Assessing the impact of unmeasured confounders for credible and reliable real-world evidence
- PMID: 32929830
- DOI: 10.1002/pds.5117
Assessing the impact of unmeasured confounders for credible and reliable real-world evidence
Abstract
Purpose: We review statistical methods for assessing the possible impact of bias due to unmeasured confounding in real world data analysis and provide detailed recommendations for choosing among the methods.
Methods: By updating an earlier systematic review, we summarize modern statistical best practices for evaluating and correcting for potential bias due to unmeasured confounding in estimating causal treatment effect from non-interventional studies.
Results: We suggest a hierarchical structure for assessing unmeasured confounding. First, for initial sensitivity analyses, we strongly recommend applying a recently developed method, the E-value, that is straightforward to apply and does not require prior knowledge or assumptions about the unmeasured confounder(s). When some such knowledge is available, the E-value could be supplemented by the rule-out or array method at this step. If these initial analyses suggest results may not be robust to unmeasured confounding, subsequent analyses could be conducted using more specialized statistical methods, which we categorize based on whether they require access to external data on the suspected unmeasured confounder(s), internal data, or no data. Other factors for choosing the subsequent sensitivity analysis methods are also introduced and discussed, including the types of unmeasured confounders and whether the subsequent sensitivity analysis is intended to provide a corrected causal treatment effect.
Conclusion: Various analytical methods have been proposed to address unmeasured confounding, but little research has discussed a structured approach to select appropriate methods in practice. In providing practical suggestions for choosing appropriate initial and, potentially, more specialized subsequent sensitivity analyses, we hope to facilitate the widespread reporting of such sensitivity analyses in non-interventional studies. The suggested approach also has the potential to inform pre-specification of sensitivity analyses before executing the analysis, and therefore increase the transparency and limit selective study reporting.
Keywords: causal inference; pharmacoepidemiology; practical recommendation; real world evidence; sensitivity analyses; unmeasured confounding.
© 2020 John Wiley & Sons Ltd.
Similar articles
-
Addressing unmeasured confounding in comparative observational research.Pharmacoepidemiol Drug Saf. 2018 Apr;27(4):373-382. doi: 10.1002/pds.4394. Epub 2018 Jan 30. Pharmacoepidemiol Drug Saf. 2018. PMID: 29383840 Review.
-
The impact of unmeasured within- and between-cluster confounding on the bias of effect estimatorsof a continuous exposure.Stat Methods Med Res. 2020 Aug;29(8):2119-2139. doi: 10.1177/0962280219883323. Epub 2019 Nov 7. Stat Methods Med Res. 2020. PMID: 31694489
-
The impact of residual and unmeasured confounding in epidemiologic studies: a simulation study.Am J Epidemiol. 2007 Sep 15;166(6):646-55. doi: 10.1093/aje/kwm165. Epub 2007 Jul 5. Am J Epidemiol. 2007. PMID: 17615092
-
Sensitivity analysis for the effects of multiple unmeasured confounders.Ann Epidemiol. 2016 Sep;26(9):605-11. doi: 10.1016/j.annepidem.2016.07.009. Epub 2016 Aug 3. Ann Epidemiol. 2016. PMID: 27576907
-
Measured and accounted-for confounding in pharmacoepidemiologic studies: Some thoughts for practitioners.Pharmacoepidemiol Drug Saf. 2021 Mar;30(3):277-282. doi: 10.1002/pds.5189. Epub 2021 Jan 6. Pharmacoepidemiol Drug Saf. 2021. PMID: 33372303 Free PMC article. Review.
Cited by
-
Assessing Additive Interactions between Protective Factors Using Relative Risk Reduction Due to Interaction.Medicina (Kaunas). 2024 Jun 26;60(7):1053. doi: 10.3390/medicina60071053. Medicina (Kaunas). 2024. PMID: 39064482 Free PMC article.
-
Assessment of the E-value in the presence of bias amplification: a simulation study.BMC Med Res Methodol. 2024 Mar 28;24(1):79. doi: 10.1186/s12874-024-02196-4. BMC Med Res Methodol. 2024. PMID: 38539082 Free PMC article.
-
Strategies to Address Current Challenges in Real-World Evidence Generation in Japan.Drugs Real World Outcomes. 2023 Jun;10(2):167-176. doi: 10.1007/s40801-023-00371-5. Epub 2023 May 13. Drugs Real World Outcomes. 2023. PMID: 37178273 Free PMC article. Review.
-
The use and misuse of risk prediction tools for clinical decision-making.Breast. 2023 Jun;69:428-430. doi: 10.1016/j.breast.2023.01.006. Epub 2023 Jan 21. Breast. 2023. PMID: 36709092 Free PMC article. No abstract available.
-
The M-Value: A Simple Sensitivity Analysis for Bias Due to Missing Data in Treatment Effect Estimates.Am J Epidemiol. 2023 Apr 6;192(4):612-620. doi: 10.1093/aje/kwac207. Am J Epidemiol. 2023. PMID: 36469493 Free PMC article.
References
REFERENCES
-
- Attempt to replicate clinical trials with real-world data generates real-world criticism, too STAT News. July 3, 2019. Available from https://www.statnews.com/2019/07/03/replicate-clinical-trials-real-world...
-
- How real world evidence was used to support approval of Ibrance for male breast cancer. The Cancer Letter. April 19, 2019. Available from https://cancerletter.com/articles/20190419_2/
-
- European Medicines Agency. Guidance for companies considering the adaptive pathways approach. 2016. Available from https://www.ema.europa.eu/en/documents/regulatory-procedural-guideline/g.... Accessed December 1, 2019.
-
- U.S. Food and Drug Administration. Framework for FDA's Real-World Evidence Dent Program 2018. https://www.fda.gov/media/120060/download. Accessed December 1, 2019.
-
- Wedam S, Fashoyin-Aje L, Bloomquist E, et al. FDA approval summary: Palbociclib for male patients with metastatic breast cancer. Clin Cancer Res. 2019;26(5):1208-1212.
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources