Medication management of stimulants in pediatric practice settings: a national perspective.

Abstract

Using a nationally representative sample of office-based physicians, the management practices of pediatricians, psychiatrists, and family practice physicians were investigated. The major aims were to determine (1) what types of services these physicians were providing to children who received stimulants, (2) what factors predicted receipt of stimulants, and (3) whether these practices were concordant or discordant with professional consensus on diagnosis and treatment of attention-deficit hyperactivity disorders (ADHD). Prescribing and management practice data from the 1995 National Ambulatory Medical Care Survey (NAMCS) were analyzed for children ages 0 to 17 years who were seen for psychiatric problems and received stimulant medication. Results indicated that 2 million visits by children were made in 1995 to psychiatrists, pediatricians, or family practitioners in which psychotropic medications were prescribed. In pediatric visits where stimulant medication was prescribed, mental health counseling was provided 47.3% of the time and psychotherapy 21.6%. Follow-up arrangements were made in 79.1% of the visits. Psychiatrists were significantly more likely to provide psychotherapy and to specify follow-up visits than were pediatricians, but less likely to provide other health counseling. Controlling for demographic and physician effects, the factors with the most significant effect on the probability of receiving stimulants were geographic region (living in the South), race (being white), receiving mental health counseling, not receiving psychotherapy, and having health insurance. Less than 50% of pediatric visits for psychiatric reasons involving stimulant medications included any form of psychosocial intervention. In 21% of these visits, no recommendations were made for follow-up care. These practices diverge from National Institutes of Health (NIH) consensus panel recommendations and association-issued practice parameters.