Our thoughts on ‘integration’ of technology appraisal guidance into NICE guidelines
We appreciate the important role NICE plays in ensuring practitioners and commissioners are provided with up-to-date, clear and comprehensive guidance to ensure they can deliver high-quality, cost-effective health care for patients. We therefore welcome the proposed process of ‘incorporation’ to support the timely production of useful and relevant guidance.
We do see a potential place for comparative assessment of costs and benefits as part of NICE guidelines, in order to provide a clear and comprehensive overview of the relative benefits and limitations of treatments within a decision space to guide prioritisation and clinical decision-making. However, we see many risks and unanswered questions associated with the ‘integration’ process as outlined in the consultation document, particularly given the proposed potential consequences of the process for existing technology appraisal guidance (i.e. withdrawal of recommendations and associated mandated funding). Our concerns relate both to the feasibility of conducting fair, robust and consistent comparative analyses of costs and benefits with available evidence to be confident that the withdrawal of technologies will improve net population health, as well as long-term potential consequences for system-wide investment and incentives for manufacturers to launch products in the UK market given the already high upfront cost of engaging in the NICE technology appraisal process. Furthermore, the proposed guideline review process itself has the potential to be time- and resource-intensive at a time where NICE has committed to a more proportionate approach to decision-making in part due to concerns around capacity and its commitment to appraise all new medicines and significant licence extensions under the voluntary scheme for branded medicines pricing, access and growth (VPAG). As a result, in our view and pending further detail on how the outlined processes and methods would be applied, there is a real risk that the proposed approach could have adverse consequences for patient access to both existing and new innovative health technologies, as well as limiting patient choice, with potential to reduce net population health in the long run.
We appreciate that the consultation documents are not designed to present the full or final detail of how the guideline integration process would be implemented, but as part of the consultation we provided comments on specific areas where we have identified risks and challenges to help inform the future development of more detailed process and methods documentation.
A summary of our overarching concerns with the proposed integration process is provided below:
The integration process may limit patient choice
Each technology in a given decision space offers unique benefits to patients, in terms of differing efficacy on a patient-by-patient basis, differing contraindications, differing mechanisms of action and differing modes of administration. Withdrawing or restricting recommendations and associated funding mandates will therefore restrict patient choice, with fewer alternative technologies available to patients who may be unsuitable for the technology deemed most cost effective during the integration process. There may also be geographical restrictions to specific technologies, in particular for highly specialised technologies, and thus withdrawing guidance could impact geographical equity of access.
It may be challenging to appraise technologies in a fair and consistent manner
A fair and consistent comparative analysis of the costs and benefits of each technology in a given decision space may be challenging, and the methods and processes outlined in the consultation document leave open a number of important questions that would influence the robustness of such analyses. For example:
- Criteria for integration – The criteria for technologies being eligible for integration are not clear, with no definition of “decision space” provided and a lack of clarity on whether technologies available under managed access agreements would be within the scope of these assessments. Furthermore, no rationale was provided for the selection of a three-year timeframe for technology appraisal guidance to be considered for the integration process; this timeframe is unlikely to result in substantial additional evidence generation in rare disease spaces or for diseases with a long disease course.
- Stakeholder input – It is unclear how stakeholders, including manufacturers, will have the opportunity to input on the integration process, for example, responding to the proposed decision problem and providing additional evidence collected on the new technology. Stakeholder input is key to ensure the process is informed by the most appropriate and up-to-date evidence available.
- Methods for comparative analysis of costs and benefits:
- It is unclear how these analyses account for the potential for differences in committee preferences across previous technology appraisals for therapies in the same decision space.
- We note that NICE piloted a ‘pathways approach’ using a central economic model under the proportionate approach to technology appraisal, but based on concerns relating to complexity and transparency, NICE have since revised their approach to make use of ‘reference models’ that can be implemented within the current single technology appraisal (STA) framework.2 This guideline integration process is likely to be subject to similar challenges if a central model would be required for guideline integration assessments.
- It is also unclear how NICE would account for the impact of experience using technologies accumulated over years of use in clinical practice, which may not be reflected in the evidence submitted during technology appraisals informing initial recommendations.
- On the one hand, RWD accrued through use in practice can provide valuable new evidence. However, its usefulness may be limited given typically shorter follow-up and the potential for known and unknown sources of bias.
- In particular, consideration would need to be given to potential bias associated with the availability of multiple interventions in a decision space. Whilst multiple technologies may receive an NICE recommendation for use in the same patient population, in practice there can be optimisation of technologies to different patient subpopulations. For example, in a recent managed access review appraisal of tisagenlecleucel for treating children with acute lymphoblastic leukaemia, it was acknowledged that the introduction of the tisagenlecleucel changed the way in which the comparator, blinatumomab, was used in the NHS, restricting its use to a population more likely to achieve a complete response and thus be eligible for curative stem cell transplant. As a result, despite the availability of more recent trial and registry data for blinatumomab, the evidence source used to inform the original appraisal was deemed most appropriate on the basis that it was the only evidence reflecting a world without tisagenlecleucel.3
- Use of decision modifiers – It is currently unclear how decision modifiers will be considered for multiple technologies, and in particular how ‘established NHS practice’ would be defined to inform the calculation of the quality-adjusted life year (QALY) shortfall for each technology, given these technologies would already be available in NHS practice at the time of guideline integration. Furthermore, there may be challenges applying fair and consistent decision-making criteria and cost-effectiveness thresholds if technologies were originally assessed under different technology appraisal frameworks (STA vs cost comparison vs highly specialised technology [HST] appraisal).
The integration of recommendations may be time- and resource-intensive and may duplicate existing NICE processes
Evaluating the clinical and cost-effectiveness of each technology in a given decision space in a fair and robust manner is likely to be a time- and resource-intensive process, which may further stretch NICE’s limited capacity to provide guidance. It’s currently unclear whether this process would require time from external assessment groups, for example. The integration process could have knock-on impacts on NICE’s ability to produce guidance and guidelines on novel topics and interventions due to resource constraints; equally, upscaling of NICE resource to meet the additional evaluation burden could in theory translate to further or increased manufacturer submission fees unless there are other external sources of funding.
The integration process may disincentivise UK product launches
NICE technology appraisals are already associated with high upfront costs – submission to NICE alone costs upwards of £100,000 to the manufacturer.4 The integration process risks withdrawal of recommendations just three years after the initial recommendation. Therefore, manufacturers may deprioritise or avoid applying for reimbursement of their product in the UK. This process therefore risks preventing patient access to new and effective technologies.