Standard dosing practice with stimulant medication often is that of taking the child's weight and estimating a dose high enough to yield a therapeutic response while minimizing repeat visits. This practice is perhaps encouraged by wishes to (1) relieve family distress as quickly as possible, (2) reduce disruptive behaviors in classrooms, (3) work in a cost and time-efficient manner. There has also been a notion that faster onset of stimulants in the bloodstream may be associated with a more powerful behavioral response; thus, taking stimulants on an empty stomach has sometimes been encouraged.
My clinical experience has been that children often resist taking stimulants. Their stated reasons include (1) it makes them feel weird or different, (2) they feel neither as entertaining nor so popular as before, (3) it's a nuisance to get to the nurse, (4) it makes them different from their friends.
It's likely that "adaptive systems" (Barkow, et al., 1992) exist in children that are comparable to those seen in wild creatures; one such system is for avoiding toxic foods. Adaptive systems are phylogenetically old and consist of several receptor-integrator-effector mechanisms working efficiently and cooperatively for a common goal that enhances survival.
Initiating medication has a striking operational similarity to procedures and results described with other species as bait poisoning (Garcia & Koelling, 1966). The paradigm involves giving a novel food and next making subject ill by injecting toxic medications later; the animal will no longer eat the novel food, even if 24 hours elapse between the novel food and the medication and even if only a single trial was given. This phenomenon is an apparent inconsistency with other kinds of learning that require immediacy of reinforcement. (Profet (1992) describes an adaptive system in mothers such that their tolerance for novel or spicy foods is reduced during pregnancy, precisely at that interval when the fetus has the greatest development of its central nervous system and greatest sensitivity to toxins.)
A possibility exists that starting with larger doses of a stimulant, especially on an empty stomach, for a more rapid therapeutic effect incurs the delayed cost of the child's abhorring his medication. Accordingly, I have recommended to physicians and parents for several years the option of starting with about half the expected behaviorally effective dose, encouraging that it be taken on a full stomach, and enlisting the child as an active collaborator in determining how much is to be given. At no time has a parent challenged either the strategy or the goal of possibly enhancing the long term acceptance by their child. Some parents seem relieved by a "go slow" approach because they, too, worry about making their child ill. (So much so that many parents seem to prefer unstudied nutritional supplements because they are seen as "natural.") Since beginning this modified approach, the only complaints I receive are from adolescents who (1) have taken stimulants in the past and don't' want to return to them or (2) still are troubled by the "being different" issues.
References
Barkow, J., Cosmides, L., Tooby, J. (Eds.) (1992) The adapted mind: Evolutionary psychology and the generation of culture. Oxford: New York.
Garcia, J.& Koelling, R. (1966) Relation of cue to consequence in avoidance learning. Psychonomic Science, 4, 123-124.
Profet, M. (1992) Pregnancy sickness as adaptation: A deterrent to maternal ingestion of teratogens. In Barkow, J., et al., op cit., pp. 327-365.