Complete learning materials
Review each topic, apply the clinical case, then complete the linked ten-question assessment.
Formulate the question
- Use PICO: patient/population, intervention or index test, comparator and outcomes. Select outcomes important to patients.
- Identify appropriate evidence: systematic reviews and randomized trials for interventions; cohort designs for prognosis; cross-sectional diagnostic accuracy studies with suitable reference standards.
- Assess applicability to the actual patient, including comorbidity and preferences.
Understand effect measures
- Relative risk = event risk in treated group / event risk in control group. Absolute risk reduction = control event risk minus treated event risk.
- Number needed to treat = 1 / absolute risk reduction (proportions), over a stated follow-up period; do not extrapolate across time horizons.
- Sensitivity and specificity are conditional on disease status; PPV/NPV depend heavily on prevalence. Likelihood ratios help update clinical probability.
Bias, confidence and limitations
- Randomization and allocation concealment limit selection bias; blinding may reduce performance or assessment bias.
- Confounding, spectrum bias, verification bias, missing data and selective reporting affect interpretation.
- Confidence intervals communicate precision; statistical significance is not equivalent to clinical importance.
Shared evidence-informed decisions
- Compare benefits, harms, burden, cost, alternatives and baseline risk.
- Use current guidelines as aids, but check quality, currency, local requirements and individual circumstances.
- Document uncertainty, patient choice and a plan for monitoring outcomes; reassess new evidence as appropriate.
Applied clinical case
Case: A trial reduces 5-year event risk from 10% to 8%. Absolute risk reduction is 2 percentage points and NNT is 50 over five years; the relative risk reduction is 20%.
Downloadable resources
Practice this lesson (10 questions)
Practice lesson questions