Diagnostic tests

Question 1

Which clinical areas are diagnostic tests applied to

Answer 1

radiological, laboratory, physiological

Question 2

What is sensitivity and specificity

Answer 2

Sensitivity=true positive rate (number of diseased individuals positive to test/total number of diseased individuals)
Specificity=false negative rate (number of disease-free patients negative to test/total number of disease free individuals)

Question 3

Positive predictive value and negative predictive value how to calculate

Answer 3

PPV==number of diseased individuals positive to test/total nummber of individuals positive to test
NPV=number of disease free individuals negative to test/total number of individuals negative to test

Question 4

Draw out a table of the relationships between the terms: sensitivity, specificity, PPV,NPV, accuracy

Answer 4

ref. notes

Question 5

How to calculate positive and negative liklihood ratios

Answer 5

PLR=sensitivity/1-specificity
NLR=1-sensitivity/specificity
If P is the prevalence of diseasePPV/(1-PPV)=P/(1-P)*PLR

Question 6

Equation relating PPV to prevalence of disease and positive likelihood ratio

Answer 6

PPV/(1-PPV)=P/(1-P)*LR+

Question 7

Purpose of predictive value and prevalence of disease

Answer 7

quantify usefulness of test

Question 8

likelihood ratio purpose

Answer 8

combines sensitivity and specificity and is used to compare usefulness of different tests

Question 9

Equation for agreement in qualitative test results

Answer 9

(observed % agreement-expected % agreement)/(100-expected % agreement)

Question 10

Choice of cut off when dichotomising continuous data

Answer 10

choose sensitivity or specificity required. Move cut point to increase sensitivity but decreasse specificity vice versa

Question 11

Bias and precision difference

Answer 11

Bias-mean difference with 95% confidence interval

Precision-standard deviation (or 95% range) of differences