A screening program for gestational diabetes mellitus (GDM) in a tertiary centre in Mumbai enrolled 1000 pregnant women at 24–28 weeks of gestation. All women underwent a 75-g oral glucose tolerance test (OGTT). The reference standard (diagnostic gold standard) was a 3-hour OGTT with strict WHO criteria. Among the 200 women who met WHO criteria for GDM, a new point-of-care HbA1c test was positive in 156 cases and negative in 44 cases. Among the 800 women without GDM, the HbA1c test was positive in 120 cases and negative in 680 cases. A pregnant woman with a positive HbA1c result comes to your clinic. What is the probability that she actually has GDM (positive predictive value)?

Explanation

## Positive Predictive Value (PPV) Calculation **Key Point:** PPV is the probability of disease given a **positive test result**. It answers: "If my test is positive, what is the chance the patient truly has the disease?" ### Formula $$\text{PPV} = \frac{\text{True Positives}}{\text{True Positives} + \text{False Positives}}$$ ### Data Organization | | GDM (by OGTT) | No GDM (by OGTT) | Total | |---|---|---|---| | **HbA1c Positive** | 156 (TP) | 120 (FP) | 276 | | **HbA1c Negative** | 44 (FN) | 680 (TN) | 724 | | **Total** | 200 | 800 | 1000 | ### Calculation $$\text{PPV} = \frac{156}{156 + 120} = \frac{156}{276} = 0.5652 \approx \mathbf{56.5\%}$$ **High-Yield:** PPV of 56.5% means that among all women with a positive HbA1c test, only about 57% actually have GDM. The remaining ~43% are false positives — a clinically important distinction that should prompt confirmatory testing before labelling a patient as GDM. **Clinical Pearl:** PPV is **test-result-centric** and **population-dependent**. It depends on disease prevalence: in this cohort, GDM prevalence is 200/1000 = 20%. A positive test in a low-prevalence population yields a lower PPV than the same test in a high-prevalence population (Park's Textbook of Preventive and Social Medicine, 25th ed.). **Why the other options are wrong:** - **65%** — This does not correspond to any standard 2×2 table calculation from the given data. - **87%** — This approximates the Negative Predictive Value (NPV = TN/(TN+FN) = 680/724 ≈ 93.9%), not PPV. - **78%** — This is the **Sensitivity** of the HbA1c test (TP/(TP+FN) = 156/200 = 78%), not PPV. Sensitivity and PPV are frequently confused; sensitivity is disease-centric, PPV is test-result-centric. **Mnemonic:** **SPIN** — **SP**ecificity rules **IN**. High specificity (few false positives) → high PPV. Here, specificity = 680/800 = 85%, which is moderate, yielding a moderate PPV of 56.5%. ## Contrast: Sensitivity vs. PPV | Metric | Formula | Interpretation | |--------|---------|----------------| | **Sensitivity** | TP/(TP+FN) = 156/200 = **78%** | Of those WITH disease, how many test positive? | | **Specificity** | TN/(TN+FP) = 680/800 = **85%** | Of those WITHOUT disease, how many test negative? | | **PPV** | TP/(TP+FP) = 156/276 = **56.5%** | Of those testing positive, how many have disease? | | **NPV** | TN/(TN+FN) = 680/724 = **93.9%** | Of those testing negative, how many are disease-free? |