PATIENT-LEVEL SIMULATIONS IN NICE TECHNOLOGY APPRAISALS: HOW OFTEN ARE THEY USED AND WHY ARE THEY CHOSEN?
Author(s)
Alasdair D. Henry, PhD1, Kristen E. Downs, MSPH2, Clare Willis, PhD1, Agota Szende, MSc, PhD1.
1Genesis Research Group, London, United Kingdom, 2Genesis Research Group, Hoboken, NJ, USA.
1Genesis Research Group, London, United Kingdom, 2Genesis Research Group, Hoboken, NJ, USA.
OBJECTIVES: Economic models for Health Technology Assessments must balance transparency and data availability. When companies anticipate scrutiny from decision-makers, they may default to traditional cohort modelling approaches, such as partitioned survival analyses or Markov models. However, patient-level simulations (PLSs), such as microsimulations and discrete event simulations, that track individual patients based on their underlying characteristics and clinical history could offer more flexibility and address shortcomings of traditional approaches. This study describes the utilisation of PLSs in NICE technology appraisals (TAs) and Highly Specialised Technology evaluations (HSTs).
METHODS: A search of NICE guidance documents was conducted. Single TAs, Multiple TAs, and HSTs describing PLSs and published between January 2022 and June 2026 were included. Evaluations were excluded from the analysis if they were withdrawn, terminated, or had been updated and replaced.
RESULTS: A total of 341 TAs and HSTs were reviewed, of which only 16 (4.7%) included a PLS. There were 13 microsimulations, 2 discrete event simulations, and 1 “probabilistic discrete-time event microsimulation model.” PLSs spanned 14 unique conditions, with osteoporosis and Pompe disease each represented by 2. Two PLSs were developed by the External Assessment Group (EAG) to address shortcomings in the company’s models, and 2 company-developed models required supplementary work following EAG feedback. Of the remaining 12 company-initiated models, all were deemed appropriate for decision-making. The most common rationales for using PLSs were complex disease pathways and patient heterogeneity. Importantly, all technologies were recommended by NICE, with 13 of 16 receiving optimised recommendations.
CONCLUSIONS: PLSs may represent an underutilised tool in NICE evaluations, with fewer than 1 in 20 TAs/HSTs utilising this approach. In disease areas with heterogeneous patient profiles and clinical outcomes dependent on patient history, PLSs could provide a more comprehensive assessment of the cost-effectiveness of new therapies.
METHODS: A search of NICE guidance documents was conducted. Single TAs, Multiple TAs, and HSTs describing PLSs and published between January 2022 and June 2026 were included. Evaluations were excluded from the analysis if they were withdrawn, terminated, or had been updated and replaced.
RESULTS: A total of 341 TAs and HSTs were reviewed, of which only 16 (4.7%) included a PLS. There were 13 microsimulations, 2 discrete event simulations, and 1 “probabilistic discrete-time event microsimulation model.” PLSs spanned 14 unique conditions, with osteoporosis and Pompe disease each represented by 2. Two PLSs were developed by the External Assessment Group (EAG) to address shortcomings in the company’s models, and 2 company-developed models required supplementary work following EAG feedback. Of the remaining 12 company-initiated models, all were deemed appropriate for decision-making. The most common rationales for using PLSs were complex disease pathways and patient heterogeneity. Importantly, all technologies were recommended by NICE, with 13 of 16 receiving optimised recommendations.
CONCLUSIONS: PLSs may represent an underutilised tool in NICE evaluations, with fewer than 1 in 20 TAs/HSTs utilising this approach. In disease areas with heterogeneous patient profiles and clinical outcomes dependent on patient history, PLSs could provide a more comprehensive assessment of the cost-effectiveness of new therapies.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
HTA371
Topic
Health Technology Assessment, Methodological & Statistical Research
Topic Subcategory
Decision & Deliberative Processes, Systems & Structure
Disease
No Additional Disease & Conditions/Specialized Treatment Areas