Practitioners'decisions of when to use ST (planned short-term) versus LT (long-term) therapy were studied via their responses to vignettes and an open-ended question about their cnteria for choosing. Using a factorial design, the effects of clients' symptoms (adjustment disorder, Dysthymic Disorder, and borderline personality disorder), gender, race (White and Black), and education (high school and college) were tested. The presence of adjustment disorder symptoms alone was able to correctly predict therapist's decision of when to use ST and LT 92% of the time. ST was recommended for adjustment disorder 95% of the time, but only 13% for Dysthymic Disorder and 9% for borderline personality. In the remainder of these cases LT was recommended. Responses to open-ended questions further support these results. Clients'race, gender, and education did not have a general effect on the choice between ST and LT, except in the dysthymia and borderline groups. There were no significant interactions between client variables. Therapist characteristics had only slight relationships to decisions made. Despite preferences to treat adjustment disorder with ST, and that 25 percent of actual respondent's chents were so diagnosed, therapists treated less than 1.5 ST cases annually.