastuzumab Yes Yes Yes Yes Yes Yes Yes

230

Generic Name US Subsidised UK Subsidised Licensed Subsidised Licensed Subsidised Licensed VANF 2007 Licensed NHS 2007 AUS PBS 2007 NZ 2007 PHARMAC 2007 2007 (6) 2007 2007 Verteporfin Yes Yes Yes Yes Yes Yes Vorinostat Yes ** Total 64 20 58 58 49 30 39 19 TABLE DA 2.5: Licensing and subsidy status of each innovative medicine by country in 2007

Table 5 Key:

* Licensed 2008 ** Licensed 2009 ~ Orphan Drug Designation # Subsidised 2008 ## Subsidised 2009 ### Subsidised 2010

Table 5 Notes:

[1] Fast Tracked by the FDA for treatment of chronic non-infectious uveitis affecting the posterior segment of the eye 231

[2] Fast Tracked by the FDA for the treatment of known or suspected cyanide poisoning [3] Fast tracked by the FDA for treatment of patent ductus arteriosis [4] Only dermal forms are listed, not ocular [5] Subsidised beginning August 2007 [6] See discussion on the effects of NICE guidance, and subsidy status in the absence of NICE guidance. [7] Funded under the Life Saving Drugs program, which operates separately from the PBS. [8] PHARMAC assessed drotrecogin alfa (activated) for inpatient hospital use in 2003 for patients with severe sepsis, with a recommendation that it is reasonable value for money in patients at high risk of death (APACHE II score of 25 or more). However, these assessments are non-binding on hospitals. [9] Only subsidised for Juvenile Idiopathic Arthritis [10] Section H, National Contract Pharmaceutical: District Health Boards may purchase Infliximab for patients out of their own funds, at a nationally agreed price

232

Licensed in non-injectable formulations, total number licensed (percentage licensed in non-injectable formulations) New Zealand Australia United Kingdom United States

A: Alimentary tract and 46, 61 (75.4%) 56, 71 (78.9%) 67, 93 (72.0%) 74, 89 (83.1%) metabolism agents B: Blood and blood forming 6, 31 (19.3%) 11, 37 (29.7%) 8, 40 (20%) 16, 44 (36.4%) organs C: Cardiovascular system agents 67, 81 (82.7%) 75, 88 (85.2%) 107, 122 (87.7%) 104, 109 (95.4%) D: Dermatologicals 33, 33 (100%) 35, 35 (100%) 55, 55 (100%) 56, 56 (100%) G: Genito urinary system agents 38, 47 (80.9%) 42, 52 (80.8%) 61, 70 (87.1%) 46, 53 (86.8%) and sex hormones H: Systemic hormonal 10, 20 (50%) 12, 27 (44.4%) 15, 31 (48.4%) 11, 17 (64.7%) preparations J: Anti-infectives for systemic 65, 123 (52.8%) 75, 138 (54.3%) 80, 128 (62.5%) 79, 124 (63.7%) use L: Anti-neoplastic and 39, 87 (44.8%) 44, 100 (44%) 47, 101 (46.5%) 39, 99 (39.4%) Immunomodulating Agents M: Musculoskeletal agents 25, 37 (67.6%) 33, 46 (71.7%) 48, 62 (77.4%) 40, 47 (85.1%) N: Nervous system agents 106, 126 (84.1%) 120, 145 (82.8%) 146, 165 (88.5%) 138, 143 (96.5%)

233

NZ licensed in Aus licensed in UK licensed in non- US licensed in non-injectable non-injectable injectable non-injectable formulations, NZ formulations, Aus formulations, UK formulations, US total number total number total number licensed total number licensed (% NZ licensed (% Aus (% UK licensed in licensed (% US licensed in non- licensed in non- non-injectable licensed in non- injectable injectable formulations to total injectable formulations to formulations to number licensed) formulations to total number total number total number licensed) licensed) licensed) P: Antiparasitic agents 7, 8 (87.5%) 16, 17 (94.1%) 12, 13 (92.3%) 15, 15 (100%) R: Respiratory system agents 31, 35 (88.6%) 31, 35 (88.6%) 42, 45 (93.3%) 48, 50 (96%) S: Sensory organ agents 47, 49 (95.9%) 53, 54 (98.1%) 56, 59 (94.9%) 46, 48 (95.8%) V: Various 9, 27 (33.3%) 8, 33 (24.2%) 12, 36 (33.3%) 7, 24 (29.2%)

TABLE DA 2.6: Licensed entities available in non-injectable formulations compared with the total number of entities in each country at ATC level 1

234

Subsidised in non-injectable formulations, total number subsidised (percentage subsidised in non-injectable formulations) New Zealand Australia United Kingdom United States VANF A: Alimentary tract and 38, 49 (77.6%) 45, 55 (81.8%) 67, 93 (72%) 29, 35 (82.9%) metabolism agents B: Blood and blood forming 6, 10 (60%) 6, 19 (31.6%) 8, 39 (20.5%) 11, 29 (37.9%) organs C: Cardiovascular system agents 63, 86 (92.6%) 66, 67 (98.5%) 107, 123 (87.0%) 48, 51 (94.1%) D: Dermatologicals 22, 22 (100%) 19, 19 (100%) 55, 55 (100%) 28, 28 (100%) G: Genito urinary system agents 25, 27 (92.6%) 29, 34 (85.3%) 61, 70 (87.1%) 21, 22 (95.4%) and sex hormones H: Systemic hormonal 8, 16 (50%) 10, 19 (52.6%) 15, 31 (48.4%) 6, 8 (75%) preparations J: Anti-infectives for systemic 49, 61 (80.3%) 59, 78 (75.6%) 81, 132 (61.4%) 57, 98 (58.2%) use L: Anti-neoplastic and 31, 65 (47.7%) 35, 75 (46.7%) 46, 96 (47.9%) 24, 59 (40.7%) Immunomodulating Agents M: Musculoskeletal agents 21, 23 (91.3%) 28, 32 (87.5%) 47, 61 (77%) 20, 25 (80%)

235

NZ subsidised in Aus subsidised in UK subsidised in US VANF non-injectable non-injectable non-injectable subsidised in formulations, NZ formulations, Aus formulations, UK non-injectable total number total number total number formulations, US subsidised (% NZ subsidised (% Aus subsidised (% UK VANF total subsidised in subsidised in non- subsidised in non- number non-injectable injectable injectable subsidised (% US formulations to formulations to formulations to total VANF subsidised total number total number number subsidised) in non-injectable subsidised) subsidised) formulations to total number subsidised) N: Nervous system agents 83, 92 (90.2%) 95, 100 (95%) 143, 162 (88.3%) 82, 86 (95.3%) P: Antiparasitic agents 2, 2 (100%) 7, 7 (100%) 12, 13 (92.3%) 7, 7 (100%) R: Respiratory system agents 22, 24 (91.7%) 23, 24 (95.8%) 42, 45 (93.3%) 19, 20 (95%) S: Sensory organ agents 32, 32 (100%) 32, 32 (100%) 56, 59 (94.9%) 23, 24 (95.8%) V: Various 6, 12 (50%) 4, 6 (66.7%) 12, 36 (33.3%) 5, 13 (38.5%) TABLE DA 2.7: Subsidised entities available in non-injectable formulations compared with the total number of entities in each country at ATC level 1

236

Median time since first Median time Difference in P value registration of entities since first median time (Wilcoxon available in non- registration of since first Rank injectable formulations all entities registration Sum test) (Days) (Days) (Days) Licensed New 9700 8936 764 0.51 Zealand Subsidised New 10914 10724 190 0.85 Zealand Licensed Australia 8699 7795 904 0.16 Subsidised 8755 8065 690 0.43 Australia Licensed United 7717 7319 398 0.11 Kingdom Subsidised United 7683 7319 364 0.12 Kingdom Licensed United 7403 6607 796 0.05 States Subsidised United 9594 8203 1391 0.04 States VANF

TABLE DA 2.8: Median time since first registration of licensed entities available in non- injectable formulation compared with all entities (collapsed by ATC code)

237