The Decision-Making Process of Payers: A Pilot Survey In The Netherlands
Recently there have been three parallel trends showing increasing data requirements for reimbursement at a central level, more decentralisation within the decision-making process, and the implementation of prescription restrictions. Until recently, the health authorities in Europe established incentives for efficient health care delivery by means of decentralisation of the health care decision-making process and the implementation of market mechanisms. At a central level, the demand for cost-effectiveness and budgetary impact data has been increasing, which already resulted in formal dossier requirements in most countries. Health insurers recently obtained increasing authority to implement policies and freedom to structure local health care organization. A positive national reimbursement decision for a new medication will not automatically mean that the drug is available for the prescribing physicians in the near future anymore. A positive reimbursement decision by the central health authorities will become a non-binding positive advice to decentralized health authorities, who may consider this advise in their decision-making process for the development of local formularies. These formularies only contain a limited number of drugs for each indication, which are available (reimbursed) for the local medical community.
The objective of this article is to depict the decision-making process of health insurers with regards to the assessment of new innovative medication for listing on their formularies. The focus of this survey is on outpatient drugs, which means that hospital-based drugs are excluded. The study was performed in The Netherlands, because health care reform in this country is advanced with regards to decentralisation of decision-making to payers.
Reimbursement of new pharmaceuticals was traditionally based on the classical or traditional registration data (efficacy, safety and quality parameters). Once a drug was registered in a country, it was automatically centrally reimbursed and available for the potential population of eligible patients. However, escalating costs have become a major concern for health care professionals, decision makers and the public, prompting the implementation of new cost containment measures over the last decade, especially for new pharmaceu-ticals (since these last few years) we can distinguish various additional data requirements, which relate especially relate to the use of the drug in daily practice. The most important new data requirements are: effectiveness, quality of life (QoL), cost-effectiveness and budgetary impact. A decision on reimbursement will be based on the value for money of a new drug, which is based on the incremental clinical benefit of the new drug compared with standard therapy. Health authorities will make a trade-off between the extra clinical benefit (therapeutical value) and the premium price of the new drug versus standard therapy. Efficacy and side effects are the main clinical criteria, but also warnings, contra-indications and route of administration may also be relevant criteria for specific indications. In addition effectiveness, which offers a picture of the actual value of a drug in daily practice, and QoL have become important criteria.
The development of prescriptions restriction is another policy tool to control the costs of pharmaceuticals. While traditionally reimbursement decisions applied to the officially registered indication, authorities have recently been imposing restrictions on the claim made for the drug. These restrictions usually relate to following a treatment protocol, to limit the number of prescribing physicians or to limit the range of indications (UK, Germany and France). The primary goal of a prescription restriction is to reduce and control the budget impact of the reimbursement of a new pharmaceutical.
These additional requirements listed above for new medications may vary by country, but this development may be considered a general trend in Europe as well as in the United States.
Until recently the health authorities in Europe started to establish incentives for efficient health care delivery by means of decentralisation of the health care decision-making process and the implementation of market mechanisms. Both traditional and recent containment measures focus specifically on the pharmaceutical drugs sector in many countries. Local authority councils have recently obtained increasing authority to implement policies and freedom to structure local health care organisation (e.g. Italy, Sweden) . A growing number of countries have recently adopted some form of purchaser-provider split. The basic idea is to create a demand side (purchasing agency) that is separate from the supply side (providers). The reforms split the system into purchasers and providers.
The decentral health authorities vary per country, but we may distinguish regional authorities (e.g. Spain, Sweden), third-party payers or health insurers (Krankenkassen, District health Authorities, etc.) and hospitals (or hospital cooperations). In this overview we will focus mainly on the third party payer and the term “payer” is used for “health insurer” in the remainder of this manuscript. A positive reimbursement decision by the central health authorities has become non-binding positive advice to decentral health authorities, who may consider this advise in their decision-making process for the development of local formularies. These formularies only contain a limited number of drugs for each indication, which are available for the local medical community. A consequence, physicians have limited prescription freedom, as they are not allowed to prescribe drugs which are not listed on the formulary.
As a result, the main target audience in Europe for pharma is increasingly becoming the total set of decentral health authorities (payers) in a country. The health care environment in Europe is likely going to be similar to the United States, where the payers (HMO and managed care organisations) are the key target audiences for the pharmaceutical industry.
The three main payers in The Netherlands were contacted for participation in this study: ACHMEA, MENZIS, and CZ. These payers have a joint market share of more than 80%. The responsible staff member for pharmacy policy was asked for participation. An interview was arranged during which the survey questions were discussed. After the interview the results of the interviews were sent to the interviewee for possible feedback. The assumption was that an innovative drug has become available, which cannot be clustered into an existing class by the central health authorities. We assumed that listing on the formulary mean full reimbursement. The interviewees were asked to rank the predefined criteria according to level of importance (from 1-5). The interviewee could also add other relevant criteria, which are not mentioned. The following predefined criteria were defined beforehand: improved efficacy versus drugs on the list, improved safety versus drugs on the list, improved ease of use versus drugs on the list, improved patient satisfaction versus drugs on the list; improved QoL patient versus drugs on the list, reduced productivity loss, improved productivity at work; drug price; impact on total annual drug budget, impact on total annual health budget, improved quality of care; cost-effectiveness data; and recommendation in clinical guidelines.
The results of the survey are shown in Table 1. Recommendation in clinical guidelines has the highest ranking for two payers, and is also mentioned by the third payer in the top 5 ranking. Cost-effectiveness data was ranked as the second most important criterion by two payers, but it was not mentioned by the third payer.
The classical criteria efficacy, safety, drug price and impact on budget were not always mentioned. Efficacy has a medium ranking of 3 by two payers. Safety and drug price were only mentioned once. Impact on drug budget was mentioned by two payers, but only with a ranking of 4. Impact on total budget was mentioned by one payer with a medium ranking of 3. Impact on productivity, ease of use and patient satisfaction were not ranked in the top 5 by any of the payers.
The survey showed that payers increasingly use cost-effectiveness data, but they state that the cost-effectiveness threshold for listing a new drug should be considered a political decision. Hence the establishment of a f20,000 per QALY threshold is not a payer’s decision, but should be set by national governments.
Including the recommendation in clinical guidelines is a very important criterion. If a new drug is not included in the clinical guidelines, payers will be hesitant to reimburse it. Physicians, however, should have appropriate time to get experience with a new drug in daily practise. If after three to five years, the drug is not yet included in the guideline, there is no rationale to keep it on the list. In addition it was noted that guidelines should be assessed on their cost-effectiveness versus standard treatment in the guidelines. Hence, temporary reimbursement is possible with new assessment after 3 years based on effectiveness in daily practise, cost-effectiveness and budget impact. Risk sharing agreements may be considered for future reimbursement in case of doubt.
At launch the main source for payers is the CFH report of a new drug, which is the national assessment report including clinical assessment, budget impact and cost-effectiveness analysis; a comment was that these reports are too technical for payers and do not correspond with daily practise. For example, actual dosing can be quite different from clinical trials.
Indirect costs due to productivity loss are not a criterion for payers because these costs do not belong to the payer’s budget. Only when large companies take corporate insurance with a payer may indirect costs be included, but the current financing system does not really create incentives for this option.
Table 2 shows results for other criteria, which are called indirect because they are not related directly to the product. Competitive benchmarking versus other health insurance companies is important for all three payers. This is important because of market mechanisms. The focus is both on quality of care, costs, and cost-effectiveness.
Attracting patients with a risk profile is important for two payers, when the disease relates to a “basic” insurance package. The budget impact is a relative criterion, because of arrangements between payers. This so-called “vereve-ningsregel” means that total annual costs per patient exceeding €20,000 will be distributed among all payers. Hence there may be a paradox to attract expensive patients in some cases. If risk patients require additional insurance, however, there is no incentive to attract at-risk patients, because so-called “vereveningsregel” only relates to “basic” insurance package.
Package deals e.g. pharma has other drugs on the list and can offer a volume discount, is considered, especially when it relates to additional insurance services. Supportive services offered by pharma, e.g. disease management programs can only be considered when they are not related to one drug and when the program, can be executed independently. In addition, most payers do their disease management programs by themselves. Risk sharing, especially when the new drug after temporary reimbursement assessment leads to moderate assessment, was mentioned as an alternative option.
Payers do not consider research projects offered by pharma, unless they can be performed fully independently.
This pilot study led to the observation that decision making has become a real multi-criteria process by payers. Decisions are not based anymore only on the classical criteria (efficacy, safety, drug price and impact on budget). In addition, the new criteria cost-effectiveness and QoL have become important. A specific important criterion is the recommendation in clinical guidelines. The classical criteria are based on registration data and drug price, whereas the new criteria cost-effectiveness, and recommendation in clinical guidelines relate especially to the use of the new drug in daily practise by the patient and the medical community. Indirect costs are not considered relevant. The medium ranking of budget impact may also be explained by the “verevening-sregel”, which means that total annual costs per patient exceeding €20,000 will be distributed among all payers.
The use of cost-effectiveness and QoL data fits with the new business strategy by payers to contract with pharmacists and other health care providers, including the recommendation in clinical guidelines. Payers are increasingly taking a broader perspective replacing the traditional budget silo mentality. Cost-effectiveness data especially is facilitating this new holistic approach, which is required in the new environment with increasing competition. Payers, however, have shorter time horizon, which may conflict with long-term time horizon in cost-effectiveness studies, especially in chronic diseases, where initial higher drugs costs will only be saved in the long-term. The emerging market mechanisms can be used to explain the critical data in the decision-making process by decentral health authorities. Payers need to find an optimum amount between cost containment and offering quality of care. Cost containment is important for generating profit, but quality of care is important for competing with other payers, as consumers/patients may switch to a competing payer.
1 J. Rovira, M. Craig. Consolidation, co-payment and selective financing: Spain remodels its healthcare system: Spectrum healthcare delivery. Decision Resources 1993;50:1-