Reference-Based Pricing Schemes

REVIEW ARTICLE

Pharmacoeconomics 2002; 20 (9): 577-591 1170-7690/02/0009-0577/$25.00/0 © Adis International Limited. All rights reserved.

Reference-Based Pricing Schemes Effect on Pharmaceutical Expenditure, Resource Utilisation and Health Outcomes Lisa L. Ioannides-Demos, Joseph E. Ibrahim and John J. McNeil Department of Epidemiology and Preventive Medicine, Monash University, Alfred Hospital, Prahran, Victoria, Australia

Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Increasing Pharmaceutical Expenditure . . . . . . . . . . . . . . . . . . . 2. Healthcare Structure and Funding for Pharmaceuticals . . . . . . . . . . 3. Reference-Based Pricing (RBP) . . . . . . . . . . . . . . . . . . . . . . . . . 4. Experience of RBP in Specific Countries According to Phase Introduced 4.1 Germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 The Netherlands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 New Zealand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 Canada (British Columbia) . . . . . . . . . . . . . . . . . . . . . . . . 4.6 Australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Arguments Against RBP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Health Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abstract

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .

577 578 578 579 580 580 582 583 584 584 585 587 588 589 589

Pharmaceutical expenditure is rising more rapidly than the general inflation rate in most advanced countries. One strategy that has been introduced to control pharmaceutical costs is reference-based pricing (RBP). Its potential is restricted to those specific segments of the drug market where several drugs (and/or their generic forms) exist without substantial evidence that any particular agent is superior. Three broad approaches have been adopted. These involve the aggregation of drugs into generic groups, related drug groups (e.g. ACE inhibitors) or drugs grouped by therapeutic indication (e.g. antihypertensives). For each drug group, a single reimbursement level or reference price is set. Drugs above the reference price require part or total payment by the patient. The experience with RBP ranges from over 10 years in Germany (involving all levels of RBP) to the more recent implementation of RBP for related drug groups in Australia. This review summarises the current state of knowledge on RBP from the published experiences in the countries where RBP has been adopted. The published systematic reviews of RBP from the countries that have implemented it suggest that RBP has been successful at temporarily capping drug prices for the RBP drug groups and achieving short term cost savings. However,

578

Ioannides-Demos et al.

other factors influencing total pharmaceutical expenditure have often occurred simultaneously and make it difficult to isolate specific effects of RBP. Further investigation is required before any valid conclusions can be drawn about the net effect of RBP on healthcare costs. RBP has withstood the initial legal challenges of pharmaceutical companies and the criticisms of some clinicians. Where the reference price is based on the lowest priced drug(s) in the group, RBP appears to be one of the few strategies likely to be effective at encouraging doctors to use the least expensive agents as first-line therapy and utilise more expensive agents in those who experience side effects or poor efficacy.

Rising expenditure in healthcare is a problem affecting most advanced countries. Medical costs in Western Europe increased by an average of 4.1% each year between 1970 and 1990, exceeding the overall rate of economic growth during the same period which increased by 2.7% annually.[1] A common method of cost containment used by government-funded healthcare systems is rationing or restricting access to new medical technology, especially pharmaceuticals since these are often seen as an easier target than health professional services. There are several systems of pharmaceutical cost containment that may be instituted. Referencebased pricing (RBP) is one system that has been used in a number of countries. This review was undertaken in response to the need for further information following the recent introduction of RBP to Australia. The aim was to summarise the current state of knowledge on RBP from the published experiences in the countries where RBP has been adopted and to determine the effect of RBP on pharmaceutical expenditure, resource utilisation and health outcomes. 1. Increasing Pharmaceutical Expenditure The main factors that increase pharmaceutical expenditure are: • price inflation • substitution of less expensive usually older treatments by more expensive generally newer therapies • increasing number of available drug interventions particularly those for long term prevention in conditions such as osteoporosis and hyperlipidaemia,  Adis International Limited. All rights reserved.

and an increasing number of drugs shown to be cost effective when compared with other interventions • increases in population numbers especially of elderly patients likely to require multiple drug therapy. Substitution of older less expensive agents with more expensive therapies has been a particularly potent driver of increased pharmaceutical costs. These new therapies are often adopted because of reports of a reduced incidence of adverse events rather than significant differences in pharmacological effectiveness. A policy of commencing all new patients on the newer agents will have a much greater influence on medication costs than the alternative strategy of initially prescribing the less expensive drug and changing to the newer agent if adverse effects are experienced. RBP is one of the few available policies likely to be effective at directing prescriber behaviour towards the latter approach. Because RBP does not influence the factors which relate to increases in the volume of drugs used it may produce, at best, a temporary interruption in the rate of increase of drug prices. 2. Healthcare Structure and Funding for Pharmaceuticals Policies have been established in a variety of countries for the purpose of containing pharmaceutical costs and expenditure and influencing drug prescribing towards a more cost effective use of therapies. In most cases, these policies have been accompanied by an expectation of a maintenance of existing health outcomes. The strategies include the restriction of publicly reimbursed drugs by posPharmacoeconomics 2002; 20 (9)

Reference-Based Pricing

579

than any other health category reaching $US90.6 billion.[12] Canada has a Patented Medicines Prices Review Board (PMPRB), an agency of the federal Canadian government, that administers a scheme to control market entry of new drugs as well as annual increases in marketed drugs. Price listing has recently been adopted by PMPRB as one point of reference in determining market entry prices. Some European countries control pharmaceutical prices directly, whereas in the United Kingdom pharmaceutical producers negotiate an allowable rate of return on all sales to the National Health Service.[13] Fixed drug expenditure budgets for doctors have been introduced in the UK and Germany.[13,14] In Canada, the PMPRB, controls the prices of new drugs under patent protection but the policies and methods used to meet the objectives of pharmaceutical cost containment are the responsibility of each province.[15]

itive lists (reimbursable drugs) or negative lists (list of non-reimbursable drugs) alone or in conjunction with reference pricing (table I). The decisions for drug inclusion or exclusion are based variously on safety, efficacy, professional opinion, cost utility, cost effectiveness and, to some extent, political decisions.[2-10] In the US, there are no universal price controls or cost-sharing policies or national drug formularies. State Medicaid programmes do have some element of price control as do private programmes. State Medicaid programmes promote generic products, and the private insurers and managed care companies have consolidated their purchasing power to receive price discounts and have implemented formulary policies, utilisation controls and disease management policies.[11] The main role of the US government is in the regulatory control of drugs entering the market. The Centers for Medicare and Medicaid Services, formerly known as the Health Care Financing Administration, oversees the Medicare and to a lesser extent the Medicaid programs. Medicaid in particular adheres to extensive utilization programs and formularies. A pricing list, developed by the Department of Veterans Affairs and adopted by the Federal Employees Health Benefit Plan, sets reimbursement rates. In 1998, spending in the US on prescription drugs grew faster

3. Reference-Based Pricing (RBP) RBP is not strictly a price control mechanism as the pharmaceutical manufacturer is free to set any price for their product. The principle of RBP is to establish a reimbursement level or reference price for a group of drugs that are regarded as generally interchangeable. There are three different

Table I. International policies for pharmaceutical cost containment Country

Positive lists

Australia

+

Canadaa

+

France

+

Germany

(+)

Negative lists

Reference-based pricing

Other

+ +

+

Control of price increases Price negotiations (PMPRB)

Ireland

Price negotiations +

+

Global budgets Price control

+

Italy

+

+

New Zealand

+

+

Norway

+

Spain Sweden

+ + (+)

+

The Netherlands

+

+

United Kingdom

+

a

+

Direct price control Price control Price regulation law on drugs Profit control, pharmaceutical budget

Varies for each province in Canada.

PMPRB = Patented Medicines Prices Review Board; + = present.

 Adis International Limited. All rights reserved.

Pharmacoeconomics 2002; 20 (9)

580

approaches to the grouping of drugs under RBP schemes. These may also be referred to as ‘phases’:[5,16] • Phase I involves the grouping of drugs which have the identical bioactive ingredient and are therefore therapeutically interchangeable i.e. generic groups (groups of different brands of the same drug). This phase is most applicable to drugs that are no longer on patent and have more than one manufacturer in direct competition. • In phase II, drugs are grouped with related but different drugs regardless of the indication (e.g. ACE inhibitors; histamine H2 receptor antagonists; HMG-CoA reductase inhibitors). • Phase III is a system in which all drugs in all classes used to treat a particular condition (e.g. hypertension) are grouped.[13] In some instances the phases have been introduced sequentially once maximal prescription cost reductions have been achieved from an existing system. Acceptance of RBP ranges from generally acceptable and least controversial for the generic grouping (phase I) followed by related-drug groupings (phase II) whilst drugs grouped by indication (phase III) is the most contested and controversial.[17] The reference price becomes the maximum price paid for the drug category and may be set at the average price for the drugs in that category or at the price of the lowest priced drug. When drugs above the reference price are prescribed, in most schemes, the difference is paid by the consumer. Drugs priced at or below the reimbursement level are fully subsidised. Although each pharmaceutical company is free to determine its own price, the reference price establishes a ceiling for the reimbursement of products in the category and raises the likelihood of products above the reference point losing their market share if they are avoided by prescribers.[5] Therefore, reference pricing is a strong constraint on companies’ pricing policies. Thus, RBP groups similar drug treatments, indirectly promotes drug(s) that are at or below the reference price and encourages price competition amongst pharmaceutical suppliers. Some drugs or drug groups, particularly new and innovative drugs,  Adis International Limited. All rights reserved.


drugs still under patent and drugs used in hospital or inpatient care, may be excluded from RBP schemes. 4. Experience of RBP in Specific Countries According to Phase Introduced The first country to implement phase I RBP was Germany; implementation occurred in 1989 (table II). Australia introduced phase I RBP in 1990, although it was not until 1994 that it was effectively established as Brand Premium Prices. By 1997, phase I RBP had been adopted in several other countries including Sweden, Denmark and Norway and was under consideration in Japan and Spain. There are differences between countries in the determinant of the reference price (table II) and the patient contribution for non-RBP drugs. In all the countries listed in table II, the patient pays the difference for drugs above the reference price[18,19] whereas in Italy the patients are required to pay for the entire price of drugs which are priced above the reference price.[20] Phase II RBP has been introduced in Germany, The Netherlands, New Zealand, the province of British Columbia in Canada, and Australia (table II). Denmark considered extending its reference price system to phase II in 1997, however, this was shelved in 1998.[18] Phase III RBP has been introduced in Germany and The Netherlands (table II). Despite the adoption of RBP by several countries there are few publications detailing the process and experiences, and there are no valid systematic evaluations of the health outcomes or economic implications especially of phase III RBP.[22,23] Most of the published information available on RBP relates to experience in Germany, British Columbia and New Zealand and the only published experience on the sequential implementation of the three levels of RBP is from Germany. 4.1 Germany

Reimbursed pharmaceutical expenditure was steadily increasing in Germany from 13.8 billion deutschmark (DM) in 1981 to DM20.6 billion in Pharmacoeconomics 2002; 20 (9)


581

Table II. Summary of reference-based pricing (RBP) Country

RBP phase

Year introduced

Determinant of reference price

Total drug expenditure

Germany

I

1989

Statistically derived average price of drugs in a category

Decreased rate of increase 1989, 1993-1995; smaller decrease in 1992

Lowest priced drug

Decreased expenditure growth rate 1997/1998, 1998/1999; increased expenditure 1998/1999, 1999/2000a

Australia

II

1992

III

1993

I

1990

II

1998

The Netherlands

III

1991

Average price of drugs in a category

Decreased expenditure growth rate 1991; increased expenditure 1992, 1993

New Zealandb

II

1992

Lowest priced drug

Decreased expenditure 1998-1999; decreased expenditure growth rate 1993-1998

Sweden

I

1993

Lowest priced drug plus 10%

Increase total pharmacy sales 1993, 1994; cost savings 1993, 1994

Denmark

I

1993

Average dosage unit price for two lowest cost products in a group

Decreased expenditure growth rate in first year

Norway

I

1993

NA

NA

Canadac (British Columbia)

II

1995

Lowest priced drug

Decreased expenditure and expenditure growth rate in 1996

a

Data from the Pharmaceutical Benefits Scheme.[21] Data were also directly obtained from the Drug Utilisation Sub-Committee, Pharmaceutical Benefits Advisory Committee, Canberra, ACT, Australia.

b

Tendering for sole subsiding of generic drugs to be the preferred brand on the Pharmaceutical Schedule since 1997.

c

Ontario has a system for generic categories which is equivalent to phase I RBP.

NA = no readily available information.

1988. Phase I RBP was introduced in 1989 with the Healthcare Reform Act (table III).[19,23,24] Following its introduction, reimbursed pharmaceutical expenditure was DM20.7 billion.[24] However, this decline in the rate of increase in reimbursed pharmaceutical expenditure was not maintained; the rate increased again in following years during a period of increasing inflation.[1] Phase II RBP was introduced in 1992 and more severe limits were put in place in 1993 (table III). Corresponding with these changes, by 1993, the sales of the seven largest research-intensive drug manufacturers declined by 16.5%, while the sales of the four largest generic drug firms increased by 36%.[1] As a result of the Healthcare Act of 1993, a fixed budget or cap was introduced for all officebased physicians where a penalty was imposed on physicians if the budget was exceeded.[14] This fixed budget may have had more of an effect on pharmaceutical cost containment than RBP as there is some evidence that office-based physicians may  Adis International Limited. All rights reserved.

have reacted by increasing referrals to hospitals and specialists.[19] Although most drugs are grouped according to phase I and II RBP criteria, there is also some grouping by comparable pharmacotherapeutic effects (phase III); this includes drug combinations. The reference price is the statistically derived average price of drugs in a category which varies with strength and package size and is adjusted annually.[22,25,26] It is applied to drug classes where there are alternatives without patent protection. Drugs not covered include pharmacy-made drugs and (since 1996) newly patented drugs, until their patent expires, with a new active ingredient that represents a therapeutic advance or produces fewer side effects.[16,18] Reference prices are established by a federal commission comprising representatives of sickness funds and physicians and finalised after conferring with drug manufacturers and pharmacists.[19,22] The price is adjusted annually taking into Pharmacoeconomics 2002; 20 (9)

582


Table III. Summary of the German healthcare reforms affecting prescription medication. Reproduced with permission from Elsevier Science[23] Date introduced

Legislation

January 1989

Healthcare reforms RBP – phase I January 1989 Increased co-payment for drugs Negative list October 1991 RBP phase II 1992 RBP – phase III 1993

January 1993

Healthcare reforms Price cut/price freeze for drugs (1993/1994) Global budgets for office-based physicians Increased co-payment for drugs Positive list 1993/1995 Introduction of health insurance card 1995

July 1997

Healthcare reforms Managed care and disease management projects Increased co-payments for drugs

the expenditure for drugs in 1991 had a more effective and lasting impact on cost containment of the pharmaceutical budget.[14,29] This is probably explained by the significant penalty the Physicians’ Association were to pay to the Sick Fund if the drug budget was exceeded.[28] The specific economic and public health impact of reference pricing using aggregate data was confounded by other factors including increases in the price of some generic drugs and patented drugs, initiatives taken to reduce drug expenditure and the reunification of Germany.[23] Giuliani and coworkers,[24] who have reviewed the German experience, consider that RBP is an effective tool for price control though they claim that further measures are required for continued cost containment. 4.2 The Netherlands

RBP = reference-based pricing.

account inflation and other factors. By January 1995, reference prices had been fixed for 156 drugs under phase I, 144 drugs in 18 phase II groups and 14 combinations involving about 120 drugs under phase III.[19] Prices of most drugs covered by reference pricing were lowered to, or near, the reference price. Initially, the purchase of drugs covered by reference prices required patients to pay the difference between the sale price and the reimbursement level. In 1993, three levels of co-payments were introduced based on drug prices and package size with an upper annual limit to protect the chronically ill and those with low incomes.[19] In 1996, prescription drugs regulated by reference prices represented 61.4% of total drug sales in Germany.[2,22] Although reference pricing has stemmed cost escalation within reference drug categories[25] and resulted in substantial savings in 1993, there are claims that compensatory increases have occurred in non-reference drug costs, patient referrals and hospital admissions.[19,27,28] There are conflicting views on the success of RBP in Germany. It is likely that the introduction in 1993 of a global budget for pharmaceutical expenditure that fixed the drug budget on the basis of  Adis International Limited. All rights reserved.

The Netherlands introduced phase III reference pricing in mid 1991. Criteria for grouping were: pharmacologically comparable active ingredients, similar indication, no clinically relevant differences, similar administration and indicated for similar age groups (e.g. non-selective β-blockers).[22,30] In principle, all drugs were grouped with the exception of drugs for which there are no comparable products on the market and innovative drugs.[25] Since 1993, new drugs are exempt and fully reimbursed only if they are therapeutically unique and are the first pharmacological option for a previously intractable condition.[18,22] Decisions on grouping drugs according to similar therapeutic effectiveness were not without controversy. There was disagreement about the therapeutic equivalence of some of the groupings e.g. the decision to group ergotamine and sumatriptan. The reimbursement limit is based on the average standard dose price within each grouping (e.g. all non-selective β-blockers, all migraine treatments) using the notion of the defined daily dose (i.e., assumed average maintenance dose per day for the drug used for the main indication in adults). The consumer pays the difference between the price of a drug and the reimbursement limit for the Pharmacoeconomics 2002; 20 (9)


drug group. The initial effect of the system was a lowering of the prices of drugs covered by RBP, an increase in the market share of less expensive generic and parallel import drugs and an increase in the price of drugs that were not clustered.[22,31] Although the overall drug expenditure increased by 11% from 1991 to 1992, the costs of drugs covered by reference prices increased less than predicted whilst the costs of nonclassified drugs increased by more than 20% per annum since 1988.[2] The high drug prices in The Netherlands compared with other Western European countries resulted in the implementation of the Price Regulation Law on Drugs in 1996. This enables the Minister of Health to set a maximum price for oral formulations based on the mean pharmacy buying prices in the four surrounding countries i.e. UK, France, Belgium and Germany.[18,30] Another strategy that was used to contain pharmaceutical costs was the availability of financial incentives for pharmacists when expensive drugs were substituted.[5,22] 4.3 New Zealand

In New Zealand, RBP was introduced in 1992. Pharmaceuticals are classified into therapeutic groups that treat the same or similar conditions (e.g. ulcer healing agents) and then into subgroups of pharmaceuticals that produce the same or similar therapeutic effect in treating the same or similar conditions (e.g. H2 antagonists). The therapeutic subgroups were established following consultation with independent medical advisors. Some drug classes are treated as one therapeutic subgroup e.g. ACE inhibitors whereas others e.g. calcium channel blockers are divided into six different therapeutic subgroups (low, medium and high potency antihypertensives, antianginals, agents for refractory angina and antihypertensive/antianginal/antiarrhythmic agents). Exemptions have been put in place where patients might have difficulties with switching medication (e.g. where switching may contribute to adverse effects or where there is a potential for drug interactions).[3,32]  Adis International Limited. All rights reserved.

583

The subsidy provided for each subgroup is equivalent to the price of the least expensive drug (e.g. the least expensive branded or unbranded H2 antagonist available). Alternatively, cross therapeutic deals may be established with manufacturers to lower the price of a product with a small market share in return for reimbursing another product with a larger market share (e.g. a 40% cut in the price of Tagamet® 1 was offered in 1996 in return for listing famciclovir). PHARMAC, the pharmaceutical agency which operates the national Pharmaceutical Schedule, considers reference pricing to have been a highly effective and powerful tool because pharmaceutical suppliers have tended to lower their prices to the subsidy level reducing expenditure for these drugs.[33,34] In the 1995/1996 financial year, savings attributed to reference pricing were estimated to be $48 million New Zealand dollars. Although the growth rate on drug spending in New Zealand has slowed, the trend has continued upward at a rate exceeding inflation. For the period March 1995 to March 1996, the growth rate was 8.15% and pharmaceutical expenditure was approximately $NZ716 million.[35] The average growth rate from 1993 to 1998 was 5%[3] with a decrease of approximately 5% between 1998 and 1999.[36] In 1998, PHARMAC negotiated with ParkeDavis and agreed to list the lipid-lowering drug atorvastatin with a cap on its expenditure after the price of quinapril was decreased by 60%, establishing the latter as the reference price for ACE inhibitors. The manufacturer of cilazapril subsequently reduced their price to the reference price of $NZ7.42. This reference price was substantially below the price paid previously for ACE inhibitors (e.g. enalapril $NZ69.00) and these two agents (quinapril and cilazapril) which had previously held 6% of the ACE inhibitor market share increased their market share to 92.4%. This transaction with Parke-Davis resulted in savings of $NZ30 million in a year and for the first time phar1 Use of brand names is for product identification purposes only and does not imply endorsement.


584

maceutical expenditure fell by approximately 5%.[3] 4.4 Sweden

In Sweden, phase I RBP was introduced in January 1993. The introduction of the scheme resulted in manufacturers cutting the price of drugs priced above the reference price in anticipation of consumers not paying the higher price. Therefore, the prices of brand name products fell close to that of the generics and the market share of original or patented products increased.[22,26,37] Savings in the national drug bill were estimated at 400 million Swedish kronor (SEK) [approximately $US52 million] for 1993 and a further SEK50 million (approximately $US6.5 million) in 1994. No further significant savings are expected with the current RBP. Despite RBP, the percentage change in year to year of pharmaceutical costs now ranges from 12.7 to 15.8% which is similar to the 10.6 to 15.3% from 1988 to 1992.[27,38] 4.5 Canada (British Columbia)

RBP was introduced in British Columbia in October 1995 in response to the doubling of the Pharmacare budget from 195 million Canadian dollars ($Can) in 1990 to $Can407 million in 1995. RBP was defined as ‘a funding methodology whereby Pharmacare2 funds those products covered under its plans at a level based on the cost of a gold standard of drug therapy for specified conditions, where the gold standard is supported by evidence-based medicine to meet the needs of the majority of patients with that condition’.[39] The basis of the policy was that if there was no scientific evidence that expensive drugs in the same therapeutic class had 2 Pharmacare is the British Columbia government-funded drug insurance plan. All drug applications requesting Pharmacare coverage are evaluated by the Pharmacoeconomic Initiative of British Columbia, a multidisciplinary volunteer expert committee whose recommendations are based on critical evaluation of pharmacoeconomic studies and are evidence based. Drug coverage is universal, free for the financially indigent and involves a small copayment which may be deductible for other patients.[9]  Adis International Limited. All rights reserved.


additional health benefits, then there was no basis for reimbursement beyond the price of the less expensive, presumably equally effective, reference drugs.[4] However, due to the time pressures to implement RBP it was not feasible to undertake independent systematic reviews of the evidence and hence decisions were based on expert opinions.[4] The aim of reference pricing was to encourage the use of the less expensive reference product as the first-line drug but to allow changes to other agents if this was necessary on account of adverse effects or unsatisfactory efficacy.[40] The Reference Based Pricing Expert Advisory Committee, a multidisciplinary expert committee consisting of an expert advisory committee made up of physicians, pharmacists, policy makers and pharmacologists, advised Pharmacare on the choice of the reference drug. Some members of the expert committee advocated evidence-based medicine but the committee saw itself as supplying timely expert advice, not reviews of evidence. RBP was initially applied to nonsteroidal anti-inflammatory drugs (NSAIDS), nitrates, H2 antagonists and omeprazole, then expanded to ACE inhibitors and calcium channel blockers. It applies to 40 of 1700 medications provided by Pharmacare.[41] Patients can choose to discuss with their physician and switch to the reference product, or pay the price difference between the drug prescribed and the agents that Pharmacare reimburse. Alternatively, the physicians can obtain a special prescribing authority where there is a special reason for the patient to take a more expensive drug. The Special Authority process which was simultaneously introduced for rapid request and approval of exemptions, initially approved exemptions at the rate of 100 per day (for anti-ulcer drugs) and this increased to 400 exemptions per day after the first three drug classes (acid-supressing agents, nitrates and NSAIDs) were affected by the policy. Exemptions were initially approved for 98% of requests which raised concerns about the process and subsequently led to modifications to the Special Authority process.[4] The administrative costs associPharmacoeconomics 2002; 20 (9)


ated with the Special Authority3 process were estimated at $Can3.7 million[26] for 7 months. Expensive media campaigns were undertaken by both the government and the pharmaceutical industry to inform the public about RBP prior to its launch. The industry’s themes were that the referencebased drugs would be the oldest and cheapest, provincial economic needs were being placed ahead of patient needs, there would be multiple adverse effects on the quality of care and health, that bureaucrats were making medical decisions instead of doctors and that there was a lack of consultation in the development of the policy. The advertisements urged people to write or call their political representatives and register their concerns. Pharmacare’s advertisements were aimed at educating the public about the policy and addressing the issues raised in the industry campaign. An assessment of seniors’ perspectives indicated that the government’s messages were more favourably received than those of the pharmaceutical industry.[42] The investigators claimed that the majority of seniors (57%) were happy with RBP (i.e., 57% thought it was good, 21% not sure, 14% thought it was both good and bad, 8% thought it was bad) although they preferred more involvement in and information on the policy change.[43] In contrast, clinicians viewed the policy as interference and a restriction to their authority.[43] In Canada, total healthcare expenditure was approximately $Can72 billion per annum in 1994 and prescription drugs accounted for just over 5% of this expenditure.[39] However, drugs represent the fastest growing proportion of total health spending in Canada. Since the introduction of the new pol3 Automatic exemptions from the Special Authority process were put in place for some medical disorders and specific medical specialties (cardiologists, nephrologists for unreferenced calcium channel blockers; rheumatologists for higher priced NSAIDs). A Special Authority process was put in place to prevent RBP of gastric drugs ‘backfiring’ by physicians opting to prescribe omeprazole rather than switching to the reference drug cimetidine.[4] This required a Special Authority for omeprazole that stated that another gastric acid–suppressing drug had been tried and failed to relieve symptoms.  Adis International Limited. All rights reserved.

585

icy, prescribing of the reference anti-ulcer drug cimetidine increased 5 times while prescribing for other drugs in the class decreased by over 50%, prescribing for sustained-release nitrates fell by more than 50% while that for the reference nitrate tripled, and prescribing of less expensive NSAIDs saved over $Can5 million (in the first year since RBP introduction).[4,15] Estimated total drug savings vary with one estimate of $Can30 million (approximately 7.5% of the drug budget) in the first year plus projected annual savings of $Can14 million (1996/ 1997 values) with the addition of antihypertensives to the policy.[4,44] Other estimates of cost savings are $Can74 million for the 2 years since the programme was launched[40] and $Can44 million per year.[41] Initial data from Pharmacare has demonstrated an increase in prescriptions for reference drugs and a decrease in non-reference drugs with an associated decrease in annual costs for drugs within reference categories from $Can42 million in the year before the introduction of RBP to $Can23.7 million in 1996.[45] Further information on the British Columbia experience is available in other articles.[46-50] 4.6 Australia

In February 1998, the Australian government introduced phase II RBP or the Therapeutic Group Premiums Scheme. The scheme was anticipated to introduce competition into pricing of medicines that were evaluated as having an equivalent therapeutic effect by independent clinical and scientific experts. A range of choice of medications is available as well as exemptions for consumers who are unable for clinical reasons to move to lower priced drugs. Exemption criteria apply where adverse effects and/or drug interactions occur or are expected to occur with all of the base-priced drugs, or the transfer to a base-priced drug would cause patient confusion resulting in problems with compliance. Doctors can apply for authority prescriptions by phone or mail. The four drug groups that were affected by the scheme were: drugs used for the treatment of hypertension and heart disease (except β-blockers) i.e. Pharmacoeconomics 2002; 20 (9)

586


3500 3000 Expenditure ($M)

PBS Pathology Specialists General practitioners

2500 2000 1500 1000 500

87 / 19 88 88 / 19 89 89 / 19 90 90 / 19 91 91 / 19 92 92 / 19 93 93 / 19 94 94 / 19 95 95 / 19 96 96 / 19 97 97 /9 8

0

19

ACE inhibitors and calcium channel blockers (except diltiazem and verapamil); cholesterol-lowering drugs [i.e., HMG-CoA reductase inhibitors (statins), except fluvastatin]; drugs for peptic ulcers; and drugs for oesophageal reflux. Although selective serotonin reuptake inhibitors were also considered for reference pricing they were excluded from the initiative on clinical grounds. For some drug groups e.g. the statin group of cholesterol-lowering drugs, all manufacturers reduced prices to a level satisfactory to the government and thereby avoided the introduction of a Therapeutic Group Premium. The patient contribution remained the same as that prior to the introduction of the scheme for medicines at the reference price. Patients who chose to stay with a higher priced drug paid an additional average premium of about $2 Australian dollars (range $A0.71 to $A4.63) which was set by manufacturers when the scheme was introduced. Only four drugs had a premium greater than $A3; these were ramipril 2.5mg and 5mg, amlodipine 10mg and cimetidine effervescent tablets. At the time of introduction of the scheme, the costs for 50% of the drugs remained the same. By August 2000, the additional cost paid by patients for the three categories that attract a Therapeutic Group Premium (ACE inhibitors, H2 antagonists and calcium channel blockers) was $A0.26 to $A7.01. In 1992/1993, healthcare expenditure in Australia was $A34.3 billion (i.e., 8.5% of Gross Domestic Product).[51] The total Pharmaceutical Benefits Scheme costs (government plus patient contribution) had increased from $A429 million in 1980/ 1981 to $A1048 in 1987/1988 and to $A2863 million in 1996/1997 (figure 1). From 1984/1985 until 1996/1997, the annual cost for the Pharmaceutical Benefit Scheme had increased by 275% whilst Medicare reimbursement for pathology had increased by 148%, specialists by 151%, general practitioners by 181% and total healthcare spending had increased by 161% (figure 1). In 1997/1998, the year the Therapeutic Group Premiums Scheme was implemented, the total Pharmaceutical Benefits Scheme (PBS) costs in-


Financial year

Fig. 1. Annual expenditure on Pharmaceutical Benefits Scheme

(PBS) drugs, pathology, specialists, general practitioners. $ = Australian dollars.

creased to $A3097 million despite the prediction of cost savings of $A560 million over four years. In 1999/2000, government expenditure for PBS prescriptions increased by 14%. Analysis of costs of therapeutic groups affected by the scheme may be a better indicator of cost savings than the aggregate pharmaceutical benefits costs. Data from the Pharmaceutical Benefits Analysis Section suggest that cost savings have occurred with therapeutic groups. The average dispensed price of ACE inhibitors on the PBS was $A31.19 for the 1998 calendar year compared with $A33.18 in 1997. Similarly, the average dispensed price of calcium channel blockers was $A25.50 in 1998 and $A26.39 in 1997. The 1999 Pharmacy Trade Report stated that the only three therapeutic categories to grow more slowly in 1998 compared with 1997 were the three categories affected by Therapeutic Group Premiums: antiulcer agents, ACE inhibitors and calcium antagonists. A decrease in the average price of these three groups was also demonstrated in the Pharmaceutical Benefits 1999/ 2000 Report on Expenditure and Prescriptions. The category that did not attract a Therapeutic Group Premium (the statins) has shown an increased expenditure since the introduction of the scheme (figure 2) and was the drug group with the highest increased cost to the government in 1999/2000. This increase in expenditure also reflects an increase in Pharmacoeconomics 2002; 20 (9)


587

volume of prescriptions due to the expansion of indications to patients with existing coronary heart disease and to patients with a blood cholesterol level of >4 mmol/L (from >5.5 mmol/L). 5. Arguments Against RBP Various concerns and claims about RBP programmes have been raised in the literature.[36,45] These include: • a slow down in the introduction of new therapeutic advances as new products are aggregated with older less expensive products and off-patent products regardless of efficacy or adverse effects • a two-tiered health system may result with unemployed and low-income earners receiving drugs at or below the reference price as they are less likely to pay the extra copayment • that the system represents an encroachment on physicians’ freedom to prescribe and may discourage individualisation of drug therapy • that it places more demands on the doctors’ time as they are required to obtain an authority to prescribe non-reference products

450 400

PBS Items ($benefit)

350 300

• that other healthcare costs may rise and patients may require close monitoring during the transition to a new drug • that it introduces a financial component to the physician-patient relationship • and that it may promote the prescribing of a drug that may not be therapeutically equivalent to other drugs in the group. The principal argument against RBP is its potential to act as a disincentive to pharmaceutical innovation. It is frequently pointed out that developments in therapeutics often take the form of small incremental advances in efficacy, tolerability and convenience. Although the individual steps in the process are small, when considered cumulatively the effect may be substantial. Therefore, processes which interfere with such development, such as the direction of funds towards generic nonresearch-based companies and the inclusion of new patented drugs in RBP therapeutic groups, may have a detrimental effect in the long-term on the investment of the pharmaceutical industry in research and innovation. A related criticism of RBP is that while the programme may decrease drug costs, overall health costs may increase since the reference product is

Simvastatin Pravastatin Fluvastatin Atorvastin Cerivastatin Total

250 200 150 100 50 0 1992

1993

1994

1995

1996

1997

1998

1999

Year

Fig. 2. Expenditure for Pharmaceutical Benefits Scheme (PBS) prescriptions for cholesterol-lowering drugs (statins) processed from January 1992 to June 2000 by the Australian Health Insurance Commission. $ = Australian dollars.



588

usually an older therapy that has largely been superseded by newer products with better patient efficacy and/or adverse effect profiles.[52] These claims are speculative and at present are unsupported by convincing data.[52] Evidence presented at the Canadian hearing in support of this proposition included a compilation of adverse effects reported by patients via a toll-free telephone which the judge ruled to be unscientific.[4] Another criticism of RBP is that one or other drug may be unfairly penalised because of poorly chosen ‘equivalent’ doses. The system depends heavily on equivalent reimbursement for equipotent therapy, since an agent may be unduly advantaged if the reference price is paid for a suboptimal dose level, especially if the prescribers then use higher doses in routine practice. Dose equivalence is typically established from comparison of commonly prescribed doses although such data may be misleading and less accurate than rigorous doseequivalency studies. For reasons such as those discussed in this section, the pharmaceutical industry has not been accepting of RBP. In British Columbia, the Pharmaceutical Manufacturers Association of Canada (PMAC) publicised its opposition using an extensive advertising campaign which included the critical review of the programme in a Canadian Cardiovascular Society position paper and an advertising campaign claiming that seniors’ health was put at risk.[4,40] The programme was also unsuccessfully challenged in both the Supreme Court and Court of Appeal. In one action, the PMAC and seven drug companies sued the province of British Columbia arguing that the Health Minister lacked authority to implement the scheme and that the scheme contravened doctors’ confidentiality provisions and the Food and Drug Act.[4,40] In New Zealand, the pharmaceutical industry claims its viability has been seriously threatened by RBP, resulting in some industries closing or downsizing and some products being withdrawn from the market or not marketed.[3] Industry has substantially decreased investment in research and development and decreased funding for continuing edu Adis International Limited. All rights reserved.


cation.[36,53] These effects are unlikely to be observed in Australia, where there is some industry protection through the Pharmaceutical Industry Investment Program which was designed to provide partial compensation to industry for the price suppression of medicines on the Pharmaceutical Benefits Scheme. 6. Health Outcomes There is limited information on the impact of RBP on health outcomes and it is generally accepted that more work is needed to provide evidence to support the introduction of these cost-containment methods. In Germany, average hospital admissions and referrals to specialists increased by 9% in 1993 compared to 1992 and by about 25% for specific drug-intensive illnesses.[14] However, this shift in health services utilisation coincided with the introduction of the health budget.[41] Whilst financial barriers have been shown to increase adverse health outcomes for the poor,[52] an assessment of the effect of RBP on other health resources in British Columbia did not demonstrate an increase in physician visits or hospitalisations for gastrointestinal bleeding after the referencing of ulcer products or in hospitalisations for syncope or myocardial infarction after reference pricing was implemented for nitrates.[40] An initial analysis of the medical services data has not shown an increase in rates of hospitalisation of seniors[41] and there are claims that net physician visits have not increased.[4] However, the use of healthcare services (general practitioner visits, specialist visits and hospital admissions) were demonstrated to be higher with the reference drug captopril compared with the other ACE inhibitors in a retrospective cohort study of 4709 Saskatchewan residents being managed for hypertension.[54] Whilst in New Zealand, inadequate clinical management was reported following the application of RBP to HMGCoA reductase inhibitors.[26,36,55] After the less potent and cheapest statin, fluvastatin, was established as the reference drug in 1997, a significant rise in lipid concentrations was demonstrated in paPharmacoeconomics 2002; 20 (9)


tients.[32,55] This was most likely due to patients, who were switched from simvastatin, receiving an inadequate dosage.[26,32,55] Six months later, doctors were advised to carefully titrate statins with weekly monitoring; the policy was then amended so patients requiring more potent statin therapy were able to receive simvastatin.[32] In Australia, a report assessing the short term impact of Therapeutic Groups Premiums (TGP) on the community and pharmacy using a random selection of approximately 100 pharmacies in three states showed good awareness by consumers of the programme, 25% of patients required further doctors visits, and an increase in patient counselling time by pharmacists.[56] Seventy-seven percent of patients required counselling and 67% required explanation of the TGP. One-off costs were $A12.35 million and ongoing costs were 11.5 cents per TGP prescription. Counselling included pharmacist time of 4.18 minutes and staff time of 2.90 minutes. 7. Discussion The stated aim of most RBP programmes is to shift prescribing to products that are less expensive but of equivalent efficacy and safety especially for first-line therapy. Most of the schemes have exceptions allowing payment for the more expensive non-reference drugs for some patients when the reference drug is not as clinically effective, well tolerated or appropriate. Exceptions also apply in some instances for new innovative drugs which may not fit into existing drug groupings. In British Columbia, some medical specialties are automatically exempted from the authority process. One commonly quoted feature of RBP is that it is not a direct restriction on the availability of drugs; individual drug prices are not set and doctors prescribing is not restricted. Rather, it is a mechanism to encourage use of the less expensive agents that are fully reimbursed as first-line therapy and to persist with these unless there are good clinical reasons to move to a more expensive agent. Policies to contain pharmaceutical costs are generally implemented without a structured evolution process to establish the effect on health out Adis International Limited. All rights reserved.

589

comes and resources beyond the pharmaceutical budget or healthcare costs. An evaluation of these policies should ideally include the effect on the quality of patient care and an assessment of the use of health professional services and hospital resources.[39] Pharmaceutical expenditure has continued to rise in countries that have implemented RBP. This is expected as RBP mainly affects only two of the major drivers that increase pharmaceutical expenditure (i.e. price inflation and substitution of older less expensive treatments by newer more expensive therapies) and applies only to a limited segment of the pharmaceutical market. However, this does not mean that RBP is not working as it has a substantial effect on controlling prices of the RBP drug groups. The introduction of a RBP system will still require consideration of complex issues, e.g. whether a new drug is sufficiently better than existing drugs to justify a higher price. With increasing use of meta-analysis of mega-trials capable of identifying very small differences in efficacy or tolerability, the simple fall back on statistically proven advantages will be too simplistic. Newer methodologies capable of expressing advantages of different therapies in a standard fashion (e.g. cost per qualityadjusted life year) are urgently needed in order to allow consistent decisions across the whole spectrum of therapeutics. 8. Conclusion Reference pricing influences only a part of the total cost of drug expenditure by influencing physicians to prescribe drugs at or below the reference price, persuading patients to accept these drugs or pay the difference, and putting pressure on the pharmaceutical companies to compete with the reference priced product. Although short term savings in drug costs have been demonstrated in some countries following the introduction of reference pricing policies, the long term savings and the success of RBP remain controversial due partly to the continued increase in pharmaceutical expenditure[23,27,39,57] and the limPharmacoeconomics 2002; 20 (9)

590


ited information on resource utilisation and health outcomes. The precise impact of reference prices is often obscured by additional forms of expenditure restraint and the use of aggregate pharmaceutical costs. Irrespective of the cost constraint system in use in any country in the western world, pharmaceutical expenditure is likely to continue to rise. This increase will be due to a range of factors including increases in drug prices and volume. The ideal pharmaceutical supply and control system should encourage cost-effective drug treatments while at the same time being flexible enough to allow exceptions for individual patients or specific patient groups when these are warranted. An evidence-based approach should be used to define criteria to demonstrate the effectiveness and safety of the reference drug. Once implemented, the outcome of the system should be evaluated for its impact on healthcare usage and costs. Acknowledgements There were no sources of funding or conflicts of interest for this article. I would like to acknowledge Professor Joel Lexchin, Toronto, Canada and Mr Peter McManus, ACT, Australia for guiding me to relevant information.

References 1. Belien P. Healthcare reform in Europe. Pharmacoeconomics 1996; 10 Suppl. 2: 94-9 2. Bloor K, Maynard A, Freemantle N. Lessons from international experience in controlling pharmaceutical expenditure III: regulating industry. BMJ 1996; 313: 33-5 3. Braae R, McNee W, Moore D. Managing pharmaceutical expenditure while increasing access. The Pharmaceutical Management Agency (PHARMAC) experience. Pharmacoeconomics 1999; 16 (6): 649-60 4. Maclure M, Potashnik TM. What is direct evidence-based policy-making: experience from the drug benefits program for seniors in British Columbia. Can J Aging 1997; 16 (S1): 132-46 5. Dickson M. The pricing of pharmaceuticals: an international comparison. Clin Ther 1992; 14 (5): 604-10 6. Therapeutic Goods Administration. Medicines regulation and the TGA. Woden (ACT): Health and Aged Care, 1999 Dec. Available from URL: http://www.health.gov.au/tga/docs/html/ medregs.htm [Accessed 2002 Jun 30] 7. Schosser R. Risk/benefit evaluation of drugs: the role of the pharmaceutical industry in Germany. Eur Surg Res 2002; 34: 203-7 8. U.S. Food and Drug Administration/Center for Drug Evaluation and Research. Drug registration and listing. Available from URL: http://www.fda.gov/cder/drls/default.htm [Accessed 2002 Jun 30]


9. Anis AH, Rahman T, Schechter MT. Using pharmacoeconomic analysis to make drug insurance coverage decisions. Pharmacoeconomics 1998; 13 (1 Pt 2): 119-26 10. Anis AH. Pharmaceutical policies in Canada: another example of federal-provincial discord. CMAJ 2000; 162: 523-6 11. Kane NM. Pharmaceutical cost containment and innovation in the United States. Health Policy 1997; 41: 71-89 12. Chartan F. US health spending accelerated in 1998. BMJ 2000; 320: 203 13. Burstall ML. The management of the cost and utilisation of pharmaceuticals in the United Kingdom. Health Policy 1997; 41: S27-43 14. von der Schulenburg J-M, Uber A. Current issues in German healthcare. Pharmacoeconomics 1997; 12 (5): 517-23 15. Vandergrift M, Kanavos P. Health policy vs industrial policy in the pharmaceutical sector: the case of Canada. Health Policy 1997; 41: 241-60 16. McLaughlin PR. Reference-based pricing of prescription drugs. Can J Cardiol 1997; 13: 31-2 17. Drummond M. Reference pricing: does it lead to cost-effectiveness of pharmaceuticals? Drug Information Association Conference on International Policies in Health Economics; 1997 Sep, Australia 18. Dickson M, Redwood H. Pharmaceutical reference prices: how do they work in practice? Pharmaceoeconomics 1998; 14: 471-9 19. Schoffski O. Consequences of implementing a drug budget for office-based physicians in Germany. Pharmacoeconomics 1996; 10 Suppl. 2: 37-47 20. Fattore G, Jommi C. The new pharmaceutical policy in Italy. Health Policy 1998; 46: 21-41 21. Health Insurance Commission. Annual report 2000-01. Available from URL: http://www.hic.gov.au/annualreport/statistics/index.htm [Accessed 2002 Jun 30] 22. Drummond M, Jonsson B, Rutten F. The role of economic evaluation in the pricing and reimbursement of medicines. Health Policy 1997; 40: 199-215 23. Schneeweiss S, Schoffski O, Selke GW. What is Germany’s experience on reference based drug pricing and the etiology of adverse health outcomes or substitution? Health Policy 1998; 44: 253-60 24. Giuliani G, Selke G, Garattini L. The German experience in reference pricing. Health Policy 1998; 44: 73-85 25. Selke GW. Reference price systems in the European community: in cost containment pricing financing of pharmaceuticals in the European community: policymaker’s view. London: LSE Health and Pharmetrica SA, 1994: 147-160 26. López-Casasnovas G, Puig-Junoy J. Review of the literature on reference pricing. Health Policy 2001; 54: 87-123 27. CCS position paper. A position paper on drug-pricing strategies for prescription pharmaceuticals in Canada. Can J Cardiol 1997; 13: 33-8 28. von der Schulenburg J-MG. Management of cost and utilization of pharmaceuticals in Germany. Health Policy 1997; 41: S45-53 29. Ulrich V, Wille E. Healthcare reform and expenditure on drugs: the German experience. Pharmacoeconomics 1996; 10 Suppl. 2: 81-8 30. de Vos CM. The 1996 pricing and reimbursement policy in The Netherlands. Pharmacoeconomics 1996; 10 Suppl. 2: 75-80 31. Rigter H. Recent public policies in the Netherlands to control pharmaceutical pricing and reimbursement. PharmacoEconomics 1994; 6 Suppl. 1: 15-21 32. Thomas M. The change of cost: reference-based pricing and the statins. Can J Cardiol 1999; 15: 535-8



33. Kletchko SL, Moore DW, Jones KL. Targeting medicines: rationalising resources in New Zealand, a preliminary paper. Wellington: PHARMAC, Pharmaceutical Management Agency Ltd, 1995 34. PHARMAC annual review 1996. Wellington: PHARMAC, Pharmaceutical Management Agency Ltd, 1996 35. PHARMAC drugscene. Wellington: PHARMAC, Pharmaceutical Management Agency Ltd, 1996 36. Martin J, Begg E. Reference pricing-is it in the public interest? N Z Med J 2000; 113: 422-5 37. Jönsson B. Pricing and reimbursement of pharmaceuticals in Sweden. PharmacoEconomics 1994; 6 Suppl. 1: 51-60 38. Ljungkvist M-O, Andersson D, Gunnarsson B. Cost and utilisation of pharmaceuticals in Sweden. Health Policy 1997; 41: S55-69 39. Boulet AP, Tessier G. Reference-based pricing in British Columbia: implications for cardiologists-an analysis. Can J Cardiol 1997; 13: 46-51 40. Mullens A. Reference-based pricing: will other provinces follow the BC lead? CMAJ 1998; 158: 239-41 41. McGregor M. Coverage of drug costs: reference-based pricing. Can J Cardiol 1998; 14: 666-8 42. Brunt JH, Chappell NL, Maclure M, et al. Assessing the effectiveness of government and industry media campaigns on seniors’ perceptions of reference-based pricing policy. J Appl Gerontol 1998; 17: 276-95 43. Mullett J, Coughlan R. Clinicians’ and seniors’ views of reference-based pricing: two sides of a coin. J Appl Gerontol 1998; 17: 296-317 44. Ministry of health fact sheet. Myths and facts about BC’s reference drug program. Victoria (BC): Ministry of Health, 1997 45. Narine L, Senathirajah M, Smith T. Evaluating reference-based pricing: initial findings and prospects. CMAJ 1999; 161: 286-8 46. Grootendorst PV, Dolovich LR, O’Brien BJ, et al. Impact of reference-based pricing of nitrates on the use and costs of anti-anginal drugs. CMAJ 2001; 165: 1011-9 47. Schneeweiss S, Soumerai SB, Glynn RJ, et al. Impact of reference-based pricing for angiotensin-converting enzyme inhibitors on drug utilisation. CMAJ 2002; 166: 737-45 48. Schneeweiss S, Walker AM, Glynn RJ, et al. Outcomes of reference pricing for angiotensin-converting-enzyme inhibitors. N Engl J Med 2002; 346: 822-9


591

49. Schneeweiss S, Maclure M, Soumerai SB. Prescription duration after copay changes in older people: methodological aspects. J Am Geriatr Soc 2002; 50: 521-5 50. Anis A. Why is calling an ACE an ACE so controversial? Evaluating reference-based pricing in British Columbia. CMAJ 2002; 166: 763-4. 51. AIHW (Australian Institute of Health and Welfare). Health expenditure, health and welfare statistical information. Canberra, ACT, Australia, 1998. Available from URL: http://www.aihw.gov.au [Accessed 2002 Jun 30] 52. Lexchin J. Controlling pharmaceutical expenditures in Canada. In: Drache D, Sullivan T, editors. Market limits in health reform: public success, private failure. London: Routledge, 1999: 292-311 53. Swinburn B, Milne RJ, Richards M, et al. Reimbursement of pharmaceuticals in New Zealand: comments on PHARMAC’s processes. N Z Med J 2000; 113: 425-7 54. Bourgault C, Elstein E, Le Lorier J, et al. Reference-based pricing of prescription drugs: exploring the equivalence of angiotensin-converting enzyme inhibitors. C MAJ 1999; 161: 255-60 55. Thomas M, Mann J. Increased thrombotic vascular events after change of statin. Lancet 1998; 352: 1830-1 56. Sirianni International Pty Ltd. Therapeutic group premiums community and pharmacy impact: an assessment of the resource impact of the government policy on community pharmacy. Pharmacy Guild research report, Australia. Melbourne: Sirianni International Pty Ltd, 1998 57. Zammit-Lucia J, Dasgupta R. Reference pricing: the European experience. Health policy review paper no. 10. London: St Mary’s Hospital Medical School Health Policy Unit, 1995

Correspondence and offprints: Dr Lisa L. Ioannides-Demos, Department of Epidemiology and Preventive Medicine, Monash University, Alfred Hospital, Commercial Road, Prahran, Victoria, 3181, Australia. E-mail: [email protected]
