This Article Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Rao, S. R. Articles by Schoenfeld, D. A. Search for Related Content PubMed PubMed Citation Articles by Rao, S. R. Articles by Schoenfeld, D. A. Related Collections Other Treatment

(Circulation. 2007;115:109-113.)
© 2007 American Heart Association, Inc.

Statistical Primer for Cardiovascular Research

Survival Methods

Sowmya R. Rao, PhD; David A. Schoenfeld, PhD

From Massachusetts General Hospital Biostatistics Center and Institute for Health Policy (S.R.R.), and Massachusetts General Hospital Biostatistics Center, Department of Biostatistics, Harvard School of Public Health (D.A.S.), Boston, Mass.

Correspondence to Sowmya R. Rao, PhD, Massachusetts General Hospital, Biostatistics Center, 50 Staniford St, Suite 560, Boston, MA 02114. E-mail srrao@partners.org

Key Words: proportional hazards models • Kaplan-Meiers estimate • survival

An extract of the first 250 words of the full text is provided, because this article has no abstract.

	Introduction

 
This article gives an overview of survival methods in medical studies. We briefly describe survival data and discuss the methods used for analysis of such data. We apply these methods to data from a clinical trial and discuss the results. Survival methods are applicable when the measure of interest is time to an event such as mortality or occurrence of disease. The concept of censoring makes survival methods unique. If a patient goes through the study without having the event, his time to the event is (right) censored, in the sense that we only know that the event happened after the last time we observed the patient. Thus, for each patient we have 2 pieces of data: The first is a time that is either the patient’s event time or the time that the patient was last followed up, and the second is an indicator that denotes whether the time is an event time or a follow-up time. Another way of thinking of censoring is to assume that each patient has an event time and a censoring time after which the patient would no longer be observed. Whenever the censoring time is less than the event time, the event time is missing. Survival methods also assume that the censoring time is unknown when it is greater than the event time.

Survival distributions are usually described in terms of 2 functions: the survival function, S(t), defined as the probability that a person survives past a specified time t; and the hazard . . . [Full Text of this Article]

This article has been cited by other articles:

				M T Streppel, M C Ocke, H C Boshuizen, F J Kok, and D Kromhout Long-term wine consumption is related to cardiovascular mortality and life expectancy independently of moderate alcohol intake: the Zutphen Study J Epidemiol Community Health, July 1, 2009; 63(7): 534 - 540. [Abstract] [Full Text] [PDF]

				C. Mehta, P. Gao, D. L. Bhatt, R. A. Harrington, S. Skerjanec, and J. H. Ware Optimizing Trial Design: Sequential, Adaptive, and Enrichment Strategies Circulation, February 3, 2009; 119(4): 597 - 605. [Full Text] [PDF]

				K. L. Lunetta Genetic Association Studies Circulation, July 1, 2008; 118(1): 96 - 101. [Full Text] [PDF]

				D. Oakes and D. R. Peterson Survival Methods: Additional Topics Circulation, June 3, 2008; 117(22): 2949 - 2955. [Full Text] [PDF]