Control of confounding in the analysis phase – an overview for clinicians

Abstract

In observational studies, control of confounding can be done in the design and analysis phases. Using examples from large health care database studies, this article provides the clinicians with an overview of standard methods in the analysis phase, such as stratification, standardization, multivariable regression analysis and propensity score (PS) methods, together with the more advanced high-dimensional propensity score (HD-PS) method. We describe the progression from simple stratification confined to the inclusion of a few potential confounders to complex modeling procedures such as the HD-PS approach by which hundreds of potential confounders are extracted from large health care databases. Stratification and standardization assist in the understanding of the data at a detailed level, while accounting for potential confounders. Incorporating several potential confounders in the analysis typically implies the choice between multivariable analysis and PS methods. Although PS methods have gained remarkable popularity in recent years, there is an ongoing discussion on the advantages and disadvantages of PS methods as compared to those of multivariable analysis. Furthermore, the HD-PS method, despite its generous inclusion of potential confounders, is also associated with potential pitfalls. All methods are dependent on the assumption of no unknown, unmeasured and residual confounding and suffer from the difficulty of identifying true confounders. Even in large health care databases, insufficient or poor data may contribute to these challenges. The trend in data collection is to compile more fine-grained data on lifestyle and severity of diseases, based on self-reporting and modern technologies. This will surely improve our ability to incorporate relevant confounders or their proxies. However, despite a remarkable development of methods that account for confounding and new data opportunities, confounding will remain a serious issue. Considering the advantages and disadvantages of different methods, we emphasize the importance of the clinical input and of the interplay between clinicians and analysts to ensure a proper analysis.

Effect of a prior cesarean delivery on pregnancy outcomes of frozen‐thawed embryo transfer: A retrospective cohort study in a freeze‐all setting

“…Based on the complete and prospectively collected data, the robustness of our results was further examined and validated with the use of propensity score matching and multiple regression analysis to control for a wide variety of relevant confounding factors. 27,31 The underlying mechanism for the inferior FET pregnancy outcomes after CD could possibly be attributed to the presence of cesarean scar defect. This defect has been reported to be 42%-61% after one CD, and occurs in 100% of patients with more than three CDs.…”

Use of Propensity Score Methodology in Contemporary High-Impact Surgical Literature

“…One of the most frequently used methods to control for confounding is multivariable regression analysis, which allows researchers to adjust for many confounding variables in 1 model. 7 Multivariable regression analysis allows for the estimation of the effect of a given exposure variable on a given outcome variable while controlling for the confounding effect of other variables. It can be used for both continuous and dichotomous variables.…”

ART use and associated factors among HIV positive caregivers of orphans and vulnerable children in Tanzania

“…These approaches (i.e. multivariable analysis and stratification) played a roles as strategies to address confounding [44][45][46][47] among others. Also, stratification by sex provides a deeper understanding of how correlates of ART use differ by gender and inform appropriate interventions that are responsive to the unique needs of each gender [48][49][50].…”