Prevalence Rate=
Number of cases of the disease/total at risk at a particular time
Incidence Rate=
(Number of new cases/total number at risk )
What is a type I error?
Type II error?
False positve
False negative
Sensitivity?
Ability to detect true disease
True positive/ (true positive and false negative)
Specificity=
The ability to detect people free of disease
True negative/(true negative and false positive)
Specificity rules disease in (SP IN, SN OUT)
Accuracy=
True positive + true Negative/ total pop
PPV=
likelihood that person with positive test actually has disease
True positive/(true positive + false positive)
Negative predictive value
person with negative test is actually disease free
True negative/ (true + false negative)
Predictave values depend on
prevalence
higher prevalence = higher PPV and lower NPV
Lower prevalence = lower PPV and higher NPV
A valid test is_____?
Specific and sensitive
Power prevents?
Type II error
ususally 0.8
larger sample = more power
Absolute risk=
incidence of disease in population
Relative risk?
incidence of disease in exposed population/incidence in population not exposed
Number needed to treat?
100/absolute risk reduction
Attributable risk?
incidence in exposed- incidence in not exposed
Odds ratio?
Can be retrospective (absolute risk is prospective)
What is the x, y axis of the ROC?
Sensitivity on y, 1-specificity on x
