Users' Guides to the Medical Literature
Guyatt G, Rennie D, Meade MO, Cook DJ
Part B Therapy
Chapter 10.1. Hypothesis Testing
Gordon Guyatt, Kameshwar Prasad, Roman Jaeschke, Deborah J. Cook, Stephen Walter
Continuous Measures of Outcome
Topics Discussed:
continuous variable, hypothesis testing, measures of outcome, null hypothesis testing
Excerpt:
"To this point, all of our examples have used outcomes such
as yes/no, heads or tails, or dying or not dying, all of
which we can express as a proportion. Often, investigators compare
the effects of 2 or more treatments using a variable such as spirometric measurements,
cardiac output, creatinine clearance, or score on a quality-of-life questionnaire.
We call such variables, in which results can take a large number
of values with small differences between those values, continuous
variables. When one compares differences between groups using
continuous outcomes, one typically asks the question whether one
can exclude chance as the explanation of a difference in means.Picture a medical school in which 2 instructors
with differing approaches teach an introductory course on medical
statistics. The instructors wish to see whether the 2 approaches lead to different results
on a final common examination. To do so, they assign the 200 medical students
in the first-year class to one instructor or the other by a process
of random allocation, through which each
student has an equal chance (50%) of being allocated to
either of the 2 instructors...."
Log in to read the full chapter:
Get full access to JAMAevidence two ways:
Subscribe to JAMAevidence
JAMAevidence is a subscription-
based website dedicated to the learning, teaching, and practicing of evidence-based medicine.
Pay Per View
Timed access to all of JAMAevidence
24 hours for $29.95
48 hours for $49.95