Pcn112 Research Trend of the Pharmacoeconomic Studies of Small-Molecule Protein Kinase Inhibitors in China

Abstract

identified published economic evaluations of cancer treatments using a model, and single technology appraisals (STAs) submitted to the National Institute for Health and Care Excellence (NICE). Searches were performed in November 2018, capturing relevant publications and STAs published since 2013 up to the date of searching. Published literature was identified using Medline and EMBASE via Ovid. The review was supplemented with methods literature known to the authors discussing cancer modelling. Results: The review identified 100 NICE STAs and 124 published studies relevant to the topic. Published studies appeared to report a greater use of discretetime state transition-models (n=102, 82%) when compared to NICE submissions. Partitioned-survival analysis (n=54, 54%) and discrete-time state-transition structures (n=41, 41%) were the main structures submitted to NICE. Justification of model structures and consideration of structural uncertainty were very limited across both the publications and the NICE submissions. Conclusions: There appears to be a stronger dominance of the partitioned-survival analysis approach in submissions to NICE, however, we believe many of the published state transition models have been incorrectly labelled and are partitioned-survival analysis. Other structures, such as a decision tree or discrete-event simulation, have also been utilised in submissions and within the published literature but only in minority of cases. Justification for the model structure was very limited, despite a recognition in the literature that model structure can greatly influence cost-effectiveness results and the validity of the economic evaluation and studies would be improved with a thorough rationale for this choice.