Comparing Machine Learning and Advanced Methods with Traditional Methods to Generate Weights in Inverse Probability of Treatment Weighting: The INFORM Study

Aug 2, 2026

Upload by Aria

Observational research helps evaluate treatments in real-world patient populations, but confounding can bias comparisons when baseline characteristics differ between treatment groups. Inverse probability of treatment weighting (IPTW) using propensity scores can reduce confounding, yet evidence on generating propensity scores with machine learning and entropy balancing remains limited. This study assessed feasibility by applying advanced models to create weights and using a weighted Cox model to estimate the sample average treatment effect. Entropy balancing weights achieved near-perfect covariate balance and comparable results to traditional logistic regression.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Purpose

Methods

Results

Conclusion

Introduction

Pragmatic and Observational Research downloaded from [https://www.dovepress.com/](https://www.dovepress.com/) For personal use only. Pragmatic and Observational Research Dovepress open access to scientific and medical research Open Access Full Text Article METHOD Comparing Machine Learning and Advanced Methods with Traditional Methods to Generate Weights in Inverse Probability of Treatment Weighting: The INFORM Study Doyoung Kwak 1 , Yuanjie Liang2 , Xu Shi 3 , Xi Tan 2 1Department of Electrical & Computer Engineering, Texas A&M University, College Station, TX, USA; 2Novo Nordisk Inc, Plainsboro, NJ, USA; 3Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA Correspondence: Xi Tan, [Email mxtz@novonordisk.com](Email mxtz@novonordisk.com) | Purpose: Observational research provides valuable insights into treatments used in patient populations in real-world settings. However, confounding is likely to occur if there are differences in patient characteristics associated with both the exposure and outcome between the groups being evaluated. One approach to reduce confounding and facilitate unbiased comparisons is inverse probability of treatment weighting (IPTW) using propensity scores. Machine learning (ML) and entropy balancing can potentially be used in generating propensity scores for IPTW, but there is limited literature on this application. We aimed to assess the feasibility of applying these methods for reducing confounding in observational studies. These methods were assessed in a study comparing cardiovascular outcomes in adults with type 2 diabetes and established atherosclerotic cardiovascular disease taking once-weekly glucagon-like peptide-1 receptor agonists or dipeptidyl peptidase-4 inhibitors.
Methods: We applied advanced methods to generate the propensity scores compared to the original logistic regression method in terms of covariate balance. After calculating weights, a weighted Cox proportional hazards model was used to calculate the sample average treatment effect. Support Vector Classification, Support Vector Regression, XGBoost, and LightGBM were the ML models used. Entropy balancing was also performed on features identified in the original cardiovascular outcomes study.
Results: Accuracy (range: 0.71 to 0.73), area under the curve (0.77 to 0.79), precision (0.53 to 0.60), recall (0.66 to 0.68), and F1 score (0.60 to 0.64) were similar between all of the advanced propensity score methods and traditional logistic regression. Among ML models, only XGBoost achieved balance in all measured baseline characteristics between the two treatment groups, closely approximating the performance of the original logistic regression. Entropy balancing weights provided the best performance among all models in balancing baseline characteristics, achieving near perfect balancing.
Conclusion: Among the advanced methods examined, entropy balancing weights performed the best for optimizing balancing and can produce similar results compared to traditional logistic regression.
Keywords: propensity score, machine learning, entropy balancing, type 2 diabetes, glucagon-like peptide-1 receptor agonists | | --- | | Introduction
Observational research offers valuable insights into patient populations exposed to different treatments in real-world settings. However, unlike in randomized clinical trials where confounding of intervention and control groups is reduced or eliminated, in real-world studies, differences inpatient characteristics associated with both the exposure and outcome between the groups being evaluated can potentially lead to underestimating or overestimating the true effect of the exposure on the outcome being measured.1,2 Commonly used methods for reducing confounding in observational studies include methods based on propensity scores such as propensity score matching and inverse probability of treatment weighting (IPTW).1,3–5
IPTW is an approach to controlling for co

FAQs

Why can observational studies produce biased treatment comparisons?

Differences in patient characteristics related to both exposure and outcome can lead to confounding, which may under- or over-estimate true treatment effects.

How does IPTW use propensity scores to address confounding?

Propensity scores estimate an individual’s probability of receiving treatment based on characteristics, and IPTW uses weights derived from these scores to create a pseudo-population with more balanced groups.

Which advanced approach performed best for covariate balance in the study?

Entropy balancing weights provided the best performance, achieving near-perfect balancing and producing results comparable to traditional logistic regression.

Aug 2, 2026

Upload by Aria