The study investigated whether a consultant-driven prereferral intervention (Mainstream Assistance Teams) may be shortened in duration, thus improving its efficiency, without reducing its effectiveness. Subjects were 60 general educators, their 60 most difficult-to-teach nonhandicapped pupils, and 22 consultants, representing 17 elementary schools in a large metropolitan school system. The teachers were assigned randomly to a short (N=24) or long (N=24) version of the prereferral intervention or to a control group (N=12). Analyses indicated that the two variants of the prereferral intervention improved teacher perceptions of their difficult-to-teach students, and decreased referrals for testing and possible special education placement. Moreover, results suggested that the short and long versions were equally effective. Implications for consultation-related activity are discussed. Contains 21 references. (Author)