ARTICLES

Pharmacotherapy Relapse Prevention in Body

Dysmorphic Disorder: A Double-Blind,
Placebo-Controlled Trial
Katharine A. Phillips, M.D., Aparna Keshaviah, Sc.M., Darin D. Dougherty, M.D., Robert L. Stout, Ph.D., William Menard, B.A.,
Sabine Wilhelm, Ph.D.

Objective: Body dysmorphic disorder is common, distressing, In phase 2, time to relapse was significantly longer with
and often severely impairing. Serotonin reuptake inhibitors escitalopram than with placebo treatment (hazard ratio=2.72,
appear efficacious, but the few existing pharmacotherapy 95% CI=1.01–8.57). Phase 2 relapse proportions were 18% for
studies were short term (#4 months), and no relapse prevention escitalopram and 40% for placebo. Among escitalopram-
studies or continuation phase studies have been conducted treated subjects, body dysmorphic disorder severity signifi-
to the authors’ knowledge. The authors report results from cantly decreased over time during the continuation phase,
the first relapse prevention study in body dysmorphic disorder. with 35.7% of subjects showing further improvement. There
were no significant group differences in body dysmorphic
Method: Adults (N=100) with DSM-IV body dysmorphic disorder disorder severity or insight, depressive symptoms, psycho-
received open-label escitalopram for 14 weeks (phase 1); 58 re- social functioning, or quality of life.
sponders were then randomized to double-blind continuation
treatmentwithescitalopramversusswitchtoplacebofor6months Conclusions: Continuation-phase escitalopram delayed
(phase 2). Reliable and valid outcome measures were utilized. time to relapse, and fewer escitalopram-treated subjects
relapsed than did placebo-treated subjects. Body dysmorphic
Results: In phase 1, 67.0% of treated subjects and 81.1% disorder severity significantly improved during 6 additional
of subjects who completed phase 1 responded to es- months of escitalopram treatment following acute response;
citalopram. Body dysmorphic disorder severity (in both the more than one-third of escitalopram-treated subjects expe-
intent-to-treat and the completer groups) and insight, de- rienced further improvement.
pressive symptoms, psychosocial functioning, and quality of
life significantly improved from baseline to end of phase 1. Am J Psychiatry 2016; 173:887–895; doi: 10.1176/appi.ajp.2016.15091243

Body dysmorphic disorder, an often severe disorder, consists desipramine (8). Four open-label trials (Ns of 15 to 30) also
of distressing or impairing preoccupations with nonexistent found that selective serotonin reuptake inhibitors (SSRIs)
or slight defects in appearance (e.g., “scarred” skin), plus are often efficacious (9–12). To our knowledge, no studies
repetitive behaviors performed in response to appearance have prospectively examined the risk of relapse following
concerns (e.g., mirror checking, excessive grooming). Body discontinuation of efficacious medication or have exam-
dysmorphic disorder has a prevalence of 1.7%22.4% (1–4), ined medication response during continuation treatment
causes substantial impairment in psychosocial functioning (following acute treatment). Such studies are needed be-
(5), and is associated with high rates of suicidality (6). cause many patients with body dysmorphic disorder
However, few pharmacotherapy studies have been receive pharmacotherapy (most often SRIs) (13) and be-
conducted in body dysmorphic disorder, and all contained cause the disorder is often chronic, necessitating longer-term
relatively small sample sizes (in the largest study, 34 patients treatment (13).
received active medication [7]). Only one double-blind, Two studies that examined relapse after SRI treatment
placebo-controlled trial (N=67) has been done (7), which were limited by small samples, lack of placebo, unblinded
reported that fluoxetine was more efficacious than placebo. evaluation of outcome, and imprecise definition of relapse
In a blinded crossover trial (N=29), the serotonin reuptake (using only the Clinical Global Impressions Scale [CGI] [14]).
inhibitor (SRI) clomipramine was more efficacious than In one, a small chart-review study from a clinical practice,

See related features: Editorial by Dr. Hollander and Mr. Hong (p. 857) and Clinical Guidance (Table of Contents)

Am J Psychiatry 173:9, September 2016 ajp.psychiatryonline.org 887

PHARMACOTHERAPY RELAPSE PREVENTION IN BODY DYSMORPHIC DISORDER

83% (N=31) of patients who discontinued an effective SRI screen for any illicit substances of abuse; mental retardation,
experienced relapse of body dysmorphic disorder (15). In dementia, brain damage, or other cognitive impairment that
the other, a 16-week open-label fluvoxamine study, six re- would interfere with participation; body image concerns
sponders who were subsequently treated in a clinical practice accounted for primarily by an eating disorder; body dys-
discontinued fluvoxamine; all relapsed from within 4 days to morphic disorder criteria not met if weight concerns were
about 2 months (11). In both studies, subjects were assessed excluded; use of psychotropic medication or herbal prepa-
only at clinic visits, thus precluding precise assessment of rations with alleged behavioral effects during the study or
time to relapse, and secondary outcomes were not assessed. for 2 weeks (6 weeks for fluoxetine) before the baseline as-
Here we report on the first prospective pharmacotherapy sessment; use of investigational medication within 3 months
relapse prevention study in body dysmorphic disorder. In of baseline or use of a depot antipsychotic within 6 months
phase 1, subjects received 14 weeks of acute open-label of baseline; previous allergic reaction to escitalopram or
escitalopram. In phase 2, responders were randomized to citalopram; current cognitive-behavioral therapy (CBT);
6 months of continued escitalopram treatment or switch presence of significant or unstable medical illness; history of
to placebo. The study’s primary aim was to compare time to a seizure disorder; females who were pregnant, breastfeed-
relapse (and relapse rates) in phase 2; we hypothesized that ing, or sexually active and not using adequate contraception;
escitalopram responders who continued escitalopram for 6 and presence of behavior that would interfere with protocol
additional months would have a longer time to relapse and a cooperation.
lower relapse rate than subjects who switched to placebo. The protocol and informed-consent documents were
Secondary and exploratory phase 2 hypotheses were approved by the hospitals’ institutional review boards. All
that subjects randomized to 6 months of continuation participants provided written informed consent after re-
escitalopram treatment would perform better than placebo- ceiving a complete description of the study. An independent
treated subjects on secondary outcome measures and would data and safety monitoring board provided oversight.
significantly improve with continued escitalopram treatment.
Phase 1 (acute open-label phase) outcomes were of in- Study Design and Procedures
terest because few pharmacotherapy studies have been done During screening, psychiatric and medical histories (using
in body dysmorphic disorder, and all used relatively small the Body Dysmorphic Disorder Form [e.g., 7]) were con-
samples (7–12). Furthermore, no study has examined re- ducted, a urine pregnancy test and a urine toxicology screen
mission (as opposed to response or improvement) with phar- were obtained, and the Structured Clinical Interview for
macotherapy. Phase 1 hypotheses were that body dysmorphic DSM-IV, Patient Version (18–20) was administered.
disorder symptoms and insight, depressive symptoms,
functioning, and quality of life would significantly improve Phase 1. Participants received open-label escitalopram mono-
after 14 weeks of open-label escitalopram, and that subjects therapy for 14 weeks: 10 mg/day during weeks 1–3, 20 mg/day
with delusional body dysmorphic disorder beliefs would be during weeks 4–6, and 30 mg/day thereafter. A lower dosage
as likely as those with nondelusional beliefs to respond to was allowed to improve tolerability. Phase 1 assessments
escitalopram. (below) were conducted at baseline, weekly during weeks 1–4,
and biweekly during weeks 6–14. Escitalopram response was
defined as $30% reduction in the BDD-YBOCS score from
METHOD
baseline that persisted for at least two consecutive assessments
Subjects and through the end of phase 1 (or through the last visit, for
This study was conducted at two academic sites: Rhode early terminators).
Island Hospital (formerly at Butler Hospital) at Brown
University and Massachusetts General Hospital. The fol- Phase 2. Nonresponders were removed from the study at the
lowing inclusion criteria were used for participants: age $18; end of phase 1. Subjects who completed phase 1, responded to
diagnosis of DSM-IV body dysmorphic disorder (including escitalopram, and agreed to continue study participation
delusional body dysmorphic disorder) for $6 months; entered phase 2. Phase 2 participants were randomly assigned
baseline total score $24 on the Yale-Brown Obsessive- to 6 additional months of treatment with continuation
Compulsive Scale Modified for Body Dysmorphic Disorder monotherapy with escitalopram or discontinuation of esci-
(BDD-YBOCS) (16, 17); score of at least moderate on the CGI talopram (via tapering by 10 mg/day per week) and substitu-
severity scale (14); ability to communicate meaningfully with tion with pill placebo. Randomization was stratified by body
the investigators; and competency to provide written in- dysmorphic disorder–related insight and by presence of major
formed consent. The following exclusion criteria were used: depressive disorder. Phase 2 assessments were conducted at
current clinically significant suicidality or a suicide attempt randomization and biweekly thereafter.
within the past year; need for inpatient or partial hospital Throughout phase 2, subjects randomized to escitalopram
treatment; current or past DSM-IV bipolar disorder or received the dosage prescribed at the end of phase 1. How-
psychotic disorder; DSM-IV alcohol or drug dependence or ever, those taking 10–20 mg/day at the end of phase 1 could
abuse in the past 3 months, or a positive result on a urine drug increase their dosage to 30 mg/day during phase 2 if

888 ajp.psychiatryonline.org Am J Psychiatry 173:9, September 2016

 PHILLIPS ET AL.

symptoms worsened before relapse; conversely, if side disorder symptoms meet full DSM-IV criteria or are in full
effects occurred at 30 mg/day, the dosage could be lowered or partial remission (13, 30). In phase 1, this rating scale was
to 20 mg/day. used to examine remission from body dysmorphic disorder,
Relapse of body dysmorphic disorder was defined as a and in phase 2 it was used to examine further improvement
50% or greater loss of BDD-YBOCS improvement that had (by one or more points) during continuation treatment with
occurred during phase 1, plus a BDD-YBOCS score .20 escitalopram.
(which corresponds to full-criteria body dysmorphic dis- Measurements were administered at all visits, except
order), plus a rating of “much worse” or “very much worse” that the functional impairment measure and the quality of
on the CGI for body dysmorphic disorder symptoms. We life questionnaire were administered only at the phase 1
developed this definition after consulting the literature baseline and termination visits and at the phase 2 midpoint
on other disorders, primarily obsessive-compulsive dis- and termination visits. Interrater reliability (based on
order (OCD), because of these disorders’ similarities (21) intraclass correlations) on the BDD-YBOCS and Brown
and because the BDD-YBOCS is derived from the YBOCS Assessment of Beliefs Scale for all scale items and total
(22, 23). However, there was no consensus on the definition score was .0.9.
of relapse in the OCD literature (24), so we developed a At all visits, vital signs were obtained, and medication
rigorous, clinically meaningful definition that was informed compliance was monitored by patient inquiry, returned pill
by this literature. Relapse criteria needed to be met for 2 count, and a drug accountability form. Any adverse physical
consecutive weeks, although for safety reasons, subjects symptoms since the last visit were rated for severity, action
could be withdrawn after meeting relapse criteria for only taken, outcome, and seriousness. Urine drug screen and preg-
1 week (and they were considered to have relapsed). Re- nancy tests were repeated at the end of phase 1.
lapsing participants were referred for doctor’s choice rescue
treatment by a psychiatrist not otherwise involved in the Data Analysis
study. Additional psychotropic medication was limited to Analyses were conducted using SAS, version 9.4 (SAS In-
zolpidem at 5–10 mg/h.s. for insomnia and sildenafil at up to stitute, Cary, N.C.), and employed a two-sided alpha=0.05
100 mg three times a week for treatment-emergent sexual type I error rate. Major variables were screened for incon-
dysfunction. sistent or abnormal values, and continuous measures were
assessed for skewness and outliers. Transformations to im-
Assessments prove normality were applied. Differences in baseline
Independent evaluators who did not provide study treatment covariates were tested using Fisher’s exact test (categorical)
administered the following reliable and valid measure- and Student’s t test (continuous). In phase 1, estimated power
ments. The BDD-YBOCS, a 12-item semistructured clinician- was .95% to detect a medium effect size for the change in
administered scale adapted from the YBOCS (16, 17), rated BDD-YBOCS score over time. Changes in scores were com-
past-week disorder severity. The CGI improvement scale, a pared with a null value using a series of Student’s t tests. Sec-
global rating scale that ranges from 1 (very much improved) to ondary analyses quantified the proportion (and 95% Wald
7 (very much worse) (14), assessed body dysmorphic disorder CIs) of subjects who were much or very much improved (CGI
symptoms and overall symptoms. The CGI for body dys- improvement score of 1 or 2) and who attained remission
morphic disorder symptoms and the overall (global) CGI from body dysmorphic disorder (Psychiatric Status Rating
were secondary outcome measures in phase 1, and the CGI for Scale score of 1 or 2) by week 14 or by an early termination
body dysmorphic disorder symptoms determined relapse in visit. Cox proportional hazards regression examined pre-
phase 2. The Brown Assessment of Beliefs Scale, a seven-item dictors of response to open-label escitalopram, with esti-
semistructured clinician-administered scale, assessed past- mated power from 76% to $90% to detect medium to large
week body dysmorphic disorder–related insight and delu- effect sizes. We planned to enroll enough subjects in phase 1
sional beliefs dimensionally (25, 26). It also classified false to obtain 58 responders for randomization in phase 2. For the
beliefs (e.g., “I look like a monster”) as delusional or non- primary aim in phase 2 of examining group differences in time
delusional. The 17-item Hamilton Depression Rating Scale to relapse (and proportion relapsing), Kaplan-Meier survival
(HAM-D) assessed current severity of depressive symptoms curves were generated, and Cox proportional hazards re-
(27). The Range of Impaired Functioning Tool assessed gression was employed (with the likelihood ratio p value
psychosocial functioning; it consists of 5- to 7-point rater- reported). In phase 2, data from assessments made after a
administered subscales for work, household duties, student patient entered rescue treatment were excluded from anal-
work, relationships with family and friends, recreation, life yses. A priori, we determined that randomization of 58
satisfaction, and global social adjustment (28). The Quality of subjects to phase 2 would yield 80% power to detect a hazard
Life Enjoyment and Satisfaction Questionnaire Short Form ratio as small as 3.0 over 6 months of follow-up. For secondary
assessed quality of life across social, leisure, household, work, analyses of change in the BDD-YBOCS score, as well as
emotional well-being, physical, and school parameters (29). changes in secondary outcome measures (the Brown As-
The Psychiatric Status Rating Scale for Body Dysmorphic sessment of Beliefs Scale, the HAM-D, the functional im-
Disorder, a 7-point scale, reflects whether body dysmorphic pairment measure, and the quality of life questionnaire), we

Am J Psychiatry 173:9, September 2016 ajp.psychiatryonline.org 889

PHARMACOTHERAPY RELAPSE PREVENTION IN BODY DYSMORPHIC DISORDER

FIGURE 1. Flow Diagram of Subject Enrollment and Progress Through the Study

Enrolled subjects (N=173)

Excluded (N=73)
Did not meet inclusion/exclusion criteria (N=37)
Declined to participate (N=19)
Unknown reasons (N=17)

Phase 1—Treated with open-label

escitalopram (N=100)

Dropped out/withdrawn (N=26):

Protocol non-adherence (N=15)
Medication-related adverse event (N=4)
Moved (N=2)
Wished to discontinue medication (N=2)
Lack of time to devote to study (N=2)
Increased suicidal ideation (N=1)

Non-responders to escitalopram (N=14)

Completed initial treatment with response

(N=60)

Phase 2—Enrolled in double-blind

treatment (N=58)

Randomized to continued
Randomized to placebo (N=30)
escitalopram (N=28)

Completed Phase 2 (N=25) Completed Phase 2 (N=24)

Dropped out/withdrawn (N=3): Dropped out/withdrawn (N=6):
Moved (N=1) Missed too many visits (N=4)
Adverse events (N=1) Began therapy (N=1)
Began therapy (N=1) Protocol non-adherence (N=1)

Analyzed (N=28) Analyzed (N=30)

employed generalized linear regression models and to compare changes in BDD-YBOCS total scores during
regressed the change score (baseline to last visit) against the phase 2 with a null value.
treatment group. For large effect sizes (d=0.8), estimated
power for a two-tailed test (alpha=0.05) was $74% for the
RESULTS
BDD-YBOCS, and for the other secondary outcome mea-
sures, using a Bonferroni-adjusted alpha of 0.0125, esti- Enrollment and Sample Description
mated power was 31%274% to detect medium to large Across the two sites, 173 subjects were enrolled (Figure 1).
effect sizes. During 6 months of continuation escitalopram Seventy-three subjects did not pass screening. One hundred
treatment, we estimated the proportion (and 95% Fisher’s subjects received initial open-label escitalopram treatment in
exact CIs) of subjects who further improved by one or more phase 1. Table 1 summarizes the participants’ demographic
points on the Psychiatric Status Rating Scale; power to and clinical characteristics; no significant site differences
detect significant improvement in 40% or more of subjects existed.
was 99% (http://web1.sph.emory.edu/cdckms/proportion-
ext-Mid-Pnew.html). We examined site differences in the Acute, Open-Label Phase (Phase 1)
effect of treatment by including site and the interaction A total of 74.0% (N=74) of subjects completed 14 weeks of
between treatment and site in regression models. Within the escitalopram treatment in phase 1. Body dysmorphic disorder
escitalopram and placebo groups, Student’s t tests were used symptom response was achieved by 67.0% (95% CI=57.8–75.7)

890 ajp.psychiatryonline.org Am J Psychiatry 173:9, September 2016

 PHILLIPS ET AL.

(N=67) of all treated subjects TABLE 1. Baseline Demographic and Clinical Characteristics, by Study Phasea
(in the intent-to-treat pop- Phase II
ulation) and by 81.1% (95% Variable Phase I (N=100) Escitalopram (N=28) Placebo (N=30)
CI=72.2–90.0) (N=60) of the
74 subjects who completed N % N % N %
phase 1. Based on CGI im- Female 64 64.0 20 71.4 20 66.7
provement scores for body Race
American Indian or Alaskan Native 2 2.0 0 0.0 1 3.3
dysmorphic disorder symp- Asian 2 2.0 1 3.6 0 0.0
toms, 71.1% (95% CI=62.1–80.2) Black/African American 7 7.0 1 3.6 1 3.3
(N=69) of the 97 subjects with a Native Hawaiian or other 0 0.0 0 0.0 0 0.0
postbaseline assessment were Pacific Islander
much or very much improved White 84 84.0 24 85.7 27 90.0
More than one race 5 5.0 2 7.1 1 3.3
by their last phase 1 visit. Full
Hispanic or Latino ethnicity 12 12.0 4 14.3 5 16.7
remission from body dysmor-
Body dysmorphic disorder age at onset 69 70.4 21 77.8 20 66.7
phic disorder was attained ,18 yearsb
by 20.0% (95% CI=12.2–27.8)
Mean SD Mean SD Mean SD
(N=20) of intent-to-treat sub-
jects and by 25.7% (95% Age (years) c
33.5 12.4 37.3 12.4 31.8 13.5
CI=15.7–35.6) (N=19) of sub- BDD-YBOCS score at randomization 10.3 7.5 13.8 6.4
a
jects who completed phase 1. b There were no significant between-group differences.
Median time to first response c Some data missing for this variable.
BDD-YBOCS=Yale-Brown Obsessive-Compulsive Scale Modified for Body Dysmorphic Disorder.
was7.9weeks(95%CI=6.9–8.9).
The mean escitalopram dose
at the last phase 1 visit was 26.2 mg/day (SD=7.2, range=5–30) clinical characteristics at randomization; no significant
(two subjects received less than 10 mg/day). differences existed by treatment. Phase 2 was completed by
Significant improvement from baseline to the last phase 1 89.3% (N=25) of escitalopram-treated subjects and by
visit was attained on the BDD-YBOCS, the Brown Assessment 80.0% (N=24) of placebo-treated subjects (n.s.).
of Beliefs Scale, the HAM-D, the functional impairment Time to relapse of body dysmorphic disorder was signifi-
measure, and the quality of life questionnaire (all p values cantly longer with escitalopram than with placebo (hazard
,0.0001). Mean BDD-YBOCS scores decreased from 32.7 ratio=2.72, 95% CI=1.01–8.57, p=0.049) (Figure 2). By the end of
(SD=5.4) at baseline to 16.9 (SD=10.2) at the last phase 1 visit phase 2, relapse proportions were 40% for the placebo group
among the 97 subjects with a postbaseline assessment (av- compared with 18% for the escitalopram group. There were
erage decrease of 48.7%), and these scores decreased to 14.9 no statistically significant between-group treatment differ-
(SD=9.3) at week 14 in the population who completed phase 1 ences on secondary outcome measures or site differences.
(average decrease of 54.4%).
Based on the CGI global improvement score, 67.0% (95%
Continuation Treatment Efficacy
CI=57.7–76.4) (N=65) of subjects were much or very much
Among the 28 subjects randomized to continuation escitalopram,
improved by their last postbaseline phase 1 visit. Response
BDD-YBOCS scores decreased significantly during phase
rates did not significantly differ between subjects with
2 (mean=4.1 points, p=0.036). Among escitalopram-treated
delusional and those with nondelusional body dysmorphic
subjects, 35.7% (95% CI=18.0–53.5) (N=10) had a further de-
disorder, but there was a trend for the 74 subjects with non-
crease in the Psychiatric Status Rating Scale score. The mean
delusional beliefs to be more likely to respond to escitalopram
escitalopramdoseattheendofphase2was28.7 mg/day (SD=4.6,
than the 26 subjects with delusional beliefs (70.3% and 57.7%,
range=7.5–30.0) (one subject received less than 10 mg/day).
respectively; hazard ratio=0.58, 95% CI=0.33–1.04, p=0.054).
The only baseline variable that predicted response of body
dysmorphic disorder was presence of a personality disorder Safety and Tolerability
(hazard ratio=0.58, 95% CI=0.36–0.94, p=0.030). Response to Phase 1. Among intent-to-treat subjects, 89% (N=89) expe-
acute open-label escitalopram did not significantly differ rienced at least one treatment-emergent adverse event (re-
by baseline body dysmorphic disorder severity or duration, gardless of relationship to treatment). Adverse events that
depressive symptoms, gender, or minority status. occurred in at least 5% of subjects and were considered at
least possibly related to study medication were fatigue
Relapse Prevention Efficacy (Phase 2) (N=37), nausea (N=30), sexual dysfunction (N=28), dry mouth
Fifty-eight participants were randomized to double-blind (N=25), insomnia (N=25), headache (N=22), appetite changes
continuation treatment with escitalopram or to discon- (N=20), agitation (N=10), flatulence (N=7), hypersomnia
tinuation of escitalopram and substitution with placebo (N=6), sweating (N=6), dizziness (N=5), yawning (N=5), and
(Figure 1). Table 1 presents participants’ demographic and irritability (N=5).

Am J Psychiatry 173:9, September 2016 ajp.psychiatryonline.org 891

PHARMACOTHERAPY RELAPSE PREVENTION IN BODY DYSMORPHIC DISORDER

FIGURE 2. Time to Relapse by Phase 2 Treatment Arm, With sizes are needed to determine whether group differences
Number of Subjects at Risk might emerge over a longer period following medication
1.0 discontinuation.
Consistent with our hypothesis, during the 6 months of
0.8 phase 2, BDD-YBOCS scores significantly improved with
Survival Probability

continued escitalopram treatment; 35.7% of subjects im-

0.6 proved by at least 1 point on the 7-point Psychiatric Status
Rating Scale. This finding is consistent with clinical obser-
0.4 vations that patients may improve further with SSRI con-
tinuation following acute response (31). Such improvement
0.2 Escitalopram has been found in studies of OCD, depression, and other
Placebo
disorders (32, 33).
0.0 Our open-label phase 1 results confirm findings from all
Escitalopram (N) 28 23 19
Placebo (N) 30 18 10 prior treatment studies, indicating that acute SRI treatment
0 5 10 15 20 25 is efficacious for a majority of patients. The present study’s
Time to Relapse (weeks from randomization) intent-to-treat response rate (67%) and degree of improve-
ment on the BDD-YBOCS are similar to those in previous,
smaller acute-phase studies (the number of SSRI-treated
Phase 2. Overall, 25.0% (N=7) of escitalopram-treated subjects ranged from 15 to 34), which had intent-to-treat
subjects and 46.7% (N=14) of placebo-treated subjects ex- response rates ranging from 53% to 83% (7, 9–12).
perienced at least one adverse event considered at least Consistent with our hypothesis and with previous re-
possibly related to study medication (Fisher’s exact p=0.11). search, body dysmorphic disorder–related insight, depres-
No serious adverse events occurred during either treatment sive symptoms, psychosocial functioning, and quality of life
phase. also significantly improved with open-label escitalopram
(9–12). The mean phase 1 escitalopram dosage (26.2 mg/day,
SD=7.2) is higher than dosages often used to treat anxiety
DISCUSSION
disorders and depression but is similar to dosages usually
Among subjects who responded to acute-phase escitalopram, recommended for OCD and those typically used for body
continued escitalopram treatment significantly delayed time dysmorphic disorder (31, 34, 35). No dose-finding studies
to relapse compared with treatment with placebo, thus have been conducted in body dysmorphic disorder; however,
confirming our primary hypothesis. Moreover, relapse oc- clinical observations and retrospective data from an obser-
curred in more than twice as many placebo-treated subjects vational study (31, 36) suggest that relatively high SRI doses
as escitalopram-treated subjects. We believe that these dif- (sometimes higher than those used in this study) are often
ferences are clinically meaningful. These findings are par- needed to effectively treat this disorder.
ticularly relevant because body dysmorphic disorder is often Our finding that individuals with delusional body
chronic (13, 30). dysmorphic disorder beliefs were as likely as those with
In the only previously published relapse study—a chart- nondelusional beliefs to respond to escitalopram is consistent
review study in a clinical practice—discontinuation of an with previous SRI studies (7–11). It is also consistent with data
effective SRI was followed by relapse (rated “much” or “very from a range of diagnostic validators indicating that delu-
much worse” on the CGI) in 83% (N=31) of cases (15), a sional and nondelusional body dysmorphic disorder consti-
considerably higher rate than in the present study (40%). tute the same disorder (37, 38), as specified in DSM-5.
However, we assessed relapse over 6 months, whereas the However, unlike previous studies, we found a trend for those
chart-review study assessed relapse between the time with nondelusional body dysmorphic disorder beliefs to be
an SRI was started and the last clinic visit, which varied more likely to respond to escitalopram.
considerably among subjects (the mean follow-up duration Limitations of the present study include statistical
was not reported). Furthermore, the two studies have power that was too limited to examine predictors of relapse
methodological differences, including different outcome or to test group differences for other reported results (e.g.,
measures and blind assessments every 2 weeks in the pres- within a type of adverse event). Analysis of multiple out-
ent study, compared with less frequent, unblinded assess- comes in phase 1 may have inflated type I error. Also, while
ments in the chart-review study. The prior study’s higher use of placebo is a gold-standard approach with many
relapse rate may also be attributable in part to patients’ methodological advantages, it does not mimic what occurs
knowledge that active medication was being discontinued, in clinical practice, where patients know whether they have
unlike the current study. discontinued a medication. Knowledge of medication dis-
Our secondary and exploratory hypothesis concerning continuation may possibly be associated with a higher
changes in clinical symptoms, functioning, and quality of life relapse rate than observed in this study. Furthermore, our
during phase 2 was not confirmed. Studies with larger sample findings may not be fully generalizable to certain clinical

892 ajp.psychiatryonline.org Am J Psychiatry 173:9, September 2016

 PHILLIPS ET AL.

Patient Perspectives
Patient 1 worsened; she relapsed after 3 months, at a BDD-YBOCS
“Ms. M,” a 53-year-old married white insurance agent, score of 32. She now thought about her appearance for 12–13
was preoccupied for 15–16 hours a day with her “asym- hours a day, and her appearance concerns caused severe
metrical” eyes and “thin” hair, which she believed looked distress and impairment in functioning (for example, she
“abnormal.” In reality, her eyes and hair looked normal. avoided 75%–80% of social activities).
These concerns were “severely distressing.” She spent many
hours a day checking her appearance in mirrors, comparing Patient 2
with others, asking her husband if she looked okay, and “Mr. N,” a single white 37-year-old electrician, had been
pulling on her eyelids to “even up” her eyes. She tried to hide preoccupied with his “nonexistent” chin, generally “ugly”
her eyes by wearing glasses and covering them with her face, and “huge” nose since age 15. He thought about his
hair. Her appearance preoccupations decreased her con- appearance for several hours a day and frequently com-
centration and productivity at work. Because she did not pared his appearance with that of others. He held his
want to be seen, she avoided most social situations, ac- head at certain angles so his perceived flaws would be
tivities such as food shopping, and physical intimacy with less noticeable. His appearance preoccupations decreased
her husband. Her distress over her appearance caused his concentration and work productivity; because he
depressed mood and frequent suicidal ideation. A bleph- felt embarrassed about how he looked, he avoided work
aroplasty, facelift, 20 sessions of psychotherapy, and treat- colleagues, work meetings, and some social situations,
ment in a partial hospital program did not improve her and he did not date. His appearance concerns caused
symptoms. Trials of fluvoxamine at 50 mg/day for less suicidal ideation. Treatment with 20 sessions of indi-
than a week and of escitalopram at 20 mg/day for 6 months, vidual psychotherapy, 10 sessions of family therapy, and
as well as fluoxetine, venlafaxine, paroxetine, citalopram, fluoxetine at 20 mg/day for 4 weeks were not beneficial.
lorazepam, and clonazepam at unknown dosages, were Treatment with 150 mg/day of sertraline for several years
not helpful, although medication adherence was poor. substantially improved his symptoms, but some symp-
At study baseline, Ms. M was diagnosed with body toms remained.
dysmorphic disorder and current major depressive dis- At study baseline, Mr. N was diagnosed with body
order. Her score on the 48-point BDD-YBOCS was 31, dysmorphic disorder and past major depressive disorder.
indicating moderate to severe body dysmorphic disorder His BDD-YBOCS score was 25 (mild to moderate body
(a cutoff point of 20 indicates current body dysmorphic dysmorphic disorder). After 12 weeks of treatment with
disorder). After 12 weeks of escitalopram, which was escitalopram, which was gradually increased to a dosage of
gradually increased to a dosage of 30 mg/day, her BDD- 30 mg/day, his BDD-YBOCS score was 4. He had no ap-
YBOCS score was 19; she now thought about her appear- pearance preoccupations or associated distress or impair-
ance for less than 1 hour a day and reported only mild ment; his score of 4 was accounted for by absent insight (he
distress and moderate impairment in functioning. She was still completely convinced that he looked “deformed
went out of the house to socialize and do errands, but her and abnormal”). After randomization to 6 months of con-
appearance concerns still diminished her work productivity. tinued escitalopram treatment, he continued to do well; at
After randomization to placebo, her symptoms gradually study termination, his BDD-YBOCS score was still 4.

settings or populations. For example, suicidality and sub- Research is also needed to investigate whether treat-
stance use disorders appear common in individuals with ment with CBT for body dysmorphic disorder will de-
body dysmorphic disorder (6, 39), yet we excluded indi- crease the risk of relapse when an effective medication is
viduals with a current or recent substance use disorder from discontinued.
this efficacy study. We also excluded more highly suicidal
individuals because of potential risks of discontinuing effi- AUTHOR AND ARTICLE INFORMATION
cacious medication, and we excluded more severely ill From Rhode Island Hospital, Butler Hospital, and the Department of
patients who required concomitant psychotherapy or a Psychiatry and Human Behavior, Alpert Medical School, Brown University,
Providence, R.I.; Massachusetts General Hospital, and the Department of
higher level of care. Study strengths include the fact that, to
Psychiatry, Harvard Medical School, Boston; and Decision Sciences In-
our knowledge, this is the first relapse prevention study in stitute, Pawtucket, R.I.
body dysmorphic disorder; it contained the largest sample
Address correspondence to Dr. Phillips ([email protected]).
of any treatment study in this disorder thus far; and it used
Presented at the third annual New England OCD Research Symposium,
randomization, placebo, and blind and rigorous assess- Orange, Conn., March 27, 2015; at the annual conference of the Anxiety
ment of outcomes at specified intervals. Additional studies and Depression Association of America, Miami, April 9–12, 2015; at the
with larger sample sizes and longer follow-up are needed. International College of Obsessive Compulsive Spectrum Disorders 10th

Am J Psychiatry 173:9, September 2016 ajp.psychiatryonline.org 893

PHARMACOTHERAPY RELAPSE PREVENTION IN BODY DYSMORPHIC DISORDER

Scientific Meeting, Miami, April 12, 2015; at Body Dysmorphic Disorder 6. Phillips KA, Menard W: Suicidality in body dysmorphic disorder: a
Research Day, Institute of Psychiatry, Psychology, and Neurosciences, prospective study. Am J Psychiatry 2006; 163:1280–1282
King’s College London, May 29, 2015; at the annual meeting of the 7. Phillips KA, Albertini RS, Rasmussen SA: A randomized placebo-
American Society of Clinical Psychopharmacology, Miami, June 22–25, controlled trial of fluoxetine in body dysmorphic disorder. Arch Gen
2015; at the 22nd annual conference of the International OCD Foundation, Psychiatry 2002; 59:381–388
Boston, July 31–Aug. 2, 2015; and at the annual meeting of the American 8. Hollander E, Allen A, Kwon J, et al: Clomipramine vs desipramine
College of Neuropsychopharmacology, Hollywood, Fla., Dec. 6–10, 2015. crossover trial in body dysmorphic disorder: selective efficacy of a
Supported by NIMH Collaborative R01 grants to Dr. Phillips (R01 serotonin reuptake inhibitor in imagined ugliness. Arch Gen Psy-
MH072917) and Dr. Wilhelm (R01 MH072854). Study medication and chiatry 1999; 56:1033–1039
matching placebo provided by Forest Laboratories. 9. Phillips KA, Najjar F: An open-label study of citalopram in body
dysmorphic disorder. J Clin Psychiatry 2003; 64:715–720
The authors thank Ashley Hart, Ph.D., Megan Kelly, Ph.D., Andri Bjornsson,
10. Phillips KA: An open-label study of escitalopram in body dysmorphic
Ph.D., Elizabeth Didie, Ph.D., Luana Marques, Ph.D., Angela Fang, Ph.D.,
disorder. Int Clin Psychopharmacol 2006; 21:177–179
Anne Chosak, Ph.D., Natalie Matheny, B.A., Stefanie Renaud, B.A., Martha
11. Phillips KA, Dwight MM, McElroy SL: Efficacy and safety of
Niemiec, B.A., Ashley Shaw, M.S., and Michelle Silverman, B.A., for per-
fluvoxamine in body dysmorphic disorder. J Clin Psychiatry 1998;
forming study ratings and/or providing study administrative support, and
59:165–171
Paul Cannistraro, M.D., Michael Hanau, M.D., and Naureen Atiullah, M.D.,
12. Perugi G, Giannotti D, Di Vaio S, et al: Fluvoxamine in the treat-
for providing relapsing patients with rescue treatment.
ment of body dysmorphic disorder (dysmorphophobia). Int Clin
ClinicalTrials.gov identifier: NCT00149799. Psychopharmacol 1996; 11:247–254
Dr. Phillips has received support from Oxford University Press (royalties), 13. Phillips KA, Menard W, Quinn E, et al: A 4-year prospective ob-
International Creative Management (royalties), American Psychiatric servational follow-up study of course and predictors of course in
Association Publishing (honoraria), Merck Manual (honoraria), Abbott body dysmorphic disorder. Psychol Med 2013; 43:1109–1117
Laboratories (presentation), AstraZeneca (presentation), Global Medical 14. Guy W: ECDEU Assessment Manual for Psychopharmacology.
Education (presentation), Janssen Research and Development (con- Washington, DC, US Department of Health, Education, and Welfare,
sultant), Transcept Pharmaceuticals (research funding), speaking hon- 1976
oraria and/or travel reimbursement from academic institutions and 15. Phillips KA, Albertini RS, Siniscalchi JM, et al: Effectiveness of
professional organizations, UpToDate (future honoraria), The Free pharmacotherapy for body dysmorphic disorder: a chart-review
Press (potential future royalties), and Guilford Press (potential future study. J Clin Psychiatry 2001; 62:721–727
royalties). Dr. Dougherty has received research support from Cyberonics, 16. Phillips KA, Hollander E, Rasmussen SA, et al: A severity rating scale
Eli Lilly, Medtronic, and Roche; and he has received honoraria or for body dysmorphic disorder: development, reliability, and validity
travel support from Insys, Johnson & Johnson, Medtronic, and Roche. of a modified version of the Yale-Brown Obsessive Compulsive Scale.
Mr. Menard has received support from Avid Radiopharmaceuticals, Psychopharmacol Bull 1997; 33:17–22
Biogen, Eli Lilly, Merck, and Transcept Pharmaceuticals. Dr. Wilhelm 17. Phillips KA, Hart AS, Menard W: Psychometric evaluation of the
is a presenter for the Massachusetts General Hospital Psychiatry Yale-Brown Obsessive-Compulsive Scale Modified for Body Dys-
Academy that specializes in educational programs that are supported morphic Disorder (BDD-YBOCS). JOCRD 2014; 3:205–208
through independent medical education grants from pharmaceutical 18. Spitzer RL, Williams JBW, Gibbon M, et al: The Structured Clinical
companies; she has received royalties from Elsevier, Guilford, New Interview for DSM-III-R (SCID): I: History, rationale, and de-
Harbinger Publications, and Oxford University Press; she has received scription. Arch Gen Psychiatry 1992; 49:624–629
research support from Novartis and honoraria from various aca- 19. Williams JBW, Gibbon M, First MB, et al: The Structured Clinical
demic institutions and foundations, including the International Interview for DSM-III-R (SCID): II. Multisite test-retest reliability.
Obsessive Compulsive Disorder Foundation and the Tourette Syn- Arch Gen Psychiatry 1992; 49:630–636
drome Association; and she has received support from the Association 20. First MB, Gibbon M, Spitzer RL, et al: Structured Clinical Interview
for Behavioral and Cognitive Therapies for her role as associate for DSM-IV Axis II Personality Disorders (SCID-II). Washington,
editor of Behavior Therapy, as well as from John Wiley & Sons for her DC, American Psychiatric Association Publishing, 1997
role as associate editor of Depression and Anxiety. Ms. Keshaviah 21. Phillips KA, Stein DJ, Rauch SL, et al: Should an obsessive-
and Dr. Stout report no financial relationships with commercial compulsive spectrum grouping of disorders be included in DSM-V?
interests. Depress Anxiety 2010; 27:528–555
22. Goodman WK, Price LH, Rasmussen SA, et al: The Yale-Brown
Received Sept. 30, 2015; revision received Nov. 26, 2015; accepted Jan. 4,
Obsessive Compulsive Scale: I. Development, use, and reliability.
2016; published online April 8, 2016.
Arch Gen Psychiatry 1989; 46:1006–1011
23. Goodman WK, Price LH, Rasmussen SA, et al: The Yale-Brown
REFERENCES Obsessive Compulsive Scale: II. Validity. Arch Gen Psychiatry 1989;
1. Rief W, Buhlmann U, Wilhelm S, et al: The prevalence of body 46:1012–1016
dysmorphic disorder: a population-based survey. Psychol Med 2006; 24. Simpson HB, Franklin ME, Cheng J, et al: Standard criteria for
36:877–885 relapse are needed in obsessive-compulsive disorder. Depress
2. Buhlmann U, Glaesmer H, Mewes R, et al: Updates on the prevalence Anxiety 2005; 21:1–8
of body dysmorphic disorder: a population-based survey. Psychiatry 25. Eisen JL, Phillips KA, Baer L, et al: The Brown Assessment of Beliefs
Res 2010; 178:171–175 Scale: reliability and validity. Am J Psychiatry 1998; 155:102–108
3. Koran LM, Abujaoude E, Large MD, et al: The prevalence of body 26. Phillips KA, Hart AS, Menard W, et al: Psychometric evaluation of
dysmorphic disorder in the United States adult population. CNS the Brown Assessment of Beliefs Scale in body dysmorphic disorder.
Spectr 2008; 13:316–322 J Nerv Ment Dis 2013; 201:640–643
4. Brohede S, Wingren G, Wijma B, et al: Prevalence of body dys- 27. Miller IW, Bishop S, Norman WH, et al: The modified Hamilton
morphic disorder among Swedish women: a population-based study. Rating Scale for Depression: reliability and validity. Psychiatry Res
Compr Psychiatry 2015; 58:108–115 1985; 14:131–142
5. Phillips KA, Quinn G, Stout RL: Functional impairment in body 28. Leon AC, Solomon DA, Mueller TI, et al: The Range of Impaired
dysmorphic disorder: a prospective, follow-up study. J Psychiatr Res Functioning Tool (LIFE-RIFT): a brief measure of functional im-
2008; 42:701–707 pairment. Psychol Med 1999; 29:869–878

894 ajp.psychiatryonline.org Am J Psychiatry 173:9, September 2016

 PHILLIPS ET AL.

29. Endicott J, Nee J, Harrison W, et al: Quality of life enjoyment and Washington, DC, American Psychiatric Association Publishing, 2015,
satisfaction questionnaire: a new measure. Psychopharmacol Bull pp 57–98
1993; 29:321–326 35. Okuda M, Simpson HB: Obsessive-compulsive disorder, in Hand-
30. Phillips KA, Pagano ME, Menard W, et al: A 12-month follow-up book on Obsessive-Compulsive and Related Disorders. Edited by
study of the course of body dysmorphic disorder. Am J Psychiatry Phillips KA, Stein DJ. Washington, DC, American Psychiatric As-
2006; 163:907–912 sociation Publishing, 2015, pp 25–56
31. Phillips KA: Understanding Body Dysmorphic Disorder: An Es- 36. Phillips KA, Pagano ME, Menard W: Pharmacotherapy for body
sential Guide. New York, Oxford University Press, 2009 dysmorphic disorder: treatment received and illness severity. Ann
32. Bech P, Lönn SL, Overø KF: Relapse prevention and residual Clin Psychiatry 2006; 18:251–257
symptoms: a closer analysis of placebo-controlled continuation studies 37. Phillips KA, Hart AS, Simpson HB, et al: Delusional versus non-
with escitalopram in major depressive disorder, generalized anxiety delusional body dysmorphic disorder: recommendations for DSM-5.
disorder, social anxiety disorder, and obsessive-compulsive disorder. CNS Spectr 2014; 19:10–20
J Clin Psychiatry 2010; 71:121–129 38. Phillips KA, Menard W, Pagano ME, et al: Delusional versus non-
33. Tollefson GD, Birkett M, Koran L, et al: Continuation treatment of delusional body dysmorphic disorder: clinical features and course of
OCD: double-blind and open-label experience with fluoxetine. J Clin illness. J Psychiatr Res 2006; 40:95–104
Psychiatry 1994; 55(Suppl:69–76):77–78 39. Grant JE, Menard W, Pagano ME, et al: Substance use disorders in
34. Phillips KA: Body dysmorphic disorder, in Handbook on Obsessive- individuals with body dysmorphic disorder. J Clin Psychiatry 2005;
Compulsive and Related Disorders. Edited by Phillips KA, Stein DJ. 66:309–316

Am J Psychiatry 173:9, September 2016 ajp.psychiatryonline.org 895