Lesson 3.2: Diagnostic Test Characteristics

Introduction

In medical practice, the ability to interpret diagnostic test results is crucial for effective patient care. This lesson will explore the characteristics of diagnostic tests, specifically focusing on sensitivity, specificity, predictive values, and likelihood ratios. Additionally, we will examine how the prevalence of a condition affects these measures and understand ROC (Receiver Operating Characteristic) curves and threshold selection.

By the end of this lesson, students will be able to:

1. Define and calculate sensitivity, specificity, predictive values, and likelihood ratios, understanding how prevalence affects these metrics.
2. Understand and apply ROC curves for determining optimal thresholds for diagnostic tests.
3. Calculate and interpret test performance measures from clinical data.
4. Explain how pretest probability impacts the interpretation of test results.
5. Familiarize with the terminology and concepts related to diagnostic test characteristics.

Sensitivity and Specificity

Definition

Sensitivity (also known as the true positive rate) measures the proportion of actual positives correctly identified by a test. It is calculated using the formula:

$$\text{Sensitivity} = \frac{\text{True Positives (TP)}}{\text{True Positives (TP)} + \text{False Negatives (FN)}}$$

Specificity (or true negative rate) measures the proportion of actual negatives that are correctly identified by the test. It is calculated as:

$$\text{Specificity} = \frac{\text{True Negatives (TN)}}{\text{True Negatives (TN)} + \text{False Positives (FP)}}$$

Worked Example

Consider a diagnostic test for a disease which has the following results in a population of 1000 individuals:

• True Positives (TP): 80
• False Negatives (FN): 20
• True Negatives (TN): 850
• False Positives (FP): 50

Calculating Sensitivity:

$$\text{Sensitivity} = \frac{TP}{TP + FN} = \frac{80}{80 + 20} = \frac{80}{100} = 0.80 \text{ or } 80\%$$

Calculating Specificity:

$$\text{Specificity} = \frac{TN}{TN + FP} = \frac{850}{850 + 50} = \frac{850}{900} = 0.944 \text{ or } 94.4\%$$

Common Misconceptions

A common misconception is that sensitivity and specificity can determine the accuracy of a test. However, they do not account for the prevalence of the disease in the population, which is crucial for understanding the real-world utility of a test.

Predictive Values

Definition

Positive Predictive Value (PPV) reflects how likely it is that a positive test result is a true positive. It is given by:

$$\text{PPV} = \frac{TP}{TP + FP}$$

Negative Predictive Value (NPV) indicates how likely it is that a negative test result is a true negative:

$$\text{NPV} = \frac{TN}{TN + FN}$$

Worked Example

Using the previous example:

Calculating Positive Predictive Value (PPV):

$$\text{PPV} = \frac{TP}{TP + FP} = \frac{80}{80 + 50} = \frac{80}{130} \approx 0.615 \text{ or } 61.5\%$$

Calculating Negative Predictive Value (NPV):

$$\text{NPV} = \frac{TN}{TN + FN} = \frac{850}{850 + 20} = \frac{850}{870} \approx 0.978 \text{ or } 97.8\%$$

Influence of Prevalence

Predictive values are heavily influenced by the prevalence of the disease in the population. As the prevalence increases, PPV increases while NPV decreases, and vice versa. This concept is crucial when interpreting diagnostic test results in different populations.

Likelihood Ratios

Definition

Likelihood Ratios (LR) provide a measure of how much a test result will change the odds of having a disease. There are two types of likelihood ratios:

• Positive Likelihood Ratio (LR+) measures how much more likely a positive test result is to occur in someone with the disease compared to someone without the disease:

$$LR+ = \frac{\text{Sensitivity}}{1 - \text{Specificity}}$$

• Negative Likelihood Ratio (LR-) measures how much less likely a negative test result is to occur in someone with the disease compared to someone without the disease:

$$LR- = \frac{1 - \text{Sensitivity}}{\text{Specificity}}$$

Worked Example

Continuing with our previous calculations:

Calculating LR+:

$$LR+ = \frac{\text{Sensitivity}}{1 - \text{Specificity}} = \frac{0.80}{1 - 0.944} = \frac{0.80}{0.056} \approx 14.29$$

Calculating LR-:

$$LR- = \frac{1 - \text{Sensitivity}}{\text{Specificity}} = \frac{1 - 0.80}{0.944} = \frac{0.20}{0.944} \approx 0.212$$

Interpretation of Likelihood Ratios

LR+ greater than 10 usually suggests that the test result significantly increases the likelihood of the condition, whereas LR- less than 0.1 suggests that the test significantly decreases the likelihood of the condition.

ROC Curves

Definition

Receiver Operating Characteristic (ROC) curves graphically represent the trade-off between sensitivity and specificity for a diagnostic test across different thresholds. An ROC curve is created by plotting the true positive rate (sensitivity) against the false positive rate (1-specificity).

Interpreting ROC Curves

The area under the ROC curve (AUC) quantifies the overall ability of the test to discriminate between outcomes. An AUC of 1 represents perfect discrimination, while an AUC of 0.5 indicates no discrimination (equivalent to random guessing).

Worked Example

Consider a test with the following characteristics:

• Sensitivity values increase as the threshold decreases.
• Specificity values decrease as the threshold decreases.

If we plot these values, we will see a curve that starts at (0,0) and ends at (1,1). The shape of the curve can help determine the ideal threshold where you balance sensitivity and specificity according to the clinical need.

Conclusion

Understanding diagnostic test characteristics is essential for interpreting and utilizing test results effectively. By mastering sensitivity, specificity, predictive values, likelihood ratios, and ROC curves, students will be equipped to make informed clinical decisions based on diagnostic testing.

Study Notes

• Sensitivity measures the true positive rate.
• Specificity measures the true negative rate.
• Predictive values depend on disease prevalence.
• Likelihood ratios help quantify the effect of test results.
• ROC curves illustrate the trade-off between sensitivity and specificity.
• The area under the ROC curve indicates test discrimination ability.