Publications

2025

1. Incidence of catheter-related bloodstream infections with sodium citrate lock therapy in adult patients receiving home parenteral nutrition: A descriptive cohort study

   Rachel Leong, Nisha J Dave, Daniel P Griffith, Anna Guo, Kirk A Easley, John R Galloway, Thomas R Ziegler, and Vivian M Zhao

   Journal of Parenteral and Enteral Nutrition, 2025

   Abs HTML PDF

   Background We determined the incidence of catheter-related bloodstream infections in adult patients requiring home parenteral nutrition (HPN) while receiving sodium citrate locks. Methods We conducted a single-center descriptive cohort study involving 38 adults who required HPN from January 1, 2020, to August 31, 2022. The exact method, assuming a Poisson distribution, was used to estimate the incidence rate of catheter-related bloodstream infections per 1000 catheter days among patients receiving sodium citrate locks. Univariate and multivariate analyses using Poisson regression and frailty models were employed to evaluate predictive factors. Results Thirty-eight patients received sodium citrate locks, with 65.8% women (mean age, 50.2 ± 14.5 years). The average length of HPN was 3.6 years. Forty-six catheter-related bloodstream infections occurred during 20,085 catheter days, demonstrating an incidence rate of 2.3 (95% confidence interval, 1.7–3.1) per 1000 catheter days. Peripheral-inserted central catheters had a higher incidence rate (3.9 per 1000 catheter days) than Hickman catheters (2.2 per 1000 catheter days), with a hazard ratio of 1.27, indicating a 27% increased risk of catheter-related bloodstream infections. Univariate and multivariate Poisson regression analyses revealed that for every 1-h increase in HPN infusion duration (h/day), the incidence rate of catheter-related bloodstream infections is expected to increase by 10%. Conclusion Catheter-related bloodstream infections are common in adult patients requiring HPN. Sodium citrate locks may help prevent these infections. Recognizing predictive factors, such as the duration of parenteral infusion, can help healthcare providers develop more effective prevention strategies.

2024

1. Average Causal Effect Estimation in DAGs with Hidden Variables: Beyond Back-Door and Front-Door Criteria

   Anna Guo, and Razieh Nabi

   arXiv preprint arXiv:2409.03962, 2024

   Abs HTML PDF Poster

   The identification theory for causal effects in directed acyclic graphs (DAGs) with hidden variables is well established, but methods for estimating and inferring functionals that extend beyond the g-formula remain underdeveloped. Previous studies have introduced semiparametric estimators for such functionals in a broad class of DAGs with hidden variables. While these estimators exhibit desirable statistical properties such as double robustness in certain cases, they also face significant limitations. Notably, they encounter substantial computational challenges, particularly involving density estimation and numerical integration for continuous variables, and their estimates may fall outside the parameter space of the target estimand. Additionally, the asymptotic properties of these estimators is underexplored, especially when integrating flexible statistical and machine learning models for nuisance functional estimations. This paper addresses these challenges by introducing novel one-step corrected plug-in and targeted minimum loss-based estimators of causal effects for a class of hidden variable DAGs that go beyond classical back-door and front-door criteria (known as the treatment primal fixability criterion in prior literature). These estimators leverage data-adaptive machine learning algorithms to minimize modeling assumptions while ensuring key statistical properties including double robustness, efficiency, boundedness within the target parameter space, and asymptotic linearity under L2(P)-rate conditions for nuisance functional estimates that yield root-n consistent causal effect estimates. To ensure our estimation methods are accessible in practice, we provide the flexCausal package in R.

2023

1. Flexible Nonparametric Inference for Causal Effects under the Front-Door Model

   Anna Guo, David Benkeser, and Razieh Nabi

   arXiv preprint arXiv:2312.10234, 2023

   Abs HTML PDF Poster Slides

   Evaluating causal treatment effects in observational studies requires addressing confounding. While the back-door criterion enables identification through adjustment for observed covariates, it fails in the presence of unmeasured confounding. The front-door criterion offers an alternative by leveraging variables that fully mediate the treatment effect and are unaffected by unmeasured confounders of the treatment-outcome pair. We develop novel one-step and targeted minimum loss-based estimators for both the average treatment effect and the average treatment effect on the treated under front-door assumptions. Our estimators are built on multiple parameterizations of the observed data distribution, including approaches that avoid modeling the mediator density entirely, and are compatible with flexible, machine learning-based nuisance estimation. We establish conditions for root-n consistency and asymptotic linearity by deriving second-order remainder bounds. We also develop flexible tests for assessing identification assumptions, including a doubly robust testing procedure, within a semiparametric extension of the front-door model that encodes generalized (Verma) independence constraints. We further show how these constraints can be leveraged to improve the efficiency of causal effect estimators. Simulation studies confirm favorable finite-sample performance, and real-data applications in education and emergency medicine illustrate the practical utility of our methods. An accompanying R package, fdcausal, implements all proposed procedures.

2. UAI

   Sufficient Identification Conditions and Semiparametric Estimation under Missing Not at Random Mechanisms

   Anna Guo, Jiwei Zhao, and Razieh Nabi

   Uncertainty in Artificial Intelligence, 2023

   Abs HTML PDF Poster

   Conducting valid statistical analyses is challenging in the presence of missing-not-at-random (MNAR) data, where the missingness mechanism is dependent on the missing values themselves even conditioned on the observed data. Here, we consider a MNAR model that generalizes several prior popular MNAR models in two ways: first, it is less restrictive in terms of statistical independence assumptions imposed on the underlying joint data distribution, and second, it allows for all variables in the observed sample to have missing values. This MNAR model corresponds to a so-called criss-cross structure considered in the literature on graphical models of missing data that prevents nonparametric identification of the entire missing data model. Nonetheless, part of the complete-data distribution remains nonparametrically identifiable. By exploiting this fact and considering a rich class of exponential family distributions, we establish sufficient conditions for identification of the complete-data distribution as well as the entire missingness mechanism. We then propose methods for testing the independence restrictions encoded in such models using odds ratio as our parameter of interest. We adopt two semiparametric approaches for estimating the odds ratio parameter and establish the corresponding asymptotic theories: one involves maximizing a conditional likelihood with order statistics and the other uses estimating equations. The utility of our methods is illustrated via simulation studies.

3. Impacts of the COVID‐19 Lockdown on Gender Inequalities in Time Spent on Paid and Unpaid Work in Singapore

   Emma Zang, Poh Lin Tan, Thomas Lyttelton, and Anna Guo

   Population and Development Review, Mar 2023

   Abs HTML PDF

   Objective: To examine the impact of the COVID-19 lockdown on gender inequalities in time spent on paid labor market work, housework, and childcare in Singapore. Background: Widespread shifts to remote work, school closures, and job losses arising from the COVID-19 pandemic have affected gender inequalities in time spent on paid and unpaid work globally. Major gaps in the literature include a lack of longitudinal data to compare time use before and during the pandemic, a lack of examination of how gender and family resources intersect to create inequalities in time use during the pandemic, and a lack of focus on potential mechanisms through which the pandemic affects time use patterns across genders. Method: We use a panel dataset of 290 married women interviewed before, during, and after the COVID-19 lockdown, and apply between-within models to examine changes in gender gaps in time use (defined as females’ time use minus males’ in this study). Results: Gender gaps in housework hours increased during and persisted after the lockdown, even as the negative gender gap in paid work hours narrowed. The gap in childcare hours expanded among households with fewer resources but decreased among households with more resources. We also find that gender ideologies and resources may have both played important roles in how the pandemic affects gender inequalities in time use. Conclusion: Our results highlight that gender and resources can interact, putting women in a vulnerable position when a pandemic strikes, especially among less-resourced households.

2022

1. Trajectories of General Health Status and Depressive Symptoms Among Persons With Cognitive Impairment in the United States

   Emma Zang, Anna Guo, Christina Pao, Nancy Lu, Bei Wu, and Terri R. Fried

   Journal of Aging and Health, Aug 2022

   Abs HTML PDF

   Objectives To identify and examine heterogeneous trajectories of general health status (GHS) and depressive symptoms (DS) among persons with cognitive impairment (PCIs). Methods: We use group-based trajectory models to study 2361 PCIs for GHS and 1927 PCIs for DS from the National Health and Aging Trends Survey 2011–2018, and apply multinomial logistic regressions to predict identified latent trajectory group memberships using individual characteristics. Results: For both GHS and DS, there were six groups of PCIs with distinct trajectories over a 7-year period. More than 40% PCIs experienced sharp declines in GHS, and 35.5% experienced persistently poor GHS. There was greater heterogeneity in DS trajectories with 55% PCIs experiencing improvement, 16.4% experiencing persistently high DS, and 30.5% experiencing deterioration. Discussion: The GHS trajectories illustrate the heavy burden of poor and declining health among PCIs. Further research is needed to understand the factors underlying stable or improving DS despite declining GHS