Epidemiology
DISEASE
PresentAbsent
TEST+ True Positives:

A
False Positives:

B
- False Negative:

C
True Negatives:

D
When you've filled out the four boxes,

 

Prevalence / Pre-Test Probability:
Sensitivity:
Specificity:
Positive Predictive Value:
Negative Predictive Value:
Positive Likelihood Ratio:
Negative Likelihood Ratio:
Positive Posttest Probability:
Negative Posttest Probability:
Prevalence / Pre-Test Probability =
      A+C      
A+B+C+D
What percentage of the population has the disease, or the likelihood of an individual patient having the disease?
Sensitivity =
  A  
A+C
How likely is the test to pick up the disease (correctly) if it is present?
Specificity =
  D  
B+D
How likely is the test to be negative (correctly) if the patient does not have the disease?
Positive Predictive Value =
  A  
A+B
What is the probability of there being disease in a patient with a positive test?
Negative Predictive Value =
  D  
C+D
What is the probability of there not being disease in a patient with a negative test?
Positive Likelihood Ratio =
  A  
  A+C  
  B  
B+D
How many times more will the test be positive in patients with the disease than in patients who do not have the disease?
Negative Likelihood Ratio =
  C  
  A+C  
  D  
B+D
How many times more will the test be negative in patients without the disease than in patients who do have the disease?
Positive Posttest Probability =
Pretest Probability x Positive Likelihood RatioWhat is the likelihood that there is actually disease in this patient?
Negative Posttest Probability =
Pretest Probability x Negative Likelihood RatioWhat is the likelihood that there is not actually disease in this patient?
Prevalence =
# of people with a disease at a point in time
      
# of people at risk (in population) at that time
How common is this condition at a given time in this population?
Incidence =
# of new cases of a disease over a time period
      
# of people at risk (in population) during that period
How often does this disease get diagnosed in this population?