The efficacy of a 15-week self-treatment program for those with bulimia nervosa was investigated. Six females with a DSM-IV diagnosis of bulimia nervosa served as participants in two concurrent multiple-baseline designs. Treatment was self-administered via a therapeutic manual and compliance was measured by the submission of weekly review questionnaires and mail-in reports. Mail-in reports included Fairburn's (1995) self-monitoring sheet, a Time Log, and author-devised Compensatory questionnaire (ACQ), and the State Self-Esteem Scale (Heatherton & Polivy, 1991). Results were examined using visual inspection and interrupted time series analysis (ITSACORR; Crosbie, 1993). At post-treatment, using the method of visual inspection, results showed that treatment exerted a modest controlling influence on binge eating and purging frequencies for 5 of 6 subjects, and that one participant became abstinent (although data was highly variable which made interpretation difficult). Data from 6-month and 1-year follow-ups showed that some, but not all, of the treatment gains were maintained. Although the compensatory skills model proposes that improvements in cognitive and behavioral coping culminate in less frequent binge eating, there was not a close temporal association between skill improvement and binge eating frequency in the current study. Instead, results might be explained by conceptualizing eating self-efficacy as a mediator of compensatory skill (i.e., improved compensatory skill leads to enhanced self-efficacy which produces reduced eating disturbance).