Diagnostic Testing

Question 1

Calculation for sensitivity

“Positivity in disease”

Answer 1

Sensitivity=TP/(TP+FN)

goes down with each FN

Question 2

Calculation of specificity

“negativity in health”

Answer 2

Specificity=TN/(TN+FP)

goes down with each false positive

Question 3

The likelihood that a negative test result is a TN rather than a FN

Answer 3

predictive value of negative

Question 4

formula for predictive value of a negative test result

Answer 4

PV-=TN/(TN+FN)
aims for 100% prediction of true negatives
goes down with each FN
related to sensitivity

Question 5

predictive value of a positive test result

Answer 5

PV+=TP/(TP+FP)

related to specificity

Question 6

calculation of prevalence (in terms of TP, TN, FP, FP)

Answer 6

(TP+FN)/(TP+FN+FP+TN)
essentially, this takes the total number of people who actually have the disease and divides it by the total number of people in the population.

Question 7

Tests with a specificity of 100%; what is the PV+?

-should you confirm or exclude the disease if it’s positive?

Answer 7

PV+=100%

confirms disease

Question 8

Tests with a sensitivity of 100%; what is the PV-?

-should you confirm or exclude the disease if it’s negative?

Answer 8

PV-=100%

exclude the disease

Question 9

what is prevalence?

Answer 9

the total number of people with disease in a population

Question 10

what happens to PV- and PV+ when there is a low prevalence of disease?

Answer 10

PV- goes up

PV+goes down

Question 11

what happens to PV- and PV+ when there is a high prevalence of disease?

Answer 11

PV- goes down

PV+goes up

Question 12

how many people have a disease if the prevalence of SLE is 1% in a population of 1000?
how many TP/FN/TN/FP if serum ANA sensitivity is 100% and specificity is 80%?

Answer 12

1000x0.01=10
10x100%sensitivity=10 TP, 0 FN
990x80%specificity=792 TN, 198 FP