Biostats and Pharmacoeconomics Flashcards
Continuous data categories
Ratio and interval data
Examples of ratio data
age, weight, height, time, blood pressure
Examples of interval data
temperature scales
Discrete data categories
nominal and ordinal
Examples of nominal data
gender, ethnicity, marital status, mortality
Examples of ordinal data
NYHA Functional Class I-IV, 0-10 pain scale
Comparing difference data (means): when is it statistically significant?
When the CI doesn’t cross 0
Comparing ratio data (RR, OR, HR): when is it statistically significant?
When the CI doesn’t cross 1
Type I errors are what?
False positives
Type 2 errors are what?
False negatives
Study power definition
Probability that a test will reject the null hypothesis
RR=1 interpretation
No difference in risk of the outcome between the groups
RR >1 interpretation
greater risk of the outcome in the treatment group
RR <1 interpretation
lower risk of the outcome in the treatment group
Wording for RR
“AS likely vs. control”
RRR definition
Indicates how much risk is REDUCED in treatment group
RRR interpretation wording
“LESS likely vs. control”
ARR definition
The reduction in risk and the incidence rate of the outcome
ARR interpretation wording
“X out of every 100 patients will benefit”
NNT definition
How many patients need to be treated before 1 patient benefits
NNT rounding rules
Round UP, you don’t want to underestimate how many people need to be treated!
NNT interpretation
“For every X patient who receives Y treatment, Z (adverse outcome) is prevented in one patient”
NNH definition
How many patients need to be treated before 1 patient gets harmed
NNH rounding rules
ROUND DOWN, you don’t want to underestimate the potential harm of an intervention!
NNH interpretation
“One case of Z (adverse outcome) is expected to occur for every X patients who take Y (treatment)”
Odds ratio definition
Used to estimate the risk of unfavorable events associated with a treatment/intervention
OR formula
AD/BC (just cross multiply and then divide)
OR interpretation
X treatment is associated with a Y% increased risk of Z (bad outcome)
HR definition
Rate at which an unfavorable event occurs within a short period of time
HR formula
HR in treatment group/HR in placebo
OR and HR interpretation: OR/HR=1
event rate is the same in the treatment and control arms, no advantage to the treatment
OR and HR interpretation: OR/HR >1
event rate in the treatment group is higher than the event rate in the control group
example: if the OR/HR is 2 for an outcome of death means that there are twice as many deaths in the treatment group
OR and HR interpretation: OR/HR <1
event rate in the treatment group is lower than the event rate in the control group
example: HR of 0.5 for an outcome of death indicates there are half as many deaths in the treatment group
Sensitivity definition
True positive! The higher the sensitivity, the better
Specificity definition
True negative!
Sensitivity formula
(A/A+C)*100%
Specificity formula
(D/B+D)*100%
Case-control study features
retrospective comparisons of cases and controls
Cohort study features
retrospective or prospective comparisons of patients with an exposure to those without it
RCT features
prospective comparison of patients who were randomly assigned to groups
Meta-analyses features
analyzes the results of multiple studies
Parametric test for continuous data with 1 group
one-sample t-test
Parametric test for continuous data with 1 group that has before and after measures
dependent/paired t-test
Parametric test for continuous data with 2 groups
Independent/unpaired student t-test
Parametric test for continuous data with ≥3 groups
ANOVA
Non-parametric test for continuous data with 1 group
Sign test
Non-parametric test for continuous data with 1 group that has before and after measures
Wilcoxon Signed Rank test
Non-parametric test for continuous data with 2 groups
Mann-Whitney test
Non-parametric test for continuous data with ≥3 groups
Kruskall-Wallis test
Discrete/Categorical test for 1 group
Chi-square test
Discrete/Categorical test for 1 group with before and after measuress
Wilcoxon Signed Rank test
Discrete/Categorical test for 2 groups
Chi-square or Fisher’s Exact test
Discrete/Categorical test for ≥3 groups
Kruskal-Wallis test
When is CMA used?
Used when 2 or more interventions have demonstrated equivalence in outcomes
What does CMA compare?
Costs of each intervention
When is CBA used?
Used to calculate and comparing benefits and costs of an intervention in terms of monetary units
What does CBA compare/determine?
If the benefits of the intervention exceed the costs of implementation; can also be used to compare multiple programs for similar or unrelated outcomes as long as the outcome measures can be converted to dollars
What does CEA compare?
Clinical effects of two or more interventions to the respective costs
What can CUA measure?
The quality of life but not the quality or utility of the years
Outcome unit of CMA
Demonstrated or assumed to be equivalent in comparative groups
Outcome unit of CBA
Dollars
Outcome unit of CEA
Natural units (life-years gained, blood pressure, % at treatment goal)
Outcome unit of CUA
QALY
