The German Institute for Quality and Efficiency in Health Care (IQWiG) published its long-awaited draft methodology on the cost-benefit analysis of drugs in January. In the second part of a two-part interview with IMS Health's monthly newsletter, Pharma Pricing & Reimbursement, IQWiG Director, Professor Peter Sawicki, continues his discussion of the proposed guidelines (click here for part 1).
IMS: Why was the efficiency frontier chosen as the methodology here, given that the expert panel was drawn from countries in which there are some common approaches to cost-effectiveness analysis? It's an unusual approach in health economics.
Peter Sawicki: The efficiency frontier method uses standard economic approaches. The panel thought that this method would fit the requirements of this country in the best way. We had workshops with the panel and they came up with this approach.
IMS: As we have discussed in part one, the focal point at the moment is how you use that technique to provide guidance on the pricing of products that are deemed to be superior. Do you have any comments on why this is a good tool for that?
PS: The chart below provides an example of the sort of situation we could show to the SK. We could demonstrate the costs on the x axis and, for example, the strokes prevented on the y axis. In this example we would say here are the interventions, A-E, and this is the position of the new intervention X at the moment, based on its price. As this falls outside of the shaded area in the chart, you need to amend its price in order to make it of comparable efficiency to other established treatments, so that it moves to the left on the chart. Whether it's to the dashed line [indicating that its efficiency relative to that of A is similar to that of A relative to that of the next most beneficial, E] or to the dotted line [where its efficiency is relative to that of A in terms of the greatest efficiency shown by any available treatment, which is that of C, relative to that of no treatment] is up to the decision maker.
Figure 1: Example of the Position of New Product X Relative to Established Products on the Efficiency Frontier
Source: PPR based on IQWiG
IMS: Where does your advice stop and where does their decision making start?
PS: It stops at our description of the relationships between the products in terms of their relative efficiency.
IMS: I ask this because there is a concern that IQWiG is getting into price setting, via this approach.
PS: What we can do is describe the costs that are involved in moving a product on that chart from one position to another. Based on these costs, you can very easily calculate a price. Whether it's the insurance company or ourselves that does that - anybody can do that.
IMS: But if you did that, wouldn't the response be that you were setting prices?
PS: No, we are not setting prices, we are just presenting a chart and you can easily draw a line and calculate the costs.
IMS: So let's look at a couple of possible scenarios. In one case, you have a category in which there is a generic. There is a concern that if a product is assessed as having an incremental additional benefit of X% to an established product (which may be a generic in this case) it will only be rewarded with a similar increment in terms of its price, relative to that of the generic. On the other hand, in an oncology indication, for example, you do not have a generic, so a similar incremental benefit would yield very real financial gains, given the likely high price of currently-available treatments.
PS: It depends on the increase in benefit. If there is a generic and the increase in benefit is large enough, then you can even go for no cost-benefit assessment with our institute. Because then, the manufacturer can claim that there is no real alternative - that the gap between the new product and the generic is too large. That is the first step.
If that is not the case, we will compare the new product to the other products on the market at their current prices. This is the case with all other types of products. For example, imagine if you put a cellphone on the market: you cannot claim that this is smaller and then look at the pricing that was in place years ago when the larger alternative first came to market and seek an increase in price relative to that price because that is not the market now. We have to compare the new product with what we have now, including the prices that we have now.
The fact is that I hope that this will push the industry more to diseases and more to developments where we will have a real step forward. Or perhaps a shift to diseases where not much has happened during the last 20 years. If we look at the y axis on the diagram, the fact is that the more beneficial the product is, the higher it will be on this axis. And the higher that is, the harder it is to improve on that and to justify a higher price with a further product. So we hope that the industry will focus on categories where we do not have so many alternatives as in the example - where there is an obvious opportunity for major improvement on what is already there. So we will see a move away from, for example, hypertension. We already have so many drugs in this field that I don't really think that pharmaceutical development in the field of hypertension, or diabetes or hypercholesterolaemia will really offer a lot of benefit for patients.
IMS: So are you saying that there is an opportunity for real gain, which will be rewarded accordingly?
PS: Yes.
IMS: In a press release issued alongside the draft methodology, you were quoted as saying: "As the cost-benefit assessment not only involves purely scientific but also normative issues, we need a societal consensus. Otherwise, decisions that are made later on the basis of specific cost-benefit assessments will not find the necessary acceptance." Perhaps you could tell us a little about what you meant by that.
PS: This is a framework for a methodology. There are certain details that we have to work on. What we want to hear is whether this is something that people think they can use to aid decision making. That is why we are engaging in the consultation on the proposed methodology.
In relation to this, we could talk for hours about the equity of distribution, the burden of diseases. In my view, the majority of opinion in Germany is critical about approaches that seek to compare health economic values across therapy areas - about having a threshold for every disease. There is a view that certain people might be harmed by that, for example elderly people or patients who are very ill because the chance for the prolongation of life is less for those patients than for others. Also certain diseases are seen to be cheaper to treat, because products in these areas were developed years ago. This contrasts with the situation in the treatment of certain cancers or transplantations, for example. To attempt to apply a single threshold to all this will cause inequity in the view of a lot of people in Germany. If we take Euro50,000 per QALY [quality-adjusted life year] as the threshold, for example, it will be easier to stay below that for some acute infectious diseases, but for chronic diseases such as HIV, or for certain cancers it may be necessary to go above this threshold. Such considerations were taken into account in relation to the steps that we had in mind to develop the methodology.
IMS: Is there any intention to look across therapeutic groups - to look at assessment that spans therapy areas?
PS: At the moment, the question is whether a product has a benefit in a single indication. We are not allowed to compare drugs across different indications or disease areas. Before we can do that, there must be a change in the law.
IMS: But this may be a possibility for the future?
PS: Of course it is. But this would require a change in the law. At the moment I don't think that it's probable that that will be changed within the next few years, but some day it may happen.
IMS: And this would involve a different approach to that outlined in the proposed methodology, wouldn't it?
PS: If and when the law was changed, then we would need a uniform measure for value. You can use QALYs or other measures, but you need a uniform measure, an integrative measure, that makes it possible to compare between diseases. I think that it is very difficult, if not impossible, to compare the outcomes of different diseases using one single value indicator.
IMS: The preamble to the draft paper suggests that the expert panel, having considered the methodologies that are being used in health economics in various contexts in other countries, found that there is no "gold standard" approach and that many such approaches do not have application to the German scenario. Were there to be a change in the law that allowed comparative analysis, would there be a need to carry out further research to find out what is appropriate for the German situation?
PS: If the Bundestag decides at some stage to establish a ceiling for expenditure at a given level of euros per QALY, for instance, we will of course follow that. But we can't define that ourselves. And I wonder how other countries have done that? There is no uniform threshold in other countries - it varies. According to studies I have read, in Holland it is €20,000 (US$31,000), while in England and Wales it is £20,000-30,000 (€26,000-€39,000; US$40,000-US$60,000). In Australia, meanwhile, it is US$50,000 (€33,000), while in the US it is US$50,000-US$100,000, depending on the payer. This suggests that there is no scientific uniform way to approach the amount of money that you want to spend.
IMS: Coming back to some shorter-term considerations: what is the timetable for the implementation of the proposed method?
PS: We will start small and we will try, perhaps in the second half of this year, to calculate an example and see how it works. The feedback from the stakeholder consultation that runs until the end of March will be reviewed by our experts and that will lead to the development of the first methodology that can be used. Then we will try it out in several dry runs or model evaluations and see how it works. After that we will modify it again.
Following this initial consultation, this is not the final methodology that will emerge and that will be used for the next 20 years. The first version of the methods will be published in summer or in early autumn of this year, and will be followed by annual updates.
IMS: Who will be involved in helping you to finalise the methods?
PS: The international expert group that developed the proposed methodology, along with health economic experts in Germany, our scientific advisory board and others. In relation to the ongoing development of our benefits methods, we have identified certain scientists in Germany who are interested in cooperation and who have time and skills to contribute and perhaps that will also be the case here. We will accept any input that helps us to improve our methods.
IMS: A concern has been voiced by the industry and other observers that there is little methodological detail in the draft and while it is made clear that this is to follow, this will only be made available once the first version of the finalised methods paper is released. By which time it will be too late for comment on the detail of the methodology.
PS: The very first draft of the paper had the detailed methodological considerations in there. But this ran to over 300 pages and the main points about the methodology being used were lost in so much textbook-style detail on approaches. So we decided to remove that because it was offering nothing additional to that already published elsewhere. It is true that there will be no further discussion about details of how the methodology will be carried out until after the publication of the preliminary protocols for the reports.
IMS: What if there was a backlash against this particular approach? Say, for example, because it isn't usually used in health economics.
PS: The critics may propose a different approach, which is better: that's fine, we would take that on board. Or the critics may say this is rubbish, this is absolutely wrong, but we do not have a better solution, and that what we need is a change of the law. If we accepted the fundamental criticism in this case, we would accept that we had failed to develop an effective methodology and would stay with the benefit analyses that we have been doing for the past three years. Then we would wait for a proposal of what to do, from the Ministry, for example. But it must be clear that the proposed approach is wrong. So far, I have not heard this.
The next step is to go through the details of the approach with the scientific community, which we will do over the course of two or three meetings [the first was scheduled for the end of February].
IMS: If it did go ahead, who would carry out the work? IQWiG alone, or with support?
PS: For each and every report we do we commission external experts and that would be done in this case.
IMS: There have been some concerns about a perceived limitation of available independent capability in health economics in Germany.
PS: If there are insufficient numbers of independent people with the necessary expertise in this field in Germany then we will go out of Germany, possibly to the German-speaking countries, Switzerland and Austria. If this is not enough, we would look at expertise elsewhere.
IMS: Taking such considerations into account and given that this is a new area for IQWiG to be working in, are you confident that you have the capability to do this, in the timeframe we have discussed?
PS: Yes, I am.
For more information on Pharma Pricing & Reimbursement, published monthly by IMS Health, contact the editor, Neil Turner at nturner@uk.imshealth.com, tel +44 (0)1223 273207.