Shu Xu
Shu Xu
Scroll
Clinical Assistant Professor of Biostatistics
-
Professional overview
-
Dr. Shu Xu’s work represents a balance of both statistical and applied aspects of quantitative methodology. Her primary quantitative interests include evaluating and developing statistical methods for longitudinal data analysis. Specifically, Dr. Xu’s research focuses on various aspects of latent growth models, missing data methods, and causal inference models.
Dr. Xu has collaborated with substance use, family, and health researchers to advance and share her knowledge of quantitative methodology and pursue a better understanding of the social sciences and public health. She has conducted research with the Family Translational Research Group at NYU and the Methodology Center at the Pennsylvania State University.
-
Education
-
BS, Psychology, East China Normal University, Shanghai, ChinaMS, Quantitative Psychology, University of California, DavisPhD, Quantitative Psychology, University of California, Davis
-
Areas of research and study
-
BiostatisticsFamily researchLongitudinal Data AnalysisMissing Data MethodsMixture ModelsQuantitative Research
-
Publications
Publications
Association of Racial Discrimination With Adiposity in Children and Adolescents
AbstractCuevas, A. G., Krobath, D. M., Rhodes-Bratton, B., Xu, S., Omolade, J. J., Perry, A. R., & Slopen, N. (n.d.).Publication year
2023Journal title
JAMA network openVolume
6Issue
7Page(s)
e2322839AbstractImportance: Childhood obesity is a major public health issue and is disproportionately prevalent among children from minority racial and ethnic groups. Personally mediated racism (commonly referred to as racial discrimination) is a known stressor that has been linked to higher body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) in adults, but little is known about the association of racial discrimination and childhood and adolescent adiposity. Objective: To assess the prospective association between self-reported experiences of racial discrimination and adiposity (BMI and waist circumference) in a large sample of children and adolescents in the Adolescent Brain Cognitive Development (ABCD) study. Design, Setting, and Participants: This cohort study used complete data from the ABCD study (2017 to 2019), involving a total of 6463 participants. The ABCD study recruited a diverse sample of youths from across the US, with rural, urban, and mountain regions. Data were analyzed from January 12 to May 17, 2023. Exposure: The child-reported Perceived Discrimination Scale was used to quantify racial discrimination, reflecting participants' perceptions of being treated unfairly by others or unaccepted by society based on their race or ethnicity. Main Outcomes and Measures: Weight, height, and waist circumference were measured by trained research assistants. BMI z scores were computed by applying the US Centers for Disease Control and Prevention's age and sex-specific reference standards for children and adolescents. Waist circumference (inches) was quantified as the mean of 3 consecutive measures. Measurements were taken from time 1 (ie, 2017 to 2019) and time 2 (ie, 2018 to 2020). Results: Of the 6463 respondents with complete data, 3090 (47.8%) were female, and the mean (SD) age was 9.95 (0.62) years. Greater racial discrimination exposure at time 1 was associated with higher BMI z score in both unadjusted (β, 0.05; 95% CI, 0.02-0.08) and adjusted regression models (β, 0.04; 95% CI, 0.01-0.08). Discrimination at time 1 was associated with higher waist circumference in unadjusted (β, 0.35; 95% CI, 0.15-0.54) and adjusted (β, 0.24; 95% CI, 0.04-0.44) models. Conclusions and Relevance: In this cohort study of children and adolescents, racial discrimination was positively associated with adiposity, quantified by BMI z score and waist circumference. Interventions to reduce exposure to racial discrimination in early life may help reduce the risk of excess weight gain across throughout life.Examining the effects of cumulative environmental stressors on Gulf Coast child and adolescent health
AbstractMeltzer, G. Y., Merdjanoff, A., Xu, V. S., Gershon, R., Emrich, C. T., & Abramson, D. (n.d.).Publication year
2023Journal title
Population and EnvironmentVolume
45Issue
3AbstractThis study examines how community-level cumulative environmental stress affects child and adolescent emotional distress and chronic health conditions both directly and indirectly through stressors at the household, family, and individual levels. Data comes from the Women and their Children’s Health (WaTCH) Study, which sought to understand the health implications of exposure to the 2010 Deepwater Horizon oil spill (DHOS) among a cohort of 596 mothers with children ages 10 to 17 in southeastern Louisiana. Community-level environmental stress was measured using a newly developed geospatial index. Household-level stressors included previous hurricane impacts, impacts of DHOS, degree of financial difficulty, and degree of housing physical decay. Family stressors included maternal depression, self-rated physical health, and degree of parenting stress. Child stress was based on perceived stress; child mental health was based on serious emotional disturbance; and child physical health was based on diagnosis of chronic illness. Structural equation modeling used weighted least squares means and variance and theta parameterization. Results showed a significant negative direct path between community-level cumulative environmental stress and child/adolescent serious emotional disturbance and chronic illness. However, the indirect relationship through household, family, and individual-level stressors was significant and positive for both child/adolescent serious emotional disturbance and chronic illness. These findings point to the centrality of the household and family in determining child and adolescent physical and mental health outcomes in communities exposed to frequent disasters and ongoing environmental stressors.Predictors of Crosscutting Patterns of Psychological Health and Family Maltreatment
Tutorial on Causal Mediation Analysis With Binary Variables: An Application to Health Psychology Research
AbstractXu, S., Coffman, D. L., Luta, G., & Niaura, R. S. (n.d.).Publication year
2023Journal title
Health PsychologyVolume
42Issue
11Page(s)
778-787AbstractMediation analysis has been widely applied to explain why and assess the extent to which an exposure or treatment has an impact on the outcome in health psychology studies. Identifying a mediator or assessing the impact of a mediator has been the focus of many scientific investigations. This tutorial aims to introduce causal mediation analysis with binary exposure, mediator, and outcome variables, with a focus on the resampling and weighting methods, under the potential outcomes framework for estimating natural direct and indirect effects. We emphasize the importance of the temporal order of the study variables and the elimination of confounding. We define the causal effects in a hypothesized causal mediation chain in the context of one exposure, one mediator, and one outcome variable, all of which are binary variables. Two commonly used and actively maintained R packages, mediation and medflex, were used to analyze a motivating example. R code examples for implementing these methods are provided.Use of electronic nicotine delivery system (ENDS) devices among U.S. Youth and adults: Findings from the Population Assessment of Tobacco and Health Study Waves 1–5
AbstractJiang, N., Xu, S., Li, L., Cleland, C. M., & Niaura, R. S. (n.d.).Publication year
2023Journal title
Addictive BehaviorsVolume
139AbstractIntroduction: Electronic nicotine delivery system (ENDS) devices evolve rapidly and impact nicotine dependence. This study described the type of ENDS devices used most frequently by U.S. youth and adults from 2013/14 to 2018/19. Methods: We analyzed Waves 1–5 data of the Population Assessment of Tobacco and Health Study. Among current ENDS users, descriptive statistics summarized the most frequently used ENDS devices (i.e., disposable cigalike, refillable cartridge, nonrefillable cartridge, tank, mod, prefilled pod, disposable pod) among youth (12–17 years), young adults (18–34 years), and older adults (≥35 years) for each wave. Results: The proportion of current ENDS users who reported they most frequently used disposable cigalikes and cartridge-based devices declined over time across all age groups. At Waves 1–4, tank was generally the most popular type for all ages and an increasing proportion of ENDS users reported they most frequently used tanks. The primary use of mods decreased among youth, and fluctuated among young and older adults. At Wave 5, prefilled pods became the dominant type (youth: 55.0%; young adults: 44.7%; older adults: 42.7%), and 4.2–10.0% of ENDS users reported using disposable pods most often. The popularity of tanks, mods, and prefilled pods was more evident in youth and young adults, and primary use of disposable pods was more common in older adults. Conclusions: The primary use of ENDS devices changed over the years and varied by age. More research is warranted to continuously monitor the characteristics of ENDS devices in youth and adults to inform product regulations and intervention efforts.Cigar Use Progression Among New Cigar Initiators: A Two-Part Growth Curve Analysis Among a Youth and Young Adult Cohort
AbstractCantrell, J., Xu, S., Kreslake, J., Liu, M., & Hair, E. (n.d.).Publication year
2022Journal title
Nicotine and Tobacco ResearchVolume
24Issue
1Page(s)
28-36AbstractIntroduction: Youth and young adults (YYAs) are at high risk of cigar use. This study's objective was to examine progression and sociodemographic differences in current cigar use and frequency among new cigar initiators. Aims and Methods: We conducted a two-part latent growth model among a nationally representative cohort of cigar initiators (aged 15-25) to examine 24-month trajectories of current cigar use and frequency (n = 1483). The cohort was recruited via address-based sampling with online data collection from 2014 to 2019 and surveyed approximately every 6 months. Results: The unconditional odds of current cigar use (ie, past 30-day use) within 6 months of initiation was 0.72 (95% confidence interval: 0.63, 0.82), corresponding to a probability of 42%. The odds of current use among recent cigar initiates declined 6 months after initiation and was followed by a stabilization in use over time. Among continued users, frequency (# days used in past 30 days) increased linearly over time but remained low (3.47 days/months at 24 months). Younger individuals, non-Hispanic African Americans, those with lower subjective financial status, and current users of cigarettes, other tobacco products and/or marijuana were at highest risk within 6 months of initiation. Males, younger users, and current cigarette smokers had the highest risk for cigar progression over time. Conclusions: This study is the first to examine longitudinal cigar use patterns among YYA cigar initiators. Findings emphasize the need for research across the cigar use spectrum and the importance of interventions targeted by age, stage of use, cigarette, other tobacco, and marijuana use and key sociodemographics to interrupt use pathways. Implications: This study is the first to examine progression of cigar use among YYAs who have newly initiated cigars. Results show a high probability of current cigar use within 6 months of initiation followed by a rapid decline and stabilization over time. Frequency increases among those who continue using cigars. Males, younger users, and current cigarette smokers had the highest risk for cigar progression over time. Findings emphasize the need for targeting interventions by age, stage of use, cigarette, other tobacco, and marijuana use and key sociodemographics to interrupt use pathways.Passive exposure to e-cigarette emissions is associated with worsened mental health
Recurrent Injecting Drug Use as a Mediator between Psychiatric Disorder and Non-Fatal Overdose
AbstractBarnes, D. M., Xu, S., Cleland, C. M., McKnight, C., & Des Jarlais, D. (n.d.).Publication year
2022Journal title
Substance Use and MisuseVolume
57Issue
8Page(s)
1248-1256AbstractBackground: Unintentional drug overdose has increased markedly in the United States. Studies document an association between psychiatric disorder and unintentional overdose; we extend this research through a preliminary test of a causal model of recurrent injection drug use mediating this relationship. Methods: In a cross-sectional study of 241 adults in New York City with a possible current substance use disorder, we conducted conventional and Imai’s mediation analyses to examine if psychiatric disorder is associated with increased prevalence of ever overdosing and if recurrent injection drug use mediates this association. Our cross-sectional data permit the first step of assessing causal models: testing if statistical associations are consistent with the model. Results: Fifty-eight percent of the sample endorsed previous psychiatric disorder diagnosis and 35.7% reported ever overdosing. Imai’s mediation analysis showed that, adjusting for covariates, the total association between psychiatric diagnosis and ever overdosing (adjusted prevalence difference [aPD] = 0.16, 95% CI 0.04–0.28) was composed of a direct effect (aPD = 0.09, 95% CI −0.03 − 0.21, p = 0.136) and an indirect effect (aPD = 0.07, 95% CI 0.02–0.13). Recurrent injecting drug use contributed to 42% (ratio of indirect effect to total effect; 95% CI 12 − 100%, p = 0.02) of the association between psychiatric diagnosis and ever overdosing. Conventional mediation analysis produced similar results. Conclusions: Our results provide a warrant for taking the necessary next step for assessing a causal model using longitudinal data, potentially providing a strong rationale for intervening on psychiatric disorders to stem overdose.Relationships Between E-cigarette Use and Subsequent Cigarette Initiation Among Adolescents in the PATH Study: an Entropy Balancing Propensity Score Analysis
The Mediating Effect of E-Cigarette Harm Perception in the Relationship between E-Cigarette Advertising Exposure and E-Cigarette Use
Youth E-Cigarette Use and Functionally Important Respiratory Symptoms: The Population Assessment of Tobacco and Health (PATH) Study Waves 3 and 4
Cannabis Use and the Onset of Cigarette and E-cigarette Use: A Prospective, Longitudinal Study among Youth in the United States
E-cigarette use, systemic inflammation, and depression
Self-report measures of coercive process in couple and parent–child dyads.
High Sensitivity and Specificity Screening for Clinically Significant Intimate Partner Violence
Psychology of wearing face masks to prevent transition of COVID-19
The lump-versus-split dilemma in couple observational coding: A multisite analysis of rapid marital interaction coding system data.
Using Security Questions to Link Participants in Longitudinal Data Collection
A Randomized, Controlled Trial of the Impact of the Couple CARE for Parents of Newborns Program on the Prevention of Intimate Partner Violence and Relationship Problems
Patterns of psychological health problems and family maltreatment among United States Air Force members
A New Look at the Psychometrics of the Parenting Scale Through the Lens of Item Response Theory
Interrater agreement statistics with skewed data: Evaluation of alternatives to Cohen's kappa
Noxious family environments in relation to adult and childhood caries
On Fitting a Multivariate Two-Part Latent Growth Model
AbstractXu, S., Blozis, S. A., & Vandewater, E. A. (n.d.).Publication year
2014Journal title
Structural Equation ModelingVolume
21Issue
1Page(s)
131-148AbstractA 2-part latent growth model can be used to analyze semicontinuous data to simultaneously study change in the probability that an individual engages in a behavior, and if engaged, change in the behavior. This article uses a Monte Carlo (MC) integration algorithm to study the interrelationships between the growth factors of 2 variables measured longitudinally where each variable can follow a 2-part latent growth model. A SAS macro implementing Mplus is developed to estimate the model to take into account the sampling uncertainty of this simulation-based computational approach. A sample of time-use data is used to show how maximum likelihood estimates can be obtained using a rectangular numerical integration method and an MC integration method.Causal Inference in Latent Class Analysis