BIRCH holds a range of in-person and online events to share recent findings and best practices.
Our Events
Methods Events
The UW BIRCH Methods Core is tasked with developing supporting research design, statistics, visualization, and results presentation with regard to BIRCH research. Methods Core faculty will host seminars and webinars to share recent findings and best practices for methodological design.
General Events
Webinars and Work in Progress Sessions will be presented by members of all BIRCH cores to present current research, provide feedback on projects, and support ongoing collaboration within the Center.
Future Events
Past Events
2024 Events
Date: February 28th, 2024
Presenters: Dorothy Mushi, Helen Jack, Kudakwashe Takarinda, Malinda Kaiyo-Utete, Anna Larsen, Joan Mutahi, Akansha Vaswani-Bye, Jane Lee and Martha Zuniga.
BIRCH hosted its second BIRCH Day at the UW Hans Rosling Center. The hybrid symposium style event drew over 30 attendees from diverse institutions including the University of Washington, Muhimbili University of Allied Sciences (MUHAS), Organization for Public Health Interventions & Development (OPHID), Kenyatta National Hospital, Entre Hermanos and AARTH. Five BIRCH-funded pilot awardees showcased their progress to date, engaging with the audience through insightful discussions and Q&A sessions. The event culminated in a feedback session with BIRCH Core Directors aimed at refining support strategies for our pilot principal investigators, yielding valuable insights.
Video link: Here
Title: Generalized Linear Mixed Effects Models for Longitudinal Analysis
Date: February 23rd, 2024
Presenter: Chongzhi Di, PhD, Professor of Biostatistics at Fred Hutchinson Cancer Center and Affiliate Professor of Biostatistics at UW.
Dr. Chongzhi Di presented this virtual workshop on linear mixed models (LMMs) and generalized linear mixed models (GLMMs) for longitudinal data analysis. This workshop discussed LMMs for continuous outcomes in the first half and then moved on to GLMMs for discrete or non-Gaussian outcomes (e.g., counts, binary data, etc.). Dr. Di discussed various aspects of LMMs and GLMMs, including model specification, estimation and inference, prediction of random effects, as well as implementations and practical considerations. Dr. Di presented GEE’s conceptual foundations, including model specification, estimation and inference, as well as implementations and practical considerations (e.g., choosing working correlations). He also introduced advanced topics including missing data and time-varying covariates in the latter part of the workshop.
Additional LMM and GLMM resources: Here
Title: Measuring Disparity and an Analytic Approach for Informing Interventions to Reduce Disparity
Date: February 22nd, 2024
Presenter: John Jackson, ScD, Assistant Professor in the Departments of Epidemiology, Biostatistics, and Mental Health at the Johns Hopkins Bloomberg School of Public Health, and core faculty in the Johns Hopkins Center for Health Equity, Center for Health Disparities Solutions, and Center for Drug Safety and Effectiveness.
Dr. John W. Jackson’s research primarily focuses on developing methodological tools for translational health equity research. This includes methods to identify high leverage targets and strategies for interventions that address health disparities, as well as methods to evaluate effects of interventions and policies, and to translate interventions to new populations and contexts, with current applications in healthcare and clinical prognosis.
During this session, Dr. Jackson focused on the following topics:
- Outlining how to incorporate equity value judgements in analytic approaches to measure and identify leverage points for reducing disparities (Dr. John W. Jackson called this “causal decomposition analysis”).
- How covariate adjustments in defining disparities and in equalizing potential determinants of disparities (decompositions) ultimately convey value judgements about what is fair and equitable in the distribution of health and its determinants.
- Various principles to guide the choice of covariates for meaningfully defining disparities and decompositions while adjusting for other covariates to account for confounding.
Video Link: Here
2023 Events
Date: October 24th, 2023
UW BIRCH hosted its very first BIRCH Day at Harborview Ninth and Jefferson Building. The symposium style workshop highlighted BIRCH’s Ending the HIV Epidemic (EHE) supplement projects. BIRCH Core directors, Supplement PIs Implementation Science experts, and Qualitative Methods experts focused on the aims, plans and timelines for the grant submissions and potential next steps. The workshop drew nearly 20 attendees representing different institutions such as the University of Washington, Public Health-Seattle and King County (PHSKC), Kaiser Foundation Research Institute, and UW/Fred Hutch Center for AIDS Research (CFAR).
Video Link: Here
Presentation Slides: Here
Title: Selecting and Designing Hybrid Effectiveness-Implementation Study Designs
Date: October 12th, 2023
Presenter: Bryan Weiner, PhD, Professor, University of Washington Department of Global Health and Department of Health Services
Dr. Bryan Weiner presented this virtual workshop on selecting and designing studies to evaluate effectiveness and implementation. Hybrid effectiveness-implementation study designs can accelerate the translation of research into practice by addressing, in the same study, questions about the effectiveness of an intervention with questions about how best to implement it. This workshop introduced hybrid study designs, offered guidance about when to use them and which type to use given the state of the science, provided hands-on experience designing a hybrid study design, and offered individual consultation and feedback.
Video Link: Here
Presentation Slides: Here
Title: Statistical analysis on treatment non-compliance and mediation analysis
Date: October 5th, 2023
Presenter: Gary Chan, PhD, Professor, University of Washington Department of Biostatistics, Department of Health Systems and Population Health, Adjunct Professor, Department of Statistics
Dr. Gary Chan presented this virtual workshop on statistical analysis on treatment non-compliance and mediation analysis. Treatment evaluation is often complicated by treatment non-compliance as the benefit of randomization is reduced. Moreover, it is often of interest to understand causal mechanisms through immediate variables along the causal pathway. Some essential evaluation concepts for randomized trials discussed in BIRCH’s previous workshop were reviewed. Dr. Chan introduced several statistical methods for handling the intricate issues posed by treatment non-compliance and mediation analysis. Regression based approaches, such as two stage least squares, Baron and Kenny’s method and Sobel test were discussed, and more contemporary frameworks such as principal stratification and causal mediation analysis were introduced.
Video Link: Here
Presentation Slides: Here
Title: Rapid Qualitative and Integrated Mixed Methods Implementation Research
Date: February 6th, 2023
Presenter: Alison Hamilton, PhD, MPH, Professor-in-Residence, UCLA Departments of Psychiatry and Biobehavioral Sciences
Dr. Alison Hamilton presented this workshop on rapid qualitative and integrated mixed methods in implementation research. Qualitative and mixed methods implementation evaluations often necessitate rapid turn-around of preliminary results, e.g., to inform implementation processes and strategies and tailoring of interventions. In this workshop, varied approaches to designing and executing rapid turn-around qualitative and mixed methods implementation evaluations were explored, with an emphasis on rapid qualitative analysis.
Examples and hands-on exercises familiarized participants with designing semi-structured, framework-driven interview guides; constructing summary templates; summarizing qualitative interview data; generating matrices; and developing products.
Video Link: https://www.youtube.com/watch?v=mBF6CxQH91M
Presentation Slides:
Relevant Readings:
Other Resources:
- UCLA CHIPTS YouTube channel: https://www.youtube.com/@uclachipts
- UW IS Resources Website: https://impsciuw.org/
- Webtool to help researchers decide which IS model to use: https://dissemination-implementation.org/
- Research Talk (where Alison’s courses are offered): https://researchtalk.com/
- COREQ: https://pubmed.ncbi.nlm.nih.gov/17872937/
- UW HIV IS Hub website: https://depts.washington.edu/uwhivishub/
- Transcription services: otter.ai, rev.com
- CFIR Website: cfirguide.org
Title: Foundations of Generalized Estimating Equations (GEE) for Longitudinal Analysis
Date: January 13th, 2023
Presenter: Chongzhi Di, PhD, Professor of Biostatistics at Fred Hutchinson Cancer Center and Affiliate Professor of Biostatistics at UW.
Dr. Chongzhi Di presented this virtual workshop on foundations of generalized estimating equations (GEE). In longitudinal data analysis, the GEE approach has been widely used since being proposed by Liang and Zeger (1986). It builds upon generalized linear models (e.g., linear regression, logistic regression) and provides a valid, efficient, and robust statistical tool for analyzing correlated or clustered data.
Dr. Di presented GEE’s conceptual foundations, including model specification, estimation and inference, as well as implementations and practical considerations (e.g., choosing working correlations). He also introduced advanced topics including missing data and time-varying covariates in the latter part of the workshop. Examples with publicly available data and R scripts were provided to illustrate its use in biomedical applications.
Video Link: Here
2022 Events
Title: Achieving Intervention EASE: The Multiphase Optimization Strategy (MOST)
Date: October 25th, 2022
Presenter: Linda M. Collins, PhD, Professor, New York University Departments of Social & Behavioral Sciences and Biostatistics
Dr. Linda M. Collins introduced an expanded methodological framework for developing, optimizing, and evaluating behavioral and biobehavioral interventions. This framework, called the multiphase optimization strategy (MOST), is a principled approach that integrates ideas from behavioral science, engineering, multivariate statistics, health economics, and decision science. MOST enables the investigator to balance intervention effectiveness, affordability, scalability, and efficiency to achieve intervention EASE. Using MOST, behavioral and biobehavioral interventions can be optimized to meet an objective chosen by the investigator. The objective may be any reasonable goal, such as an intervention that offers the best expected outcome achievable without exceeding a specified upper limit on implementation cost or time. MOST relies heavily on resource management by strategic choice of highly efficient experimental designs. Recent advances include an approach to identifying value-efficient interventions. Dr. Collins proposed that MOST offers several benefits, including more rapid long-run improvement of interventions, without requiring a dramatic increase in research resources.
Video Link: Here
Title: Treatment Effects Evaluation in Observational Studies: Why do we have different methods?
Date: September 26th, 2022
Presenter: Gary Chan, PhD, Professor, University of Washington Department of Biostatistics, Department of Health Systems and Population Health, Adjunct Professor, Department of Statistics
Dr. Gary Chan presented this virtual workshop on treatment effects evaluation. Many methods and concepts exist for estimating treatment effect using observational studies: propensity scores, regression adjustments, predictive margins, weighting, matching, stratification, double robust estimation, covariate balancing, etc. There is considerable difficulty for an applied researcher to navigate the methodological zoo. Dr. Chan presented the conceptual foundations, described the pros and cons, and made some recommendations. Examples were provided using R.
Video Link: Here
Title: So you want to do a systematic review? An introduction to systematic review methodology and practicalities
Date: February 25th, 2022
Presenter: Diana Louden, MLib, Life Sciences Librarian, University of Washington Libraries
Diana Nelson Louden presented on features of systematic and scoping reviews, formulating research questions, developing protocols and search strategies, using tools to manage the study selection process, and reporting results. Case studies of HIV systematic reviews were provided by Anjuli Wagner, PhD, MPH, Assistant Professor of Global Health, and Mira Reichman, BA, Clinical Psychology Graduate Student, to illustrate how these principles play out in practice.
Video Link: Here
Title: Cluster Randomized Trials and the Stepped Wedge Design
Date: February 11th, 2022
Presenters: Jim Hughes, PhD, MS, Professor, University of Washington Statistics; Emily Voldal, PhD, University of Washington Biostatistics.
Dr. Jim Hughes, PhD, MS, and Dr. Emily Voldal, PhD, covered cluster randomized trials and stepped wedge designs, followed by useful examples. The first portion of the workshop was didactic and the second portion focused on conducting these analyses in R.
Video link: Here
Title: Measuring Surrogacy in Clinical Research with An Application to Studying Surrogate Markers for HIV Treatment-as-Prevention
Date: January 14th, 2022
Presenter name: Ying Qing Chen, Professor of Medicine, Stanford Prevention Research Center
Dr. Ying Qing Chen discussed measuring surrogacy in clinical research. Dr. Chen reviewed the main statistical methods for evaluating surrogate markers, and suggested a new measure, the so-called “population surrogacy fraction of treatment effect,” or simply the rho-measure, in the setting of clinical trials. The rho-measure carries an appealing population impact interpretation. Dr. Chen applied the new measure along with other prominent measures to the HIV Prevention Trial Network 052 Study, a landmark trial for HIV/AIDS treatment-as-prevention.
Video link: Here