Epi Flashcards
Cross sectional study is called a prevalence study because
It looks at one snapshot in time
Attributable risk =
Incidence in exposed - incidence in unexposed
Nnt=
1/attributable risk
Relative risk=
Incidence in exposed/incidence in unexposed
Odds ratio=
Odds that diseased person is exposed/odds that non diseased person is exposed
Ppv=
Probability that pt with positive test result has the disease (a/a+b) where b is the false positives
Npv=
Probability that pt with negative test result didn’t have the disease (d/c+d) where c is the false negatives
When prevalence goes down, what happens to ppv and npv
Ppv goes down
Npv goes up
What are case control studies
Looks at pp with and without dz and looks back to see whether they had an exposure or not
P value of under ___ is statistically significant
0.05
Type I(a) error
Probability of concluding there is a difference in treatment effects between groups when there is not
Type II (B) error
Probability of concluding there is no difference in treatment effects when in fact a difference exists
Power is the
Probability that a study will find a statistically sig difference when one is truly there
Power=
1-type II(b) error
Primary prevention
Preventative measures to decrease incidence of dz (exercise to prevent cv dz)
Secondary prevention
Identifying dz early (eg routine mammograms)
Tertiary prevention
Measures to decrease morbidity and mortality (tx of dz)
Most common cause of cancer death
Lung
Most prevalent cancers
Prostate and breast, then lung second and crc third
Leading cause of death
Heart dz both sexes, followed by cancer
1-specificity=
False positive rate
Define npv and ppv
Ppv= prob of having the disease of the test is positive (nothing to do with sensitivity)
Npv=prob of not having dz of test is negative
Difference between case control and retrospective cohort studies
Retrospective cohort: review past risk factors and compare resulting incidence of disease
Case control: look at diseased vs non diseased cases and compare past risk factor frequency
Raising the cutoff point in a test will ___ specificity and __ sensitivity
Increase
Decrease
Increase in prevalence ___ ppv And ___ npv
Increase ppv
Decrease npv
Decrease in prevalence ___ ppv And ___ npv
Decrease ppv
Increase npv
Two small z test and t test are used to compare
Two means
Chi squared test is used to compare
Proportions (eg observed as proportion of expected)