A beloved college statistics professor walked into class one day and announced, “Statistics is the science that says that if your head is on fire and your feet are on ice then, on average, you are quite comfortable.”
We roared with laughter at this hyperbolic reductionism, but she had introduced our class of statistics novices to an important concept: exclusively focusing on averages can be misleading when variance around that point estimate is substantial. This almost self-evident notion has important implications for contemporary cardiovascular science. Contemporary clinical trials randomize tens of thousands of patients to new treatments and are designed to precisely and accurately estimate the average effect across the trial-eligible population. But clinicians are seldom concerned about the average effect; rather, what we need is an estimate of effectiveness and safety for the individual patient in our care or for a subgroup of trial-eligible patients with similar characteristics. The development of pragmatic trials, access to large repositories of real-world data, and advances in statistical techniques are only now helping us measure and interpret heterogeneity in the effectiveness and safety of novel interventions.
Loccoh E, Kazi DS. Coronary Artery Calcium Score–Guided Statin Therapy for Primary Prevention in African American Adults—One Size Does Not Fit All. JAMA Cardiol. Published online May 13, 2020. doi:10.1001/jamacardio.2020.1260
Coronavirus Resource Center
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: