Factors predictive of relapse in cases of bulimia nervosa

American Family Physician,  Nov 1, 1994  

• E-mail
• Print
• Link

Studies indicate that the relapse rate in patients with bulimia nervosa ranges from 30 to 40 percent in the first six to eight months after recovery. Definitions for recovery differ but may be defined as a period of at least two months without binge eating and purging, and relapse may be defined as a return of symptoms at least weekly for a minimum of two months.

Olmsted and associates studied the rate of relapse in patients with bulimia nervosa and attempted to identify indexes of functioning before or after treatment that predicted the subsequent relapse.

Participants in the study were 48 women who met criteria for bulimia nervosa in the revised third edition of the Diagnostic and Statistical Manual of Mental Disorders and who were available for follow-up two years after completion of treatment in a day hospital program. Symptom control was defined as no more than one episode of binge eating or vomiting per week in the month before discharge from the treatment program. Patients who binged or vomited two or more times weekly for three months were considered to have returned to the full syndrome. AU of the patients who did not meet this criterion for relapse over the follow-up period were included in the no-relapse group.

Of the 48 patients in the study, 31.3 percent relapsed within the two-year follow-up period. A total of 16.7 percent of the patients surpassed the threshold level of symptoms within three months of leaving the treatment program. By the end of the sixth month, the relapse rate was 25.0 percent and the majority of the relapses had already occurred. The duration of relapse varied from three to 24 months, and 66.7 percent of the patients who relapsed would have met the diagnostic criteria for bulimia nervosa at least 50 percent of the time.

Among the demographic variables, age was the only significant predictor of relapse and accounted for 8 percent of the variance between patients who relapsed and those who did not. Among the pretreatment variables, greater frequency of vomiting and higher score on the bulimia subscale of an eating attitudes test were the only significant predictors of relapse; frequency of vomiting was the strongest predictor. Binge eating frequency, self-esteem, depression and social adjustment were not significant predictors of relapse.

Among post-treatment variables, vomiting frequency was the strongest predictor of relapse. A higher score on the interpersonal distrust subscale of an eating disorder inventory also was a significant predictor. The prognostic usefulness of being able to trust others may be specific to this study program.

The authors conclude that in patients with bulimia nervosa, residual symptoms of vomiting associated with partial remmission are a significant risk factor for relapse.

COPYRIGHT 1994 American Academy of Family Physicians
COPYRIGHT 2008 Gale, Cengage Learning