2- test characteristics Flashcards
Define sensitivity of a test
Proportion of those with disease that it correctly identifies
Define specifity
Proportion of those w/o disease that it correctly identifies
Define and calculate positive predictive value (PPV)
Probability that pt with positive result actually has the disease
TP/TP+FP
Define and calculate negative predictive value (NPV)
Probability that pt with negative result doesn’t have the disease
TN/TN+FN
Calculate PPV
sensitivityprevalence/sensprev+(1-spec)(1-prev)
Describe the concepts of FP, FN, TP, TN
FP- test positive, no disease
FN- test negative, disease
TP- test positive, disease
TN- test negative, no disease
Define diagnostic testing
To confirm or exclude presence of disease
Define Screening test
Assess risk of disease in an asymptomatic pt to determine need for further confirmatory tests
Define prognostic tools
Assess risk of further disease and need for preventative measures
Tests also used to monitor progress
Screening tests vs diagnostic tests
ST
Purpose: detect potential disease indicators
Target pop: Large no of asymptomatic, at risk
Test method: Simple for pts and staff, cheap
Positive result threshold- high sens and spec
Positive result: indicate disease suspicion. Used in combo with other risk factors- warrant confirmation
Cost: cheap, benefits justify costs, large no of people screened to identify small no of cases
DT
Purpose: establish presence/absence of disease
Target pop: Symptomatic pts, or asympt with +ve screening tests
Test methods: Invasive, expensive, justifiable
+ve result threshold: higher spec, accuracy and precision
Positive result: definite diagnosis
Cost: higher, justified to establish diagnosis
eg TTG tests- coeliac disease, high sens and spec
Bayes likelihood ratio
Likelihood a +ve test expected in pt with disease compared to pt without disease
Post test odds
Pretest odds* LR
