scholarly journals Evaluating the performance of propensity score matching based approaches in individual patient data meta-analysis

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Fatema Tuj Johara ◽  
Andrea Benedetti ◽  
Robert Platt ◽  
Dick Menzies ◽  
Piret Viiklepp ◽  
...  
Keyword(s):  
Propensity Score ◽  
Propensity Score Matching ◽  
Propensity Scores ◽  
Individual Patient Data ◽  
Meta Analysis ◽  
Random Effect ◽  
Multidrug Resistant ◽  
Patient Data ◽  
Propensity Score Model ◽  
Score Model

Abstract Background Individual-patient data meta-analysis (IPD-MA) is an increasingly popular approach because of its analytical benefits. IPD-MA of observational studies must overcome the problem of confounding, otherwise biased estimates of treatment effect may be obtained. One approach to reducing confounding bias could be the use of propensity score matching (PSM). IPD-MA can be considered as two-stage clustered data (patients within studies) and propensity score matching can be implemented within studies, across studies, and combining both. Methods This article focuses on implementation of four PSM-based approaches for the analysis of data structure that exploit IPD-MA in two ways: (i) estimation of propensity score model using single-level or random-effects logistic regression; and (ii) matching of propensity scores (PS) across studies, within studies or preferential-within studies. We investigated the performance of these approaches through a simulation study, which considers an IPD-MA that examined the success of different treatments for multidrug-resistant tuberculosis (MDR-TB). The simulation parameters were varied according to three treatment prevalences (according to studies, 50% and 30%), three levels of heterogeneity between studies (low, moderate and high) and three levels of pooled odds ratio (1, 1.5, 3). Results All approaches showed greater biases at the higher levels of heterogeneity regardless of the choices of treatment prevalences. However, matching of propensity scores using within-study and preferential-within study reported better performance compared to matching across studies when treatment prevalence varied across-studies. For fixed prevalences, a random-effect propensity score model to estimate propensity scores followed by matching of propensity scores across-studies achieved lower biases compared to other PSM-based approaches. Conclusions Propensity score matching has wide application in health research while only limited literature is available on the implementation of PSM methods in IPD-MA, and until now methodological performance of PSM methods have not been examined. We believe, this work offers an intuition to the applied researcher for the choice of the PSM-based approaches.

scholarly journals Causal Subclassification Tree Algorithm and Robust Causal Effect Estimation via Subclassification

2020 ◽  
Vol 10 (1) ◽  
pp. 40
Author(s):  
Tomoshige Nakamura ◽  
Mihoko Minami
Keyword(s):  
Propensity Score ◽  
Propensity Scores ◽  
Causal Effect ◽  
Model Misspecification ◽  
Robust Estimator ◽  
Propensity Score Model ◽  
Confounding Variables ◽  
Score Model ◽  
Effect Estimation ◽  
Weighting Methods

In observational studies, the existence of confounding variables should be attended to, and propensity score weighting methods are often used to eliminate their e ects. Although many causal estimators have been proposed based on propensity scores, these estimators generally assume that the propensity scores are properly estimated. However, researchers have found that even a slight misspecification of the propensity score model can result in a bias of estimated treatment effects. Model misspecification problems may occur in practice, and hence, using a robust estimator for causal effect is recommended. One such estimator is a subclassification estimator. Wang, Zhang, Richardson, & Zhou (2020) presented the conditions necessary for subclassification estimators to have $\sqrt{N}$-consistency and to be asymptotically well-defined and suggested an idea how to construct subclasses.

Hepatocellular carcinoma recurrence after direct-acting antiviral therapy: an individual patient data meta-analysis

Gut ◽  
2021 ◽  
pp. gutjnl-2020-323663
Author(s):  
Victor Sapena ◽  
Marco Enea ◽  
Ferran Torres ◽  
Ciro Celsa ◽  
Jose Rios ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Propensity Score ◽  
Individual Patient Data ◽  
Meta Analysis ◽  
Recurrence Risk ◽  
Patient Data ◽  
Death Risk ◽  
Significant Difference ◽  
Direct Acting ◽  
Hcc Recurrence

ObjectiveThe benefit of direct-acting antivirals (DAAs) against HCV following successful treatment of hepatocellular carcinoma (HCC) remains controversial. This meta-analysis of individual patient data assessed HCC recurrence risk following DAA administration.DesignWe pooled the data of 977 consecutive patients from 21 studies of HCV-related cirrhosis and HCC, who achieved complete radiological response after surgical/locoregional treatments and received DAAs (DAA group). Recurrence or death risk was expressed as HCC recurrence or death per 100 person-years (100PY). Propensity score-matched patients from the ITA.LI.CA. cohort (n=328) served as DAA-unexposed controls (no-DAA group). Risk factors for HCC recurrence were identified using random-effects Poisson.ResultsRecurrence rate and death risk per 100PY in DAA-treated patients were 20 (95% CI 13.9 to 29.8, I2=74.6%) and 5.7 (2.5 to 15.3, I2=54.3), respectively. Predictive factors for recurrence were alpha-fetoprotein logarithm (relative risk (RR)=1.11, 95% CI 1.03 to 1.19; p=0.01, per 1 log of ng/mL), HCC recurrence history pre-DAA initiation (RR=1.11, 95% CI 1.07 to 1.16; p<0.001), performance status (2 vs 0, RR=4.35, 95% CI 1.54 to 11.11; 2 vs 1, RR=3.7, 95% CI 1.3 to 11.11; p=0.01) and tumour burden pre-HCC treatment (multifocal vs solitary nodule, RR=1.75, 95% CI 1.25 to 2.43; p<0.001). No significant difference was observed in RR between the DAA-exposed and DAA-unexposed groups in propensity score-matched patients (RR=0.64, 95% CI 0.37 to 1.1; p=0.1).ConclusionEffects of DAA exposure on HCC recurrence risk remain inconclusive. Active clinical and radiological follow-up of patients with HCC after HCV eradication with DAA is justified.

scholarly journals Treatment and outcomes in children with multidrug-resistant tuberculosis: A systematic review and individual patient data meta-analysis

PLoS Medicine ◽  
2018 ◽  
Vol 15 (7) ◽  
pp. e1002591 ◽  
Author(s):  
Elizabeth P. Harausz ◽  
Anthony J. Garcia-Prats ◽  
Stephanie Law ◽  
H. Simon Schaaf ◽  
Tamara Kredo ◽  
...  
Keyword(s):  
Systematic Review ◽  
Individual Patient Data ◽  
Meta Analysis ◽  
Multidrug Resistant ◽  
Patient Data ◽  
Resistant Tuberculosis ◽  
Multidrug Resistant Tuberculosis

scholarly journals Is central obesity associated with diabetic retinopathy in Chinese individuals? An exploratory study

2019 ◽  
Vol 47 (11) ◽  
pp. 5601-5612
Author(s):  
Jian-Bo Zhou ◽  
Jing Yuan ◽  
Xing-Yao Tang ◽  
Wei Zhao ◽  
Fu-Qiang Luo ◽  
...  
Keyword(s):  
Diabetic Retinopathy ◽  
Propensity Score ◽  
Propensity Score Matching ◽  
Central Obesity ◽  
Propensity Scores ◽  
Meta Analysis ◽  
Before And After ◽  
Independent Association ◽  
Two Stages ◽  
The Relationship

Objective To our knowledge, the independent association between central obesity, defined by waist circumference (WC) or waist-to-hip ratio (WHR), and diabetic retinopathy (DR) remains unknown in Chinese individuals. Method The study was conducted in two stages. First, the relationship between WC or WHR and DR was estimated in a case-control set (DR vs. non-DR) for the whole population before and after propensity score matching. Subsequently, a systematic review and meta-analysis was performed on evidence from the literature to validate the relationship. Results Of 511 eligible patients, DR (N = 156) and non-DR (N = 156) patients with similar propensity scores were included in the propensity score matching analyses. Central obesity (defined by WC) was associated with risk of DR (odds ratio [OR] 1.07, 95% confidence interval [95% CI] (1.03–1.10). The meta-analysis showed that central obesity significantly increased the risk of DR by 12% (OR 1.12, 95% CI 1.02–1.22). Analysis of data from 18 studies showed a significant association between continuous body mass index and risk of proliferative DR (OR 0.95, 95% CI 0.93–0.98; I2 = 50%). Conclusion Central obesity, particularly as defined by WC, is associated with the risk of DR in the Chinese population.

scholarly journals Estimating treatment importance in multidrug‐resistant tuberculosis using Targeted Learning: An observational individual patient data network meta‐analysis

Biometrics ◽  
2020 ◽  
Vol 76 (3) ◽  
pp. 1007-1016
Author(s):  
Guanbo Wang ◽  
Mireille E. Schnitzer ◽  
Dick Menzies ◽  
Piret Viiklepp ◽  
Timothy H. Holtz ◽  
...  
Keyword(s):  
Individual Patient Data ◽  
Meta Analysis ◽  
Multidrug Resistant ◽  
Patient Data ◽  
Data Network ◽  
Resistant Tuberculosis ◽  
Multidrug Resistant Tuberculosis

Magnetic Resonance Imaging Improves Breast Screening Sensitivity in BRCA Mutation Carriers Age ≥ 50 Years: Evidence From an Individual Patient Data Meta-Analysis

2015 ◽  
Vol 33 (4) ◽  
pp. 349-356 ◽  
Author(s):  
Xuan-Anh Phi ◽  
Nehmat Houssami ◽  
Inge-Marie Obdeijn ◽  
Ellen Warner ◽  
Francesco Sardanelli ◽  
...  
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Breast Screening ◽  
Individual Patient Data ◽  
Meta Analysis ◽  
Random Effect ◽  
Patient Data ◽  
Resonance Imaging ◽  
Mutation Carriers ◽  
Screening Sensitivity

Purpose There is no consensus on whether magnetic resonance imaging (MRI) should be included in breast screening protocols for women with BRCA1/2 mutations age ≥ 50 years. Therefore, we investigated the evidence on age-related screening accuracy in women with BRCA1/2 mutations using individual patient data (IPD) meta-analysis. Patients and Methods IPD were pooled from six high-risk screening trials including women with BRCA1/2 mutations who had completed at least one screening round with both MRI and mammography. A generalized linear mixed model with repeated measurements and a random effect of studies estimated sensitivity and specificity of MRI, mammography, and the combination in all women and specifically in those age ≥ 50 years. Results Pooled analysis showed that in women age ≥ 50 years, screening sensitivity was not different from that in women age < 50 years, whereas screening specificity was. In women age ≥ 50 years, combining MRI and mammography significantly increased screening sensitivity compared with mammography alone (94.1%; 95% CI, 77.7% to 98.7% v 38.1%; 95% CI, 22.4% to 56.7%; P < .001). The combination was not significantly more sensitive than MRI alone (94.1%; 95% CI, 77.7% to 98.7% v 84.4%; 95% CI, 61.8% to 94.8%; P = .28). Combining MRI and mammography in women age ≥ 50 years resulted in sensitivity similar to that in women age < 50 years (94.1%; 95% CI, 77.7% to 98.7% v 93.2%; 95% CI, 79.3% to 98%; P = .79). Conclusion Addition of MRI to mammography for screening BRCA1/2 mutation carriers age ≥ 50 years improves screening sensitivity by a magnitude similar to that observed in younger women. Limiting screening MRI in BRCA1/2 carriers age ≥ 50 years should be reconsidered.

scholarly journals Chasing Balance and Other Recommendations for Improving Nonparametric Propensity Score Models

2017 ◽  
Vol 5 (2) ◽  
Author(s):  
Beth Ann Griffin ◽  
Daniel F. McCaffrey ◽  
Daniel Almirall ◽  
Lane F. Burgette ◽  
Claude Messan Setodji
Keyword(s):  
Propensity Score ◽  
Propensity Scores ◽  
Model Fit ◽  
Estimation Model ◽  
Propensity Score Model ◽  
Treatment Groups ◽  
Covariate Balance ◽  
Learning Technique ◽  
Accurate Treatment ◽  
Score Model

Abstract:In this article, we carefully examine two important implementation issues when estimating propensity scores using generalized boosted models (GBM), a promising machine learning technique. First, we examine which of the following methods for tuning GBM lead to better covariate balance and inferences about causal effects: pursuing covariate balance between the treatment groups or tuning the propensity score model on the basis of a model fit criterion. Second, we examine how well GBM can handle irrelevant covariates that are included in the estimation model. We find that chasing balance rather than model fit when estimating propensity scores yielded better covariate balance and more accurate treatment effect estimates. Additionally, we find that adding irrelevant covariates to GBM increased imbalance and bias in the treatment effects. The findings from this paper have useful implications for other work focused on improving methods for estimating propensity scores.

Sign in / Sign up

Export Citation Format

Share Document