DISTINGUISHING RECOVERY FROM REMISSION IN A COHORT OF BULIMIC WOMEN -HOW SHOULD ASYMPTOMATIC PERIODS BE DESCRIBED

Background: Empirical definitions of remission and recovery from eatin g disorders are needed to understand outcome data and compare results across studies. Method: 106 treatment-seeking women with bulimia nervo sa, who had abstained from binging and purging for at least 4 weeks, w ere followed prospectively. Relapse was defined as at least 4 consecut ive weeks of either binging and purging weekly or hinging two or more times per week, regardless of purging. Recovery was differentiated fro m remission based on the probability of relapse. The minimum number of weeks after which the risk of relapse leveled off was used as the cut -off to distinguish between the two outcomes. Kaplan-Meier methods wer e used to estimate the weekly probability of relapse. Results: When de fining remission as at least 4 weeks of being asymptomatic, a quarter of the women relapsed within 11 weeks. By 37 weeks, only 49% of the wo men remained asymptomatic (95% CI, 41-61). The probability of relapse was substantial for approximately a year after a woman ceased to binge and purge. Conclusion: Bulimia nervosa is an episodic disorder. As a conservative approach, periods of being asymptomatic that last less th an 1 year should be labeled as remissions, not recoveries. (C) 1997 El sevier Science Inc.