UW 1 Flashcards
What does a two-sample t test measure
Compares 2 means
What does a two-sample Z test measure
Compares 2 means
What does ANOVA measure
Compares 3 or more means
What is Chi-square test for?
Compares percentages or proportions
What is meta analysis used for
Pooling data from several studies to analyze
Big statistical power
Which type of prevention’s goals are to prevent recurrence and slow progression
Tertiary Prevention
Which type of prevention decreases disease prevalence
Secondary Prevention
Which type of prevention decreases disease incidence
Primary Prevention
What kind of prevention: Health education programs promoting healthy lifestyles
Primary Prevention
What kind of prevention: Community BP screening
Secondary Prevention
What kind of prevention:
Exercise program prescribed to pts recovering from MI
Tertiary Prevention
What does NNT represent
The number of people needed to treat to prevent one case
Calculation for NNT
Inverse of Incidence Rate
1 / Incidence Rate
Calculation for NNH
1 / Prevalence
Top 3 reasons for Infant Mortality
- Birth defects
- Low birth weight/Prematurity
- SIDS
Calculation for infant Mortality rate
Infant deaths / Live births X 1,000
Maternal Mortality Rate
Maternal deaths / Live births X 100,000
Case Fatality Rate
Deaths / Cases X 100
Proportionate Mortality Rate (PMR)
Deaths from specified cause / Total deaths X 100
Increased specificity effect on PPV
Increased PPV
Increased sensitivity effect on NPV
Increased NPV
Effect of increased prevalence on NPV and PPV
Increased prevalence =
Increased PPV
Decreased NPV
Effect of decreased prevalence on NPV and PPV
Decreased Prevalence =
Decreased PPV
Increased NPV
In Bimodal curve, effect of decreasing cutoff point
Higher TP and Sensitivity
Lower Specificity
In Bimodal curve, effect of increasing cutoff point
Lower FP and Sensitivity
Higher Specificity
Cross sectional study records prevalence or incidence?
Prevalence
Chi Squared test
One time point
Case control study is prospective or retrospective
Retrospective
Can assess Odds Ratio
Cohort study measures what
Incidence
Relative Risk or Attributable Risk
Is Prospective
What is Attributable Risk
How many more cases in one group?
Incidence rate of exposed group minus incidence rate of unexposed group
Odds Ratio looks at
odds of getting a dz w exposure to risk factor v nonexposure
Equation for Odds Ratio
OR = AD/BC
Precision
Reliability
Accuracy
Validity
Measuring patient satisfaction w leading questions
Measurement bias
How to avoid pygmalion effect or experimenter expectancy bias
Double blind design
Late look bias
Individuals w severe dz are less likely to be uncovered in a survey b/c they die first
What does p value measure
Strength or significance of data against null hypothesis
What does small p value achieve
Statistical significance
If p < 0.05
Reject null hypothesis
Type I error
Rejects null hypothesis when it is true
Type II Error
Fails to reject null hypothesis when it is false
What increases power
Increasing sample size
T test
Compares the means of two groups from a single nominal variable
