Evidence Based Medicine

Front 1

Context formulation

Back 1

Cognitive representation of a problem that frames or constrains the solution. Eg: a disease/syndrome

Front 2

Heuristics predisposing to cognitive errors (3)

Back 2

Representativeness
Availability
Anchoring or Conservatism

Front 3

Representativeness heuristic

Back 3

Overestimate likelihood of disease because of its resemblance to the symptoms of that disease

Front 4

Availability heuristic

Back 4

Overestimate likelihood of disease because of vividness of recall, or underestimate based on difficulty of recall

Front 5

Anchoring or Conservatism heuristic

Back 5

Failure to revise diagnositc probabilities as much as Bayes' theorum would imply

Front 6

False alarm rate

Back 6

Given a positive test, the likelihood of not having the disease

Front 7

False reassurance rate

Back 7

Given a negative test, the likelihood of having the disease

Front 8

Likelihood ratio positive

Back 8

Sensitivity/(1-Specificity) OR
True pos/Fals pos

Front 9

Likelihood ratio negative

Back 9

(1-Sensitivity)/Specificity

Front 10

Lead-time bias

Back 10

Screening appears to increase survival because patients are found to have disease at an earlier point in time, even though the natural history of the disease is unaltered

Front 11

Length time bias

Back 11

(length of the interval over which the disease is asymptomatic and there for screenable)
Screening is more likely to detect patients w/ slower progressing disease, so it will apear that screening improves survival (since it is not getting the aggressive disease)

Front 12

Overdiagnosis

Back 12

Screening detects diseases that would not become clinically significant, thereby increasing the apparent prognostic effects of the screen

Front 13

Treatment threshold

Back 13

Probability of disease above which you should treat and below which you should not treat