The aim of the present study was two-fold: to assess the long-term effects of biofeedback on migraine, and to determine the extent to which additional treatment after biofeedback treatment is completed can affect follow-up results. Ninety-six women whose migraine had been treated through biofeedback either 2-3 years (shorter follow-up) or 6-7 years (longer follow-up) earlier were divided into four groups according to how long ago they had received their biofeedback treatment and whether or not they had received any additional treatment (whatever the nature) since the posttreatment assessment: shorter follow-up without additional treatment (n = 24), shorter follow-up with additional treatment (n = 22), longer follow-up without additional treatment (n = 24), and longer follow-up with additional treatment (n = 26). Subjects were asked to keep a headache diary for five weeks, as they had done during earlier assessments. Overall, results showed that migraine activity was significantly less at follow-up than at pretreatment. However, whether the observed long-term benefits could be attributed to biofeedback was unclear because medication was found to be as much in use at follow-up as at pretreatment. Results also indicated that 51% of the subjects did seek additional treatment between posttreatment and follow-up. Contrary to expectations, however, additional treatment was not associated with better therapeutic benefits at follow-up. The research and the clinical implications of those findings are discussed.