Term
|
Definition
| Patient with a certain disease or exposure compared to a similar group without the disease or exposure |
|
|
Term
|
Definition
Two groups compared over time
-One group with “exposure”, the other without the “exposure” |
|
|
Term
|
Definition
•Best uses
-when exposures can’t be controlled
-when outcomes occur infrequently
-when RCT is not ethical |
|
|
Term
|
Definition
•Best uses
-Rare conditions
-Diseases or conditions that may take a long time to develop |
|
|
Term
|
Definition
A treatment group is compared to a control group
Group members are assigned randomly |
|
|
Term
|
Definition
•Best uses
-Drug therapies
-Medical treatments |
|
|
Term
|
Definition
| The “traditional” or “classic” review |
|
|
Term
|
Definition
Reproducible methods to find and select articles to be included
-Should include both inclusion and exclusion criteria
-Quality of studies evaluated
-Evidence summarized
-Findings interpreted |
|
|
Term
|
Definition
-Numeric data from separate studies combined in meta analysis
-Uses statistical/mathematical methods to combine numerical data from studies |
|
|
Term
|
Definition
| a comparison group receives a standard treatment |
|
|
Term
|
Definition
| subjects don’t know what treatment they received |
|
|
Term
|
Definition
| subjects and investigators don’t know what treatment they received |
|
|
Term
| INTENTION TO TREAT ANALYSIS |
|
Definition
| pts are analyzed in the groups they were randomized not the treatment they received |
|
|
Term
|
Definition
| control subjects receive a placebo (inactive pill) or “sham” treatment |
|
|
Term
|
Definition
| subjects receiving different treatments are matched to balance confounding variables (e.g. sex, age), results analyzed in terms of differences between pairs |
|
|
Term
|
Definition
P- population
I- intervention
C- comparison
O- outcome |
|
|
Term
HOW TO EVALUATE THE EQUALITY OF A STUDY:
PP-ICONS |
|
Definition
Problem
Patient or Population
Intervention
Comparison
Outcome
Number of Subjects
Statistics |
|
|
Term
|
Definition
used to determine “significance” of results
-Assume that there is no difference between the groups (null hypothesis)
-What is the chance that the effect we see is due to chance alone (e.g. really no difference)?
-Traditional cut off 0.05 i.e. less than 1/20 chance results are false |
|
|
Term
|
Definition
| proportion of pts responding to control treatment |
|
|
Term
| Experimental event rate (EER) |
|
Definition
| proportion of pts responding to experimental treatment |
|
|
Term
|
Definition
| a ratio of the incidence of a health outcome in treatment vs. control |
|
|
Term
| Number needed to treat (NNT) |
|
Definition
| number of pts who must be treated to prevent one adverse event |
|
|
Term
| Number needed to harm (NNH) |
|
Definition
| number of pts who must receive and intervention to be harmed |
|
|
Term
|
Definition
| ability of test to identify those with disease (SNOUT) |
|
|
Term
|
Definition
| ability of test to identify those who don’t have disease (SPIN) |
|
|
Term
|
Definition
| systemic errors in the way the study was designed, executed, or interpreted |
|
|
