Audience: Of interest to HTA producers.
What you’ll learn: Participants will: (i) better understand confounding and confounder summary scores; (ii) be able to estimate propensity scores and disease risk scores; (iii) know their strengths and limits; (iv) have discussed their application within HTA.
- Dr. Jason Guertin, Université Laval
- Dr. Mina Tadrous, Women’s College Hospital
Abstract: Interest in real-world evidence (RWE) using real-world data (RWD) for HTAs has recently skyrocketed. RWD can provide researchers and decision-makers insights on the utilization, effectiveness, and cost of technologies; such studies are often referred to as observational studies. Despite their strengths and value, observational studies are prone to numerous types of biases, namely confounding bias. These biases, if uncontrolled for, may lead to erroneous results and mislead researchers and decision-makers. The purpose of this workshop is to introduce participants to two methodologies, propensity scores and disease risk scores, which are frequently used to adjust for confounding within observational studies. Although both methodologies were developed more than 30 years ago, their use and applications remain unclear. Using practical examples based on empirical and simulated data, participants will be exposed to the strengths and weaknesses of the two methodologies and will learn when and how to use them within comparative effectiveness studies and economic evaluations based on RWD.