Session Chair: Neal Goldstein, Drexel University
Increasingly data mined from the electronic health record (EHR) are being used in epidemiological research. But more data does not equate to better quality research. In this workshop, we will cover the basics of working with EHRs and designing valid epidemiological analyses. The workshop will be a mix of didactic lecture and interactive group exercises. Participants are requested to provide planned or active research questions in advance, as these will form the basis of breakout group exercises.
Lecture topics will include:
1. Designing and analyzing epidemiological studies using EHR data for both inpatient and outpatient settings.
2. Obtaining data from the EHR, including data export, linkage, and variable manipulation (e.g. parsing data from free text).
3. Architecture of the EHR and terminology/data standards.
4. Understanding the clinical population and how this relates to a target/general population.
5. Common pitfalls in working with EHR data and resources for additional reference.
Audience: Researchers interested in EHR data, including proposed and active research projects; students and trainees to seasoned investigators welcome.
Increased interactions with diverse peers enhance students’ educational experiences and bring measurable improvements in learning outcomes for all. Diversity also contributes to the scientific rigor of our scholarship and are necessary for the longevity and robustness of our discipline. Positive classroom climates and teaching practices have been shown to improve persistence and academic and emotional development among diverse students. As instructors, we have a responsibility to level the playing field, so that every student has an equal opportunity to master the learning objectives in our courses.
Building on the wealth of scholarship produced by our colleagues in the social sciences this half-day workshop will employ an active learning approach to developing inclusive classrooms and curricula in Epidemiology (e.g. lectures followed by small group discussions and revising existing syllabi). The following domains are key to the infusion of inclusivity in our courses 1) minding the privilege gap between our students and ourselves when developing our courses, 2) acknowledging and confronting implicit biases, and 3) mitigating stereotype threat in our classrooms. The workshop will feature several faculty serving as presenters and facilitators including: Yvette Cozier (Boston University), Chanelle Howe (Brown University), Sharon Schwartz (Columbia University), Renee Johnson (Johns Hopkins University), Sophie Godley (Boston University), Shawnita Sealy-Jefferson (Ohio State University), Candice Belanoff (Boston University), Seth Prins (Columbia University) and Anjum Hajat (University of Washington). It has been developed in conjunction with the SER Diversity and Inclusion committee.
This workshop applies the Causal Roadmap to estimate the causal effects with multiple intervention variables, such as the cumulative effect of an exposure over time, controlled direct effects, and effects on survival-type outcomes with right-censoring. We will cover longitudinal causal models, identification in the presence of time-dependent confounding; and estimation of joint treatment effects using G-computation, inverse probability weighting (IPW), and targeted maximum likelihood estimation (TMLE). During the workshop session, participants will work through the Roadmap using an applied example and implement these estimators with the ltmle R package. Prior training in causal inference in a single time-point setting is recommended, but not required.
The workshop will consider sensitivity analysis for different forms of bias in epidemiology. It will begin with confounding, focusing on a new metric to evaluate sensitivity to unmeasured confounding called the E-value. The E-value is the minimum strength of association, on the risk ratio scale, that an unmeasured confounder would need to have with both the exposure and the outcome, conditional on the measured covariates, to fully explain away the exposure-outcome association. E-value calculations for risk ratios, outcomes differences, odds ratios, and hazard ratios will be discussed. The E-value can be calculated in a straightforward way from study results and its use could help unify assessment of unmeasured confounding. The workshop will proceed by describing very recent analogous easy-to-implement approaches to also address differential measurement error and selection bias. We will conclude by presenting recent extensions allowing sensitivity analysis for all three forms of bias. The methods, taken as a whole, will constitute a straightforward comprehensive approach to bias analysis.
Session Chair: Steve Mooney, University of Washington
This workshop will introduce participants to the R statistical computing platform for use in epidemiologic analysis. It is not intended to transform untested novices into R wizards in a mere half-day; rather, the goal will be to introduce the conceptual underpinnings, tools, and external resources that participants will need to overcome barriers to using R that they might encounter on their own, later. The material is designed for epidemiologists who are already familiar performing analyses using other statistical software (e.g. SAS/Stata/SPSS) but who have no first-hand experience with the R language. More specifically, the course will cover 1) basic R syntax, 2) importing data, 3) constructing, cleaning, and manipulating data objects, 4) loading and using external packages, 5) simple statistical modeling, and 6) graphics. Participants must bring a laptop with R installed; the instructor will be available by email beforehand to assist with R installation if difficulties arise.