Literature DB >> 26436519

Adjusting for Confounding in Early Postlaunch Settings: Going Beyond Logistic Regression Models.

Amand F Schmidt1, Olaf H Klungel, Rolf H H Groenwold.   

Abstract

BACKGROUND: Postlaunch data on medical treatments can be analyzed to explore adverse events or relative effectiveness in real-life settings. These analyses are often complicated by the number of potential confounders and the possibility of model misspecification.
METHODS: We conducted a simulation study to compare the performance of logistic regression, propensity score, disease risk score, and stabilized inverse probability weighting methods to adjust for confounding. Model misspecification was induced in the independent derivation dataset. We evaluated performance using relative bias confidence interval coverage of the true effect, among other metrics.
RESULTS: At low events per coefficient (1.0 and 0.5), the logistic regression estimates had a large relative bias (greater than -100%). Bias of the disease risk score estimates was at most 13.48% and 18.83%. For the propensity score model, this was 8.74% and >100%, respectively. At events per coefficient of 1.0 and 0.5, inverse probability weighting frequently failed or reduced to a crude regression, resulting in biases of -8.49% and 24.55%. Coverage of logistic regression estimates became less than the nominal level at events per coefficient ≤5. For the disease risk score, inverse probability weighting, and propensity score, coverage became less than nominal at events per coefficient ≤2.5, ≤1.0, and ≤1.0, respectively. Bias of misspecified disease risk score models was 16.55%.
CONCLUSION: In settings with low events/exposed subjects per coefficient, disease risk score methods can be useful alternatives to logistic regression models, especially when propensity score models cannot be used. Despite better performance of disease risk score methods than logistic regression and propensity score models in small events per coefficient settings, bias, and coverage still deviated from nominal.

Entities:  

Mesh:

Year:  2016        PMID: 26436519     DOI: 10.1097/EDE.0000000000000388

Source DB:  PubMed          Journal:  Epidemiology        ISSN: 1044-3983            Impact factor:   4.822


  4 in total

1.  Control of confounding in the analysis phase - an overview for clinicians.

Authors:  Johnny Kahlert; Sigrid Bjerge Gribsholt; Henrik Gammelager; Olaf M Dekkers; George Luta
Journal:  Clin Epidemiol       Date:  2017-03-31       Impact factor: 4.790

2.  Using Real-World Data in Health Technology Assessment (HTA) Practice: A Comparative Study of Five HTA Agencies.

Authors:  Amr Makady; Ard van Veelen; Páll Jonsson; Owen Moseley; Anne D'Andon; Anthonius de Boer; Hans Hillege; Olaf Klungel; Wim Goettsch
Journal:  Pharmacoeconomics       Date:  2018-03       Impact factor: 4.981

3.  Comparison of variance estimators for meta-analysis of instrumental variable estimates.

Authors:  A F Schmidt; A D Hingorani; B J Jefferis; J White; Rhh Groenwold; F Dudbridge
Journal:  Int J Epidemiol       Date:  2016-12-01       Impact factor: 7.196

Review 4.  Clinical Perspectives on Targeting Therapies for Personalized Medicine.

Authors:  Donald R J Singer; Zoulikha M Zaïr
Journal:  Adv Protein Chem Struct Biol       Date:  2015-12-29       Impact factor: 3.507

  4 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.