Placing Trials in Context Using Bayesian Analysis

JAMAPublished 15 March 1995

James M. Brophy

DOI10.1001/jama.1995.03520350053028

Source URL

Citations151

SJR quartileQ1

SJR score5.35

SNIP10.71

TL;DR

This analysis suggests that the clinical superiority of tissue-type plasminogen activator over streptokinase remains uncertain, and the usefulness of the Bayesian approach is demonstrated using the results of the recent GUSTO study of various thrombolytic strategies in acute myocardial infarction.

Abstract

Standard statistical analyses of randomized clinical trials fail to provide a direct assessment of which treatment is superior or the probability of a clinically meaningful difference. A Bayesian analysis permits the calculation of the probability that a treatment is superior based on the observed data and prior beliefs. The subjectivity of prior beliefs in the Bayesian approach is not a liability, but rather explicitly allows different opinions to be formally expressed and evaluated. The usefulness of this approach is demonstrated using the results of the recent GUSTO study of various thrombolytic strategies in acute myocardial infarction. This analysis suggests that the clinical superiority of tissue-type plasminogen activator over streptokinase remains uncertain.

Keywords

MathematicsMedicineEconomics, Econometrics and Finance

New England Journal of MedicineAn International Randomized Trial Comparing Four Thrombolytic Strategies for Acute Myocardial Infarction

4,059 Citations1993The GUSTO Investigators

The findings of this large-scale trial indicate that accelerated t-PA given with intravenous heparin provides a survival benefit over previous standard thrombolytic regimens.

View PDF

New England Journal of MedicineThe Importance of Beta, the Type II Error and Sample Size in the Design and Interpretation of the Randomized Control Trial

1,641 Citations1978Jennie A. Freiman, Thomas C. Chalmers +2 more

The LancetIn-hospital mortality and clinical course of 20 891 patients with suspected acute myocardial infarction randomised between alteplase and streptokinase with or without heparin

620 Citations1990The International Study Group

In a study with 2 x 2 factorial design, 20,891 patients with suspected acute myocardial infarction of less than 6 h duration were randomly allocated to alteplase (recombinant tissue plasminogen activator, tPA) or streptokinase (SK) and to subcutaneous Heparin, beginning 12 h after the start of thrombolytic therapy or no heparin.

Infectious Diseases NewsletterStatistical analysis and the illusion of objectivity

420 Citations1988

It is argued that reaching sensible conclusions from statistical analysis of these data may require subjective input, and to acknowledge the subjectivity inherent in the interpretation of data is to recognize the central role of statistical analysis as a formal mechanism by which new knowledge can be integrated with exist ing knowledge.

JAMAAre All Significant P Values Created Equal?

199 Citations1987Warren S. Browner

Knowing the P value and power, or the confidence interval, for the results of a research study is necessary but insufficient: the reader must estimate the prior probability that the research hypothesis is true.

Annals of Internal MedicineClinical Trials and Statistical Verdicts: Probable Grounds for Appeal

138 Citations1983George Diamond, JAMES S. FORRESTER

Classic analysis is most misleading when the hypothesis in question is already unlikely to be true, when the baseline event rate is low, or when the observed differences are small.

Annals of Internal MedicineLarge-Scale Trials of Thrombolytic Therapy for Acute Myocardial Infarction: GISSI-2, ISIS-3, and GUSTO-1

58 Citations1993Paul M. Ridker, Christopher O'Donnell +2 more

Three large-scale randomized trials directly compared the thrombolytic agents most commonly used in the United States show, at most, small absolute differences among agents, in terms of both lives saved and major complications (including hemorrhagic stroke).

Statistics in MedicineReporting bayesian analyses of clinical trials

44 Citations1993Michael D. Hughes

This work focuses on two types of prior that deserve consideration, the first of which is the non-informative prior giving standardized likelihood distributions as post-trial probability distributions and the second which has a spike of probability mass at the point of no treatment effect.

Annals of Internal MedicineA Response to “Holding GUSTO Up to the Light”

27 Citations1994Paul M. Ridker, Christopher J. O'Donnell +2 more

The GustO investigators dispute the statement that no thrombolytic regimen in the GUSTO trial showed any mortality advantage for patients treated 4 or more hours after symptom onset, a critical issue because most patients with acute myocardial infarction in the United States come to the hospital during this time frame.