Clinical Assessment Methods Flashcards

A test detects 90 of 100 patients with disease and is negative in 160 of 200 patients without disease. What is its specificity?

Answer: 80%

Specificity = TN/(TN+FP) = 160/200 = 80%.

As disease prevalence in a tested population decreases, what happens to the positive predictive value of a test with fixed sensitivity and specificity?

Answer: It decreases

Lower prevalence yields proportionally more false positives, lowering PPV.

A reference interval is typically defined as the central what percentage of results from a healthy reference population?

Answer: 95%

Reference intervals usually span the central 95% (mean ±2 SD for Gaussian data).

Which statistic best compares the precision of two methods that have very different mean values?

Answer: Coefficient of variation

CV expresses SD as a percentage of the mean, allowing comparison across concentrations.

Which plot is most useful for visualizing agreement between a new and a reference method by graphing the difference against the average?

Answer: Bland-Altman plot

A Bland-Altman plot graphs differences versus means to reveal bias and limits of agreement.

The smallest concentration of analyte that can be reliably distinguished from zero is called the:

Answer: Limit of detection

The limit of detection is the lowest amount distinguishable from a blank.

A receiver operating characteristic (ROC) curve plots which two values against each other?

Answer: Sensitivity (true positive rate) vs 1 − specificity (false positive rate)

ROC curves plot true positive rate against false positive rate at varying cutoffs.