A screening program for gestational diabetes mellitus (GDM) in a tertiary hospital in Mumbai enrolls 5000 pregnant women at 24–28 weeks of gestation. A new point-of-care glucose meter is evaluated against the gold standard 75-g oral glucose tolerance test (OGTT). Among 800 women with confirmed GDM on OGTT, the glucose meter correctly identified 760 cases. Among 4200 women without GDM on OGTT, the glucose meter correctly ruled out 3948 cases. What is the specificity of the glucose meter for GDM screening?

Explanation

## Calculating Specificity **Key Point:** Specificity is the probability that a negative test result correctly identifies the absence of disease. It is calculated as: Specificity = TN / (TN + FP), where TN = true negatives and FP = false positives. ### Data Extraction from the Vignette From the problem statement: - **Disease present (GDM confirmed on OGTT):** 800 women - Correctly identified by glucose meter (TP) = 760 - Missed by glucose meter (FN) = 800 − 760 = 40 - **Disease absent (no GDM on OGTT):** 4200 women - Correctly ruled out by glucose meter (TN) = 3948 - Incorrectly identified as GDM (FP) = 4200 − 3948 = 252 ### Specificity Calculation $$\text{Specificity} = \frac{\text{TN}}{\text{TN} + \text{FP}} = \frac{3948}{3948 + 252} = \frac{3948}{4200}$$ $$\text{Specificity} = 0.94 = 94\%$$ ### 2×2 Contingency Table | | GDM Present (OGTT+) | GDM Absent (OGTT−) | Total | |---|---|---|---| | **Glucose Meter +** | 760 (TP) | 252 (FP) | 1012 | | **Glucose Meter −** | 40 (FN) | 3948 (TN) | 3988 | | **Total** | 800 | 4200 | 5000 | **High-Yield:** Specificity is calculated only from the disease-absent group (the denominator includes only those who truly do NOT have the disease). It answers the question: "Of all people without the disease, how many does the test correctly identify as negative?" ### Related Test Characteristics $$\text{Sensitivity} = \frac{\text{TP}}{\text{TP} + \text{FN}} = \frac{760}{800} = 95\%$$ $$\text{Positive Predictive Value (PPV)} = \frac{\text{TP}}{\text{TP} + \text{FP}} = \frac{760}{1012} = 75.1\%$$ $$\text{Negative Predictive Value (NPV)} = \frac{\text{TN}}{\text{TN} + \text{FN}} = \frac{3948}{3988} = 99\%$$ **Clinical Pearl:** This glucose meter has high sensitivity (95%) and high specificity (94%), making it suitable for screening. The very high NPV (99%) means a negative test is very reassuring and can safely exclude GDM in most cases. **Mnemonic:** **SPIN and SNOUT** - **SPIN:** Specificity rule-IN (high specificity test that is positive rules IN the disease) - **SNOUT:** Sensitivity rule-OUT (high sensitivity test that is negative rules OUT the disease) This meter has both high sensitivity and specificity, so it performs well in both directions.