Research Methodology, Biostatistics, and Evidence-Based Practice
Core Concepts
Understand study designs (cohort, case-control, cross-sectional, RCT), measures of association (odds ratio, relative risk, hazard ratio), and measures of diagnostic accuracy (sensitivity, specificity, PPV, NPV, likelihood ratios). Know p-values, confidence intervals, and the difference between statistical and clinical significance. Recognize bias types (selection, information, confounding) and the levels of evidence.
Common Traps
- Confusing odds ratio with relative risk; OR approximates RR only when the outcome is rare.
- Misinterpreting PPV and NPV because they depend on prevalence.
- Treating a p-value as a measure of effect size.
- Ignoring confidence intervals and relying only on p-values.
Rule of Thumb
When evaluating a diagnostic test, remember that sensitivity and specificity are prevalence-independent, but PPV and NPV are not. A worked example: a test with 90% sensitivity and 95% specificity in a population with 1% prevalence yields a low PPV despite good sensitivity; increasing prevalence raises PPV. For therapy questions, use the number needed to treat (NNT = 1/ARR).
Evidence-Based Practice
Appraise studies for validity, results, and applicability; prefer systematic reviews and RCTs for therapy questions.