Assessing the Validity and Reliability of Diagnostic and Screening Tests

Vocabulary flashcards covering diagnostic and screening test validity, sensitivity, specificity, predictive values, and testing strategies.

Validity of a Screening Test

The ability of a screening test to correctly distinguish between who has a disease and who does not.

Sensitivity

The ability of a test to correctly identify those who have the disease, calculated as $\frac{\text{TP}}{\text{TP} + \text{FN}}$.

Specificity

The ability of a test to correctly identify those who do not have the disease, calculated as $\frac{\text{TN}}{\text{TN} + \text{FP}}$.

Gold Standard

An external source of truth regarding the disease status of individuals in a population (such as a tumor biopsy or cardiac catheterization) used to evaluate test performance.

Bimodal Distribution

A continuous data distribution that exhibits two distinct peaks, such as the age distribution of reported hepatitis C cases in Massachusetts in 2009.

Unimodal Distribution

A continuous data distribution with a single peak, such as low-density lipoprotein cholesterol (LDL-C) concentration levels.

True Positive (TP)

An individual who actually has the disease and receives a positive screening test result.

False Positive (FP)

An individual who does not have the disease but receives a positive screening test result, leading to unnecessary follow-up procedures and anxiety.

False Negative (FN)

An individual who has the disease but receives a negative screening test result, which can prevent critical early intervention.

True Negative (TN)

An individual who does not have the disease and receives a negative screening test result.

Sensitivity-Specificity Trade-off

The inverse relationship in tests of continuous variables where lowering the cutoff point increases sensitivity but decreases specificity, while raising the cutoff point increases specificity but decreases sensitivity.

Sequential (Two-Stage) Screening

A screening strategy where a less expensive or invasive test is conducted first, and positive screeners are recalled for a second test, resulting in a net loss in sensitivity and a net gain in specificity.

Simultaneous (Parallel) Testing

A testing strategy where multiple tests are performed at the same time and a positive result on any test classifies the individual as positive, increasing net sensitivity and decreasing net specificity.

Net Sensitivity

The overall proportion of diseased individuals identified as positive when multiple tests are used, calculated as $\frac{\text{Test A only} + \text{Test B only} + \text{Both Tests}}{\text{Total Diseased}}$.

Net Specificity

The overall proportion of non-diseased individuals identified as negative by all tests in simultaneous testing, calculated as $\frac{\text{Negative on Both Tests}}{\text{Total Non-diseased}}$.

Positive Predictive Value (PPV)

The proportion of individuals testing positive who actually have the disease, calculated as $\frac{\text{TP}}{\text{TP} + \text{FP}}$.

Negative Predictive Value (NPV)

The proportion of individuals testing negative who actually do not have the disease, calculated as $\frac{\text{TN}}{\text{TN} + \text{FN}}$.

Prevalence Impact on PPV

The direct relationship where an increase in disease prevalence in a screened population leads to an increase in the Positive Predictive Value (PPV).

Screening $2 \times 2$ Matrix

A $2 \times 2$ table cross-tabulating test results (positive/negative) against true disease status (diseased/non-diseased) used to derive sensitivity, specificity, PPV, and NPV.