Critical Assessment of Research in Addiction (2025-2026 Recording)-Critical Assessment of Research in Addiction Recording

Video Transcription

Video Summary

The lecture, given by epidemiologist Dr. Ryan Cook of OHSU, focused on how to critically assess addiction research and how clinicians contribute to strong study design. He outlined a stepwise process: start with a clear research aim, translate it into a specific, testable hypothesis, choose a study design that matches the question, collect the right data, and then analyze it appropriately. He emphasized that addiction medicine clinicians are essential partners in research because they understand the clinical realities and can identify the key variables needed for valid studies.<br /><br />Dr. Cook reviewed major study designs, from commentaries and case reports to cross-sectional, cohort, case-control, and randomized controlled trials (RCTs), explaining the strengths and limitations of each. He highlighted that observational studies are often necessary in addiction because many exposures cannot be randomized ethically, but they are vulnerable to bias. He also discussed evidence synthesis through systematic reviews and meta-analyses.<br /><br />A major portion of the talk covered bias, especially confounding, selection bias, and information bias. He explained that confounding is a common challenge in addiction research and cannot be solved by statistical tests alone; it must be addressed in study design and variable selection. He also discussed selection bias, especially loss to follow-up, and information bias from imperfect measurement tools like urine drug screens.<br /><br />Finally, he distinguished between confounders, mediators, and effect modifiers, stressing that each plays a different role in analysis. His main message was to be skeptical but not pessimistic when reading addiction research, and to rely on team science, careful planning, retention efforts, and clinically informed interpretation.

Keywords

addiction research

epidemiology

study design

hypothesis

observational studies

randomized controlled trials

confounding

selection bias

information bias

systematic review

meta-analysis

cohort study

case-control study

clinical research

urine drug screens