The Gold Standard for Evidence
Human clinical trials represent the essential final validation step for health interventions. Phase I trials (20-100 subjects) prioritise safety and tolerability, establishing maximum tolerated doses. Phase II trials (100-500 subjects) assess efficacy signals. Phase III trials (1,000-5,000 subjects) represent definitive efficacy and safety assessment, typically randomised controlled trials with pre-specified primary endpoints and adequate statistical power.
Endpoint types determine what outcomes trials measure. Primary endpoints are pre-specified outcomes determining trial success; these should be clinically relevant hard endpoints measuring patient-centred outcomes (symptom resolution, mortality, hospitalisation). Secondary endpoints measure additional outcomes, generating hypotheses. Surrogate endpoints predict but don't guarantee clinical benefit.
Intent-to-treat (ITT) versus per-protocol (PP) analysis distinctions influence interpretation. ITT preserves randomisation benefits but dilutes intervention effects. PP clarifies 'true' intervention effects but introduces bias. Statistical significance (p-value < 0.05) differs fundamentally from clinical significance.