Causal Mediation Analysis Training: Methods and Applications Using Health Data

COVID-19 UPDATE: THE 2020 Causal Mediation Analysis Training WILL BE HELD REMOTELY VIA LIVE-STREAM, August 12-14 BEGINNING AT 10AM EDT.


Causal Mediation Analysis and Causal Inference TrainingCapacity reached! Join the waitlist for the next live-stream, virtual Causal Mediation Analysis Training: August 12-14, 2020


The Causal Mediation Analysis Training is a 3-day intensive boot camp of seminars and hands-on analytical sessions to provide an overview of concepts and data analysis methods used to investigate mediating mechanisms.


Subscribe for updates on registration and scholarship dates, deadlines, and announcements.



Summer 2020 dates: Live-stream, online training August 12-14, 2020; 10:00am - 5:00pm EDT

Mediation analysis is an emerging field in causal inference relevant for comparative effectiveness research, evaluating and improving policy recommendations, and explaining biological mechanisms. Training in the potential outcomes framework for causal inference is important to understand the assumptions required for valid mediation analyses. This course will equip participants with foundational concepts and cutting edge statistical tools to investigate mediating mechanisms.

This three-day intensive course will cover some of the recent developments in causal mediation analysis and provide practical tools to implement these techniques and assess the mechanisms and pathways by which causal effects operate. Led by a team of experts in causal mediation techniques at Columbia University, this course will integrate lectures and discussion with hands-on computer lab sessions using R and Stata. The course will cover the relationship between traditional methods for mediation in environmental health, epidemiology, and the social sciences and new methods in causal inference using a wide variety of examples to illustrate the techniques and approaches. We will discuss 1) when the standard approaches to mediation analysis are valid for dichotomous, continuous, and time-to-event outcomes, 2) alternative mediation analysis techniques when the standard approaches will not work, using ideas from causal inference and natural direct and indirect effects 3) the no-unmeasured confounding assumptions needed to identify these effects, and 4) how regression approaches for mediation analysis can be extended in the presence of multiple mediators.

By the end of the workshop, participants will be familiar with the following topics:

  • Understand when traditional methods for mediation fail
  • Articulate concepts about mediation under the counterfactual framework and assumptions for identification
  • Formulate and apply regression approaches for mediation for single and multiple mediators
  • Develop facility with use of software for mediation and interpretation of software output

Investigators at all career stages are welcome to attend, and we particularly encourage trainees and early-stage investigators to participate.


  1. Each participant must be familiar with linear and logistic regression.
  2. Each participant must have experience with programming in Stata and/or R. Data labs will alternate using platforms to show methods.
  3. Although the instructors will provide an overview of the fundamentals of causal inference (potential outcomes, directed acyclic graphs, and marginal structural models), we invite the participants to read chapters 1-7, 11, and 12 of Hernán MA, Robins JM (2020). Causal Inference: What If. Boca Raton: Chapman & Hall/CRC (free).
  4. Each participant is required to bring a personal laptop with R/RStudio installed prior to the first day of the workshop, as all lab sessions will be done on your personal laptop. R is available for free download and installation on Mac, PC, and Linux devices. You will receive a Stata license for the duration of the training.


Linda Valeri, PhD, Mailman School of Public Health, Columbia University. Linda Valeri, is Assistant Professor of Biostatistics at Columbia University Mailman School of Public Health. Dr. Valeri is an expert in causal inference with a focus on statistical methods for causal mediation analysis, measurement error, and missing data. She is interested in translating statistical methods in public health to improve our understanding of mental health, environmental determinants of health, and health disparities. Dr. Valeri is also a passionate teacher. In the past five years she has been teaching full semester as well as short courses on causal mediation analysis at premier academic institutions such as Columbia University, Harvard University, University of Michigan and Erasmus Universiteit Rotterdam (Netherlands).

Caleb Miles, PhD, Mailman School of Public Health, Columbia University. Caleb Miles is an Assistant Professor of Biostatistics at Columbia University Mailman School of Public Health. Dr. Miles works on developing semiparametric methods for mediation analysis and other branches of causal inference, and applying them to problems in public health. His applied work has largely been in HIV/AIDS, and he has more recently begun to work on psychiatric applications. He is currently working on methods to account for measurement error in mediation analysis, and developing hypothesis tests for mediated effects with improved power.


Kosuke Imai, PhD, Harvard University. Kosuke Imai is a Professor in the Department of Government and the Department of Statistics at Harvard University. He is also an affiliate of the Institute for Quantitative Social Science where his primary office is located. Before moving to Harvard in 2018, Dr. Imai taught at Princeton University for 15 years where he was the founding director of the Program in Statistics and Machine Learning. He specializes in the development of statistical methods and their applications to social science research and is the author of Quantitative Social Science: An Introduction (Princeton University Press, 2017). Outside of Harvard, Dr. Imai served as the President of the Society for Political Methodology from 2017 to 2019. He is also Professor of Visiting Status in the Graduate Schools of Law and Politics at The University of Tokyo.


Mark van der Laan, PhD, University of California, Berkeley, Keynote Speaker 2019.


Training scholarships are available for the Causal Mediation Training.


COVID-19 Update: The Causal Mediation Analysis Training will no longer take place in person due to the COVID-19 pandemic. The Training will instead be a live-stream, remote training that takes place over live, online video on August 12-14, 2020 from 10am EDT - 5pm EDT. Please note this training is not a self-paced, pre-recorded online training. 


This training provided really comprehensive instruction on causal mediation, which was enhanced by the hands-on lab. Overall, I felt like the gained the tools to implement causal mediation techniques in my own work. - Faculty member at University of Washington School of Public Health, 2019 

Both instructors did a great job in pacing the material and explaining starting from the foundation. There were also some great examples that were being explored.Kaylee H., Statistician at Weill Cornell Medicine, 2019 

The format of the training with theory on Day 1, and practicum on day 2-3 afternoon was great; very useful for adult learning.  The instructors were very patient with the wide range of skill levels of the trainees balancing the depth of teaching with the assistance in basic skills. - Vidhu T., Assistant Professor at Columbia University, 2019

The workshop was informative, the lecturers were knowledgeable and the hands-on component (lab) facilitated application of theoretical principles to techniques. - Lillian P, Postdoctoral research fellow at Columbia University, 2019

This was a highly useful, immediately applicable training led by excellent instructors. They provided an effective overview of underlying concepts (counterfactuals, directed acyclic graphs, marginal structural models), and in-depth instruction on various methods for assessing mediation, as well as opportunities to practice applying these methods using multiple statistical software programs. - Sarah L., Epidemiologist


COVID-19 Update: With the training being offered virtually instead of in-person, we are passing along any and all costs saved to attendees.

 Early-Bird Rate (through 6/15/20 6/22/20)Regular Rate (6/16/20 6/23/20-8/1/20)Columbia Discount*
Student/Postdoc/Trainee  $1,495 $1,095 $1,695 $1,24510%
Faculty/Academic Staff/Non-Profit Organizations$1,695 $1,245$1,895 $1,39510%
Corporate/For-Profit Organizations​$1,895 $1,395$1,995 $1,545NA

*Columbia Discount: This discount is valid for any active student, postdoc, staff, or faculty at Columbia University. To access Columbia discount, email for instructions and specify if you are paying by credit card, or internal transfer within Columbia.

Invoice Payment and Group Registrations: If you would prefer to pay by invoice/check, or would like to pay for a group of registrants, please email with details.

Registration Fee: Fee includes course material, breakfast, lunch, and refreshment breaks. Course material will be available to all students after the workshop. Lodging and transportation are not included.  

Cancellations: For summer 2020, no administrative fees will be assessed due to the evolving COVID-19 situation. Cancellation notices must be received via email at least 14 days prior to the workshop start date in order to receive a full refund. Please email your cancellation notice to Due to workshop capacity and preparation, we regret that we are unable to refund registration fees for cancellations after these dates, unless a new COVID-19 restriction is implemented that impedes virtual attendance, in which case any registration cancellation <14 days prior to a training related to COVID-19 restriction beyond your control (institutional policy, shift in work responsibilities, etc.) will be fully refunded and no administrative fee will be assessed. Because of the significant resources required to develop these trainings, you will be asked to submit supporting documentation (e.g. employer email notice, local regulations, etc.) for any COVID-19 related cancellation <14 days before a given training. 

If you are unable to attend the training, we encourage you to send a substitute within the same registration category. Please inform us of the substitute via email at least one week prior to the training to include them on attendee communications, updated registration forms, and materials. Should the substitute fall within a different registration category your credit card will be credited/charged respectively. Please email substitute inquiries to In the event Columbia must cancel the event, your registration fee will be fully refunded. 





Want updates on new Training details or registration deadlines? Subscribe here.

Questions? Email the CMA team at

The Causal Mediation Analysis Training is hosted by Columbia University's SHARP Program in the Mailman School of Public Health.