Does The Nice Diagnostics Assessment Programme Use An Empirical Icer Threshold?

Abstract

Technologies used for diagnosis, monitoring, risk stratification and screening are an integral part of efficient healthcare systems. The NICE diagnostics assessment programme (DAP) was established in 2009 to promote consistent adoption of innovative and cost-effective diagnostics in the UK NHS. The objective of this study was to establish the empirical cost-effectiveness threshold for diagnostics evaluated by the DAP and to assess the importance of the ICER in decision-making. All NICE DAP assessments published before June 2017 were included; the guidance document and diagnostics assessment report (DAR) were reviewed. The type of technology, disease area, adoption decision, rationale for decision, ICER and research recommendations were extracted. Based on the 27 DAP assessments reviewed, there were 17 positive, 9 negative and 11 neutral recommendations (insufficient evidence). Multiple technologies were often evaluated within the same DAP assessment, and separate recommendations were sometimes given for different technologies or indications. Approved technologies were dominant, associated with QALY gains at costs of up to £85,300/QALY or associated with QALY losses at cost-savings from £4,324/QALY lost. Low ICERs, cost-savings and non-health related benefits were quoted as reasons for approval. Rejected technologies had on average higher ICERs (median: £1,468,175, range: £4,148–£15,000,000) compared to approved technologies (median: £21,127, range: £319–£85,300), but could also be dominant. Rejection decisions were often due to high ICERs, insufficient evidence or poor robustness of the model; the latter two reasons were grounds for rejection despite a cost-effective ICER. The DAP assessments reviewed were often limited by insufficient evidence and high levels of decision uncertainty. The ICER threshold of £20,000–£30,000 seemed to be mostly adhered to when recommending diagnostic technologies. However, the ICER was not the only determining factor, as non-health related benefits, including patient preferences, convenience and anxiety, were also considered by the committee when approving technologies despite high ICERs.