Biostatistics Flashcards
Mean is preferred for ____ data that is ____ ____
Continuous, normally distributed
Median is preferred for _____ data or _____ that is _____
Ordinal, continuous, skewed
Mode is preferred for ____ data
Nominal
When the null hypothesis is rejected in error
Type I error
Type I error is avoided by…
Alpha < 0.05
False positive
Type I error
False negative
Type II error
Null hypothesis is rejected when it should have been accepted
Type II error
Type II error avoided with…
Larger sample size
Probability that a test will reject the null hypothesis correctly
Power
Ratio of risk in the exposed (treatment) group divided by risk in the control group
Relative risk (RR)
Number of subjects in group with an unfavorable event divided by total number of subjects in group
Risk
RR = 1
No difference
RR < 1
Lower risk in treatment group
RR > 1
Greater risk in treatment group
How much risk is reduced in treatment group v. control group
Relative risk reduction (RRR)
RRR
1 - RR
(%risk in control) - (%risk in treatment)
Absolute risk reduction (ARR)
1 / (risk in control) - (risk in treatment)
NNT
For NNT round ___, for NNH round ___
Up; down
Rate at which an unfavorable event occurs within a short period of time
Hazard ratio (HR)
OR Formula
AD/BC
If 1 group with continuous data
One-sample t-test
If 1 group with categorical/discrete data
Chi-square test
If 1 group with before/after measures with continuous data
Dependent/paired t-test
If 1 group with before/after measures with categorical/discrete data
Wilcoxon Signed-Rank test
If 2 groups with continuous data
Independent/unpaired student t-test
If 2 groups with categorical/discrete data
Chi-square or Fisher’s exact test or Mann-Whitney
If 3+ groups with continuous data
ANOVA
If 3+ groups with categorical/discrete data
Kruskal-Wallis test
of pts who test positive out of those who actually have the condition
Sensitivity
of pts who test negative out of those who actually do not have the condition
Specificity
Includes data for all patients originally allocated to each treatment group even if the patient didn’t complete the trial
Intention-to-treat
Analysis of subset of trial population who completed the study according to the protocol
Per protocol
In a forest plot, if the line crosses the center line, this means
No statistical significance
In a forest plot, if the line is to the left of the center line, this means…
Statistically significant benefit
In a forest plot, if the line is to the right of the center line, this means…
Statistically significant harm
Retrospective comparisons of cases (patients with disease) & controls (without disease
Case-control
Retrospective or prospective comparisons of patients with an exposure to those without
Cohort
Prospective comparison of patients who were randomly assigned to groups
RCT
Analyzes results of multiple studies
Meta-analysis
Change in costs & outcomes when 2 treatment alternatives are compared
Incremental cost-effectiveness ratios
Used when 2+ interventions have demonstrated equivalence in outcomes & costs of each intervention are being compared
Cost-minimization analysis
Comparing benefits & costs of an intervention in terms of monetary units
Cost-benefit analysis
Used to compare clinical effects of 2+ interventions to the respective costs
Cost-effectiveness analysis
Specialized form of CEA that uses QALYs and DALYs
Cost-utility analysis