Are you ready for the Danish Medicines Council 2.0? – The QALY
The implementation of a “QALY model” in the Danish Medicines Council (DMC) will have a significant impact on the assessment process.
The quality adjusted life-year (QALY) is primarily used to guide health-care resource allocation decisions. It is primarily used in countries such as UK, Norway, Sweden, Finland, Australia, and Canada. It is considered standard practice to use QALYs as the health outcome in health economic evaluations and within health technology assessment (HTA).
In this short overview, we will focus on QALYs within the context of societal resource allocation, and how QALYs can help inform the DMC decisions.
How will a “QALY model” impact the DMC process?
The implementation of QALYs may lead to more informed decisions and potentially greater transparency in the assessment process. It will also allow the HTA to have one single unit of benefit by which all drugs can be judged fairly. Additionally, it will pose different requirements for both the DMC and the applicants.
Because QALYs are estimated through health economic models and are primarily relevant in relation to cost-effectiveness, there will be a much greater focus on the economic analysis than in the current process. Therefore, it will be even more important to submit robust health economic models that can demonstrate the value of an intervention, specifically in the Danish context. However, there will still be a strong focus on the clinical assessment, and it is important that the health economic model reflects the data observed in the clinical study, where available.
The process may move towards the approach used in other “QALY-countries”. However, we still expect a “Danish model”, that requires a somewhat bespoke approach.
So what is a QALY?
The concept is actually very simple. Some interventions impact only survival, some impact only health-related quality-of-life (HRQoL), and some impact both. The quality-adjusted life-year (QALY) is designed to capture the impact in any of these scenarios.
The QALY is a generic valuation of health benefits. In the QALY it is assumed that health is a function of the length of life and the HRQoL.
The basic concept is that individuals move through health states over time and that each health state has a value (utility weight) attached to it. The utility weight is normally elicited from a representative sample of the general population.
Health states are valued on a scale, where the upper end of the scale is defined as perfect health, with a value of 1, and being dead is considered to be equivalent to 0. Some health states may be regarded as worse than death and have negative scores. The value scale has interval scale properties, i.e. a gain from 0.4 to 0.6 is equally as valuable as a gain from 0.6 to 0.8. This property allows for aggregation of QALY changes.
The expected survival (exemplified by “years lived” in the illustration) is weighted by the health state utilities (“Health-related quality-of-life” in the illustration) to yield the number of QALYs (the area between the two curves in the illustration).
Why are QALYs relevant?
When allocating scarce resources within fixed budgets, opportunity costs are crucial. In a fixed budget health care system, where increased costs will displace other health care services already provided, the opportunity cost is measured as the health loss resulting from the displacement of activities to fund the selected intervention. To estimate the opportunity cost across the healthcare system, a generic effect measure of health is therefore needed.
Therefore, the purpose of the QALY is to be generic to inform resource allocation decisions. This property comes with a cost – lack of disease specificity (Link).
How can opportunity cost be assessed?
Opportunity costs can be assessed through cost-utility analyses (CUAs), i.e. using QALYs as the effect outcome in the cost-effectiveness analysis. The main result of a CUA is the “incremental cost-effectiveness ratio” (ICER), which can be calculated by dividing the incremental cost (difference in costs) with the incremental QALYs (difference in QALYs) between competing healthcare interventions. The ICER is an expression of what we can expect to pay for the anticipated health and survival gain.
It is important to note that QALYs are not directly available from clinical trials. QALYs are synthesized by using the HRQoL and survival (often in combination with external data) and modelling this over the relevant time horizon. Models, independent of the outcome of interest, often rely on several assumptions – assumptions that are necessary to translate the information from the clinical data to meaningful outcomes for decision making. These assumptions can be tested in the models and it is standard practice to do so.
What is the role of QALYs in decision making?
The use of QALYs has often been associated with a great deal of controversy. Therefore, it is important to illustrate the role of QALYs in the decision-making process.
The QALY does:
- Inform decisions through ICERs by estimating how much health we can expect to achieve for the costs we spend (if we choose to spend it)
The QALY does not:
- Decide on anything – decisions often rely on several other aspects besides just cost-effectiveness
- Incorporate issues of equity, fairness, and severity. Some countries have included these elements as additional decision criteria
Why is the DMC moving away from the current assessment model?
The DMC is currently using a form of multi-criteria decision analysis (MCDA) inspired by the German AMNOG process, where benefits associated with an intervention is expressed on a categorical 7 step scale ranging from negative benefit to major benefit. There are several problems with the current process from a decision-making perspective. This is not optimal since the role of the DMC is to make decisions. We will exemplify some of these problems in the following:
There is an upper limit to how great of a benefit an intervention can provide – the “Major benefit” category. To give an example: One product provides in a life year gain of 2 years and another product provides in a life year gain of 10 years. Both are classified as having a major benefit, however, the absolute benefit gain is obviously different. This should be reflected in different WTPs for the two products. However, the categories are the basis of the negotiation process and subsequent DMC decision, and therefore it is difficult to account for these differences in the current process.
The QALY has interval scale properties and the upper limit of the potential QALY gain will only be limited by the expected life duration from the onset of the intervention.
This also relates somewhat to the issue presented in problem one; the same category does not necessarily reflect the same health gain for different products. The benefit levels are determined by how well the product performs for selected outcome measures. The outcome measures are often generic e.g. overall survival, progression-free survival, adverse events, and HRQoL. However, the outcome measures can also be disease specific. The benefit of this approach is that it may greater at capturing differences in efficacy for the specific disease than the QALY. However, it is an issue in a decision-making context since the actual health gain may vary greatly for the category.
The QALY is a generic measure and therefore not as sensitive as the disease-specific measures. However, a QALY can be directly interpreted regardless of the specific disease area. This is more helpful in a decision-making context.
The benefit categories are primarily based on relative differences in efficacy, i.e. hazard ratios, risk ratios, and odds ratios. Absolute differences are also a criterium in the process, however, these act only as a binary measure for whether the efficacy difference can be considered relevant. Relative differences are intuitively more difficult to interpret and use in a decision-making context than absolute differences.
The QALY is a measure of absolute health gain and is, therefore, easier to interpret in a decision-making context.
The benefit categories are based on truncated time horizons. The benefit categories rely on the follow-up period in the clinical trials, which most of the time include a high degree of censoring. Statistical extrapolations of the clinical outcomes are not conducted in the clinical assessment, which is a problem, because resource allocation decisions need to incorporate the future benefit beyond the clinical trial.
Since QALYs are absolute health gains, the assessment relies on extrapolating the health benefits beyond the limited time horizon in the clinical trials. CUA models extrapolate the costs and QALYs to the relevant time horizon, which is often a lifetime horizon. Using only a trial time horizon would often result in very limited QALY gains and very high ICERs.
Outcomes and benefit categories are not weighted systematically. When determining the relevant outcomes for an assessment of a particular product, the DMC determines which outcomes are critical and which outcomes are less important. This process is standard MCDAs, where the combined weight of the weight added to each attribute should sum to 1. However, in the DMC process, this is a narrative weighting, and adding in additional outcomes may not necessarily reduce the weight of other outcomes. This opens up for a “pick and choose” approach, which limits transparency in the decision-making process.
In the QALY, the weight (utility) is decided by the population and is an inherent part of the QALY. Therefore, when using QALYs, there is no need to make assessments about the weight of the different outcomes.
Are you ready for the DMC version 2.0?
In Nordic HTA, we have great experience in developing and adapting cost-utility models, and we have a strong track record of successful clinical and health economic applications to the DMC. Please feel free to contact us if you would like to know more about how we can support you to be ready for the DMC version 2.0.