Analysis of covariance: A useful tool for the pharmacologist to reduce variation and improve precision using fewer animals. (28th April 2022)

• Record Type: Journal Article
• Authors: Stanley, Philip
• Other Names: Stephens Gary guestEditor.; Shukla Arun guestEditor.
• Is Part Of: British journal of pharmacology. Volume 179:Number 14(2022)
• Page Start: 3645
• Page End: 3650
• Publication Date: 2022-04-28
• Subjects: Pharmacology -- Periodicals; Chemotherapy -- Periodicals; Drug Therapy -- Periodicals; Pharmacology -- Periodicals; 615.1
• DOI: 10.1111/bph.15855
• Languages: English
• ISSNs: 0007-1188

Abstract

Abstract : Commonly employed methods for reducing unwanted variation in pharmacology studies, such as data normalisation to baseline or control values, are suboptimal and potentially detrimental. This article highlights the value of using a technique called analysis of covariance (ANCOVA) to incorporate supplementary measurements in the analysis to improve the precision of the treatment comparisons and reduce the required sample size. This technique is not new, but unfortunately, it has been under‐utilised in the statistical analysis of pharmacological data. ANCOVA is an extension of the methods of analysis of variance (ANOVA) and simply requires that we take additional measurements that are statistically related to the response measurements but which are themselves unaffected by the treatments being studied. Most often, in pharmacological studies, these additional measurements would be the baseline or pre‐dose reading corresponding to the response measurement. As many pharmacologists routinely make use of baseline readings in their analysis through normalisation, this article describes a more appropriate method to account for the variation in the baseline readings.