“Regression to the mean” is a very important concept with significant clinical implications. Most chronic diseases have a fluctuating clinical course. For example, patients with chronic back pain will have pain that varies from day to day and can be exacerbated by certain postures and/or activities. Patients will usually contact a physician when their symptoms are most severe. Therefore if nothing were to be done, it is likely but not certain,that the symptoms would improve over time as the severity reduces or regresses to its mean (average) level. It is therefore not uncommon for patients’ symptoms to improve by the time they see their physician. This does not mean that no intervention should be done. However it does make it difficult to determine what if any effect an intervention is actually having on a clinical outcome. Regression to the mean, as well as the powerful placebo effect (the topic of another future blog), are factors that can exert significant effects that are difficult to control for. One way to do so is to design rigorous clinical double blind randomized controlled trials that subject groups of patients that are statistically similar to protocols that differ in a specific manner without the patients or evaluators being aware of this treatment difference. By doing so, any difference in outcome can be attributed to the treatment given since the effect of other factors should average out. Regression to the mean is a very important concept to keep in mind when evaluating and treating patients with chronic diseases whose clinical manifestations fluctuate significantly over time.
A.G. Barnett, J.C. van der Pols, and A.J. Dobson. Regression to the mean: what it is and how to deal with it. The International Journal of Epidemiology: 34: 215-220. 2004.
- Ariely. Predictably Irrational, Revised and Expanded Version: The Hidden Froces That Shape Our Decisions. Harper Collins. 2009.