Objective. Biofeedback is considered an effective treatment for anal constipa tion, but a substantial proportion of patients fail to improve. Our aim was to i dentify the key predictors of outcome using a comprehensive standardized evaluat ion of anorectal function. Material and methods. We retrospectively analysed the clinical and physiological data of 148 patients consecutively treated for const ipation due to functional outlet obstruction by biofeedback. Clinical evaluation was performed by means of a structured questionnaire. Anorectal evaluation incl uded anal pressure, neural reflexes, defecatory dynamics, rectal compliance, rec tal sensitivity and balloon expulsion test. Biofeedback treatment was performed using a manometric technique. The clinical response to biofeedback treatment was evaluated as good (improvement of constipation) or poor (no improvement or wors ening). Results. Of the 148 patients included, 112 (86 F, 26 M; age range 8- 67 years) were followed- up for between 1 and 44 months, and 66% had a good res ponse to treatment. The response depended on the severity of the defecatory dysf unction. Thus, lack of anal relaxation during straining and inability to evacuat e a 1 ml intrarectal balloon were inversely related to physiological variables r elated to therapeutic success. Among the 49 patients with absent anal relaxation , 51% had a good response to treatment (versus 78% in patients with partial relaxation; p < 0.01), and among the 29 patients with failed balloon expulsion, 48% responded to treatment (versus 74% in patients able to evacuate ≥ 1 ml intrarectal balloon; p < 0.05). Conclusions. Even in the presence of negative p redictors, biofeedback is a valuable treatment option in a substantial proportio n of constipated patients. Objective. Biofeedback is considered an effective treatment for anal constipation, but a substantial proportion of patients fail to improve. Our aim was to i dentify the key predictors of outcome using a consistent standardized evaluation of anorectal function. Material and methods. We retrospectively analysed the clinical and physiological data of 148 patients consecutively treated for const ipation due to functional outlet obstruction by biofeedback. Clinical evaluation was performed by means of a structured questionnaire. Anorectal evaluation incl uded anal pressure, neural reflexes, defecatory dynamics, rectal compliance, rec tal sensitivity and balloon expulsion test. Biofeedback treatment was performed using a manometric technique. The clinical response to biofeedback treatment was evaluated as good (improvement of constipation) or poor (no improvement or worsening). Results. Of the 148 patients included, 112 (86 F, 26 M; age range 8- 67 years) were followed-up for between 1 and 44 months, and 66% had a good res ponse to treatment. The response depended on the severity of the defecatory dysfunction. Thus, lack of anal relaxation during straining and inability to evacuate ea 1 ml intrarectal balloon were inversely related to physiology Among the 49 patients with absent anal relaxation, 51% had a good response to treatment (versus 78% in patients with partial relaxation; p <0.01), and among the 29 patients with failed balloon expulsion, 48 % responded to treatment (with 74% in patients able to evacuate ≥ 1 ml intrarectal balloon; p <0.05). Conclusions. Even in the presence of negative p redictors, biofeedback is a valuable treatment option in a substantial proportions of constipated patients.