Diagnostics 1 - sensitivity, specificity & likelihood ratios

Question 1

gaussian (normal) distribution

Answer 1

in normally distributed population, determined by proximity to the mean
67% within 1 sd of the mean
95% within 2 sd of the mean
99.7% within 3 sd of the mean

Question 2

non gaussian distribution

Answer 2

mean not necessarily the best way to determine normality
use median as reference point instead
rather than standard deviations, use percentiles

Question 3

100% sensitivity

Answer 3

every single person with the disease will get a positive result

Question 4

false positive

Answer 4

healthy person with an incorrect positive result

Question 5

100% specificity

Answer 5

every single person without the disease will get a negative result

Question 6

false negative

Answer 6

person with the disease incorrectly told they do not have the disease

Question 7

2x2 grid

Answer 7

fracture present fracture absent
test + true positive (A) false positive (B)
test - false negative (C) true negative (D)

sensitivity = A/(A+C)
specificity = D/(B+D)
positive predictive value = A/(A+B)
negative predictive value = D/(C+D)
LR + = A/(A+C) / B/(B+D)
LR - = C/(A+C) / D/(B+D)

1 - specificity = false positive rate = B/(B+D)

1 - D/(B+D) = B/(B+D)

Question 8

likelihood ratios

Answer 8

likelihood of result meaning disease vs. no disease

LR+ 1 = test useless
LR+&raquo_space;1 = test useful to rule disease in
LR- 1 = useless
LR- «1 = useful to rule disease out (ideally 0)