REBM Flashcards
How do you calculate risk?
What does it mean
Describe a cross-secional study
Descrieb a cohort study
Describe a case-control study
Describe an intervention study
What is the difference between prevalence and incidence and what the two variational forms of incidence
How do you calculate prevelence?
How do you calculate rate?
What does it mean?
How do you caculate total person time at risk?
How do you calculate prevelence odds and risk odds?
Calculate the sensitivity
Sens is the percentage of identified positives aka true positive to total reference positives
40/50*100 = 80%
sens - sensitive = postive
Calculate the specificity
Spec is the percentage of those without the disease correctly identified aka true negative to total reference negative
550/600*100 = 91.7%
Specky = negative
Calculate the positive predictive value
If you receive a positive test what is the chance you have the disease? True positives to total index positives
40/90*100 = 44.4%
Calculate the nagative predictive value?
If you receive a negative test what is the chance you don’t have the disease? True negatives to total index negatives
550/560*100 = 98.2%
Calculate the likelyhodd ratio for positive results?
Sensitivity/(100-Specificity) – Another way to phrase this is it takes the % of true positives and divides it by the % of false positives. It is therefore dependant on the accuracy of the test for positives results and the percentage of people falsely identified as having disease. If this ratio is greater than one, then the positive test result is associated with disease.
80/(100-91.7) = 12 >1 therefore the test result is associated with disease. In other words people with the disease are 12x more likely to get a positive result
Calculate the likelyhood ratio for negative results?
(100-sensitivity)/Specificity - another way to phrase this is takes the % of false negatives and divides it by the % of true negatives. It is therefore dependent on the accuracy of the test for negative results and the percentage of people falsely identified as not having disease. If this ratio is less than one, then the negative result is associated with not having disease
(100-80)/91.7 = 0.22 < 1 therefore a negative result is associated with not having the disease. In other words people without the disease are 5 times more likely to get a negative result
WHat does p-value imply?
How do you calculate risk ratio and risk difference?
How do yu calculate a QALY?
Length of life multiplied by quality of life weight
How do you calculate incremental cost effectiveness ratio and what is it used for?
It would be used to compare a new intervention to an old one in a cost benifit analysis
