About

Jennifer Bobb, PhD, is an affiliate associate professor in biostatistics at the University of Washington and an associate editor of the journal Biostatistics. She aims to apply rigorous statistical methods to address important problems in public health, with a focus on statistical issues that arise when data not originally collected for research, such as administrative claims data or electronic health records, are used for scientific questions relevant to clinical practice and health policy. At Kaiser Permanente Washington Health Research Institute (KPWHRI), she collaborates with scientists across various research areas, including mental and behavioral health and social determinants of health. Dr. Bobb provides statistical leadership on pragmatic clinical trials, including evaluating models for treating opioid use disorder and depression, and increasing medication treatment for opioid use disorders within primary care settings. She has developed statistical guidance to address methodological challenges in pragmatic trials leveraging EHR data. Her expertise in environmental biostatistics has led her to conduct large-scale epidemiological investigations on the health effects of exposure to extreme heat and air pollution, developing flexible modeling approaches for multi-pollutant mixtures and publicly available software. Recently, she has collaborated on the Moving to Health study, exploring how changes in the built environment influence long-term weight and diabetes management, which has led to new methodologies addressing spatial confounding and time-varying health impacts. Dr. Bobb holds a PhD in biostatistics from Johns Hopkins Bloomberg School of Public Health and completed a postdoctoral fellowship at Harvard T.H. Chan School of Public Health. She has served on various professional committees, including the American Statistical Association, and received the 2024 COPSS Emerging Leader Award.

Research topics

• Medicine
• Family medicine
• Internal medicine
• Gerontology
• Psychiatry
• Environmental health
• Demography
• Nursing
• Emergency medicine

Selected publications

• Statistical Design of Pragmatic Trials Using Electronic Health Record Data when Outcome Assessments are Uncontrolled and Irregular

  ArXiv.org · 2026-05-08

  articleOpen access1st authorCorresponding

  Pragmatic trials increasingly define outcomes using real-world data such as electronic health records, where assessments are collected during routine care rather than at fixed timepoints. Consequently, these uncontrolled assessments may be irregular, sparse, and affected by the intervention (intervention-dependent assessments), which can lead to biased treatment effect estimates. We developed a simulation study to inform the statistical approach for trials with uncontrolled assessments, which we applied to the MI-CARE pragmatic trial. Using a pre-trial cohort mimicking eligibility and outcome measurement, we estimated assessment frequency and timing and combined these estimates with assumptions about how the intervention effects might impact assessment. We simulated sparse and intervention-dependent assessments and compared single-measure approaches with longitudinal models using all scores. Under intervention-dependent assessments, we found that naive methods such as using the best score or using a randomly selected score without adjusting for measurement timing produced substantial bias. Models that adjusted flexibly for the follow-up timing estimated time-point specific or time-averaged treatment effects without bias. Simulation results informed the selection of the statistical approach for the MI-CARE trial. Among unbiased methods, the most powerful was a linear mixed model with exponential correlation structure, adjustment for time since baseline, and a time-varying intervention effect to estimate the intervention effect at the end of the intervention window. Future studies can use pre-trial data to conduct a simulation study tailored to the trial's data features to inform the analytic approach. Trials with uncontrolled assessments should consider the potential for intervention-dependent assessments and select an appropriate method to avoid bias.

  PublisherOA PDF

• Increased risk of type I errors for detecting heterogeneity of treatment effects in cluster-randomized trials using mixed-effect models

  BMC Medical Research Methodology · 2026-01-12

  articleOpen accessSenior author

  Evaluating heterogeneity of treatment effects (HTE) across subgroups is common in both randomized trials and observational studies. Although several statistical challenges of HTE analyses including low statistical power and multiple comparisons are widely acknowledged, issues specific to clustered data, including cluster randomized trials (CRTs), have received less attention. For testing interactions in linear mixed-effects models (LMM), Barr et al. (2013) suggested that: random slopes for interaction terms should be studied. In this paper, we explore the impact of model misspecification, including generalized LMM (GLMM) with or without random slopes, and provide recommendations for conducting inference for HTE across subgroups in CRTs. We conducted a simulation study to evaluate the performance of common analytic approaches for testing the presence of HTE for continuous, binary, and count outcomes: generalized linear mixed models (GLMM) and generalized estimating equations (GEE) including interaction terms between treatment and subgroup. Several simulation scenarios covered broad range of scenarios in CRTs, for example, small to a large number of clusters, small to moderate cluster-specific random slopes for subgroup. The performance metric was the empirical type I error rate compared to a nominal level. We applied the analytical methods to a real-world CRT using the count outcome utilization of healthcare from the motivating Primary Care Opioid Use Disorder treatment (PROUD) trial. We found that standard GLMM analyses that assume a common correlation of participants within clusters can lead to severely elevated type 1 error rates of up to 47.2% compared to the 5% nominal level if the within-cluster correlation varies across subgroups. A maximal GLMM, which allows subgroup-specific within-cluster correlations, achieved the nominal type 1 error rate, as did GEE (though rates were slightly elevated even with as many as 50 clusters). Applying the methods to the real-world CRT, we found a large impact of the model specification on inference. We recommend that HTE analyses using the maximal GLMM account for within-subgroup correlation to avoid anti-conservative inference. For Wald t-testing of HTE in small sample clusters, appropriate small sample correction methods should be considered based on the outcome data type.

  PublisherDOI

• Additional file 1 of Increased risk of type I errors for detecting heterogeneity of treatment effects in cluster-randomized trials using mixed-effect models

  Figshare · 2026-01-01

  articleOpen accessSenior author

  Supplementary Material 1.

  PublisherDOI

• Substance Use Disorder Care Transitions: Study Protocol for a Mixed-Methods Study Applying Multi-State Models and Qualitative Interviews to Comprehensively Characterize Care in a US Healthcare System (Preprint)

  JMIR Research Protocols · 2026-02-25

  articleOpen access
  PublisherDOI

• Variability in medication treatment of opioid use disorder in primary care: Comparison of PROUD trial intervention clinics and other exemplar clinics

  Journal of Substance Use and Addiction Treatment · 2026-01-01

  articleOpen access
  PublisherOA PDFDOI

• Association of unhealthy alcohol use reported in routine outpatient screening with 30-day hospital readmission risk

  Journal of Substance Use and Addiction Treatment · 2026-04-24

  article
  PublisherDOI

• Statistical Design of Pragmatic Trials Using Electronic Health Record Data when Outcome Assessments are Uncontrolled and Irregular

  arXiv (Cornell University) · 2026-05-08

  preprintOpen access1st authorCorresponding

  Pragmatic trials increasingly define outcomes using real-world data such as electronic health records, where assessments are collected during routine care rather than at fixed timepoints. Consequently, these uncontrolled assessments may be irregular, sparse, and affected by the intervention (intervention-dependent assessments), which can lead to biased treatment effect estimates. We developed a simulation study to inform the statistical approach for trials with uncontrolled assessments, which we applied to the MI-CARE pragmatic trial. Using a pre-trial cohort mimicking eligibility and outcome measurement, we estimated assessment frequency and timing and combined these estimates with assumptions about how the intervention effects might impact assessment. We simulated sparse and intervention-dependent assessments and compared single-measure approaches with longitudinal models using all scores. Under intervention-dependent assessments, we found that naive methods such as using the best score or using a randomly selected score without adjusting for measurement timing produced substantial bias. Models that adjusted flexibly for the follow-up timing estimated time-point specific or time-averaged treatment effects without bias. Simulation results informed the selection of the statistical approach for the MI-CARE trial. Among unbiased methods, the most powerful was a linear mixed model with exponential correlation structure, adjustment for time since baseline, and a time-varying intervention effect to estimate the intervention effect at the end of the intervention window. Future studies can use pre-trial data to conduct a simulation study tailored to the trial's data features to inform the analytic approach. Trials with uncontrolled assessments should consider the potential for intervention-dependent assessments and select an appropriate method to avoid bias.

  PublisherDOI

• Economic cost of strategic implementation approaches to increase uptake of digital therapeutics for substance use disorders in a large integrated health system

  PLOS Digital Health · 2026-01-08

  articleOpen access

  Evidence-based digital therapeutics are a promising approach for the scale-up of substance use disorder (SUD) treatments. Despite demonstrated efficacy, utilization of digital therapeutics is low. Strategic implementation approaches have potential for increasing digital therapeutic use. Applicability to health systems depends, in part, on the economic costs. The objective of this study was to describe implementation and intervention costs of implementation strategies to increase uptake of an evidence-based digital treatment for SUD. We conducted an economic evaluation alongside a hybrid type III cluster-randomized trial within a large integrated health system. All clinics implemented a standard implementation (SI) strategy, and clinics were assigned using 2x2 factorial randomization to additionally receive practice facilitation (PF) and/or health coaching (HC). Implementation costs included the cost of time devoted to implementation activities and direct operating costs. Time devoted to implementation activities was ascertained through structured meeting logs and time use surveys. Operating costs were captured using project budget reports. Intervention costs included expenses for prescriptions and healthcare encounters related to the digital therapeutic, measured using electronic health record data. Univariate statistics were calculated for cost estimates with comparisons presented by trial arm, implementation activity, staff role and study month. Analyses were conducted from a health system perspective. Twenty-one primary care sites participated in the trial. Over the 50-month study period, the total cost of all implementation activities was $748,088. Implementation costs per clinic were highest in the SI + PF + HC arm ($48,029), followed by SI + HC ($36,544), SI + PF ($30,665) and SI alone ($24,774). Intervention costs were highest in the SI + PF + HC arm ($18,051), followed by SI + PF ($11,492), SI + HC ($967) and SI alone ($1,879). Findings from this study can guide health systems by informing the economic investment required to employ implementation strategies demonstrated to increase uptake of evidence-based practices for behavioral health conditions. Trial Registration: NCT05160233.

  PublisherDOI

• Additional file 1 of Increased risk of type I errors for detecting heterogeneity of treatment effects in cluster-randomized trials using mixed-effect models

  Open MIND · 2026-01-01

  articleSenior author

  Supplementary Material 1.

  DOI

• Substance Use Disorder Care Transitions: Study Protocol for a Mixed-Methods Study Applying Multi-State Models and Qualitative Interviews to Comprehensively Characterize Care in a US Healthcare System (Preprint)

  2026-02-25

  articleOpen access

  <sec> <title>BACKGROUND</title> Many effective substance use disorder (SUD) care pathways exist in healthcare systems. However, patients with SUD often report poor care experiences particularly regarding timely follow-up, clinician and general satisfaction ratings, and care communication. Because SUD care pathways involve transitions across clinicians and venues of care, adequate treatment requires coordination across clinicians and settings to ensure unmet needs are addressed, and appropriate SUD care is delivered. Surprisingly little is known about the real-world care pathways patients engage in once identified as having SUD. </sec> <sec> <title>OBJECTIVE</title> This study protocol describes research that will comprehensively characterize the pathways of care utilized by those who obtain care for SUD and compare the quality and effectiveness of these care pathways. Specific Aims are to (1) Apply multi-state models (MSM) to characterize the spectrum of care transitions among patients with SUD within a large integrated health system, (2) Generate data-driven insights to improve care delivery for SUD using estimated MSMs, and (3) Observe and explore patient and clinician experiences with care transitions across common care pathways using qualitative methods. </sec> <sec> <title>METHODS</title> Quantitative data sources will include electronic healthcare records, insurance claims, self-reported measures of substance use and SUD symptoms, and death data from an integrated healthcare system in Washington State. To identify care pathways, we will apply continuous time, multi-state modeling methods to empirically observe the longitudinal course of SUD care transitions undertaken by patients over time; each “state” or occurrence of care will be characterized by the intervention received (e.g., evaluation or assessment, behavioral treatment or counseling, pharmacotherapy) and setting of care (e.g., outpatient, intensive outpatient, inpatient or residential). The estimated parameters from the fitted MSMs will be transformed or interpreted to characterize SUD care quality, such as wait times for SUD visits and receiving an adequate psychotherapy dose. To compare effectiveness of identified pathways, we will examine terminal states including death, remission of SUD, and loss to follow-up. Finally, employing a mixed methods design, we will purposively sample and interview patients engaged in the empirically derived pathways to understand their care experiences; we will also observe and interview clinicians to elucidate health system factors that facilitate or hinder their ability to link patients to SUD care. </sec> <sec> <title>RESULTS</title> This study was funded in June 2025 and received IRB approval on July 17, 2025. We expect preliminary data collection for quantitative analyses will be complete by May 2026 and final data collection will be complete by November 2027. We expect qualitative data collection will be complete by November 2029. </sec> <sec> <title>CONCLUSIONS</title> This study will identify and compare the quality and effectiveness of common pathways that patients take when they obtain treatment for SUD and provide decision makers with information on how to better organize SUD care delivery. </sec> <sec> <title>CLINICALTRIAL</title> N/A </sec>

  PublisherDOI

Frequent coauthors

• Katharine A. Bradley

  118 shared

• Gwen T. Lapham

  Kaiser Permanente Washington Health Research Institute

  85 shared

• Theresa E. Matson

  Kaiser Permanente Washington Health Research Institute

  61 shared

• Joseph E. Glass

  Kaiser Permanente Washington Health Research Institute

  56 shared

• Emily C. Williams

  VA Puget Sound Health Care System

  56 shared

• Andrew J. Saxon

  Marin Community Foundation

  53 shared

• Julia H. Arnsten

  Montefiore Medical Center

  46 shared

• Denise M. Boudreau

  Kaiser Permanente Washington Health Research Institute

  38 shared

Awards & honors

• Emerging Leader Award from the Committee of Presidents of St…
• COPSS Emerging Leader Award (2024)