ISSN:
1432-1041
Keywords:
blood cholesterol
;
hypolipidemic drugs
;
regression to the mean
;
clinical trial design
;
beforeafter measurements
Source:
Springer Online Journal Archives 1860-2000
Topics:
Chemistry and Pharmacology
,
Medicine
Notes:
Summary The phenomenon of regression to the mean is widespread. It may affect any quantitative biological data in which there is “within subject” variability. It is demonstrated, as an example in practice, by a shift from high to lower values with time in a selected subset of patients with an abnormally high blood cholesterol. The phenomenon will influence experiments involving before-after measurements of a continuous variable. An incorrect conclusion may be reached if investigators do not take regression to the mean into account when designing a clinical trial of hypolipidemic therapy. The reality of the phenomenon is illustrated by a prospective study.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1007/BF00561905
Permalink