Construction Of Microsimulation-Based Dynamic Decision Analysis Model For Rheumatoid Arthritis And Health Economic Evaluation

Objective:(1)This study aims to establish a mapping formula by constructing a mapping model between the Health Assessment Questionnaire-Disability Index(HAQ-DI)and health utility(Five-Level Euro Qol Five-Dimensional Questionnaire,EQ-5D-5L).This method facilitates the later health economic evaluation to calculate utilities for patients with rheumatoid arthritis(RA).(2)By constructing a microsimulation model of RA and conducting a health economic evaluation of the dynamic treatment sequences,this study identifies treatment strategies that could be cost-effective and affordable for both patients and the government.Methods:(1)The patients with RA over 18 years old were selected from a tertiary hospital in Hunan province from August 2020 to July 2021 by using convenience sampling.The general information survey,Chinese version of EQ-5D-5L,and Chinese version of HAQ-DI scale were used to investigate the relationship between HAQ-DI and health utility.Beta regression and multivariate Probit regression models were constructed based on the survey data.And then choose the most optimal model between them.Finally,the mapping formula between HAQ-DI score and health utility value based on the RA population in China was developed.(2)This study constructed a microsimulation model of RA through R software.Based on the mapping model and microsimulation model of RA,a health economic evaluation was conducted among different treatment strategies.Specifically,cost-effectiveness analyses compared conventional synthetic disease-modifying antirheumatic drugs(cs DMARDs)and biological DMARDs(b DMARDs)were performed for patients with RA after methotrexate(MTX)failure.The incremental cost,incremental quality-adjusted life years(QALY)and incremental costeffectiveness ratio(ICER)were compared among different treatment sequences to select the most cost-effective treatment strategy.One-way sensitivity analysis,probabilistic sensitivity analysis(PSA),and scenario analysis were performed to evaluate the uncertainty of input parameters and the robustness of the model.Results:(1)A total of 228 RA patients were included in the cross-sectional study.The mean value of HAQ-DI and pain Visual Analogue Scale(VAS)score of patients was 1.296(SD,0.822)and 54.78(SD,23.987),respectively,indicating moderate disability.The mean value of EQ-5D-5L and EQ-VAS score was 0.485(SD,0.343)and 63.27(SD,11.13).The results showed that the Beta regression model with both HAQ-DI score and pain VAS score as covariables was the optimal mapping model.By constructing Beta regression model,mapping formulas were generated.Therefore,EQ-5D-5L(utility value)could be estimated through HAQ-DI score and finally applied to RA health economic evaluation model.(2)Triple therapy(MTX + sulfasalazine + hydroxychloroquine)was associated with lower costs and QALYs than b DMARDs treatment sequences.The ICERs of those b DMARDs sequences ranging from598,308 to 714,700 RMB/QALY,higher than the WTP threshold in China(213,891 RMB/QALY).The baseline Disease Activity Score-28(DAS28)impacted the model outcomes the most.Scenario analyses indicated that adding triple therapy to b DMARDs treatment sequences as a first-,second-,third-,or fourth-line therapy is very cost-effective,at a WTP of 70,871RMB/QALY.Conclusions: From the Chinese payer perspective,compared with b DMARDs treatment sequences,triple therapy is estimated to be costeffective for patients with active RA,at a current WTP threshold of212,627 RMB/QALY.Furthermore,a very cost-effective level was obtained regardless of the position of triple therapy within the b DMARDs treatment sequence compared with sequences not including triple therapy.