HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Gary L. Grunkemeier Ruyun Jin Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Grunkemeier, G. L. Articles by Jin, R. Search for Related Content PubMed PubMed Citation Articles by Grunkemeier, G. L. Articles by Jin, R. Related Collections Education

Ann Thorac Surg 2007;83:1934-1939
© 2007 The Society of Thoracic Surgeons

---

The Statistician's Page

Power and Sample Size: How Many Patients Do I Need?

Providence Health & Services, Portland, Oregon

^_* Address correspondence to Dr Jin, 9205 SW Barnes Rd., Suite 33, Portland, OR 97225 (Email: ruyun.jin@providence.org).

The first 300 words of the full text of this article appear below.

	Introduction

 
In our combined consulting experience (42 statistician-years), the two most frequently asked questions are: (1) "Is this result statistically significant?" and (2) "How many patients do I need for my study?" In this expository article, we will attempt to provide an understanding of how the statistical significance of a completed study is determined and how, by adding the notion of power, sample sizes for a future study can be estimated. We start with a test of statistical significance, because that is the number 1 question, and it also introduces some of the concepts needed to answer question number 2. We will emphasize graphical portrayals of these concepts, and relegate statistical details to Appendix 1. Appendix 2 lists some examples of software available for sample size calculation.

	Completed Study

 
There is current interest in postoperative atrial fibrillation (AF), and also current controversy about the role of aprotinin in cardiac surgery. As an illustrative data set, we will use all 1,730 isolated heart valve replacements at Providence St. Vincent Medical Center, Portland, Oregon from 1998 to 2005 in patients without preoperative AF. The numbers of valves implanted with and without the use of aprotinin were about equal (835 and 895, respectively), but the percentage of aprotinin usage increased with time. The incidence of new AF was 21.3% (178 of 835) with aprotinin and 25.9% (232 of 895) without aprotinin, a difference of –4.6%. This result is not risk-adjusted, and it is used for demonstration purposes only. Yet the results agree remarkably with a recent article on thoracic aortic surgery that showed a lower incidence of post-surgical arrhythmia (mostly AF) with the use of aprotinin (20%) compared with the matched controls (25%) [1].

Determining Statistical Significance
The statistical significance of this observed difference is determined as follows. We cannot prove that the true difference is . . . [Full Text of this Article]

This article has been cited by other articles:

				M. A. Mohammed and J. J. Deeks In the Context of Performance Monitoring, the Caterpillar Plot Should Be Mothballed in Favor of the Funnel Plot Ann. Thorac. Surg., July 1, 2008; 86(1): 348 - 348. [Full Text] [PDF]