Drug Prices and Value for Money: Title and subTitle BreakThe Australian Pharmaceutical Benefits Scheme

In this issue of JAMA, Pearson and Rawlins¹ describe the work of the National Institute for Health and Clinical Excellence (NICE), the main health technology assessment agency in the United Kingdom. NICE provides comprehensive advice, including assessment of efficacy and comparative cost-effectiveness of medical interventions and publishes clinical guidelines.² This ambitious project is the most prominent and far reaching of a number of international initiatives, including health technology assessment programs in Canada and Spain^{3 - 4} and pharmaceutical cost-effectiveness evaluations linked to reimbursement decisions in Australia, the Netherlands, and Canada.^{5 - 7} The relevance to the United States has increased with the Medicare Prescription Drug Improvement and Modernization Act of 2003.⁸ This act introduces a voluntary program of prescription drug coverage but the US government will not get directly involved in price negotiation, leaving implementation of the drug plans to managed care organizations and pharmacy benefit managers. These organizations will carry a degree of financial risk and will be interested in achieving maximum value for money, thus increasing the relevance of pharmacoeconomic analyses.

The Pharmaceutical Benefits Scheme (PBS) was established in 1953 to guarantee Australians access to subsidized essential medicines. By the early 1990s, the list of subsidized drugs had expanded to cover most common treatable conditions and the PBS had become a central feature of Medicare, the universal health care program operated by the Australian commonwealth.⁵ The Pharmaceutical Benefits Advisory Committee (PBAC) makes recommendations to the commonwealth minister for Health and Ageing about listing drugs on the PBS. Since 1993, the PBAC has made a detailed assessment of the comparative cost-effectiveness of all new medicines.⁹ The recommendations include which drugs should be subsidized, the prices at which they appear to be “cost-effective,” and what restrictions should be applied to their use (Table). Unlike NICE, the PBAC’s decisions have a direct budgetary impact and the estimated total cost of a new drug is part of the committee’s deliberations.⁹ The PBAC functions largely in a reactive capacity, reviewing submissions from the pharmaceutical industry. In contrast, NICE is proactive and sets its own agenda, which is aligned with national health priorities in the United Kingdom.²

The guidance provided by NICE is based on independent health technology reviews, and stakeholders are involved at each stage. The board of NICE, the executive staff, and the chair of the Appraisals Committee are appointed by the Secretary of State for Health, but the institute is a Special Health Authority, which gives it greater independence than the PBAC,² whose members are not government employees but are appointed by the federal health minister and serve at the minister’s discretion. The importance of this was illustrated in 2001 when the minister discharged the committee, following persistent lobbying by the pharmaceutical industry.¹³

Pearson and Rawlins note industry’s view that a negative appraisal of a pharmaceutical product by NICE could damage global sales. However, most public criticisms of NICE have come from economists, academics, and clinicians rather than from pharmaceutical companies.^{14 - 17} In contrast, the Australian PBS has been under relentless pressure from international pharmaceutical manufacturers. In this, US manufacturers have enjoyed the support of their government. The US Department of Commerce has criticized Australia (and other members of the Organization for Economic Co-operation and Development) for using reference pricing and monopsonistic buying powers to negotiate prices for patented drugs that are sold at substantially lower prices than those in the United States.¹⁸ Australian prices are published on a searchable Web site (http://www1.health.gov.au/pbs/index.htm), making them accessible to other bodies involved in negotiations with industry. Pharmaceutical companies have attempted to weaken the price-setting role of the PBAC, including by reviews of the committee’s roles and functions, political lobbying, legal challenges, and a series of measures in the Australia–United States Free Trade Agreement, including the introduction of a decision review mechanism.^{13 ,19}

NICE sets its own agenda and commissions evaluations of new health technologies. Consequently, it has an open approach to consultation and releases full details of its guidance, including the reasoning behind its recommendations.² The PBAC is bound by the secrecy provisions of the Australian National Health Act and submissions from pharmaceutical companies are treated as “commercial in confidence.” This has limited the capacity of the committee to publicize the reasons for its decisions and has made it difficult for the committee to defend itself against criticism after rejection of prominent drugs. However, this situation is changing. The current committee has won concessions from the pharmaceutical industry about disclosure and reasons for positive recommendations and rejections are now provided on the committee’s Web site.²⁰

NICE and the PBAC have similar approaches to evaluating the cost-effectiveness of drugs, relying on systematic reviews, incremental cost-effectiveness ratios, and modeling. The main difference is that NICE commissions its evaluations; in Australia the pharmaceutical companies perform the analyses, which are evaluated by the health department and its consultants prior to consideration by the PBAC (Table). The key measure used by NICE is the cost per quality-adjusted life-year (QALY) gained by the new technology. The committee has relied more on cost per life-year gained and calculations based on surrogate or intermediate outcomes (eg, incremental cost of achieving a specified reduction in the Hamilton Rating Scale for Depression). This is because every drug must be evaluated. In many cases, there are neither data on the effects of treatment on survival nor an acceptable measure of quality of life at the time listing decisions are made.

Pearson and Rawlins¹ state that NICE works with a band of cost-effectiveness ratios up to $45 900 per QALY (exchange rate at press: AU $1 = US $0.75); above this, rejection of a new technology is increasingly likely. Likewise, the PBAC does not work with an explicit cost-effectiveness threshold. However, there is a relationship between the cost per QALY gained and the probability of rejection of a drug. Between 1994 and 2003, the highest cost per QALY at which a drug was recommended for listing by the committee was $52 400.^{2 ,21} Above this value 9 (82%) of 11 applications were rejected or withdrawn by the manufacturer and 2 were given only conditional approval (subject to price reductions). This compares with rejection (or withdrawal) of 41 (46%) of 89 listing applications with an estimated cost per QALY gained below $52 000. It appears that NICE and the PBAC have similar cost per QALY thresholds for rejection. However, the PBAC rejects a higher proportion of drug applications with lower cost per QALY values. There are several possible explanations. The decision-making process is quite adversarial, and PBAC members may have reservations about the quality of data assembled by pharmaceutical companies and their consultants. These data have been shown to be error prone.²² The PBAC has a direct responsibility for estimating budgetary impacts and recommending prices, and this may make committee members more risk averse. Both bodies consider factors other than the cost-effectiveness ratio. For instance, the PBAC accounts for the magnitude of the clinical benefit, the availability of alternatives, and the “rule of rescue.”^{9 ,23}

The nature of the recommendations made by NICE and the PBAC also differ. NICE is more likely to provide guidance on the use of a class of drugs (eg, glitazones), while the PBAC is concerned with the availability price and total cost of individual agents, based on manufacturers’ specific requests. This can lead to confusion. For instance, in Australia pioglitazone and rosiglitazone have different PBS restrictions. When drugs have very high costs, the PBAC can enter agreements with the different stakeholders leading to specific eligibility criteria, stopping rules based on clinical response, and financial risk sharing with manufacturers.²⁴

Pearson and Rawlins¹ highlight the apparent contradiction that a drug may appear cost-effective, based on its estimated cost per QALY gained, but be “unaffordable” because of the prevalence or incidence of the condition being treated. However, if the cost-effectiveness threshold has been set appropriately the technology should be affordable. If not, then other factors must have come into play, such as use of the drug outside of its intended clinical indications or structural problems with the financing of the relevant health care program.

Criticisms of NICE and the PBAC have differed. Economists have chided NICE for its apparent readiness to recommend new health technologies with high cost-effectiveness ratios without sufficient consideration of the likelihood of creating inflationary pressures on the health care budget. Critics consider that NICE should contribute more to effective prioritization of resources.^{16 ,25} There has also been concern about the uptake of NICE guidance, which in one study appeared to be variable and influenced, at least in part, by costs.²⁶ The PBAC, on the other hand, has been widely viewed as too tough with an initial rejection of more than half of submissions from industry and a final rejection rate of around one third.²¹ Other criticisms of both bodies are that they do not review older, potentially cost-ineffective technologies.¹⁶ The PBAC agenda is generally set by pharmaceutical companies, which make submissions comparing new products with drugs that may have been listed for many years and have unknown cost-effectiveness ratios. Although a new drug may have an attractive incremental cost-effectiveness ratio compared with the older product, both may represent poor value for money compared with nondrug treatments. For NICE, the agenda is set by a political process that has similarly concentrated on the new, rather than removing potentially redundant and cost-ineffective technologies. In the medium to longer term, both NICE and the PBAC will have to broaden their consideration to include assessing older technologies that perhaps should be discontinued.

The experiences of NICE and the Australian PBAC show that decisions about funding new drugs can be based on formal measures of cost-effectiveness. Both organizations have demonstrated a resolve to recommend against drugs with very high cost-effectiveness ratios, expressed as cost per QALY gained. Funding decisions should not be based solely on a cost-effectiveness measure; other factors are important, including the quality of underlying evidence, the magnitude of the clinical benefit, and the availability of alternative treatments. Decisions to fund a new treatment need not be a simple yes or no. The PBAC has used cost-effectiveness measures as the basis of price negotiations. Prior approvals for a drug subsidy can be linked to pretreatment disease severity measures; the attainment of defined clinical response criteria can be used as the basis of continued subsidy, and manufacturers can participate in financial risk sharing. Combinations of these approaches can enable selective access to drugs that might otherwise be prohibitively expensive.