One of the primary objectives of CAN-AIM is to develop and validate new methods that will enhance the analyses of prospective studies of drug safety and effectiveness.

In 2011-2014, the methodology experts among our team members, together with our trainees and collaborators, have continued active research on the development, validation and applications of novel statistical and epidemiological methods directly relevant to CAN-AIM objectives.

Specifically, our recent research activities focused mostly on two major methodological challenges, essential for the validity, accuracy and/or efficiency of the analyses of prospective, observational, longitudinal post-marketing studies of real-life safety and effectiveness of medications: (i) improvement of the accuracy of drug exposure modeling; and (ii) controlling for biases due to different sources of confounding.

In addition, we have initiated research, with our trainees, on more specialized, and relatively less studied, methodological issues frequently encountered in real-life studies of adverse or intended effects of drugs, related to (iii) reducing residual confounding, due to mis-modeling the confounder effects; (iv) exposure measurement errors; and (v) accounting for measurement problems or uncertainty regarding the outcomes.


  1. Flexible modeling of time-varying drug use:
    1. Abrahamowicz M, Beauchamp M-E, Sylvestre M-P. Comparison of alternative models for linking drug exposure with adverse effects.  Statistics in Medicine. 2012 May 20;31(11-12):1014–1030.
  2. Extending  the weighted cumulative exposure (WCE) method to marginal structural models: (Cumulative effects of time-varying exposures )
    1. Xiao Y, Abrahamowicz M, Moodie EEM, Weber R, Young J. Flexible marginal structural models for estimating the cumulative effect of a time-dependent treatment on the hazard: reassessing the cardiovascular risks of didanosine treatment in the Swiss HIV cohort study. Journal of the American Statistical Association. 2014 Jun;109(506):455-464.
    2. Xiao Y, Abrahamowicz M, Moodie EEM. Comparison of approaches to weight truncation for marginal structural Cox models. Epidemiologic Methods. 2013 May;2(1):1-20.
    3. Kyle RP, Moodie EEM, Klein MB, Abrahamowicz M. Correcting for measurement error in time-varying covariates in marginal structural models. American Journal of Epidemiology. 2016 Aug;184(3):249-258.
    4. Burne RM, Abrahamowicz M. Adjustment for time-dependent unmeasured confounders in marginal structural Cox models using validation sample data. Statistical Methods in Medical Research. 2019 Feb;28(2):357-371.
    5. Bally M, Beauchamp M-E, Abrahamowicz M, Nadeau L, Brophy JM. Risk of acute myocardial infarction with real-world NSAIDs depends on dose and timing of exposure. Pharmacoepidemiology & Drug Safety. 2018 Jan;27(1):69-77.
    6. Danieli C, Abrahamowicz M. Competing risks modeling of cumulative effects of time-varying drug exposures. Statistical Methods in Medical Research. 2019;28(1):248-262.
  3. Improving efficiency of instrumental variables (IV) corrections for unmeasured confounding:
    1. Abrahamowicz M, Beauchamp M-E, Ionescu-Ittu R, Delaney JCA, Pilote L.  Reducing the variance of the prescribing preference-based instrumental variable estimates of the treatment effect.  American Journal of Epidemiology.  2011 Aug 15;174(4):494-502. 
    2. Ionescu-Ittu R, Abrahamowicz M, Pilote L. Treatment effect estimates varied depending on the definition of the provider prescribing preference-based instrumental variables. Journal of Clinical Epidemiology. 2012 Feb; 65(2):155-162. 
    3. Liu A, Abrahamowicz M, Siemiatycki J. Conditions for confounding of interactions. Pharmacoepidemiology & Drug Safety. 2016 Mar; 25(3):287-296. 
    4. Burne RM, Abrahamowicz M. Martingale residual-based method to control for confounders measured only in a validation sample in time-to-event analysis. Statistics in Medicine. 2016 Nov; 35(25):4588-4606. 
    5. Delcoigne B, Colzani E, Prochazka M, Gagliardi G, Hall P, Abrahamowicz M, Czene K, Reilly M. Breaking the matching in nested case-control data offered several advantages for risk estimation. Journal of Clinical Epidemiology. 2017 Feb; 82:79-86. 
  4. Validating multi-stage approaches to test for treatment effect modifications and for subgroup-specific treatment effects: 
    1. Abrahamowicz M, Beauchamp ME, Fournier P, Dumont A. Evidence of subgroup-specific treatment effect in the absence of an overall effect: is there really a contradiction? Pharmacoepidemiology and Drug Safety. 2013 Nov;22(11):1178-88.
  5. Extending the weighted cumulative exposure (WCE) method to active pharmaco-vigillance: 
    1. van Gaalen R, Abrahamowicz M, Buckeridge DL. The impact of exposure model misspecification on signal detection in prospective pharmacovigilance. Pharmacoepidemiology and Drug Safety. 2015 May;24(5):456-67.
  6. Novel “Missing cause’ approach (alternative to IV’s) to control for unmeasured confounding:
    1. Abrahamowicz M, Bjerre L, Beauchamp M-E, LeLorier J, Burne R. The missing cause approach to unmeasured confounding in pharmacoepidemiology. Statistics in Medicine. 2016 Mar;35(7):1001-16.
  7. Multiple exposure-risk models:
    1. van Gaalen RD, Abrahamowicz M, Buckeridge DL. Using multiple pharmacovigilance models improves the timeliness of signal detection in simulated prospective surveillance. Drug Safety. 2017 Nov;40(11):1119-1129.
    2. Kollhorst B, Abrahamowicz M, Pigeot I. The proportion of all previous patients was a potential instrument for patients’ actual prescriptions of nonsteroidal anti-inflammatory drugs. Journal of Clinical Epidemiology. 2016 Jan; 69:96-106.