What Do Anesthesiologists Know about p Values, Confidence Intervals, and Correlations: A Pilot Survey

Background. Statistical methods form the basis for clinical decision-making in evidence-based anesthesia. Data on the knowledge of anesthesiologists about statistics are lacking. This pilot study aims to provide a first impression of the anesthesiologists’ understanding of commonly used concepts in...

Full description

Saved in:
Bibliographic Details
Main Authors: Patrick Schober, Sebastiaan M. Bossers, Phi-Vu Dong, Christa Boer, Lothar A. Schwarte
Format: Article
Language:English
Published: Wiley 2017-01-01
Series:Anesthesiology Research and Practice
Online Access:http://dx.doi.org/10.1155/2017/4201289
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Background. Statistical methods form the basis for clinical decision-making in evidence-based anesthesia. Data on the knowledge of anesthesiologists about statistics are lacking. This pilot study aims to provide a first impression of the anesthesiologists’ understanding of commonly used concepts in statistics. Methods. A cross-sectional pilot survey was performed at a major international anesthesia conference. The questionnaire consisted of three basic multiple-choice questions on the topics “p value,” “confidence interval,” and “correlation.” Results of the questions are reported as percentage of correct answers (95% confidence interval). Results. 65 questionnaires were analyzed. Forty participants were male, and mean age was 40 (standard deviation: 10) years. The question addressing the p value was correctly answered by 15% (95% CI: 8 to 27%) of respondents. The question concerning the 95% confidence interval was answered correctly by 28% (95% CI: 18 to 40%) of participants. For the question about correlation, a correct answer was given by 52% (95% CI: 40 to 64%). None of the participants answered all questions correctly, and 19 participants provided a wrong answer to all questions. Conclusions. Anesthesiologists seem to demonstrate a poor understanding of statistical key concepts. Further studies are needed to address statistical knowledge gaps among anesthesiologists more comprehensively.
ISSN:1687-6962
1687-6970