PRM127 - ECONOMIC MODELLING CONSIDERATIONS IN RARE DISEASES: A TARGETED REVIEW OF HIGHLY SPECIALISED TECHNOLOGY APPRAISALS

Abstract

Economic evaluation of rare diseases is associated with methodological challenges. This study will assess models submitted under the National Institute for Health and Care Excellence (NICE) highly specialised technologies (HST) programme to identify challenges associated with rare disease model development. A targeted review of the NICE website was conducted on 7th June 2018. All published HST appraisals were included. Data were extracted from identified studies regarding modelling approach, patient population, interventions, results, NICE recommendation, and modelling challenges cited by the evidence review group (ERG). A total of 7 published HST appraisals were identified. All 7 HST appraisals received a positive recommendation. A further 10 HST guidelines are in development. Of 7 published HST appraisals identified, 6 presented cost-consequence analyses while only 1 presented a fully incremental cost-utility analysis. All models employed a Markov state-transition structure. The primary challenge noted by the ERG was a lack of high-quality randomised clinical trial data, cited in 6 of 7 published HST appraisals. Further common challenges included use of surrogate outcomes in clinical trials, a lack of robust utility data, and uncertainty surrounding the natural history of the disease. Moreover, it is often unclear how best to reflect societal benefits associated with new treatments for rare diseases. The review identified barriers to achieving traditional cost-effectiveness thresholds, including high drug acquisition costs and the use of specialist administration centres. Non-randomised clinical trial data obtained from small samples will lead to substantial model input uncertainty. The use of surrogate outcomes in trials leads to the requirement for extrapolation to endpoints pertinent to the model time horizon. Such challenges should be explicitly recognised as it is often impossible to follow best practice when conducting economic evaluations in rare diseases. Patient and clinician involvement throughout model development is key to bridge such data gaps.