Sensitivity, Specificity, & Predictive Value

In the context of Laboratory Operations, these statistical parameters describe the diagnostic utility of a test method. While QC (Mean/SD) tells us if the instrument is working correctly (analytical performance), Sensitivity and Specificity tell us if the test is clinically useful for diagnosing disease (clinical performance). These concepts answer the question: “How good is this test at distinguishing between healthy and diseased patients?”

The Four Outcomes

To calculate these parameters, we compare the test result against a “Gold Standard” (the absolute truth, often determined by biopsy or comprehensive clinical evaluation). This creates a 2x2 contingency table:

1. True Positive (TP): Patient has the disease, and the test is Positive
2. True Negative (TN): Patient does not have the disease, and the test is Negative
3. False Positive (FP): Patient does not have the disease, but the test is Positive (Type I Error). “False Alarm.”
4. False Negative (FN): Patient has the disease, but the test is Negative (Type II Error). “Missed Diagnosis.”

Diagnostic Sensitivity

Sensitivity measures the ability of a test to correctly identify those with the disease. It is the “Positivity in Disease.”

• The Question: “If the patient has the disease, what is the probability that the test will be positive?”
• Formula \[ \text{Sensitivity \%} = \frac{\text{True Positives (TP)}}{\text{Total with Disease (TP + FN)}} \times 100 \]
• Clinical Utility
  • Tests with High Sensitivity: have very few False Negatives. They rarely “miss” a case
  • Screening Tests: High sensitivity is required for screening (e.g., D-Dimer for DVT/PE). You want to catch everyone who might have the clot. A negative D-Dimer effectively rules out a clot because the test is so sensitive it rarely misses one
  • Mnemonic: SnNout – A highly Sensitive test, when Negative, rules out disease

Diagnostic Specificity

Specificity measures the ability of a test to correctly identify those without the disease. It is the “Negativity in Health.”

• The Question: “If the patient is healthy, what is the probability that the test will be negative?”
• Formula \[ \text{Specificity \%} = \frac{\text{True Negatives (TN)}}{\text{Total Healthy (TN + FP)}} \times 100 \]
• Clinical Utility
  • Tests with High Specificity: have very few False Positives. They rarely give a “false alarm.”
  • Confirmatory Tests: High specificity is required for confirmation. For example, the Sickle Solubility test is a screening test (sensitive), but Hemoglobin Electrophoresis is the specific confirmatory test. A positive electrophoresis confirms the diagnosis because it doesn’t cross-react with other things
  • Mnemonic: SpPin – A highly Specific test, when Positive, rules in disease

Predictive Values

While Sensitivity and Specificity are inherent properties of the test method itself (fixed values), Predictive Values change depending on the Prevalence of the disease in the population being tested. These are the values most useful to the clinician treating an individual patient

Positive Predictive Value (PPV)

• The Question: “The patient just tested positive. What is the chance they actually have the disease?”
• Formula \[ \text{PPV \%} = \frac{\text{True Positives (TP)}}{\text{Total Positives (TP + FP)}} \times 100 \]
• Influence of Prevalence: If you use a test in a population where the disease is rare (low prevalence), the PPV drops significantly because the False Positives will outnumber the True Positives. This is why we don’t screen the general public for rare leukemias; a positive result would more likely be a statistical error than cancer

Negative Predictive Value (NPV)

• The Question: “The patient just tested negative. What is the chance they are actually healthy?”
• Formula \[ \text{NPV \%} = \frac{\text{True Negatives (TN)}}{\text{Total Negatives (TN + FN)}} \times 100 \]
• Influence of Prevalence: In a population with low disease prevalence, the NPV is extremely high (because almost everyone is negative anyway). High NPV provides confidence that a negative result is truly negative

Summary Table

Parameter	Formula	Purpose	Mnemonic
Sensitivity	\(TP / (TP + FN)\)	Screening (Don’t miss it)	SnNout (Negative Rules Out)
Specificity	\(TN / (TN + FP)\)	Confirmation (Be sure)	SpPin (Positive Rules In)
PPV	\(TP / (TP + FP)\)	Patient Probability (If +)	Dependent on Prevalence
NPV	\(TN / (TN + FN)\)	Patient Probability (If -)	Dependent on Prevalence