noemi kreifChange photo
    Add Contact Information
    Regression, propensity score (PS) and double-robust (DR) methods can reduce selection bias when estimating average treatment effects (ATEs). Economic evaluations of health care interventions exemplify complex data structures, in that the... more
    Regression, propensity score (PS) and double-robust (DR) methods can reduce selection bias when estimating average treatment effects (ATEs). Economic evaluations of health care interventions exemplify complex data structures, in that the covariate–endpoint relationships tend to be highly non-linear, with highly skewed cost and health outcome endpoints. When either the regression or PS model is correct, DR methods can provide unbiased, efficient estimates of ATEs, but generally the specification of both models is unknown. Regression-adjusted matching can also protect against bias from model misspecification, but has not been compared to DR methods. This paper compares regression-adjusted matching to selected DR methods (weighted regression and augmented inverse probability of treatment weighting) as well as to regression and PS methods for addressing selection bias in cost-effectiveness analyses (CEA). We contrast the methods in a CEA of a pharmaceutical intervention, where there are extreme estimated PSs, hence unstable inverse probability of treatment (IPT) weights. The case study motivates a simulation which considers settings with functional form misspecification in the PS and endpoint regression models (e.g. cost model with log instead of identity link), stable and unstable PS weights. We find that in the realistic setting of unstable IPT weights and misspecifications to the PS and regression models, regression-adjusted matching reports less bias than DR methods. We conclude that regression-adjusted matching is a relatively robust method for estimating ATEs in applications with complex data structures exemplified by CEA.
    Research Interests:
    Upload File
    Statistical approaches for estimating treatment effectiveness commonly model the endpoint, or the propensity score, using parametric regressions such as generalised linear models. Misspecification of these models can lead to biased... more
    Statistical approaches for estimating treatment effectiveness commonly model the endpoint, or the propensity score, using parametric regressions such as generalised linear models. Misspecification of these models can lead to biased parameter estimates. We compare two approaches that combine the propensity score and the endpoint regression, and can make weaker modelling assumptions, by using machine learning approaches to estimate the regression function and the propensity score. Targeted maximum likelihood estimation is a double-robust method designed to reduce bias in the estimate of the parameter of interest. Bias-corrected matching reduces bias due to covariate imbalance between matched pairs by using regression predictions. We illustrate the methods in an evaluation of different types of hip prosthesis on the health-related quality of life of patients with osteoarthritis. We undertake a simulation study, grounded in the case study, to compare the relative bias, efficiency and confidence interval coverage of the methods. We consider data generating processes with non-linear functional form relationships, normal and non-normal endpoints. We find that across the circumstances considered, bias-corrected matching generally reported less bias, but higher variance than targeted maximum likelihood estimation. When either targeted maximum likelihood estimation or bias-corrected matching incorporated machine learning, bias was much reduced, compared to using misspecified parametric models.
    Research Interests:
    Upload File
    Many cost-effectiveness analyses (CEAs) use data from observational studies. Statistical methods can only address selection bias if they make plausible assumptions. No quality assessment tool is available for appraising CEAs that use... more
    Many cost-effectiveness analyses (CEAs) use data from observational studies. Statistical methods can only address selection bias if they make plausible assumptions. No quality assessment tool is available for appraising CEAs that use observational studies. We developed a new checklist to assess statistical methods for addressing selection bias in CEAs that use observational data.

    The checklist criteria were informed by a conceptual review and applied in a systematic review of economic evaluations. Criteria included whether the study assessed the ‘no unobserved confounding’ assumption, overlap of baseline covariates between the treatment groups and the specification of the regression models. The checklist also considered structural uncertainty from the choice of statistical approach.

    We found 81 studies that met the inclusion criteria: studies tended to use regression (51%), matching on individual covariates (25%) or matching on the propensity score (22%). Most studies (77%) did not assess the ‘no observed confounding’ assumption, and few studies (16%) fully considered structural uncertainty from the choice of statistical approach.

    We conclude that published CEAs do not assess the main assumptions behind statistical methods for addressing selection bias. This checklist can raise awareness about the assumptions behind statistical methods for addressing selection bias and can complement existing method guidelines for CEAs.
    Upload File
    Upload File
    Upload File
    Objectives: To identify the pattern of the risk of death over long-term in unresectable hepatocellular carcinoma by determining the appropriate distribution to extrapolate overall survival and to assess the role of the Weibull... more
    Objectives: To identify the pattern of the risk of death over long-term in unresectable hepatocellular carcinoma by determining the appropriate distribution to extrapolate overall survival and to assess the role of the Weibull distribution as the standard survival model in oncology. Research Design and Methods: To select the appropriate distribution three types of data sources have been analyzed. Patient level data from two randomized controlled trials and published Kaplan-Meier curves from a systematic literature review provided short term follow -up data. They were supplemented with patient level data, with long term follow-up from the National Cancer Institute, New South Wales. Published Kaplan-Meier curves were read in and a time-to event dataset was created. Distributions were fitted to the data from the different sources separately. Their fit was assessed visually and compared using statistical criteria based on log-likelihood, the Akaike information criterion (AIC), and the Bayesian information criterion (BIC). Results: Based on both published and patient-level, and both short- and long-term follow-up data, the Weibull distribution, used very often in cost-effectiveness models in oncology, does not seem to offer a good fit in hepatocellular carcinoma among the different survival models. The best fitting distribution appears to be the lognormal, with loglogistic as the second-best fitting function. Results were consistent between the different sources of data. Conclusions: In unresectable hepatocellular carcinoma, the Weibull model, which is often treated at the gold standard, does not appear to be appropriate based on different sources of data (two clinical trials, a retrospective database and published Kaplan-Meier curves). Lognormal distribution seems to be the most appropriate distribution for extrapolating overall survival.
    OBJECTIVE: The objective of this study was to assess the quality of life and drug costs associated with switching from any ongoing antipsychotic treatment to once-daily extended release quetiapine fumarate (quetiapine XR) in patients with... more
    OBJECTIVE: The objective of this study was to assess the quality of life and drug costs associated with switching from any ongoing antipsychotic treatment to once-daily extended release quetiapine fumarate (quetiapine XR) in patients with schizophrenia.METHODS: This assessment was based on data collected during a 12-week study in patients with schizophrenia (n = 477) who switched from their current antipsychotic due to insufficient efficacy or poor tolerability to a flexible dose of quetiapine XR. Patients were assigned utilities based on their Positive and Negative Syndrome Scale (PANSS) scores and the presence of adverse events by applying the methods of Lenert et al.1. Quality adjusted life year (QALY) gains were calculated assuming a linear change of utility between two consecutive visits. Incremental costs were calculated by comparing the hypothetical mean drug cost (assuming patients stay on previous treatment) with the actual mean cost of quetiapine XR based on European prices.RESULTS: Patients who completed the study (n = 279) increased their average utility by 0.116, corresponding to a QALY gain of 0.0207. For the total sample, the mean utility increased by 0.09, reflecting a QALY gain of 0.0170. The additional costs for quetiapine XR per QALY gained varied from approximately 16,000 euro to 24,000 euro. Notably, this is a non-comparative study; therefore, no conclusions can be reached regarding the relative impact of switching to quetiapine XR compared with other antipsychotics. Further limitations included the short trial duration on which the utility estimates are based, and the restriction of cost data to drug costs alone. Furthermore, in a 'real world' scenario, it is to be expected that other drug regimens might be introduced during periods of treatment failure.CONCLUSION: This analysis demonstrates that patients with schizophrenia who switch their antipsychotic medication to quetiapine XR because of insufficient efficacy or poor tolerability benefit from significant QALY gains at a reasonable increase in drug cost.
    BACKGROUND: In the 'Arimidex', Tamoxifen Alone or in Combination (ATAC) trial, the aromatase inhibitor (AI) anastrozole had a significantly better efficacy and safety profile than tamoxifen as initial adjuvant therapy for hormone... more
    BACKGROUND: In the 'Arimidex', Tamoxifen Alone or in Combination (ATAC) trial, the aromatase inhibitor (AI) anastrozole had a significantly better efficacy and safety profile than tamoxifen as initial adjuvant therapy for hormone receptor-positive (HR+) early breast cancer (EBC) in postmenopausal patients. To compare the combined long-term clinical and economic benefits, we carried out a cost-effectiveness analysis (CEA) of anastrozole versus tamoxifen based on the data of the 100month analysis of the ATAC trial from the perspective of the German public health insurance.PATIENTS AND METHODS: A Markov model with a 25-year time horizon was developed using the 100-month analysis of the ATAC trial as well as data obtained from published literature and expert opinion.RESULTS: Adjuvant treatment of EBC with anastrozole achieved an additional 0.32 quality-adjusted life-years (QALYs) gained per patient compared with tamoxifen, at an additional cost of D 6819 per patient. Thus, the incremental cost effectiveness of anastrozole versus tamoxifen at 25 years was D 21,069 ($30,717) per QALY gained.CONCLUSIONS: This is the first CEA of an AI that is based on extended follow-up data, taking into account the carryover effect of anastrozole, which maintains the efficacy benefits beyond therapy completion after 5 years. Adjuvant treatment with anastrozole for postmenopausal women with HR+ EBC is a cost-effective alternative to tamoxifen.Copyright 2010 S. Karger AG, Basel.
    Introduction Sunitinib, an oral, multitargeted receptor tyrosine kinase inhibitor, delays disease progression, with a median overall survival (OS) of more than 2 years, improves quality of life and is becoming the first-line standard of... more
    Introduction Sunitinib, an oral, multitargeted receptor tyrosine kinase inhibitor, delays disease progression, with a median overall survival (OS) of more than 2 years, improves quality of life and is becoming the first-line standard of care for metastatic renal carcinoma (mRCC). Purpose To assess the economic value of sunitinib as first-line therapy in mRCC within the Spanish healthcare system. Methods An adapted Markov model with a 10-year time horizon was used to analyse the cost effectiveness of sunitinib vs. sorafenib (SFN) and bevacizumab/interferon-α (BEV/IFN) as first-line mRCC therapy from the Spanish third-party payer perspective. Progression-free survival (PFS) and OS data from sunitinib, SFN and BEV/IFN pivotal trials were extrapolated to project survival and costs in 6-week cycles. Results, in progression-free life-years (PFLY), life years (LY) and quality-adjusted life-years (QALY) gained, expressed as incremental cost-effectiveness ratios (ICER) with costs and benefits discounted annually at 3%, were obtained using deterministic and probabilistic analyses. Results Sunitinib was more effective and less costly than both SFN (gains of 0.52 PFLY, 0.16 LY, 0.17 QALY) and BEV/IFN (gains of 0.19 PFLY, 0.23 LY, 0.16 QALY) with average cost savings/patients of €,124 and €23,218, respectively. Using a willingness-to-pay (WTP) threshold of €50,000/QALY, sunitinib achieved an incremental net benefit (INB) of €9,717 and €31,211 compared with SFN and BEV/IFN, respectively. At this WTP, the probability of sunitinib providing the highest INB was 75%. Conclusion Our analysis suggests that sunitinib is a cost-effective alternative to other targeted therapies as first-line mRCC therapy in the Spanish healthcare setting.
    Objectives Endometriosis presents with significant pain as the most common symptom. Generic health measures can allow comparisons across diseases or populations. However, the Medical Outcomes Study Short Form 36 (SF-36) has not been... more
    Objectives Endometriosis presents with significant pain as the most common symptom. Generic health measures can allow comparisons across diseases or populations. However, the Medical Outcomes Study Short Form 36 (SF-36) has not been validated for this disease. The goal of this study was to validate the SF-36 (version 2) for endometriosis. Methods Using data from two clinical trials (N = 252 and 198) of treatment for endometriosis, a full complement of psychometric analyses was performed. Additional instruments included a pain visual analog scale (VAS); a physician-completed questionnaire based on patient interview (modified Biberoglu and Behrman—B&B); clinical global impression of change (CGI-C); and patient satisfaction with treatment. Results Bodily pain (BP) and the Physical Component Summary Score (PCS) were correlated with the pain VAS at baseline and over time and the B&B at baseline and end of study. In addition, those who had the greatest change in BP and PCS also reported the greatest change on CGI-C and patient satisfaction with treatment. Other subscales showed smaller, but significant, correlations with change in the pain VAS, CGI-C, and patient satisfaction with treatment. Conclusions The SF-36—particularly BP and the PCS—appears to be a valid and responsive measure for endometriosis and its treatment.

    Join noemi and 21,197,470 other researchers on Academia.edu

    not now
    30 39 34 33 36 38 31 35 40 32
    Academia © 2015