Washout Period

A washout period is a designated interval of time in a clinical trial during which participants discontinue a previous treatment and allow its effects to dissipate before beginning a new treatment or entering a study intervention phase. Washout periods are most commonly used in crossover trial designs, where the same participants receive multiple treatments in sequence, and in trials where participants must stop existing medications before starting the study intervention. The purpose of the washout is to eliminate carryover effects, ensuring that the observed outcomes during the subsequent treatment period reflect only the effects of the new treatment and not residual effects from the prior treatment. A common rule of thumb sets the washout at about five elimination half lives of the previous treatment (and of any active metabolites), lengthened when the drug's pharmacodynamic effects last longer.

Without an adequate washout period, treatment effects from a prior intervention may carry over into the next study period, biasing the comparison between treatments. In a crossover trial, if a drug with a long duration of action is followed immediately by a second treatment, the measured outcomes during the second period will reflect a mixture of both treatments, making it difficult to attribute effects to either one. Carryover bias can inflate or attenuate treatment differences and produce misleading conclusions about relative efficacy or safety. In parallel group trials that require participants to stop existing medications before randomization, an inadequate washout may result in withdrawal effects being confused with adverse events from the study drug, or residual efficacy from the prior medication may mask the true treatment difference. The design and justification of the washout period is therefore a critical methodological consideration that directly affects the internal validity of the trial.

The washout period begins after the last dose of the previous treatment and ends before the first dose of the next treatment. During this time, participants typically receive no active treatment (though placebo may be administered to maintain blinding) while the drug and its active metabolites are eliminated from the body. The required duration depends on the pharmacokinetic properties of the drug, particularly its elimination half life. For a drug eliminated by first order kinetics, about 97% has been eliminated after five half lives (96.9%), which is why five half lives is a common rule of thumb for pharmacokinetic washout. However, pharmacodynamic effects may persist longer than the drug itself, particularly for medications that modify receptor expression, gene transcription, or other biological processes with slow recovery kinetics. In such cases, the washout must be guided by pharmacodynamic data rather than pharmacokinetics alone. Researchers should measure the relevant outcome at the end of the washout period to check that participants have returned close to their baseline values before starting the next treatment.

Use a washout period in crossover trials to prevent carryover effects between treatment periods. Use a run in washout before randomization in parallel group trials when participants are taking medications that could interact with or confound the study intervention. Washout periods are essential when studying drugs with long half lives, when outcomes are particularly sensitive to residual drug effects, or when the mechanism of action involves slow onset or slow offset biological changes. The washout duration should be specified in the study protocol and justified based on pharmacokinetic and pharmacodynamic data, referenced literature, or regulatory guidance.

A common mistake is using a washout period that is too short, allowing carryover effects to bias the subsequent treatment period. This is particularly problematic for drugs with active metabolites that have longer half lives than the parent compound. Another error is failing to consider pharmacodynamic persistence, relying solely on pharmacokinetic half life when the biological effects of the drug last much longer than its presence in the bloodstream. In some cases, the washout period may cause ethical concerns if participants with serious conditions must forgo treatment for an extended period, potentially leading to disease worsening, participant withdrawal, or selection bias toward less severe cases. Researchers sometimes treat a non significant statistical test for carryover as proof that the washout was adequate, although such tests have low power, and the once common two stage analysis (testing for carryover, then analyzing only first period data if the test is significant) is now discredited because it gives biased results. Not measuring the baseline value at the end of the washout (before starting the next treatment) removes a direct check that the washout achieved its purpose.

A crossover trial compares two antihypertensive medications in 60 patients with mild hypertension. Each patient receives drug A for four weeks, undergoes a two week washout period, and then receives drug B for four weeks (with the sequence reversed for half the participants). The longer of the two plasma half lives is 12 hours (drug A), so two weeks equal 28 half lives of drug A, far beyond the five half lives of the usual rule of thumb, and neither drug is expected to remain in meaningful amounts. Blood pressure is measured at the end of each four week treatment period and at the end of the washout period. The washout blood pressure readings are close to participants' pre treatment baseline values (mean difference from original baseline: 1.2 mmHg), which supports the chosen washout duration. A test for carryover effects in the crossover analysis is not significant (p = 0.72), which is consistent with an adequate washout but does not by itself show that carryover was absent, because such tests have low power; the case for the washout rests mainly on the drugs' half lives and the washout blood pressure readings.