University of British Columbia

M. Ehsan Karim

Associate Professor in Health Data Science

  • School of Population and Public Health, University of British Columbia
  • Scientist, Centre for Advancing Health Outcomes, St. Paul’s Hospital
  • Associate Member, UBC Department of Statistics

He develops causal inference and machine learning methods that turn large health administrative databases into reliable real-world evidence.

Portrait of Dr. M. Ehsan Karim
MSFHR / Health Research BC Scholar 2018–2023 · AMS UBC OER Champion 2024–2025
Peer-reviewed publications
130+
PI research funding
$1.29M
Trainees supervised
36
Invited presentations
55
Books & open textbooks
6
Citations
2,400+
Biography

About

Dr. M. Ehsan Karim is an Associate Professor in Health Data Science in the School of Population and Public Health at the University of British Columbia, a Scientist at the Centre for Advancing Health Outcomes (St. Paul’s Hospital), and an Associate Member of UBC Statistics. He held a Michael Smith Health Research BC Scholar Award from 2018 to 2023 and was promoted to Associate Professor with tenure in July 2025.

He earned his PhD in Statistics from UBC, supported by an MS Canada studentship, and completed postdoctoral training in epidemiology and biostatistics at McGill University with a fellowship from the Canadian Network for Observational Drug Effect Studies (CNODES).

His research program develops and applies causal inference and machine learning methods to answer real-world comparative effectiveness questions from non-randomized data, with a central focus on large health administrative databases and multiple sclerosis as a key application area. An advocate for open education, he authored the open-access textbook Advanced Epidemiological Methods, teaches UBC’s PhD capstone methods course, and was named a UBC OER Champion for 2024–2025.

Announcements

News & Highlights

  • July 2025

    Promoted to Associate Professor with tenure, UBC School of Population and Public Health.

  • 2025–2029

    Co-leads the CIHR-funded STRIVE-STBBI skills training network ($800,000) for real-world intervention evaluation.

  • February 2025

    Appointed Associate Editor of Pharmacoepidemiology and Drug Safety.

  • 2025

    Received the Editorial Contribution Award and Author Service Award from BMC Medical Research Methodology.

  • 2024–2025

    Named AMS UBC OER Champion for open educational resource contributions.

  • May 2026

    Presenting the “AI and LLMs — Research Perspective” workshop at the Society for Epidemiologic Research Annual Meeting.

Research Program

Research Themes

  • Causal Inference Methods

    Develops and evaluates methods for estimating treatment effects from non-randomized data — including targeted maximum likelihood estimation (TMLE) with single and double cross-fitting, marginal structural models, and propensity score approaches. His work tackles time-dependent confounding, treatment-confounder feedback, and immortal time bias in longitudinal studies.

  • Machine Learning & AI for Health

    Investigates when machine learning genuinely improves causal estimation — from Super Learner weight estimation to deep-learning propensity score architectures such as Dragonnet and autoencoders. The goal is rigorous, verifiable use of AI in health research rather than black-box prediction.

  • Real-World Evidence & Pharmacoepidemiology

    Advances the methodology of real-world evidence generation, including high-dimensional propensity score (hdPS) techniques and their machine learning and doubly robust extensions for residual confounding control. This program also addresses non-adherence and per-protocol effects in pragmatic clinical trials.

  • Health Administrative Data

    Builds statistical strategies for the scale and messiness of population-level administrative databases — proxy adjustment, missing data, survival prediction, and disease-specific comorbidity indices — so routinely collected health data can support trustworthy inference.

  • Multiple Sclerosis Applications

    Applies these methods where they matter: comparative effectiveness of MS therapies, an MS-specific Comorbidity Summary Index (MSCSI), and risk prediction algorithms for the MS prodrome to enable earlier detection. Supported by CIHR and MS Canada.

  • Survey & Longitudinal Methods

    Develops guidance for analyzing nationally representative complex surveys (Canadian, US, and international) and longitudinal cohorts, including propensity score analysis in the complex survey context and small-sample inference for stepped-wedge designs.

Funded by Canadian Institutes of Health Research (CIHR) · Natural Sciences and Engineering Research Council of Canada (NSERC) · MS Canada · Michael Smith Health Research BC · BC SUPPORT Unit

Scholarship

Selected Publications

A selection from 130+ peer-reviewed articles spanning causal inference methodology, machine learning for health, and high-impact applied work.

  1. Towards robust causal inference in epidemiological research: employing double cross-fit TMLE in right heart catheterization data

    Mondol MH, Karim ME.

    American Journal of Epidemiology 194(10), 2813–2819, 2025 Causal inference

  2. Buprenorphine/naloxone vs methadone for the treatment of opioid use disorder

    Nosyk B, Min JE, Homayra F, et al. (incl. Karim ME).

    JAMA 2024 High-impact applied

  3. Key considerations for choosing a statistical method to deal with incomplete treatment adherence in pragmatic trials

    Hossain MB, Karim ME.

    Pharmaceutical Statistics 22(1), 205–231, 2023 Pragmatic trials

  4. Recommendations for the use of propensity score methods in multiple sclerosis research

    Simoneau G, Pellegrini F, Debray T, et al., Karim ME.

    Multiple Sclerosis Journal 28(9), 1467–1480, 2022 Guidance Senior author

  5. Dealing with treatment-confounder feedback and sparse follow-up in longitudinal studies: application of a marginal structural model in a multiple sclerosis cohort

    Karim ME, Tremlett H, Zhu F, Petkau J, Kingwell E.

    American Journal of Epidemiology 190(5), 908–917, 2021 Causal inference

  6. Can we train machine learning methods to outperform the high-dimensional propensity score algorithm?

    Karim ME, Pang M, Platt RW.

    Epidemiology 29(2), 191–198, 2018 Machine learning

  7. Estimating inverse probability weights using super learner when weight-model specification is unknown in a marginal structural Cox model context

    Karim ME, Platt RW, and the BeAMS study group.

    Statistics in Medicine 36(13), 2032–2047, 2017 Machine learning

  8. Comparison of statistical approaches for dealing with immortal time bias in drug effectiveness studies

    Karim ME, Gustafson P, Petkau J, Tremlett H, and the BeAMS study group.

    American Journal of Epidemiology 184(4), 325–335, 2016 Methods

  9. Marginal structural Cox models for estimating the association between beta-interferon exposure and disease progression in a multiple sclerosis cohort

    Karim ME, Gustafson P, Petkau J, et al.

    American Journal of Epidemiology 180(2), 160–171, 2014 SER Top-10 Article of 2014

  10. Association between use of interferon beta and progression of disability in patients with relapsing-remitting multiple sclerosis

    Shirani A, Zhao Y, Karim ME, et al.

    JAMA 308(3), 247–256, 2012 High-impact applied

130+ peer-reviewed publications · 2,400+ citations · h-index 25 (Google Scholar, May 2026)

134 peer-reviewed journal articles, plus 13 refereed conference proceedings.

View all on Google Scholar →
Open Scholarship

Books & Software

Books & open textbooks

Research software

  • svyTable1 Survey-weighted descriptive statistics and diagnostic tables R package · 2025
  • ReSliceTMLE Resampling-based targeted maximum likelihood estimation R package · 2025
  • TB Mortality Risk Calculator Mortality risk prediction for people diagnosed with TB R Shiny app · 2024
  • Crossfit Sample splitting (cross-fit) for TMLE in causal inference R package · 2023
  • simMSM / genMSM / iptw Simulation and weighting tools for marginal structural models R & Stata · 2020
Education

Teaching

  • SPPH 604 — Application of Advanced Epidemiological Methods

    Developer & sole instructor · PhD level · 3 credits

    UBC’s PhD capstone methods course, developed by Dr. Karim and taught since 2018. Covers causal inference, propensity scores, mediation analysis, machine learning, survey data analysis, and missing data — with 65 contact hours of hands-on tutorials bridging the methodological gap for SPPH doctoral trainees.

  • SPPH 381H — Health Data Science: AI and Knowledge Translation

    Developer · 400-level undergraduate

    A comprehensive toolkit for applying data science to health research, featuring “Zero to AI Co-Pilot” pedagogy, a verification-first assessment framework for auditing AI-generated code, and OCAP®/CARE integration for Indigenous data sovereignty.

  • MEDI 504A — Emerging Topics in Experimental Medicine

    Co-instructor · Graduate · 1.5 credits

    Part of a Faculty of Medicine health data science initiative, spanning multi-omics, machine learning and AI in clinical research, and large administrative datasets in population health.

Open education

Most of Dr. Karim’s course content is published as Open Educational Resources, freely available to UBC students and a global audience — including the Advanced Epidemiological Methods textbook and the ML4PH machine learning materials. His OER work has been supported by UBC OER Fund Implementation and Affordability Grants (~$50,000).

Recognition

  • AMS UBC OER Champion Award, 2024–2025 — recognizing open educational resource contributions
  • Faculty of Medicine Merit Awards (teaching, research and service), four consecutive years, 2021–2024
  • More than a dozen conference workshops and webinars delivered nationally and internationally, from SER and the Statistical Society of Canada to R/Medicine
Mentorship

Trainees & Mentorship

Graduate students
22
Postdoctoral fellow
1
Undergraduate trainees
13

Dr. Karim supervises graduate students in both the School of Population and Public Health and the Department of Statistics, building their skills in data science, causal inference, and health analytics. His trainees have completed theses spanning machine learning for survival prediction, the multiple sclerosis prodrome, comorbidity index development, and methods for pragmatic trials — and have gone on to positions at Harvard, McGill, McMaster, and the BC Centre for Disease Control.

Prospective students

Dr. Karim supervises a small number of graduate students in SPPH and Statistics, and takes on new trainees only as capacity and funding allow. Admission is through the SPPH and Statistics graduate programs; contacting him is not a substitute for a formal application.

If you would like to be considered, email ehsan.karim@ubc.ca with a brief note on your research interests and attach:

  • your CV, including a list of publications
  • an unofficial transcript
  • a list of references
Support & Recognition

Funding & Awards

CAD $1.29M in research funding as principal investigator since 2017.

Selected grants as Principal Investigator
Funder · Program Years Project Amount
CIHR · HIV/AIDS & STBBI Clinical Trials Network, Phase 2 2025–2029 Skills Training for Real-world InterVention Evaluation in STBBIs (STRIVE-STBBI) $800,000
Michael Smith Foundation for Health Research · Scholar Award 2018–2023 A causal inference framework for analyzing large administrative health care databases, with a focus on multiple sclerosis $450,000
MS Canada · Discovery Research Grant 2023–2026 Development and validation of the Multiple Sclerosis-specific Comorbidity Summary Index (MSCSI) $199,859
CIHR · Project Grant 2024–2025 A risk prediction algorithm for prodromal multiple sclerosis $153,000
NSERC · Discovery Grant 2018–2025 Improving causal inference methods in statistics for analyzing big data $126,000
BC SUPPORT Unit · Real-World Clinical Trials Methods Cluster 2018–2021 Developing and evaluating causal inference methods for pragmatic trials $100,000
MS Canada · Catalyst Research Grant 2024–2026 Reducing residual confounding in MS research: a machine learning approach $29,976

Selected awards & distinctions

  • Michael Smith Foundation for Health Research Scholar Award, $450,000 (2018–2023)
  • AMS UBC OER Champion Award (2024–2025)
  • Faculty of Medicine Merit Award, UBC — four consecutive years (2021, 2022, 2023, 2024)
  • First-authored article named among the top 10 “2014 Articles of the Year” in the American Journal of Epidemiology — Society for Epidemiologic Research (2015)
  • Editorial Contribution Award and Author Service Award, BMC Medical Research Methodology (2025)
  • Exemplary Reviewer (top 20), Epidemiology (2020)
  • Best Reviewer, Pharmacoepidemiology and Drug Safety (2015 and 2017)
  • Outstanding Academic Performance Award, UBC Faculty of Medicine (2019, 2020)
Academic Community

Service

Editorial roles

  • Associate Editor, Pharmacoepidemiology and Drug Safety (2025–present)
  • Editorial Board Member, BMC Medical Research Methodology (2024–2025)
  • Associate Editor, Journal of Statistical Research (2020–present)

Peer review

Reviewer since 2013 for more than 60 journals in biostatistics, epidemiology, and public health — including Biometrics, Statistics in Medicine, Epidemiology, American Journal of Epidemiology, and BMJ. Recognized as Exemplary Reviewer (Epidemiology, 2020) and Best Reviewer (Pharmacoepidemiology and Drug Safety, 2015, 2017).

Grant panels

CIHR Project Grant committees (2023–2025) and College of Reviewers (2024–2028) · NSERC Discovery Grants, Mathematics and Statistics (2018–2024) · MS Canada · Dutch Research Council (NWO) · MS Research Australia · Health Research Council of New Zealand · Fonds de recherche du Québec.

Conference leadership

Frequent chair and organizer of invited sessions on causal inference and machine learning at JSM, the Statistical Society of Canada, CSEB, and ICSA-Canada.

Correspondence

Contact

ehsan.karim@ubc.ca

Get in touch — email is the fastest way to reach him.

Mailing address

School of Population and Public Health, UBC
216 – 2206 East Mall
Vancouver, BC V6T 1Z3, Canada