RXRonghui Xu

Ronghui Xu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Ronghui Xu has a total of 122 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Evolving Treatment Patterns and Clinical Outcomes in Children With Newly Diagnosed Crohn's Disease and Ulcerative Colitis in the United States, 2007–2023

    by Michelle M. Gonzalez, Dhruv Ahuja, Yuchen Qi, Ronghui Xu, Vipul Jairath, Shane W. Goodwin, Ola Olen, William A. Faubion, Michael D. Kappelman, Siddharth Singh

    INTRODUCTION: We examined treatment patterns and clinical outcomes in a contemporary cohort of children with newly diagnosed inflammatory bowel disease (IBD) in a large, commercially insured US population. METHODS: Using the OptumLabs Data Warehouse, we created an inception cohort of physician-diagnosed IBD in children aged 6–17 years between 2007 and 23. We examined cumulative incidence of hospitalization, surgery, safety events, and medication use in patients with Crohn's disease (CD) and ulcerative colitis (UC), overall and stratified by diagnostic era, race and ethnicity, sex, and age at diagnosis. RESULTS: Among 787 children with newly diagnosed CD (age at diagnosis, 13.4 ± 2.9 year; 57% boys; 82% White, 5% Black, 7% Hispanic) followed over mean 4.6 ± 3.0 years, 5-year risk of IBD-related hospitalization and major surgery was 30% and 10%, respectively. In children diagnosed in 2019–23, 1-year cumulative use of corticosteroids and advanced therapy was 46% and 75%, respectively; advanced therapy use increased significantly over time (aHR, 3.43 [95% CI, 2.71–4.32] for 2019–23 vs 2007–12), whereas corticosteroid use declined (0.67 [0.53–0.84]). Risk of IBD-related hospitalization was lower in children diagnosed between 2019 and 23 (0.65 [0.43–0.97]) compared with those diagnosed between 2007–12; surgical risk did not decline significantly, with confidence intervals compatible with a clinically meaningful reduction. Among 399 children with newly diagnosed UC (age at diagnosis, 14.0 ± 3.0 year; 53% boys; 78% White, 5% Black, 7% Hispanic) followed over 4.5 ± 3.0 years, 5-year risk of IBD-related hospitalization and colectomy was 36% and 8%, respectively. Advanced therapy use increased significantly over time (4.46 [2.85–6.98] for 2019–23 vs 2007–12), with no significant change in hospitalization or surgical risk across eras. No consistent racial and ethnic, age, or sex differences in advanced therapy use or clinical outcomes were identified in either CD or UC. DISCUSSION: In a contemporary cohort of commercially insured children with newly diagnosed IBD, early advanced therapy use has increased markedly over time. Surgical risk did not decline significantly in either disease; however, the study was not powered to exclude a clinically meaningful reduction, and these findings are hypothesis-generating rather than conclusive.

  • Comparative Safety of Advanced Therapies in Patients With Ulcerative Colitis: An Administrative Claims-Based Study

    by Dhruv Ahuja, Claudia Dziegielewski, Kuan-Hung Yeh, Sagar B. Patel, Shane W. Goodwin, Christopher Ma, Ashwin N. Ananthakrishnan, Namrata Singh, Vipul Jairath, Ronghui Xu, et al.

    INTRODUCTION: We conducted a retrospective cohort study comparing the safety of advanced therapies in patients with ulcerative colitis (UC). METHODS: Using an administrative claims database (OptumLabs Data Warehouse), we identified patients with UC who initiated treatment with tumor necrosis factor-α (TNF) antagonists, anti-integrin agents, anti-interleukin, Janus kinase inhibitors (JAK), or sphingosine-1 phosphate receptor modulators between 2016 and 2022 and had followed up for at least 1 year before and after treatment initiation. We compared the risk of serious infections, venous thromboembolism, and major adverse cardiovascular events through multinomial propensity score-based inverse probability weighting, with propensity scores estimated through generalized boosted models that accounted for disease characteristics, healthcare utilization, comorbidities, and previous and concomitant medications, and the competing risk of mortality. We calculated cause-specific hazard ratios (HR) and 95% confidence intervals for multiple treatment comparisons. RESULTS: We included 9,430 patients with UC treated with TNF antagonists (n = 4,111), anti-integrins (n = 3,165), anti-ILs (n = 1,342), JAK inhibitors (n = 701), or sphingosine-1 phosphate receptor modulators (n = 111), followed over median 27 months. After adjusting for confounding variables, anti-ILs were associated with a lower risk of serious infections compared with TNF antagonists (HR, 0.66 [95% confidence interval, 0.51–0.87]) and anti-integrins (HR, 0.75 [0.57–0.98]). JAK inhibitors were associated with a lower risk of serious infections compared with TNF antagonists (HR, 0.66 [0.46–0.94]). The incidence of venous thromboembolism (incidence rate, 1.4–2.0 per 100 py) and major adverse cardiovascular event (0.5–1.0 per 100 py) was very low, without any significant differences across agents. DISCUSSION: In a real-world cohort of patients with UC, anti-ILs and JAK inhibitors were associated with a lower risk of serious infections and may offer net benefit, especially in patients with previous TNF antagonist exposure.


  • Open Access

    Birth Outcomes in Women Who Have Taken Vedolizumab in Pregnancy: Results From the Vedolizumab Pregnancy Exposure Registry

    by Christina D. Chambers, Diana L. Johnson, Yunjun Luo, Ronghui Xu, Margaret P. Adam, Stephen R. Braddock, Kenneth Lyons Jones

    OBJECTIVES: There are limited data on the safety of vedolizumab in pregnancy for the treatment of Crohn's disease or ulcerative colitis. Between 2015 and 2022, the Organization of Teratology Information Specialists conducted a prospective, observational pregnancy registry study with 275 pregnant women residing in the United States or Canada. METHODS: Women were enrolled in 1 of 3 cohorts: vedolizumab-exposed (N = 99); disease-matched unexposed to vedolizumab, but treated with another biologic in pregnancy (N = 76); or unexposed with no chronic health conditions (N = 100). Women and their infants were followed up to 1 year postpartum with maternal interviews, questionnaires, medical records abstraction, and a subset of infants who received a physical examination. Study outcomes were major structural birth defects, minor birth defects, pregnancy loss, preterm delivery, prenatal and postnatal growth deficiency, serious or opportunistic infections, malignancies, and developmental milestones. RESULTS: In the overall registry, 17 of 275 pregnancies (6.2%) were lost to follow-up. Among pregnancies ending in at least 1 live born infant, 7 of 94 (7.4%) in the vedolizumab-exposed cohort compared with 4 of 71 (5.6%) in the disease-matched cohort had a major birth defect (adjusted risk ratio 1.07, 95% confidence interval [CI] 0.33, 3.52). Compared with the disease-matched cohort, women in the vedolizumab-exposed group were not statistically significantly more likely to experience spontaneous abortion (adjusted hazard ratio 1.01, 95% CI 0.17, 5.89). Women in the vedolizumab-exposed group were slightly but not significantly more likely to deliver preterm (adjusted hazard ratio 1.58, 95% CI 0.65, 3.82). CONCLUSIONS: No significant increased risks were noted with vedolizumab exposure for any of the other study outcomes. These data add reassuring evidence in support of the safety of vedolizumab in pregnancy.


  • Doubly robust estimation under a possibly misspecified marginal structural Cox model

    by Jiyu Luo, Denise Rava, Jelena Bradic, Ronghui Xu

    Summary: In this article we consider the marginal structural Cox model, which has been widely used to analyse observational studies with survival outcomes. The standard inverse probability weighting method under the model hinges on a propensity score model for the treatment assignment and a censoring model that incorporates both the treatment and the covariates. In such settings model misspecification can often occur, and the Cox regression model’s non-collapsibility has historically posed challenges when striving to guard against model misspecification through augmentation. We introduce a novel joint augmentation to the martingale-based full-data estimating functions and develop rate double robustness, which allows the use of machine learning and nonparametric methods to overcome the challenges of non-collapsibility. We closely examine its theoretical properties to guarantee root-$ n $ inference for the estimand. The estimator extends naturally to estimating a time-average treatment effect when the proportional hazards assumption fails, and we show that it satisfies both the assumption-lean and the well-specification criteria in the context of a causal estimand for censoring survival data; that is, it is a functional of the potential outcome distributions only and does not depend on the treatment assignment mechanism, the covariate distribution or the censoring mechanism. The martingale-based augmentation approach is also applicable to many semiparametric failure time models. Finally, its application to a dataset provides insights into the impact of mid-life alcohol consumption on mortality in later life.


  • Doubly robust estimation under covariate-induced dependent left truncation

    by Yuyao Wang, Andrew Ying, Ronghui Xu

    Summary: In prevalent cohort studies with follow-up, the time-to-event outcome is subject to left truncation leading to selection bias. For estimation of the distribution of the time to event, conventional methods adjusting for left truncation tend to rely on the quasi-independence assumption that the truncation time and the event time are independent on the observed region. This assumption is violated when there is dependence between the truncation time and the event time, possibly induced by measured covariates. Inverse probability of truncation weighting can be used in this case, but it is sensitive to misspecification of the truncation model. In this work, we apply semiparametric theory to find the efficient influence curve of the expectation of an arbitrarily transformed survival time in the presence of covariate-induced dependent left truncation. We then use it to construct estimators that are shown to enjoy double-robustness properties. Our work represents the first attempt to construct doubly robust estimators in the presence of left truncation, which does not fall under the established framework of coarsened data where doubly robust approaches were developed. We provide technical conditions for the asymptotic properties that appear to not have been carefully examined in the literature for time-to-event data, and study the estimators via extensive simulation. We apply the estimators to two datasets from practice, with different right-censoring patterns.


  • Marginal Structural Illness-Death Models for Semi-competing Risks Data

    by Yiran Zhang, Andrew Ying, Steve Edland, Lon White, Ronghui Xu


  • Open Access

    Doubly Robust Estimation under Possibly Misspecified Marginal Structural Cox Model

    by Jiyu Luo, Denise Rava, Jelena Bradic, Ronghui Xu

    In this paper we address the challenges posed by non-proportional hazards and informative censoring, offering a path toward more meaningful causal inference conclusions. We start from the marginal structural Cox model, which has been widely used for analyzing observational studies with survival outcomes, and typically relies on the inverse probability weighting method. The latter hinges upon a propensity score model for the treatment assignment, and a censoring model which incorporates both the treatment and the covariates. In such settings, model misspecification can occur quite effortlessly, and the Cox regression model's non-collapsibility has historically posed challenges when striving to guard against model misspecification through augmentation. We introduce an augmented inverse probability weighted estimator which, enriched with doubly robust properties, paves the way for integrating machine learning and a plethora of nonparametric methods, effectively overcoming the challenges of non-collapsibility. The estimator extends naturally to estimating a time-average treatment effect when the proportional hazards assumption fails. We closely examine its theoretical and practical performance, showing that it satisfies both the assumption-lean and the well-specification criteria discussed in the recent literature. Finally, its application to a dataset reveals insights into the impact of mid-life alcohol consumption on mortality in later life.

  • Open Access

    On Defense of the Hazard Ratio

    by Andrew Ying, Ronghui Xu

    In this short communication, we describe the recent debate on whether the hazard function should be used for causal inference in time-to-event studies and consider three different potential outcomes frameworks (by Rubin, Robins, and Pearl, respectively) as well as use the single-world intervention graph to show mathematically that the hazard function has causal interpretations under all three frameworks. In addition, we argue that the hazard ratio over time can provide a useful interpretation in practical settings.


  • Doubly robust estimation of the hazard difference for competing risks data

    by Denise Rava, Ronghui Xu

    We consider the conditional treatment effect for competing risks data in observational studies. We derive the efficient score for the treatment effect using modern semiparametric theory, as well as two doubly robust scores with respect to (1) the assumed propensity score for treatment and the censoring model, and (2) the outcome models for the competing risks. An important property regarding the estimators is rate double robustness, in addition to the classical model double robustness. Rate double robustness enables the use of machine learning and nonparametric methods in order to estimate the nuisance parameters, while preserving the root‐ asymptotic normality of the estimated treatment effect for inferential purposes. We study the performance of the estimators using simulation. The estimators are applied to the data from a cohort of Japanese men in Hawaii followed since 1960s in order to study the effect of mid‐life drinking behavior on late life cognitive outcomes. The approaches developed in this article are implemented in the R package “HazardDiff”.


  • Measures of explained variation under the mixture cure model for survival data

    by Yingwei Peng, Yuyao Wang, Ronghui Xu

    Explained variation is well understood under linear regression models and has been extended to models for survival data. In this article, we consider the mixture cure models. We propose two approaches to define explained variation under the mixture cure models, one based on the Kullback-Leibler information gain and the other based on residual sum of squares. We show that the proposed measures have desired properties as measures of explained variation, similar to those under other regression models. A simulation study is conducted to demonstrate the properties of the proposed measures. They are also applied to real data analyses to illustrate the use of explained variation.