Propensity Score Weighting Using Overlap Weights: A New Method Applied to Regorafenib Clinical Data and a Cost-Effectiveness Analysis

Abstract

Background: In situations of markedly different population characteristics and weak population overlap, inverse propensity score (PS) weights suffer from extreme values. The new propensity score weighting method using overlap weights (PSOW) overcomes this limitation by estimating the overlap population at the point of highest mutual overlap, thus may be preferred to other balancing methods (trimming, target, or inverse weights) in some situations.Objectives: To evaluate the performance of PSOW with regorafenib effectiveness data from previously treated patients with metastatic colorectal cancer based on the Czech national registry data (regorafenib) and a global phase 3 randomized clinical trial (RCT) (placebo). The second goal was to assess the cost-effectiveness of regorafenib versus placebo.Methods: Individual data on progression-free survival (PFS)/overall survival (OS) were balanced via PSOW for age, sex, Eastern Cooperative Oncology Group performance status, number of treatment lines, metastatic colorectal cancer location, KRAS mutation, and time from metastases estimated using logistic regression. The weighted Kaplan-Meier PFS/OS curves were used in a 3-state partitioned survival model. The R code is provided.Results: In comparison with target or inverse PS weights, PSOW showed remarkable performance measured by effective sample size and PS weight distribution or extreme weights despite the weak overlap between the registry and RCT. In the registry or RCT cohort, regorafenib provided better survival compared with the RCT. The new PSOW hazard ratio for OS was 0.53 (RCT: 0.79), which is conservative compared with inverse or target weights with a hazard ratio of 0.44 and 0.27, respectively. Conclusion:This is the first use of PSOW for clinical data and cost-effectiveness analysis. It is promising in cases of weak or small population overlap and makes pharmacoeconomic modeling, in such cases, feasible.

Effectiveness of surgical fixation for rib fractures in relation to its timing: a retrospective Japanese nationwide study

“…We used the "overlap weights" method to compare the groups because there were substantial differences between them and some extremely high or low propensity scores [18,19].…”

Improvement of LV Reverse Remodeling Using Dynamic Programming of Fusion-Optimized Atrioventricular Intervals in Cardiac Resynchronization Therapy

“…Multivariable regression models were used to explore the outcomes between BiV+SyncAV and BiV groups when adjusted for clinical covariates, QLV/QRS. Because it is a retrospective study, we also conducted analyses with two propensity score weighting methods to reduce the effects of confounders: inverse probability weighting (IPTW) and overlap weighting (OW) ( 20 , 21 ). The individual propensity score of BiV+SyncAV was estimated via a multivariable logistic regression model, including all characteristics.…”

“…Clinical symptoms, 6-min walk distances (6MWT), Minnesota Living with Heart Failure (MLHF) score, NYHA class, the threshold of pacing, electrode impedance, electrocardiographic (20,21). The individual propensity score of BiV+SyncAV was estimated via a multivariable logistic regression model, including all characteristics.…”

Hydroxychloroquine might reduce risk of incident endometriosis in patients with systemic lupus erythematosus: A retrospective population-based cohort study

“…In addition, we conducted further survey into the impact of medication on endometriosis and made sensitivity analysis through Propensity score overlap weighting and inverse propensity score weighting to reduce the bias (Table 3). 22 In Table 3, patients in the high dosage group using medication with immunosuppressive drugs and corticosteroid have lower risk of endometriosis development (HR, 0.267; 95% CI 0.072-0.997 and HR, 0.355; 95% CI 0.135-0.934 respectively).…”