NNT, NNH, sensitivity, specificity, relative risk, odds ratio, study design hierarchy — every biostatistics concept tested on the NAPLEX with formulas and worked examples.
| Measure | Formula | Clinical Meaning |
|---|---|---|
| ARR | Control rate - Treatment rate | Absolute difference in event rates |
| RRR | ARR / Control rate | Proportional reduction in risk (can be misleading) |
| NNT | 1 / ARR | Patients treated to prevent 1 event (lower = better) |
| NNH | 1 / ARI | Patients treated before 1 is harmed (higher = safer) |
| Sensitivity | TP / (TP + FN) | Detects true positives. SnNout (rules OUT) |
| Specificity | TN / (TN + FP) | Detects true negatives. SpPin (rules IN) |
| PPV | TP / (TP + FP) | Probability of disease given positive test |
| NPV | TN / (TN + FN) | Probability of no disease given negative test |
| Relative Risk | Risk exposed / Risk unexposed | Used in cohort studies. RR=1 means no association |
| Odds Ratio | (a/c) / (b/d) from 2x2 table | Used in case-control studies. OR ≈ RR when disease rare |
Key NAPLEX concept: Statistical significance (p < 0.05) does NOT equal clinical significance. A drug may reduce MI risk with p = 0.02 but ARR of 0.3% (NNT = 333). Always evaluate ARR and NNT — not just p-value or RRR — when judging clinical value. Wide confidence intervals indicate imprecision regardless of p-value.
This is a commonly tested NAPLEX concept: sensitivity and specificity are properties of the test and do NOT change with prevalence. However, PPV and NPV DO change with prevalence. As prevalence increases, PPV increases (more true positives) and NPV decreases. In low-prevalence populations, even a highly sensitive/specific test has poor PPV — many positives will be false positives.
PharmacyExam.com includes dedicated biostatistics questions with full clinical context explanations.
Explore PharmacyExam →