Article

A Placebo-Controlled Study of Fluoxetine

in Continued Treatment of Bulimia Nervosa
After Successful Acute Fluoxetine Treatment

Steven J. Romano, M.D.
Katherine A. Halmi, M.D.
Neena P. Sarkar, Ph.D.
Stephanie C. Koke, M.S.
Julia S. Lee, M.S.
Objective: The efficacy of fluoxetine in
the acute management of bulimia nervosa
is well established; however, few con-
trolled studies have examined whether
continuation of pharmacotherapy pro-
vides protection from relapse. This study
compared the efficacy and safety of treat-
ment with fluoxetine versus placebo in
preventing relapse of bulimia nervosa dur-
ing a 52-week period after successful acute
fluoxetine therapy.
Method: Patients who met DSM-IV crite-
ria for bulimia nervosa, purging type,
were assigned to single-blind treatment
with 60 mg/day of fluoxetine. After 8
weeks of treatment, patients were consid-
ered responders if they experienced a de-
crease ≥ 50% from baseline in the fre-
quency of vomiting episodes during 1 of
the 2 preceding weeks. Responders were
randomly assigned to receive 60 mg/day
of fluoxetine or placebo and were moni-
tored for relapse for up to 52 weeks. Pa-
tients met relapse criteria if they experi-
enced a return to the baseline vomiting
frequency that persisted for 2 consecutive
weeks.
Results: Of the 232 patients who entered
the acute phase, 150 patients (65%) met
response criteria and were randomly as-
signed to receive fluoxetine (N=76) or pla-
cebo (N=74). Fluoxetine-treated patients
exhibited a longer time to relapse than
placebo-treated patients. Quantitative
analysis of other efficacy measures, in-
cluding frequency of vomiting episodes,
frequency of binge eating episodes, Clini-
cal Global Impression severity and im-
provement scores, the patient’s global im-
pression score, and Yale-Brown-Cornell
Eating Disorder Scale score, indicated that
the efficacy of fluoxetine treatment was
statistically superior, compared to pla-
cebo. There were no clinically relevant
differences in safety between groups. At-
trition in this study was high, especially in
the first 3 months after random assign-
ment to treatment groups.
Conclusions: Continued treatment with
fluoxetine in patients with bulimia ner-
vosa who responded to acute treatment
with fluoxetine improved outcome and
decreased the likelihood of relapse.

(Am J Psychiatry 2002; 159:96–102)

B ulimia nervosa, characterized by uncontrolled

binge-eating and self-induced purging or other compen-
satory behaviors to prevent weight gain, is associated with
substantial rates of psychiatric and medical comorbidity
(1). Its lifetime prevalence is estimated to be as high as
2.8% (2). Although the causal mechanisms underlying bu-
limia nervosa are not completely understood, evidence
suggests that it is associated with abnormalities in central
serotonergic function. Women with bulimia nervosa have
a blunted prolactin response and a normal cortisol re-
sponse to the postsynaptic serotonergic agonist m-chloro-
phenylpiperazine given orally and to the serotonin pre-
cursor L -tryptophan given intravenously, suggesting
alteration of receptor activity at or above the hypothala-
mus (3). A blunted prolactin response to fenfluramine has
also been demonstrated (4). Since satiety in response to
food intake is mediated in part by serotonergic input to
the ventromedial hypothalamus, a defect in serotonin
function may produce impaired recognition of satiety,
thus contributing to binge eating (5).
A number of antidepressants have been effective in the
short-term treatment of bulimia nervosa. Overall, medica-
tion treatment decreases binge frequency by an average of
56%, compared with an average decrease of 11% with pla-
cebo treatment (6). Decreases in the frequency of self-in-
duced vomiting appear to be similar. Tricyclic antidepres-
sants, monoamine oxidase inhibitors, bupropion, and
trazodone have all been shown to be therapeutically supe-
rior to placebo (7–16). Fluoxetine has proven to be effec-
tive in reducing bulimic symptoms in two 8-week double-
blind trials (17, 18) and in one longer term, 16-week dou-
ble-blind trial (19).
Fewer data are available regarding the long-term effi-
cacy of pharmacotherapy, as well as its impact on relapse
prevention. Uncontrolled follow-up studies have sug-
gested that patients with bulimia nervosa often remain
symptomatic for 4–6 years, although they may maintain

96 Am J Psychiatry 159:1, January 2002


improvements achieved during treatment (6). A review of
eight studies with follow-up periods of 6 months to 6 years
showed that approximately one-third of the patients who
improved acutely experienced a resurgence of bulimic
symptoms. The probability of relapse appears to be high-
est during the first year after recovery, with the risk declin-
ing after 4 years (20).
The present study was designed to prospectively evalu-
ate the efficacy and safety of treatment with fluoxetine
versus placebo in preventing relapse of bulimia nervosa
over a 52-week double-blind treatment period after re-
sponse to acute fluoxetine treatment.
Method
Study Population
The study was conducted at 16 sites in the United States. All in-
vestigators were psychiatrists or physicians specializing in psychi-
atry. The protocol was approved by the institutional review board
at each site. Patients’ participation in the study was voluntary, and
all patients gave written informed consent before enrollment.
The study population included male and female outpatients, at
least 18 years old, with a psychiatric diagnosis of bulimia nervosa,
purging type, as defined by the DSM-IV criteria. Purging must
have included self-induced vomiting. Patients were excluded if
they had participated in a prior fluoxetine study or taken fluoxe-
tine within 5 weeks before enrollment or if they had been previ-
ously treated with 60 mg/day of fluoxetine for longer than 8
weeks. Patients were excluded if they had coexisting schizophre-
nia or bipolar disorder, mood-congruent or mood-incongruent
psychotic features, serious suicidal risk, organic brain disease, a
history of seizures, a medically unstable condition, or a history of
an alcohol and/or other substance abuse disorder within 3
months before enrollment. Patients who had used a monoamine
oxidase inhibitor within 2 weeks before enrollment or had used
other investigational drugs or psychoactive medications within 4
weeks before enrollment were excluded. Patients who had re-
ceived cognitive behavior therapy within 4 weeks of enrollment or
who planned to begin a structured, focused therapy at any time
during the study were excluded.
Study Design
After a 1-week no-therapy screening phase, patients who con-
tinued to meet entry criteria were assigned to single-blind treat-
ment with 60 mg/day of fluoxetine. Dosage adjustment, at the cli-
nician’s discretion, was allowed during the first 2 weeks of
treatment to accommodate patients who were initially unable to
tolerate 60 mg/day of fluoxetine. After 8 weeks of acute treatment,
responders were randomly assigned to receive 60 mg/day of flu-
oxetine or placebo for up to 52 weeks. Nonresponders and pa-
tients who were unable to tolerate a dose of 60 mg/day were dis-
continued from the study.
Study medication was packaged in blister packs and contained
either 20-mg fluoxetine capsules or matching placebo capsules.
At each visit, the patient returned the blister packs so that the re-
maining capsules could be counted. Patients who missed their
medication for 5 consecutive days or who failed to attend visits
within the stated periods were deemed noncompliant and were
withdrawn from the study. The treatment group to which the pa-
tient was assigned was determined by a computer-generated ran-
domization sequence. No one at the study site had access to the
study randomization list. Emergency labels, generated by a com-
puter drug-labeling system, were available to the investigator.
These labels, which revealed the patient’s treatment group, were
Am J Psychiatry 159:1, January 2002
ROMANO, HALMI, SARKAR, ET AL.
opened only if the choice of follow-up treatment depended on the
patient’s therapy assignment or in medical emergencies.
Study Procedures
During the screening phase and the 8-week single-blind acute
therapy phase, patients were seen by the investigators each week.
During the 52-week, double-blind therapy period, visits occurred
at 2-week intervals during the first 8 weeks and at 4-week inter-
vals thereafter.
Efficacy was measured by the change in the number of vomit-
ing and binge-eating episodes per week and by ratings on the Eat-
ing Disorder Inventory (21), Yale-Brown-Cornell Eating Disorder
Scale (22), and Clinical Global Impression (CGI) severity and im-
provement scales (23) and by the patient’s global impression,
which uses a 7-point scale identical to the Clinical Global Impres-
sion improvement scale. The 17-item Hamilton Depression Rat-
ing Scale (24) was used to measure the presence and severity of
depression. Information about bulimic behaviors was obtained
by using a daily diary in which patients recorded their episodes of
binge eating, vomiting, and other purging behaviors.
The CGI severity measure was obtained at the beginning of
acute therapy (visit 2). The Eating Disorder Inventory, Yale-
Brown-Cornell Eating Disorder Scale, CGI improvement scale,
the patient’s global impression, and Hamilton depression scale
were completed at the beginning of acute therapy (visit 2), at the
end of acute therapy (visit 10), and at 4-week intervals throughout
the double-blind relapse-prevention period. The patients’ diaries
were collected at each visit. Data on adverse events were collected
by open-ended questioning about general well-being and prob-
lems with medication and were recorded at each visit regardless
of the perceived relationship to therapy.
Statistical Analyses
The primary efficacy measure was the change in the number of
vomiting episodes per week. This measure was also used to deter-
mine the patient’s eligibility for continuation into the double-
blind therapy period and to assess relapse during the double-blind
period. To be designated a treatment responder at the end of the
acute therapy period, patients must have experienced a decrease
of ≥50% in the frequency of vomiting episodes during at least 1 of
the 2 preceding weeks, compared to the baseline (screening) week.
Patients were considered relapsers during the double-blind period
if they experienced a return to the baseline frequency of vomiting
for 2 consecutive weeks. For ethical reasons, patients who experi-
enced a return to the baseline frequency of vomiting for a single
week and whose condition had deteriorated significantly could
also be designated as relapsers at the clinician’s discretion.
Time-to-relapse curves were estimated for each treatment
group by using the product-limit (Kaplan-Meier) method for the
data for all patients with at least one postrandomization mea-
surement. A two-sided log-rank test was used to compare the
time-to-relapse distributions. Confidence limits for estimated cu-
mulative relapse rates were constructed for the 3- , 5- , and 12-
month time points by using Greenwood’s estimate of the variance
(25), and the difference in estimated relapse rates was tested at
these time points by using the proportions test proposed by Gray
and Tsiatis (26). In addition, a Cox proportional hazards model
stratified by investigative site was used to assess the relationship
between baseline covariates (frequency of vomiting and binge
eating) and time to relapse.
Endpoint analysis of continuous efficacy and safety data (vital
signs and laboratory data) evaluated change from randomization
to the last visit in the postrandomization period (last observation
carried forward). An analysis of variance model with treatment,
investigator, and the treatment-by-investigator interaction was
used to compute the summary statistics for normally distributed
measurements. A rank-transformed nonparametric model pro-
97
FLUOXETINE FOR BULIMIA

TABLE 1. Characteristics of Patients With Bulimia Nervosa FIGURE 1. Kaplan-Meier Estimates of the Percentage of
Before Acute Treatment With Fluoxetine and at Random Patients With Bulimia Nervosa Remaining Relapse Free in
Assignment to Continued Treatment or Placebo After a Randomized, Placebo-Controlled Study of Continued
Acute Response to Fluoxetinea Treatment After Acute Response to Fluoxetinea
Patients Patients
Receiving Receiving Patients Remaining at Risk
Fluoxetine Placebo 0 months 3 months 6 months 12 months
Characteristic (N=76) (N=74) Fluoxetine
N N (N=76) (N=42) (N=26) (N=13)
Characteristics at baseline Placebo
Sex (N=74) (N=16) (N=10) (N=6)
Female 74 73 100
Male 2 1
Ethnicity

Remaining Relapse Free

Percentage of Patients
Caucasian 71 65 80
Other 5 9

Mean SD Mean SD 60

Age (years)b 29.5 7.0 30.0 9.3

Body mass index 22.5 3.9 23.0 3.8 40
Age at diagnosis 25.3 7.7 26.3 9.3
Vomiting episodes/weekb 12.1 8.7 14.0 11.7
Binge eating episodes/weekb 10.3 7.7 12.5 10.1 20
Clinical Global Impression severity score 4.5 0.6 4.5 0.7
Hamilton Depression Rating Scale score 10.5 6.1 10.5 5.9
Eating Disorder Inventory total score 76.6 26.9 78.4 29.9 0
Yale-Brown-Cornell Eating Disorder 0 100 200 300 400
Scale score 18.8 4.1 18.3 5.1 Days From Randomization
Characteristics at random assignment to Relapse or Discontinuation
Vomiting episodes/week 4.1 5.5 4.5 6.1
Binge eating episodes/week 3.0 4.8 3.9 5.1 a Two-sided log-rank test applied to the Kaplan-Meier survival func-
Clinical Global Impression severity tion (χ2=5.79, df=1, p<0.02).
score 2.9 1.0 2.9 0.9
Hamilton Depression Rating Scale
scoreb 4.6 3.9 6.1 5.3 apy. The treatment groups were comparable at baseline
Eating Disorder Inventory total score 37.0 22.0 39.1 27.2 with respect to age, gender, race, body mass index, and
Yale-Brown-Cornell Eating Disorder
Scale score 9.4 4.8 9.4 5.4 severity of illness. At random assignment to treatment
a No statistically significant differences between treatment groups. groups, the patients’ severity of illness was comparable.
Baseline age was determined at the beginning of a 1-week no- The majority of the patients were women and Caucasian
therapy screening phase (visit 1); all other baseline measurements (Table 1).
were assessed at the end of the no-therapy screening phase
(visit 2). Of the 232 patients who received single-blind acute
b Unequal variance.
therapy, 94 patients (40.5%) had comorbid depression
(Hamilton depression scale total ≥12). There was no sig-
posed by Akritas et al. (27) was used for measurements that failed nificant difference in response rates between depressed
the normal assumption. Data from three sites with two or fewer and nondepressed patients (62.8% versus 65.7%) (χ2=0.21,
patients per treatment group were pooled together for the analy-
df=1, p=0.65). Of the 232 patients who received single-
sis of continuous efficacy and safety.
Data on patients’ characteristics were summarized at baseline
blind acute therapy, 41 patients (17.7%) had complete re-
and at random assignment to the treatment groups. Treatment mission of symptoms, that is, no vomiting episodes during
differences were assessed with Fisher’s exact test for categorical the final week of acute therapy.
variables and Student’s t test for continuous variables. Reasons
for discontinuation from the study and adverse events that first Time to Relapse
occurred or worsened during double-blind therapy were com-
pared between treatment groups by using Fisher’s exact test. The primary hypothesis, that fluoxetine treatment pro-
Statistical significance was set at p<0.05. All statistical tests longs the time to relapse, was assessed by a log-rank test
were two-sided. applied to the Kaplan-Meier survival function. The test
was statistically significant (χ2=5.79, df=1, p<0.02). The
Results survival curve shows that most of the relapses among pla-
cebo-treated patients occurred during the first 3 months
Acute Response after randomization (Figure 1). Compared with the pla-
Of the 265 patients who participated in psychological cebo group, the fluoxetine-treated group had a lower
and physical screening, 33 did not meet entry criteria and number of relapses during the first 3 months and a more
were excluded. Of the 232 patients who received single- gradually decreasing probability of remaining relapse free
blind acute therapy, 150 patients (64.7%) met response cri- (Figure 1). The estimated relapse rates and differences in
teria and were randomly assigned to double-blind ther- relapse rates (the rate for the placebo group minus the rate

98 Am J Psychiatry 159:1, January 2002

 ROMANO, HALMI, SARKAR, ET AL.

TABLE 2. Estimated Relapse Rates of Patients With Bulimia Nervosa at 3, 6, and 12 Months of a Randomized, Placebo-
Controlled Study of Continued Treatment After Acute Response to Fluoxetine
Patients Receiving Fluoxetine Patients Receiving Placebo
Estimated Estimated
Time N Relapse Rate (%) 95% CI (%) N Relapse Rate (%) 95% CI (%) Difference (%)a 95% CI (%) p
3 months 42 19 9 to 29 17 37 23 to 51 18 1 to 35 <0.04
6 months 26 29 16 to 42 9 43 26 to 59 14 –7 to 35 0.19
12 months 11 33 19 to 48 5 51 30 to 71 18 –8 to 42 0.17
a Analysis of differences conducted with the proportions test proposed by Gray and Tsiatis (26).

TABLE 3. Mean Change in Efficacy Measures from Randomization to Endpoint for Patients With Bulimia Nervosa in a
Placebo-Controlled Study of Continued Treatment After Acute Response to Fluoxetine
Efficacy Measure Patients Receiving Fluoxetine Patients Receiving Placebo Analysis
Mean SD Na Mean SD Na χ2b df p

Vomiting episodes/week 2.92 7.08 74 4.82 8.43 71 10.34 1 0.001

Binge eating episodes/week 2.47 6.58 74 4.11 6.70 71 5.91 1 <0.02

Mean SD Na Mean SD Na Fc df p
Clinical Global Impression
Severity score 0.45 1.33 75 0.97 1.21 71 8.43 1, 118 0.004
Improvement score 0.77 1.43 75 1.37 1.39 71 7.13 1, 118 0.009
Patient’s global impression score 0.72 1.54 75 1.37 1.49 71 4.94 1, 118 <0.03
Hamilton Depression Rating Scale total score 2.03 5.66 75 3.23 6.60 71 1.74 1, 118 0.19
Eating Disorder Inventory total score 7.79 25.49 71 17.41 24.45 69 2.30 1, 113 0.13
Yale-Brown-Cornell Eating Disorder Scale
Total score 2.92 7.91 63 7.38 6.80 60 9.75 1, 95 0.002
Preoccupation score 1.53 3.82 64 3.63 3.74 60 7.28 1, 96 0.008
Ritual score 1.35 4.51 63 3.75 3.79 60 8.51 1, 95 0.004
a Number of patients with a measurement at the randomization visit and at least one postrandomization visit. Last observation was carried
forward.
b Asymptotic chi-square distribution from rank-transformed nonparametric model by Akritas et al. (27).
c Type III sum of squares analysis of variance with treatment, investigator, and treatment-by-investigator interaction.

for the fluoxetine group) were computed at 3- , 6- , and 12-
month time points (Table 2). The difference in relapse
rates was statistically significant at the 3-month time
point and remained approximately constant from 3–12
months, although the difference was not statistically sig-
nificant at the 6- and 12-month time points because of the
high attrition rate in the study. The Cox proportional haz-
ards model indicated that the relapse rate did not depend
on the baseline frequency of vomiting or binge eating.
Endpoint Analysis of Change in Efficacy Measures
Analysis of mean change in efficacy measures from ran-
domization to endpoint shows that both the fluoxetine-
treated group and the placebo-treated group worsened
over time in each measure. However, statistically signifi-
cant differences favoring fluoxetine were observed for
vomiting episodes, binge-eating episodes, CGI severity
score, CGI improvement score, patient’s global impression
score, and Yale-Brown-Cornell Eating Disorder Scale total
score, including statistically significant differences in both
the preoccupation and ritual subtotals (Table 3).
Safety
Adverse events that were first reported or that worsened
during double-blind therapy were compared between
treatment groups. The only event with a statistically signif-
icant difference in frequency between groups was rhinitis
(31.6% for the fluoxetine group, compared to 16.2% for the
placebo group) (p<0.04, Fisher’s exact test). The rate of dis-
continuation because of an adverse event was also similar
between treatment groups, and the only adverse event
leading to discontinuation in ≥2% of patients was unin-
tended pregnancy (2.6% for the fluoxetine group, com-
pared to 4.1% for the placebo group).
There were no statistically significant differences be-
tween groups in change from randomization to endpoint
for systolic or diastolic blood pressure, heart rate, temper-
ature, or weight. Although there were statistically signifi-
cant differences in mean changes in certain laboratory
analytes from baseline to endpoint (magnesium, bicar-
bonate, and urine pH; all with a negative change in the flu-
oxetine-treated group and a positive change in the pla-
cebo-treated group), the mean endpoint values were not
indicative of drug-related toxicity.
Reasons for discontinuation were analyzed by time in-
tervals after randomization (0–3 months, 3–6 months, and
6–12 months) (Table 4). Although the overall number of
patients who discontinued for various reasons was similar
among treatment groups, there were some interesting
trends that appeared to be time-related. In the first 3
months of the study, a large number of patients (11 fluoxe-
tine-treated patients and 20 placebo-treated patients) dis-
continued for reasons of either patient decision or physi-
cian decision. Further inspection of individual study
comments for these patients revealed that eight fluoxe-

Am J Psychiatry 159:1, January 2002 99

FLUOXETINE FOR BULIMIA

TABLE 4. Reasons for Attrition of Patients With Bulimia Within the group randomly assigned to continued fluoxe-
Nervosa From a Randomized, Placebo-Controlled Study of tine treatment, the comparable relapse rates among de-
Continued Treatment After Acute Response to Fluoxetine
pressed patients and nondepressed patients (24.1% versus
Reason for Attrition and Months After Randomization
21.2%) suggest that comorbid depression is not a predic-
Treatment Group 0–3 3–6 6–12 Total
tor of relapse among responders to acute treatment for
Adverse event
Fluoxetine 2 2 0 4 bulimia nervosa.
Placebo 3 0 0 3 The majority of patients who received placebo relapsed
Lost to follow-up
Fluoxetine 4 2 3 9
within the first 3 months after randomization, in contrast
Placebo 6 3 1 10 to the more gradual loss of fluoxetine-treated patients due
Patient decision to relapse during the course of the trial. This finding sug-
Fluoxetine 8 4 6 18
Placebo 14 4 2 20
gests that there may be a subgroup of pharmacologically
Physician decision responsive bulimic patients who are highly sensitive to
Fluoxetine 3 0 0 3 medication discontinuation. Future research may help
Placebo 6 1 0 7
Protocol variance
elucidate traits of this subgroup and potentially determine
Fluoxetine 1 1 2 4 clinical strategies. It is noteworthy that the pharmacoki-
Placebo 1 0 0 1 netic profile of fluoxetine and norfluoxetine, its active me-
Relapse
Fluoxetine 12 2 3 17
tabolite, parallels this observed pattern of relapse, as ap-
Placebo 20 0 2 22 proximately 5–6 weeks are required for drug clearance.
Noncompliance To our knowledge, the present study is the largest pla-
Fluoxetine 3 3 2 8
Placebo 5 0 0 5 cebo-controlled, double-blind trial evaluating relapse
All reasons prevention associated with maintenance pharmacother-
Fluoxetine 33 14 16 63 apy for bulimia nervosa. However, the study had several
Placebo 55 8 5 68
limitations. This study utilized patient diaries to assess
the primary outcome variable of weekly episodes of vom-
tine-treated patients and 15 placebo-treated patients ap- iting. Although no independent method to assess the reli-
peared to discontinue because of poorer than expected ef- ability of these self-reports was used, one would expect
ficacy in this time period. None of these patients met the a equal distribution of unreliable measures across treat-
priori relapse criteria and were considered censored in the ment groups. It is important to note that vomiting is re-
time-to-relapse analysis. ported with greater clarity than is binge eating, the latter
Relapse as a reason for discontinuation occurred in 17 tending to be assessed more variably by patients. Al-
fluoxetine-treated patients and 22 placebo-treated pa- though limited by the accuracy and veracity of self-report,
tients. Within the fluoxetine-treated group, there was no and in light of the absence of a more objective index of
significant difference in relapse rates between depressed measurement, this method of recording is an improve-
and nondepressed patients (24.1% versus 21.2%) (χ2=0.09, ment over retrospective account. Patient diaries have
df=1, p=0.77). been used extensively in studies assessing bulimia ner-
vosa, including the large multicenter fluoxetine trial ref-
Discussion erenced earlier (18).
Determining operational definitions of response and re-
The results of the single-blind therapy phase confirm the lapse in bulimia nervosa trials represents a challenge, and
efficacy of fluoxetine in the acute management of bulimia efforts in this area may be viewed as evolving. On the basis
nervosa, with the response rate of 64.7% consistent with of the results of the large multicenter trial (18), which re-
previous reports (6). The remission rate of 17.7% is also ported a median reduction of 56% in weekly vomiting epi-
consistent with literature reports, which cite low absti- sodes in the group who received 60 mg/day of fluoxetine
nence rates during acute therapy (28). The comparable re- and which calculated response as ≥50% improvement in
sponse rates among depressed patients and nondepressed weekly vomiting episodes, a reduction of at least 50% was
patients (62.8% versus 65.7%) further suggest that the ab- felt to represent a clinically meaningful improvement to a
sence or presence of comorbid depression does not affect single treatment intervention. Similarly, a sustained re-
response in patients with bulimia nervosa, consistent with turn to the frequency of vomiting experienced at the time
results reported in a previous multicenter study (18). of acute study entry was considered an appropriate crite-
During 52 weeks of double-blind relapse prevention rion for relapse. Clearly, the definition of response used in
monitoring, patients who were randomly assigned to con- this study allows for persistence of symptoms, even for the
tinued treatment with fluoxetine had a significantly longer maintenance of syndromal criteria by some patients.
time to relapse than patients who received placebo. Anal- However, considering the negative consequences, both
yses of mean change from randomization to endpoint for psychological and medical, of persistent bulimic behavior,
each primary and secondary efficacy measure provided a 50% reduction in frequency of vomiting episodes bene-
further evidence of the beneficial effect of fluoxetine. fits clinical status.

100 Am J Psychiatry 159:1, January 2002


Attrition during the relapse prevention phase of the
trial was high in both treatment groups, although there
was a trend for more fluoxetine-treated patients to com-
plete the protocol. The largest amount of patient attrition
occurred in the first 3 months after randomization. Re-
lapse, patients’ decisions, and physicians’ decisions were
the most common reasons for discontinuation. Investiga-
tors’ study notes for the patients who discontinued par-
ticipation because of the patient’s decision or the physi-
cian’s decision revealed that eight of the fluoxetine-
treated patients and 15 of the placebo-treated patients
were discontinued for worsening disease condition. None
of these patients met the formal relapse criteria and were
therefore considered as censored observations in the
time-to-relapse analysis. It is important to note, even in
light of the large attrition rate, that the estimated differ-
ence in relapse rates at 1 year was virtually the same as the
estimated difference after 3 months. This finding indi-
cates that the observed benefit of fluoxetine treatment
occurs in the first 3 months and that this cumulative ben-
efit remains constant after this time period. As clinical ex-
perience underscores the difficulty of maintaining pa-
tients with bulimia in longer-term therapy, the extended
period in which fluoxetine-treated patients remained in
the study, even in light of the noted attrition rate, suggests
the benefit of continued treatment.
Although fluoxetine treatment was associated with a
significantly lower rate of relapse, there was worsening
over time on all measures of efficacy, reflecting symptom-
atic regression. This finding suggests that fluoxetine alone
may not be an adequate treatment after acute response in
most patients and that additional management strategies
may be required to augment or to sustain initial improve-
ment. Several studies have addressed the utility of com-
bined approaches in an effort to establish effective treat-
ment regimens (17, 29–32). Fichter and associates (17),
Goldbloom and associates (29), and Mitchell and associ-
ates (30) were unable to show greater improvement with
the combination of pharmacologic therapy and counsel-
ing than with counseling alone. Agras and associates (31)
studied 71 patients randomly allocated to treatment with
desipramine, cognitive behavior therapy, and a combina-
tion of the two interventions. At 16 weeks, both cognitive
behavior therapy and combined therapy were superior to
medication alone in reducing binge-eating and purging.
Continuing cognitive behavior therapy appeared to pre-
vent relapse for up to 72 weeks in the patients who discon-
tinued medication treatment. In addition, 77% of the pa-
tients with combined therapy achieved abstinence from
binge eating and purging, compared with 54% of those re-
ceiving cognitive behavior therapy alone. Walsh and asso-
ciates (32), in a placebo-controlled study of 120 women,
found that cognitive behavior therapy plus medication
Am J Psychiatry 159:1, January 2002
ROMANO, HALMI, SARKAR, ET AL.
(desipramine followed by fluoxetine if the desipramine
was poorly tolerated or ineffective) was superior to medi-
cation alone in reducing binge eating and vomiting, but
supportive psychotherapy plus medication was not.
Mitchell and associates (30), in a review of controlled trials
of pharmacotherapy and psychotherapy in the treatment
of bulimia nervosa, hypothesized that a longer course of
pharmacologic therapy in combination with counseling
would be of greatest benefit. Given the high attrition rate
observed in this fluoxetine trial and the known benefits of
cognitive behavior therapy, the latter should play a pri-
mary role in both acute and continuation treatment of pa-
tients with bulimia.
To our knowledge, the current study represents the larg-
est relapse prevention trial to date assessing a pharma-
cotherapy treatment for bulimia nervosa. This study dem-
onstrated that continued treatment with fluoxetine in
patients who responded to acute therapy was well toler-
ated and associated with a significant reduction in the
likelihood of relapse during a 52-week monitoring period.
An important finding, which is consistent with clinical ex-
perience and reported research results, was the gradual
worsening of symptom severity in both the fluoxetine-
treated and placebo-treated groups. Although fluoxetine
treatment was statistically and clinically superior to pla-
cebo treatment in multiple measures of outcome, the lat-
ter observation suggests the benefit of a multimodal
approach. Additional outcome data on the long-term ben-
efits of combined therapeutic approaches in improving
the likelihood of sustaining initial gains and in reducing
the rate of relapse should influence future treatment
guidelines. The findings of this study suggest that clini-
cians should consider providing acute responders to flu-
oxetine with maintenance fluoxetine treatment for up to 1
year, with psychotherapeutic interventions implemented
on the basis of individual need.
Acknowledgments
This study was completed with the participation of the following
investigators: Anne Becker, M.D., Massachusetts General Hospi-
tal, Boston; Barton Blinder, M.D., Ph.D., Newport Beach, Calif.;
Harry Brandt, M.D., Center for Eating Disorders, Towson, Md.;
Lynn Cunningham, M.D., Vine Street Clinical Research, Spring-
field, Ill.; Brenda Erikson, M.D., University of New Mexico, Al-
buquerque; James Ferguson, M.D., Pharmacology Research Cor-
poration, Salt Lake City; Tana Grady, M.D., Duke University,
Durham, N.C.; Harry Gwirtsman, M.D., Vanderbilt University
Medical Center, Nashville, Tenn.; James Hudson, M.D., McLean
Hospital, Belmont, Mass.; Walter Kaye, M.D., University of Pitts-
burgh Western Psychiatric Institute, Pittsburgh; John Lauriello,
M.D., University of New Mexico, Albuquerque; Russell Marx,
M.D., San Luis Rey Hospital, Encinitas, Calif.; Pauline Powers,
M.D., University of South Florida College of Medicine, Tampa; Jef-
frey Simon, M.D., Northbrooke Research Center, Brown Deer,
Wis.; B. Timothy Walsh, M.D., New York Psychiatric Institute, New
York; and Kathyrn Zerbe, M.D., Menninger Clinic, Topeka, Kan.
101
FLUOXETINE FOR BULIMIA
Presented in part at the 11th Congress of the European College of
Neuropsychopharmacology, Oct. 31–Nov. 4, 1998, Paris, and the
37th annual meeting of the American College of Neuropsychophar-
macology, Dec. 14–18, 1998, Las Croabas, Puerto Rico. Received Feb.
1, 2001; revision received June 4, 2001; accepted July 31, 2001. From
Pfizer Inc, New York; Weill Medical College of Cornell University,
White Plains, NY; Lilly Research Laboratories, Eli Lilly and Company,
Indianapolis; and Cleveland Clinic, Cleveland. Address reprint re-
quests to Ms. Koke, Lilly Research Laboratories, Lilly Corporate Center
2200, Indianapolis, IN 46285; skoke@lilly.com (e-mail).
Supported by a clinical research grant from Eli Lilly and Company.
The authors thank Roy Tamura of Eli Lilly and Company for statisti-
cal assistance.
References
1. Herzog DB, Nussbaum KM, Marmor AK: Comorbidity and out-
come in eating disorders. Psychiatr Clin North Am 1996; 19:
843–859
2. Kendler KS, MacLean C, Neale M, Kessler R, Heath A, Eaves L:
The genetic epidemiology of bulimia nervosa. Am J Psychiatry
1991; 148:1627–1637
3. Brewerton TD, Mueller EA, Lesem MD, Brandt HA, Quearry B,
George DT, Murphy DL, Jimerson DC: Neuroendocrine re-
sponses to m-chlorophenylpiperazine and L-tryptophan in bu-
limia. Arch Gen Psychiatry 1992; 49:852–861
4. McBride PA, Anderson GM, Khait VD, Sunday SR, Halmi KA: Se-
rotonergic responsivity in eating disorders. Psychopharmacol
Bull 1991; 27:365–372
5. Leibowitz SF: The role of serotonin in eating disorders. Drugs
1990; 39(suppl 3):33–48
6. Jimerson DC, Herzog DB: Pharmacologic approaches in the
treatment of eating disorders. Harv Rev Psychiatry 1993; 1:82–
93
7. Pope HG Jr, Hudson JI, Jonas JM, Yurgelun-Todd D: Bulimia
treated with imipramine: a placebo-controlled, double-blind
study. Am J Psychiatry 1983; 140:554–558
8. Hughes PL, Wells LA, Cunningham CJ, Ilstrup DM: Treating bu-
limia with desipramine: a placebo-controlled double-blind
study. Arch Gen Psychiatry 1986; 43:182–186
9. Agras WS, Dorian BG, Kirkley BG, Arnow B, Bachman J: Imi-
pramine in the treatment of bulimia: a double-blind con-
trolled study. Int J Eat Disord 1987; 6:29–38
10. Kaplan AS, Garfinkel PE, Garner DM: Bulimia treated with car-
bamazepine and imipramine, in 1987 Annual Meeting New
Research Program. Washington, DC, American Psychiatric As-
sociation, 1987, p 127
11. Price WA, Babai MR: Antidepressant drug therapy for bulimia:
current status revisited (letter). J Clin Psychiatry 1987; 48:385
12. Barlow J, Blouin J, Blouin A, Perez E: Treatment of bulimia with
desipramine: a double-blind crossover study. Can J Psychiatry
1988; 33:129–133
13. Horne RL, Ferguson JM, Pope HG Jr, Hudson JI, Lineberry CG,
Ascher J, Cato A: Treatment of bulimia with bupropion: a mul-
ticenter controlled trial. J Clin Psychiatry 1988; 49:262–266
14. Walsh BT, Gladis M, Rose SP, Stewart JW, Stetner F, Glassman
AH: Phenelzine vs placebo in 50 patients with bulimia. Arch
Gen Psychiatry 1988; 45:471–475
102
15. Kennedy SH, Piran N, Warsh JJ, Prendergast P, Mainprize E,
Whynot C, Garfinkel PE: A trial of isocarboxazid in the treat-
ment of bulimia nervosa. J Clin Psychopharmacol 1988; 8:391–
396
16. Pope HG Jr, Keck PE Jr, McElroy SM, Hudson JI: A placebo-con-
trolled study of trazodone in bulimia nervosa. J Clin Psycho-
pharmacol 1989; 9:254–259
17. Fichter MM, Leibl K, Rief W, Brunner E, Schmidt-Auberger S, En-
gel RR: Fluoxetine versus placebo: a double-blind study with
bulimic inpatients undergoing intensive psychotherapy. Phar-
macopsychiatry 1991; 24:1–7
18. Fluoxetine Bulimia Nervosa Collaborative Study Group: Fluox-
etine in the treatment of bulimia nervosa: a multicenter, pla-
cebo-controlled, double-blind trial. Arch Gen Psychiatry 1992;
49:139–147
19. Goldstein DJ, Wilson MG, Thompson VL, Potvin JH, Rampey AH
(Fluoxetine Bulimia Nervosa Research Group): Long-term flu-
oxetine treatment of bulimia nervosa. Br J Psychiatry 1995;
166:660–666
20. Keel PK, Mitchell JE: Outcome in bulimia nervosa. Am J Psychi-
atry 1997; 154:313–321
21. Garner DM, Olmsted MP, Polivy J: Eating Disorder Inventory.
Odessa, Fla, Psychological Assessment Resources, 1983
22. Mazure CM, Halmi KA, Sunday SR, Romano SJ, Einhorn AM: The
Yale-Brown-Cornell Eating Disorder Scale: development, use,
reliability, and validity. J Psychiatr Res 1994; 28:425–445
23. Guy W (ed): ECDEU Assessment Manual for Psychopharmacol-
ogy: Publication ADM 76-338. Washington, DC, US Department
of Health, Education, and Welfare, 1976, pp 218–222
24. Hamilton M: Development of a rating scale for primary depres-
sive illness. Br J Soc Clin Psychol 1967; 6:278–296
25. Greenwood M: The Natural Duration of Cancer: Reports on
Public Health and Medical Subjects 33. London, Her Majesty’s
Stationery Office, 1926, pp 1–26
26. Gray RJ, Tsiatis AA: A linear rank test for use when the main in-
terest is in difference in cure rates. Biometrics 1989; 45:899–
904
27. Akritas MG, Arnold SF, Brunner E: Nonparametric hypotheses
and rank statistics for unbalanced factorial designs. J Am Statis-
tical Assoc 1997; 92:258–265
28. Mitchell JE, Raymond N, Specker S: A review of the controlled
trials of pharmacotherapy and psychotherapy in the treatment
of bulimia nervosa. Int J Eat Disord 1993; 14:229–247
29. Goldbloom DS, Olmsted M, Davis R, Clewes J, Heinmaa M,
Rockert W, Shaw B: A randomized controlled trial of fluoxetine
and cognitive behavioral therapy for bulimia nervosa: short-
term outcome. Behav Res Ther 1997; 35:803–811
30. Mitchell JE, Pyle RL, Eckert ED, Hatsukami D, Pomeroy C, Zim-
merman R: Comparison study of antidepressants and struc-
tured intensive group psychotherapy in the treatment of bu-
limia nervosa. Arch Gen Psychiatry 1990; 47:149–157
31. Agras WS, Rossiter EM, Arnow B, Schneider JA, Telch CF, Rae-
burn SD, Bruce B, Perl M, Koran LM: Pharmacologic and cogni-
tive-behavioral treatment for bulimia nervosa: a controlled
comparison. Am J Psychiatry 1992; 149:82–87
32. Walsh BT, Wilson GT, Loeb KL, Devlin MJ, Pike KM, Roose SP,
Fleiss J, Waternaux C: Medication and psychotherapy in the
treatment of bulimia nervosa. Am J Psychiatry 1997; 154:523–
531
Am J Psychiatry 159:1, January 2002