This report evaluates the benefits of health economic assessments of pharmaceutical products used in inpatient care.
The Swedish Agency for Health and Care Services Analysis (Vårdanalys) has been commissioned by the Swedish Government (S2012/3857/FS) to evaluate the benefits of health economic assessments of pharmaceutical products used in inpatient care (so-called hospital pharmaceuticals, Swedish “klinikläkemedel”). This interim report evaluates the effectiveness of the processes involved in the health economic assessments, hereinafter referred to as the “hospital pharmaceuticals project”. In conjunction with the final report, which will be presented to the Government in April 2014, Vårdanalys will consider whether a more knowledge-based and equal healthcare provision has been achieved. It will also produce an overall assessment of the benefits of health economic assessments of hospital pharmaceuticals.
For hospital pharmaceuticals – i.e. drugs used in inpatient care and ordered from hospital pharmacies for the clinic or unit in which they are used – no previous health economic assessment has been carried out at national level, and nor have any processes been developed for price negotiations with the pharmaceutical companies. Instead, these assessments and processes were carried out by each individual county council, or in partnerships between various county councils. In November 2010, the Swedish Dental and Pharmaceutical Benefits Agency (TLV) was commissioned by the Swedish Government to set up a pilot scheme whereby the public health authorities would be given the opportunity to apply to the agency for health economic assessments. The hospital pharmaceuticals project was started in January 2011, and a final report on the pilot scheme will be presented in December 2014. The initiative in terms of which pharmaceutical products are to be assessed by TLV will be taken by a national coordinating group within the Swedish Association of Local Authorities and Regions (SALAR, or SKL in Swedish) – “Nya läkemedelsterapier” (NLT, “New pharmaceutical product therapies”) – which was formed at SALAR at the request of the directors of Sweden’s county councils. Once the NLT group has selected the hospital pharmaceuticals, TLV will request background data from the pharmaceutical companies involved, perform a health economic assessment and then put together a knowledge base that will illustrate the cost-effectiveness of the medicinal product. From the knowledge base, the NLT group will then produce recommendations that will be communicated to the county councils and NLT. NLT may also hold discussions with the pharmaceutical companies about price reductions. However, the county councils will decide independently whether or not they wish to enter into agreements and decisions about the use of these pharmaceutical products.
PURPOSE AND ISSUES
The purpose of the report is to illustrate the effectiveness of the processes involved in the health economic assessments. The hospital pharmaceuticals project is a pilot scheme that is undergoing continuous development. The report by Vårdanalys therefore illustrates the process as it has been developed so far.
According to the assignment, we will be dealing with the following aspects of the hospital pharmaceuticals project:
• Ensuring that the initiation process from the New Pharmaceutical Product Therapies (NLT) Group has been coordinated and prioritised in an effective manner
• The duration of the period taken for case processing
• Obtaining information and supplying data. For example, has TLV been given access to the material from the pharmaceutical industry that the authority requires in order to be able to produce a knowledge base?
• The quality of the documentation obtained from the industry, and TLV’s cost-effectiveness calculations
• NLT’s communication with the county councils, and its recommendations for a knowledge base
• How the pharmaceutical companies regard TLV’s work, and what they consider to be the benefit of the knowledge base presented by the authority, as well as the costs associated with the production of these knowledge bases for the companies
In conjunction with the final report, Vårdanalys will illustrate the effects of the health economic assessments and consider whether a more knowledge-driven and equal use of pharmaceutical products has been achieved. Vårdanalys will also illustrate the effects as far as patients and the county councils are concerned. This interim report makes no recommendations; instead, an overall assessment and recommendations as to how the operation can be further developed with a view to ensuring good healthcare will be provided in the final report.
The method used in the assessment is essentially a qualitative one. Vårdanalys has conducted interviews with representatives from TLV, NLT, pharmaceutical companies, patient organisations and a selection of county councils. Vårdanalys has also collated and analysed process documents (23–26), TLV’s knowledge bases (17–22, 27, 28) and statistics relating to processing times and sales. In order to assess the quality of the documentation obtained from the industry and TLV’s cost-effectiveness calculations, an analysis has also been carried out on the health economic data. As part of the quality assurance process, Vårdanalys has obtained the views of a reference group consisting of a health economist and representatives from patient organisations, county councils and the pharmaceutical industry. Representatives from TLV, NLT, the county councils interviewed, patient organisations and pharmaceutical companies have also been given the opportunity to examine the facts in the report.
- NLT was not designed for its present mission
NLT is part of SALAR. In other words, it is an association, and is therefore not subject to some of the regulations that apply to the county councils, among others – for example, the Swedish Freedom of the Press Act (1949:105), the Swedish Public Access to Information and Secrecy Act (2009:400) and the Swedish Health and Medical Services Act (1982:763). NLT’s mission has developed over time, from one of providing support to the county councils in the introduction of a small number of new pharmaceutical products to a significantly more extensive role in the hospital pharmaceuticals project. TLV’s appointment of NLT as a suitable player changed NLT’s mission into one that includes prioritising which pharmaceutical products TLV should assess. NLT is also the recipient of the knowledge bases that TLV is developing, which may form the basis for NLT’s recommendations. In certain cases, NLT also holds discussions with the pharmaceutical companies about pricing, though NLT itself emphasises that its mandate in discussions about pricing is unclear. Vårdanalys considers that the present organisation was not designed for this increase in responsibility, and that the definition of NLT’s mission in the hospital pharmaceuticals project is unclear. NLT itself points out this issue, as well as indicating that the group is not organised, staffed or financed for its expanding mission.
Vårdanalys considers that NLT’s mission in the hospital pharmaceuticals project has grown as a pragmatic solution, designed to bring about national coordination among the county councils on the issues involved. There has been no other natural player in the existing structure to take on this role. However, we consider that NLT’s current role and mandate are unclear, and that its organisation is not suited to its mission. One weakness with the NLT structure is that its role and mandate are entirely built on the trust and desire to help of the county councils, rather than being built in a way that ensures it functions over time. Another failing is that several sets of regulations relating, among other things, to transparency, visibility and fundamental principles within the field of healthcare – such as the prioritisation ruling by the Swedish Government – need not be applied by NLT as it is a collaborative body within SALAR. Even though it is possible for NLT to apply the legislation (as indeed it has stated that it does), there is a major difference between choosing to apply the rules and being obliged to do so. Among other things, there is no clear accountability, while the incentive to comply fully with the legislation is perhaps weaker than it would otherwise have been. Other players also indicate some uncertainty as to whether or not NLT applies the rules.
- The hospital pharmaceuticals project is characterised by a lack of transparency
Vårdanalys considers that the hospital pharmaceuticals project is characterised throughout by a lack of transparency. The areas of deficiency relate above all to information and the visibility of the work of TLV and NLT. NLT provides no information about the selection criteria for, and reasoning behind, the pharmaceutical products nominated for health economic assessment. Neither TLV nor NLT provides information as to when the knowledge bases or recommendations are expected to be ready, or as to the current phase of the investigation process at any particular time. Patients and pharmaceutical companies are experiencing a lack of information as to when to expect NLT’s recommendations. It is also not apparent on what grounds NLT makes its recommendations, nor what has been taken into consideration when making these. NLT’s recommendations are also worded very briefly and contain no justifications. As NLT is an association, there are no statutory requirements for it to justify its decisions, nor to provide for transparency and visibility in its processes and decision-making. Given, however, that NLT’s recommendations are used as the basis for decisions on prioritisation within the healthcare service, it is important for any decisions made to be openly justified.
Conversely, TLV is covered by the regulations governing public access and secrecy. In the opinion of TLV, information at the authority is covered by rules of confidentiality that preclude it from providing information in any way other than as it does currently. This, too, therefore affects visibility and transparency on the project.
- Secrecy and secret agreements make communication and visibility more difficult to achieve, though they can lead to reductions in prices
To a certain extent, the rules on secrecy and the decision to keep the data confidential make communication between the parties in the hospital pharmaceuticals project more difficult. In the opinion of TLV, these prevent TLV from informing NLT whether an application already exists within the benefits system for the nominated pharmaceutical product. NLT is, however, able to ask the companies about this, while the companies themselves can also choose to inform NLT. It is also considered that secrecy regulations prevent the county councils from finding out what information has been sent to TLV from the pharmaceutical companies, and about TLV’s calculations – which may have provided a more in-depth picture of the cost-effectiveness of the medicinal product.
The proposed agreements that have been reached so far following pricing discussions between NLT and the pharmaceutical companies are considered confidential at NLT and are then subject to secrecy regulations at the county councils. The reason behind the companies’ wish to keep the agreements secret is that a reference price model applies in many countries. A reduction in the official price in Sweden would therefore have a domino effect on the price in other countries, many of which are bigger and more important markets than Sweden. The secrecy regulations also mean that the number of people who are party to the content of the agreements is limited, further reducing visibility. The lack of transparency in the agreements also increases the risk of corruption. On the other hand, the benefit of secrecy rules is that the pharmaceutical companies are able to agree to larger price reductions as these are not made visible, while the county councils can use pharmaceutical products that may not, perhaps, otherwise have been available to patients.
- Processing times are long
Processing times on the hospital pharmaceuticals project affect patients primarily when the county councils choose to postpone the use of pharmaceutical products while awaiting TLV’s knowledge base and NLT’s recommendations. This is a problem as the assessments relate to treatments that may have a great effect on patients’ quality of life and chances of survival.
The processing times in the various stages of the process include initiation by NLT, the time taken by the companies in producing the material, TLV’s health economic assessments, NLT’s recommendations, and the decision by the county councils on the use of the pharmaceutical product. As only two pharmaceutical products have been through the entire process, it is not possible to assess the length of the processing time in its entirety. Vårdanalys notes, however, that the generation of health economic assessments and recommendations has been a lengthy process. The time-consuming elements that have contributed to this include the completion of material by the pharmaceutical companies for TLV, the recruitment of external experts for TLV, and the time taken for NLT to make its recommendations. The health economic knowledge bases have taken between 8 and 17 months to produce. For most products in the hospital pharmaceuticals project, NLT has still not published any updated recommendations based on the knowledge bases, despite the fact that for some pharmaceutical products it has been more than six months since these were published. One explanation for the long processing times is that this is a pilot scheme that has taken time to start up. However, for the pharmaceutical products that Vårdanalys has analysed we have still not seen any trend towards shorter processing times.
- There is a demand for TLV’s knowledge bases, but there is clear potential for improvement
The pharmaceutical companies and county councils that have been interviewed by Vårdanalys are for the most part positive towards health economic assessments of hospital pharmaceuticals and consider that TLV’s knowledge bases are of high quality. Vårdanalys is of the same opinion and considers that TLV’s knowledge bases have been written in a clear and logical way, which helps to improve understanding of the health economic assessments. According to the pharmaceutical companies and NLT, the partnership and communication with TLV have also worked well.
However, the companies do criticise TLV’s forward planning, and the fact that TLV has not always reported deviating opinions on the part of the companies in the knowledge bases. Both patient organisations and pharmaceutical companies criticise TLV for having chosen, in certain cases, experts who either were not impartial or were insufficiently knowledgeable. TLV has, however, taken measures to develop the process by expanding its search for experts.
Vårdanalys has identified a number of areas of improvement for TLV’s health economic assessments. For example, TLV is not transparent in its reporting of indirect costs – including production losses in the case of patients with impairment or loss of working capacity. Nor does TLV communicate the degree of uncertainty in the analyses with sufficient clarity. There is also no uniform method for systematised reviews. Such a method would reduce the risk of failing to identify important aspects that require closer analysis. It is also not apparent how TLV validates the information it receives from the pharmaceutical companies, and the extent to which TLV seeks information from other sources. If excessive reliance is placed on information from the companies, there is a risk that TLV will lose out on other perspectives that would complete the picture.
- There is insufficient patient involvement
The selection process – the knowledge bases, recommendations and agreements produced – has a major influence on which pharmaceutical products are offered to the patients. Many players are involved, but patients and patient organisations have no clear role or part in the process. Patient organisations have been invited to information meetings with TLV and NLT on only a few occasions. Insufficient information is provided to patients and the public. It is difficult to obtain a general view of the various stages in the process, and a number of patient organisations are unclear as to the responsibility of the various players.
- There is a need to coordinate national initiatives
The introduction of new pharmaceutical products is an important issue, and many parallel initiatives in this area are taking place, including within the framework of the national pharmaceutical products strategy. The initiatives that we consider to be of particular relevance to the introduction of new pharmaceutical products are: the ordered introduction of new pharmaceutical products, the streamlining of the procurement of pharmaceutical products, the ordered introduction of new cancer drugs, the Swedish Pharmaceutical and Pharmacy Inquiry, and the continued analysis by Vårdanalys. Vårdanalys takes the view that it is important for the Swedish Government to adopt a comprehensive approach to these efforts and for there to be coordination between the various initiatives. By making the most of the insights provided by the various parties, we believe there is the potential to develop an effective process for the assessment of hospital pharmaceuticals.
- The hospital pharmaceuticals project has the potential to create value but is relatively demanding on resources
Both the county councils and the pharmaceutical companies acknowledge the value of access to health economic knowledge bases. Health economic assessments of hospital pharmaceuticals can also result in the more extensive use of cost-effective pharmaceutical products. This benefits patients, county councils and the community in general if it means that more effective use can be made of the resources of the healthcare services.
The new process involves significantly more players than previously, which will demand greater resources. TLV has increased its staffing levels in order to manage the new mission. The companies will need to provide the resources to produce the health economic material that TLV is requesting. The companies also risk a loss of sales where the county councils postpone the use of the pharmaceutical product while awaiting TLV’s knowledge base, or following a negative recommendation from NLT. For the companies, the health economic assessment means that pharmaceutical products will be evaluated by TLV, NLT and the county councils, thus entailing the use of more resources than was previously the case.
NLT’s undertakings within the hospital pharmaceuticals project will lay claim to significant resources on the part of the group’s members. As far as the county councils are concerned, the secret agreements will lead to an increase in administration, as these are regarded as being complicated. Each individual county council will need to make a decision about the agreements, and they will often have to monitor the use in order to have access to discounts. On the other hand, this monitoring may help by providing increased knowledge of which patients respond to the various treatments. A reduction in price may also free up budgetary space for the county councils, allowing pharmaceutical products to be made available that perhaps would otherwise not have been used.
Central coordination of assessments of pharmaceutical product use has advantages for the county councils. By raising price discussions to a national level, the county councils will have the opportunity to achieve potentially better terms with the pharmaceutical companies. This is particularly the case for the small county councils, who are in a weaker negotiating position.
An overall assessment of the benefit of health economic assessments in relation to the increase in resources required will be carried out in the final report by Vårdanalys, where the effects of the pilot scheme will also be illustrated.