It's not that bad: error introduced by oral stimulants in salivary cortisol research.

Evidence that oral stimulants can produce interference effects in salivary cortisol assays has led to advice to avoid their use. However, in studies with young children, the use of these saliva-producing substances increases compliance with collection procedures. Four experiments are described to examine the effects of two commonly used stimulants, SweetTarts and Koolaid. Across these experiments, interference effects produced by different quantities of these stimulants (0.025, 0.1, and 0.2 g) and those produced in two commonly used assays, DELFIA and EIA, were explored. The impact of using cotton rolls soaked with an oral stimulant prior to saliva collection was also examined. In general, oral stimulants did not affect the rank ordering of cortisol values, as the results for stimulant-treated samples were highly correlated with those of untreated samples (approximately .90 for serially collected samples and .95 for aliquots of the same saliva pool). Depending on which assay was used, however, oral stimulants increased or decreased the cortisol levels reported, with effects sizes in the small to medium range. Thus, oral stimulants should not be used with only a portion of the subjects in a study, nor should researchers assay stimulant-treated samples from the same study using different assays. When used sparingly, oral stimulants can be employed without compromising the quality of salivary cortisol data. Copyright 2005 Wiley Periodicals, Inc.