NICE proposes simplifying how medicines and medical devices are selected for evaluation.

12 Oct 2020 11:00 AM

As part of the work underway to review the methods and processes NICE uses to develop guidance on medicines, medical devices and diagnostics, NICE has launched a public consultation on proposals for changing how it selects the topics it will develop guidance on.

The proposals clarify the criteria that would see a device or diagnostic selected for NICE guidance development. In particular, these include where the costs and impacts are expected to be significantly cost incurring or cost saving, or there is uncertainty about the likely cost or the impact it would have on the healthcare system.

For medicines, the proposals confirm the commitment in the 2019 Voluntary Scheme for Branded Medicines Pricing and Access that NICE will appraise all new active substances and significant licence extensions for existing medicines, except where there is a clear rationale not to do so (for example, if the topic is a duplicate or has a significant overlap with an existing topic or policy arrangement).

Similarly, all new or significantly modified interventional procedures that will protect patient safety will be selected if they are available to the NHS or independent sector, or about to be used outside of formal research.

This approach would move away from the 15 criteria currently used to select topics for evaluation by NICE’s Centre for Health Technology Evaluation and provide a clearer and simpler process.

Helen Knight, programme director for Technology Appraisals and Highly Specialised Technologies at NICE, recently said:

“Topic selection plays an important role in the development of NICE guidance and is designed to ensure that the guidance we produce is on topics that support healthcare professionals and others to provide care of the best possible quality.

“These proposals will ensure we can continue to meet these ambitions at a time of unprecedented change in the healthcare system.”

Other key proposals include: