IQWiG and the gold standard “randomization”
About the author
Prof. Matthias P. Schönermark, M.D., Ph.D.
Founder and Managing Director
Founder and Managing Director
Fon: +49 511 64 68 14 – 0
Fax: +49 511 64 68 14 18
Fax: +49 511 64 68 14 18
Randomization describes the random allocation of study participants to the interventional and control group of the study and is used to neutralize potential confounders. There are often ethical objections to the randomization principle, which, for example, relate to the equipoise. In this sense, the ethical prerequisite for setting up a RCT is the balance between the study interventions. However, from the IQWiGs point of view, the equipoise and therefore the ethical acceptability should also be considered in all other studies and data sources. In addition, the definition of an equilibrium on closer inspection can lead to challenges, because who determines the balance of what and based on which criteria?
The evaluation of the equipoise of medical interventions refers to a balanced overall assessment with regard to the benefit aspects and the damage potential. However, this assessment may vary from different perspectives, among other things because the terms benefit and damage, but also the weighting of the single parameters (e.g. benefits in overall survival vs. disadvantages in health-related quality of life) have not yet been clearly established. The equilibrium can therefore be determined only with difficulty externally but is usually based on an individual assessment of the respective study participant, after he has been informed about the alternatives according to the current medical knowledge. The consent to participate in a RCT or another comparative study is then only possible, if the subject does not prefer any of the interventions in order to not restrict the interpretability of the results and if there are no other incentives associated with the consent. The ethical acceptability of a RCT is thus argued on the basis of a collective equipoise. An imbalance between the interventions and thus the question of the justifiability of the study applies, according to this argument, not only for RCTs, but rather for any kind of comparisons.
However, according to the IQWiG, criteria that could be used to describe the likelihood of an imbalance, regardless of individual (collective) opinion, can be defined as follows: There is a relevant difference between the alternatives in one or more patient-relevant outcomes, which has already been established with sufficient certainty.
This combination of conditions is not based on the compilation of an overall balance of operationalized and individual benefit and damage potentials but makes it possible to make an evaluation independent of indication and person. For example, the balance does not need to be questioned due to marginal differences in surrogate parameters, which have been determined with sufficient certainty, and were assessed differently. It is also possible to counteract the influence of the individual assessment by different types of representation, for example the type of scaling (absolute vs. relative).
On this basis, the IQWiG defines only two cases in which the conduction of a RCT is not necessary or justifiable: It can be argued against the conduction of a RCT, if relevant differences in patient-relevant outcomes have already been established with sufficient certainty. This is the case, for example, if several RCTs (usually two high-quality clinical trials) have been conducted for a given intervention, rendering further examinations obsolete, or the rare case, if a new medical intervention has such dramatic effects, e.g. in overall survival, so further evaluation is unjustifiable, since the question of benefit has already been clearly answered.
However, in (almost) all other cases, setting up a RCT to clarify open issues, according to IQWiG, is the most appropriate method and should be used as standard in the benefit assessment. The ethical objections regarding randomizations in RCTs can be invalidated, since the described rationale also applies to most other comparative studies and data sources. On closer inspection, the determination of an equipoise is – partly because of individual perspectives – associated with challenges. For this reason, the combination of three aspects (relevant differences, patient-relevant outcomes, sufficient certainty) should be used for a non-indication and person-independent assessment of a potential imbalance.
As an established strategy consultancy in the healthcare sector with a focus on accompanying the market access of new pharmaceutical products, in particular within the framework of the early benefit assessment, the SKC consulting company intensively deals with the current methodological standards and argumentation of the IQWiG and G-BA in order to anticipate together with their clients – as part of an overall strategy – any pitfalls, and also address them at an early stage.
BY Prof. Matthias P. Schönermark, M.D., Ph.D., Founder and Managing Director of SKC Beratungsgesellschaft mbH and Thora Mrosowsky, M. Sc. Health Economics, SKC Beratungsgesellschaft mbH
Sources:
Deutsches Ärzteblatt, Jahrgang 115, Blatt 3 (German only)