Introduction
A clinical trial is any study in humans designed to discover or verify the clinical, pharmacokinetic or pharmacodynamic effects of an investigational medicine, to establish its efficacy and safety.
Trial designs
- Parallel group: each patient receives one treatment.
- Cross-over: each patient receives both treatments in random order with a wash-out; fewer subjects needed; only for stable chronic conditions.
- Factorial: tests two or more treatments at once (A, B, A+B, neither).
- Open (non-blinded), single-blind (patient unaware) or double-blind (patient and investigator unaware).
Reducing bias
Placebo
A placebo looks identical to the test drug but has no active substance. It separates the drug's real pharmacodynamic effect from psychological effects and from the natural course of the disease. Adverse reactions also occur on placebo: in a losartan trial, headache occurred in 14% on the drug but 17% on placebo, so the drug did not cause it.
Blinding
Prevents expectations of patients and investigators from influencing results and their assessment.
Randomisation
Each subject has an equal chance of entering any group. It balances known and unknown risk factors, removes the investigator's influence on allocation, and makes statistical tests valid.
A randomised, double-blind, placebo-controlled trial is the strongest design for proving that a drug, and not chance, bias or expectation, causes the effect.