Statistics and Biostatistics seminar series
Room: M3 3127
Considerations when Discussing Causal Concepts with Applied Researchers
Many biostatisticians collaborate with applied researchers who have varied knowledge of statistics. Even among those with great understanding, discussion of causal inference concepts can be quite tricky.
In this talk, I will discuss the lifecourse of a series of related projects, each of which required discussing a misconception of a common causal concept:
1) drawing causal inferences from randomized experiments versus observational studies; 2) defining the target estimand; and 3) trade-offs between matched studies and inverse probability weighting. The motivating example will be a series of projects designed to study the feasibility and effectiveness of integrating Omron VitalSightTM, a blood pressure monitoring device that automatically transmits blood pressure readings to the electronic health record, into the health system to assess feasibility and initial effects in
(1) uncontrolled (last two office blood pressures ≥140/90 mm Hg) and (2) general (diagnosed hypertension or last office blood pressure ≥140/90 mmHg) hypertension patient populations.