Am J Health-Syst Pharm
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 QUICK SEARCH:   [advanced]


     


This Article
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Inciardi, J.
Right arrow Articles by McMahon, K
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Inciardi, J.
Right arrow Articles by McMahon, K
American Journal of Health-System Pharmacy, Vol 59, Issue 10, 968-972
Copyright © 2002 by American Society of Health-System Pharmacists


Articles

Using correspondence analysis in pharmacy practice

JF Inciardi, T Stijnen, and K McMahon


Correspondence analysis (CA) and some of its uses in pharmacy practice are described. CA is a multivariate and graphic form of exploratory data analysis that reduces a multidimensional contingency table to a two-dimensional plot with minimal loss of information. An exploratory association between medication use and the number and severity of falls among the elderly living at home is used to illustrate the process of CA. Row profiles are constructed by dividing the count data in each cell by the total of the corresponding row, converting frequency data into relative frequencies. A graph of the data can be made by recalculating the frequencies to illustrate how the data might appear in a three-dimensional space with the axes defined by the three categories of falling (minimum, moderate, and major). For example, the number of patients using central nervous system (CNS) agents is plotted using the relative frequencies as vectors to locate a position in the three-dimensional space. All drug profiles and the average profile will lie exactly in a triangular plane defined by the terminal points of each axis. In this manner, data with three dimensions can be projected onto a two-dimensional space. Transferring the vectors named by pharmacologic class onto the plot requires creating two axes that cross at a common point or the origin. This point locates the average profile and defines an important reference for making comparisons. CA revealed that CNS agents may be associated with moderate to major falls, psychotherapeutic agents with moderate falls, and anticoagulants with moderate to minimum falls. CA has widespread uses in pharmacy practice. It can identify patients at risk for serious but preventable drug-related complications, enabling pharmacists to allocate pharmacy resources to the areas in most need and suggest hypotheses for future research. Clinicians can also use CA to analyze vast amounts of patient-related data to uncover hard-to-detect associations. The use of CA in pharmacy practice will allow new strategies for improved patient care to be more readily appreciated and implemented.
 






HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Copyright © 2002 by the American Society of Health-System Pharmacists.