Antiasthma agents are usually evaluated in children only after they ha
ve been developed in adults. Homogeneous well-defined study population
s should be selected on the basis of age, respiratory function test re
sults, symptoms, and drug use. Because the severity of asthma varies w
idely across patients and over time in a given patient, studies of ant
iasthma drugs should be long-term. Delivery of inhaled drugs should be
done in a well-standardized manner (method of inhalation, delivery sy
stem, dose). Whenever possible, the evaluation criteria should include
a parameter reflecting ventilatory function (peak flow rate, forced e
xpiratory volume in one second). The marked placebo effect seen in ast
hma requires use of a controlled design.