Written by Alec Miners, Head of Health Economics
The National Institute for Health and Care Excellence’s (NICE’s) new processes and methods guide includes several important changes to the way submitted economic evaluations should be conducted (1). While the headline change is arguably the removal of the end-of-life criteria and the introduction of a new severity modifier, the document contains other interesting changes. One of these is the requirement (request?) to include a description of the conceptual modelling process. Section 4.6.3 states:
‘The conceptual model development process used to inform the choice of model structure should be transparent and justified. This should include details of expert involvement in this process (for example, number of experts, details of their involvement, how they were chosen). It is not enough to state that the chosen model structure has previously been used in published model reports or accepted in submissions to NICE.’
While organisations such as ISPOR (2) have recognised the importance of conceptual modelling for some time, we suspect that the idea will be unfamiliar to many.
What is conceptual modelling and what does it involve?
The literature on conceptual modelling in healthcare is relatively limited (see McMeekin for a helpful review (3)). While an agreed definition doesn’t exist, it’s reasonable to say that it’s a process that starts prior to the model coding stage, typically involving qualitative research and/or literature reviewing. Tappenden refers to two stages, in which ‘problem-orientated’ and then ‘design-orientated’ conceptual models are built (4). In the problem-orientated model process, important issues are determined such as the choice of comparator technologies, the perspective, and time horizon. In a NICE context, most of these elements are formally set at the scoping stage and outlined in the methods guide.
It is much more likely that Section 4.6.3 is emphasising the importance of the design-orientated step, in which factors such as the actual model structure, the choice of (health) events to include, the factors that determine their likelihood, and the type of modelling approach to use are chosen. The process often involves the development of a graphical representation of the decision problem in the real-world, providing an easily understandable representation for stakeholders. In cost-utility analysis, this can comprise a graphical representation of the impact of the health condition on morbidity (i.e. health-related quality of life [HRQoL]), mortality and cost; the mechanisms by which these elements are affected, and the impact of an effective treatment.
In many ways, documenting these decisions has always been considered good modelling practice. But does the explicit reference to conceptual modelling in the new guide suggest something more formal than what was previously expected by NICE? It would be difficult to argue otherwise, yet few recommendations on the required methods or depth of this process are provided – only that more details as to how expert opinion was included are to be provided.
Whatever the precise nature of conceptual modelling, its appearance in the methods guide seems to reflect a desire by NICE not to be tied down by precedent model structures. This is not to say that precedent will no longer be important, but to be reliant on it could become increasingly risky, particularly when disease processes are complicated, poorly understood, or where there are competing paradigms. This is speculative of course, and perhaps the importance of including a conceptual model stage will be lowered by committees because of other additional and existing submission requirements. Perhaps many organisations submitting models to NICE already include a model conceptualising process, but do not currently label it as such or document it in much detail? It may also be that for some disease areas, model design choices are limited, meaning that a conceptualisation process is likely to be brief.
At Source, we are continually thinking through the implications of a number of NICE’s new requirements, including Section 4.6.3. If you would like to discuss them, or any HTA-related issue, do please get in touch.
In the meantime, is this new requirement important or a hill of beans? Let us know your experiences or thoughts on this topic here.
References
1. National Institute for Health and Care Excellence. NICE health technology evaluations: the manual. 2022. Available at: https://www.nice.org.uk/process/pmg36/chapter/how-to-use-this-manual
2. Roberts et al. Value Health. 2012;15(6):804-11.
3. McMeekin NJM. PhD thesis. 2921. 2021. Available at: https://theses.gla.ac.uk/82570/
4. Tappenden. HEDS Discussion Paper 12/05. 2012.
