Psychiatry Research 170 (2009) 36

Contents lists available at ScienceDirect

Psychiatry Research
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s

Review article

Cross-cutting issues and future directions for the OCD spectrum

Eric Hollander a,, Suah Kim b, Ashley Braun b, Daphne Simeon b, Joseph Zohar c
a
b
c

Psychiatry, Monteore Medical Center, University Hospital of Albert Einstein College of Medicine, Bronx, NY, United States
Psychiatry, Mount Sinai School of Medicine, New York, NY, United States
Department of Psychiatry, Chaim Sheba Medical Center, Tel Hashomer, Israel

a r t i c l e

i n f o

Article history:
Received 3 January 2008
Received in revised form 17 March 2008
Accepted 26 July 2008
Keywords:
Obsessive-compulsive disorder
Obsessive-compulsive spectrum disorders
Genetics
Cross-species models
Impulse control disorders

a b s t r a c t
The research planning agenda for DSM-V examined possible similarities in phenomenology, comorbidity,
familial and genetic features, brain circuitry, and treatment response between obsessive-compulsive disorder
(OCD) and several related disorders that are characterized by repetitive thoughts or behaviors. Such data
support a re-examination of the DSM-IV-TR classication of OCD and the anxiety disorders, with possible
inclusion of a group of obsessive-compulsive spectrum disorders (OCSDs) in DSM-V. Various disorders were
systematically examined for inclusion in such a grouping, and later a smaller number were determined to meet
threshold criteria for inclusion in the OCSDs. The disorders that were originally examined included OCD,
obsessive-compulsive personality disorder (OCPD), Tourette's syndrome (TS) and other tic disorders,
Sydenham's chorea, Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal
Infections (PANDAS), trichotillomania (TTM), body dysmorphic disorder (BDD), autism, eating disorders,
Huntington's and Parkinson's disease, impulse control disorders, as well as substance and behavioral
addictions. Certain disorders such as BDD, OCPD, TS, and TTM share many commonalities with OCD in
phenomenology, comorbidity, familial and genetic features, brain circuitry, and treatment response. Other
disorders, such as the impulse control disorders (ICDs) share some common features with OCD, but also differ
in many ways as well. The articles presented in this issue of Psychiatry Research are a result of this international
collaboration, which examined diagnostic and classication issues of OCSDs for DSM-V in a conference titled
The Future of Psychiatric Diagnosis: Rening the Research Agenda: Obsessive-Compulsive Behavior
Spectrum held in June 2006 at the American Psychiatric Association's headquarters in Arlington, VA.
2008 Elsevier Ireland Ltd. All rights reserved.

Contents
1.
2.
3.
4.
5.
6.
7.
8.
9.

Introduction . . . . . . . . . . . . . . . . . . . . . . .
Genetics and OCD nosology . . . . . . . . . . . . . . .
Cross-species models of OCD-spectrum disorders . . . . .
OCD and impulse control disorders . . . . . . . . . . . .
Tourette's disorder, trichotillomania, and OCD . . . . . . .
Autism and Parkinson's disease and OCD . . . . . . . . .
Schizophrenia and OCD . . . . . . . . . . . . . . . . .
Methodological issues in the obsessive-compulsive spectrum
Conclusion . . . . . . . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

1. Introduction
The research planning agenda for DSM-V examined possible
similarities in phenomenology, comorbidity, familial and genetic fea Corresponding author. Department of Psychiatry, Child Psychiatry Annex, Monteore Medical Center, University Hospital of Albert Einstein College of Medicine, 111 E.
210th Street, Bronx, NY 10467-2490, United States.
E-mail address: eholland@monteore.org (E. Hollander).
0165-1781/$ see front matter 2008 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.psychres.2008.07.015

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

3
4
4
4
5
5
5
5
6

tures, brain circuitry, and treatment response between obsessivecompulsive disorder (OCD) and several related disorders that are
characterized by repetitive thoughts or behaviors. Such data support a
re-examination of the DSM-IV-TR classication of OCD and the anxiety
disorders, with possible inclusion of a group of obsessive-compulsive
spectrum disorders (OCSDs) in DSM-V. Various disorders were systematically examined for inclusion in such a grouping, and later
a smaller number were determined to meet threshold criteria for
inclusion in the OCSDs. The disorders that were originally examined

E. Hollander et al. / Psychiatry Research 170 (2009) 36

included OCD, obsessive-compulsive personality disorder (OCPD),

Tourette's syndrome (TS) and other tic disorders, Sydenham's chorea,
Pediatric Autoimmune Neuropsychiatric Disorders Associated with
Streptococcal Infections (PANDAS), trichotillomania (TTM), body
dysmorphic disorder (BDD), autism, eating disorders, Huntington's
and Parkinson's disease, impulse control disorders, as well as substance and behavioral addictions.
Certain disorders such as BDD, OCPD, TS, and TTM share many
commonalities with OCD in phenomenology, comorbidity, familial
and genetic features, brain circuitry, and treatment response. Other
disorders, such as the impulse control disorders (ICDs) share some
common features with OCD, but also differ in many ways as well. The
articles presented in this issue of Psychiatry Research are a result of
this international collaboration, which examined diagnostic and classication issues of OCSDs for DSM-V in a conference titled The Future
of Psychiatric Diagnosis: Rening the Research Agenda: ObsessiveCompulsive Behavior Spectrum held in June 2006 at the American
Psychiatric Association's headquarters in Arlington, VA.
2. Genetics and OCD nosology
Nicolini et al. provide insights into the genetics and familial factors
of OCD and related disorders. Family studies of OCD generally show
that the prevalence of OCD is signicantly higher in relatives, especially
in the presence of comorbid tics and earlier age of onset, and also
depends on the types of obsessions and compulsions exhibited by
probands. Twin studies, although few in number, have suggested some
signicant genetic inuence in the heritability of OC symptoms. The
most studied candidate genes for OCD include catechol-O-methyltransferase (COMT), monoamine oxidase-A, dopamine transporter
(DAT) and dopamine receptors DRD1, DRD2, DRD3, and DRD4, the
serotonin transporter (SERT) and g-HT2A and 5HT1B, and the glutamate NMDA subunit receptor genes with the most promising genes
being 5HTTLPR and 5HT1B from the serotonin system, and GRIN2B and
SLC1A1 from the glutamate system. Family studies on the whole have
supported the heterogeneity of OCD and the identication of subgroups of patients including early onset, sex-typing, symptom clustering, and treatment response. There may also be some evidence of
a single major gene contributing to some OCD subtypes such as
symmetry and ordering, eating disorders, early age of onset, and sexspecic subtypes.
Genome scans in the future may hold promising linkage studies on
the entire genome and involve extensive phenotypic analyses for OCD.
Future research should strive to develop comparable data on the
genetic ndings in all disorders on the OCSD, as current ndings
for other spectrum disorders are less comprehensive than those of
OCD proper. Further developments in genetic studies would allow
for prediction or risk models that would address both genetic and
environmental risk factors and also help enhance OCD diagnostic
criteria, which may then be followed by preventative strategies for
high-risk children. Although family studies give insight into important
correlational relationships, causation cannot be inferred using such
methods. These limitations should be considered when designing
future research.
3. Cross-species models of OCD-spectrum disorders
Boulougouris, Chamberlain, and Robbins describe animal modeling of OCSDs as occurring on two levelsthe etiological level and
the symptomatic level. It is probable that several candidate genes
contribute to OCD vulnerability so that difculties in modeling arise.
Furthermore, behavioral symptoms simulated in mice may not wholly
reect the subtleties of the disorder. Still, advances in understanding
the neural substrates of OCD and efcacy of pharmacologic treatment
may validate animal models. Ethological and behavioral models have
identied several repetitive behaviors in animals suggestive of OCD

such as acral lick dermatitis, hair-pulling, barbering, cribbing, and

wheel-running, and have tested the effects of drug therapy. Of note,
clomipramine appeared to alleviate acral lick dermatitis and other
repetitive behaviors thought to be stress-induced coping responses
similarly seen in OCD. It is difcult to know, however, how closely
related these animal stereotypies are to OCD rituals. Exploring what
mechanisms are involved in turning habit into compulsions in these
animals is of interest. Genetic models of OCD have involved the hoxb8
mutant seen in excessive grooming and manipulations of DA and 5-HT
functioning that have elicited similar behaviors from repetitive jumping to chewing. Boulougouris and colleagues also describe signal
attenuation and extinction as behavioral models of OCD, whereby the
animal receives poor feedback that a behavior has been completed
leading to perseverative compulsions that can be reduced by drugs
typically used to treat OCD. Neuroimaging studies in humans point to
the orbitofrontal cortex (OFC) and the caudate nucleus in OCD with
probable dopaminergic and serotonergic mediation, and lesions in the
OFC promote repetitive behaviors observed in rats. Stop-signal
inhibition, in particular, has been linked to dysfunction of the lateral
OFC in humans with OCD and in homologous brain regions in rats.
4. OCD and impulse control disorders
Potenza, Koran, and Pallanti describe both commonalities and
differences in the clinical, phenomenological, and biological features
between intermittent explosive disorder (IED), pathological gambling
(PG), and OCD, with the overarching similarity being an inability to
resist repetitive behaviors that may be harmful to the self or others. A
distinct contrast between these disorders is the ego-dystonic nature of
OCD versus the ego-syntonic nature of impulse control disorders
(ICDs). However, the repetitive behaviors in ICDs may change over
time and become less pleasure-driven and more driven to alleviate
distress, thereby resembling OCD.
From a phenomenological point of view, although IED and OCD are
similar in their intrusiveness and repetitiveness, the behaviors seen in
IED are not intended to reduce anxiety and are not in response to
obsessions, as compulsions are in OCD. Other differences include a
higher prevalence rate of IED than OCD, IED predominance in males,
and an association of IED with marriage and low educational level.
Although few genetic studies of IED have been conducted, one study
has shown a possible linkage with the 5HT 1B receptor gene, which has
not been associated with OCD. While 5HT systems and the orbitofrontal cortex have been linked with both IED and OCD, it appears that
their involvement differs; administration of 5HT1 and 5HT2 receptor
agonists and fenuramine worsened OC symptoms and increased
prolactin release in OCD patients, whereas these drugs blunted prolactin response in IED patients. Also, increased activity in the ventromedial prefrontal cortex has been observed in OCD, while decreased
activity was implicated in IED. Likewise, decreased activity in the
frontostriatal circuitry has been observed in PG. Serotonin reuptake
inhibitors (SRIs) are effective in both IED and OCD, although doses
used to treat the former tend to be lower. Although the data are less
clear in PG, SRIs may be somewhat effective in treating PG, especially
when there is a comorbid anxiety disorder. Mood stabilizers appear
effective in treating IED and PG, while they have been less effective as
augmentation strategies in OCD. Antipsychotic agents may show some
efcacy in treating OCD, but not PG.
Behavioral therapy approaches also differ between OCD, IED, and
PG, with exposure and response prevention being used in OCD
treatment; cognitive behavior therapy (CBT), group and family therapy
showing some effectiveness in treating IED, and CBT, motivational
therapy or Gamblers' Anonymous being helpful in PG. PG and OCD are
similarly characterized by repetitive behaviors that usually contribute
to poorer life functioning, but while these behaviors are aimed at
reducing anxiety in OCD sufferers, the gambling behavior is at rst
pleasure-driven in PG. Although PG may share some phenomenological

E. Hollander et al. / Psychiatry Research 170 (2009) 36

features with OCD, it may be more closely related to substance abuse and
dependence. Some studies also suggest greater comorbidity between PG
and several other disorders than with OCD. Multiple allelic variants have
been implicated in PG with a common gene, the 5HT transporter, being
associated with both PG and OCD. However, the types of 5HT transporter
alleles in these two disorders differ the short allele linked with PG and
the long allele linked with OCD. As ndings are mixed and scarce, further
studies are needed to examine the phenomenology, neurobiology, and
treatment response of ICDs and OCD to better understand these
heterogeneous disorders and to allow for better diagnostic measures
and clinical care.
5. Tourette's disorder, trichotillomania, and OCD
Ferrao, Miguel, and Stein compare the phenomenology, psychobiology, and treatment response of OCD, Tourette's syndrome (TS),
and trichotillomania (TTM) in considering their reclassication into a
spectrum of related disorders. While compulsive behaviors observed
in OCD without TS are responses to obsessive thoughts, the repetitive
behaviors in TS and OCD with tics are usually exhibited to alleviate
unpleasant sensations. Similarly, while the repetitive behavior of TTM
exclusively involves hair-pulling, this behavior often follows high
anxiety and results in lowered anxiety. The authors point out that OCD
combined with vocal or motor tics exhibits similar frequency of
repetitive behaviors as TS but greater symptom frequency than OCD,
and lies between these two disorders in phenomenological features
such as comorbidity, symptom onset, and frequency of somatic
obsessions. These ndings support the idea that OCD and tic disorders
may lie on a continuum.
Family studies show a greater prevalence of OC symptoms and OCD
in relatives of TS sufferers, as well as higher rates of tics or TS in family
members of OCD patients, when compared with healthy controls.
Likewise, TTM and OCD are more common in relatives of TTM
probands. OCD and TS may also show common genetic vulnerabilities,
namely those involved in the serotonin, dopamine, and glutamate
systems. Although fewer genetic ndings are present in TTM, this
disorder has exhibited treatment response with dopamine blockers.
Some similarities in the neurobiology of OCD and TS include involvement of the corticostriatal and thalamic circuits, and irregular
dopamine transporter and receptor densities.
Although antipsychotic agents are traditionally used to treat TS,
while SRIs are used to treat OCD, some studies have shown that
augmentation therapies involving these two classes of drugs are
effective in treating both OCD and TS. The ndings are mixed, but a
review of clinical trials involving antipsychotic augmentation with SRIs
has suggested better treatment response in those with OCD and tics.
There are considerably fewer reports of pharmacological treatments for
TTM than OCD, and SRIs have been more effective in open-label trials
than in controlled studies. Treatment strategies involving psychotherapy, electroconvulsive therapy, neurosurgery, deep brain stimulation,
transcranial magnetic stimulation, and immunological interventions for
TS and OCD have shown fewer similarities and require further research
for areas of possible overlap. Psychotherapy has shown some efcacy in
treating these disorders as well. Both TTM and TS symptoms have been
shown to be alleviated somewhat by using the CBT technique of habit
reversal, while for OCD the most effective psychotherapy techniques
have been shown to be exposure and response prevention.
6. Autism and Parkinson's disease and OCD
Hollander, Wang, Braun, and Marsh discuss two neurological disorders that encompass OC features: autism and Parkinson's disease
(PD). Although autism, a developmental disorder, and PD, a degenerative disorder, may at rst appear dissimilar, both disorders may be
characterized by repetitive behaviors and impulsive behaviors, and
similar processes may occur in both a developmental and a degene-

rative disorder. Autism spectrum disorders and OCD are similarly

characterized by rigid observance of routines and rituals, and some
autistic patients also report having obsessions and compulsions. Studies on OC symptoms (OCS) in PD are mixed. However, a condition in
PD called punding, which may be due to excess dopaminergic therapy
in PD, resembles the repetitive behaviors seen in OCD and is also often
anxiety-reducing. By contrast, punding is not rigid, does not occur in
response to obsessions, and is not aimed at averting unpleasant events.
Although autism and OCD share some similar comorbidities, autism is
also signicantly comorbid with seizures, epileptiform EEG abnormalities, mental retardation, genetic disorders, and speech and language
disorders. Findings on comorbidities in OCS with PD are limited, but
punding is associated with ICDs, psychosis, and excessive use of
dopaminergic medications resembling an addiction, as well as motoric
side effects such as extreme on-off uctuations and dyskinesias.
Autism is similar to early onset OCD in its predominance in boys,
prolonged course of illness, and association with tics. The course of
illness of OCS in PD and punding is unclear, although in certain cases,
antiparkinsonian treatments appear to have an inuence on the course
of the disease. Autism is highly heritable, and studies have suggested
associations between OCD or OC behaviors in parents of autistic
children and repetitive behavior scales. Genetic studies show mixed
ndings for candidate genes common between OCD and autism,
although there has been some evidence for the involvement of the
serotonin gene SLC6A/5-HTT in both disorders. Family and genetic
studies are scarce and less clear for OCS or ICDs in PD, but functional
polymorphisms of genes regulating the dopamine system are of
interest for comorbid ICDs. The corticostriatalthalamic circuitry is
associated with OCD, and the repetitive behaviors in autism, but the
nature of the abnormalities varies across the disorders. Important
neural similarities between autism, PD, and OCD involve basal ganglia
and frontal lobe dysfunction.
7. Schizophrenia and OCD
Schizo-OCD refers to clinical presentation of both schizophrenia
and OCD symptoms in a patient. This dual diagnosis can be perplexing
with regards to neurofunctional alteration, given that OCD is typically
characterized by hyperfrontality whereas schizophrenia is marked by
hypofrontality. Several studies have attempted to compare schizophrenia, OCD, and Schizo-OCD using various cognitive tasks, but
results are inconclusive. Pallanti et al. investigated cognitive event
related potentials (ERP) measured during a discriminative response
task (DRT) in Schizo-OCD patients, compared with patients with OCD
without psychotic features, patients with schizophrenia without OCD,
and healthy controls. A total of 11 Schizo-OCD patients, 16 OCD patients, 14 schizophrenic patients, and 12 healthy controls participated.
When comparing the ERP results between groups, the Schizo-OCD
group was found to exhibit a unique abnormal pattern compared with
the OCD only, schizophrenia only, and healthy control groups. These
results may have important implications for the reconceptualization
of OCD subtypes and the obsessive-compulsive spectrum. Further
studies are needed to replicate these preliminary ndings.
8. Methodological issues in the obsessive-compulsive spectrum
McKay and Neziroglu discuss methodological and statistical approaches to classifying disorders in the OCSD. In their article they
elucidate concerns regarding the heterogeneity of conditions in this
spectrum, and particularly discuss the differences in classifying a
disorder as an OCD subtype versus a distinct disorder in the OCSD. An
OCD subtype must demonstrate the core symptoms and phenomenology of obsessions and compulsions, and must lack other prominent
symptoms which might distinguish it from OCD proper. However,
simply classifying subtypes based on symptom similarity is limited by
the absence of a hierarchical, theory-based approach to understanding

E. Hollander et al. / Psychiatry Research 170 (2009) 36

and classifying subtypes. A broadly accepted commonality between

disorders of compulsivity and impulsivity is the presence of decits
in the behavioral inhibition system. However, reliance on decit
models in the OCSD is not desirable because, with this method, one
malfunctioning area is being used to classify a large, heterogeneous
group of disorders.
There are several methodological and statistical concerns in
identifying conditions in the OCSD, such as the lack of effect size
reporting in the OCSD literature. Recently the use of statistical equivalence testing has offered advances in comparing correspondence for
two independent groups (e.g., two disorders along the proposed OCSD).
This method would aid in determining whether a condition is a possible
OCD subtype, or a condition in the OCSD.
In preparation for DSM-V, a dimensional model approach to the
OCSD has been proposed which would ease screening and diagnosis.
Since by denition, a dimensional model is based on degree of variance
rather than a yes/no categorical response, this method would be
particularly appropriate for the OCSD. Other approaches to determining whether a disorder is a subtype or condition include cluster
analysis or multivariate taxometric analyses. Each of these methods
has advantages and disadvantages which are discussed in the McKay
and Neziroglu article in this issue of Psychiatry Research.
9. Conclusion
There is growing interest in the scientic community to explore
the relationships between OCSDs and OCD based on commonalities of
phenomenology, comorbidity, course of illness, brain circuitry, familial

and genetic factors, and treatment response. The research planning conference on OCSDs aimed to bring an international group of
scientists together to gather empirical research that may inform
classication for future DSM efforts. The articles in this issue of Psychiatry Research present some of the ndings related to genetics and
OCD nosology, cross-species models of OCSDs, and the relationships
between OCD and ICDs, Tourette's syndrome, trichotillomania, autism,
and Parkinson's disease. Although ndings are mixed in comparing
the domains listed above, future directions in research should
examine OCSDs based on endophenotypic features.
Endophenotyping efforts should include the following features: (1)
clarication of OCD symptom dimensions; (2) clarication of inclusion
criteria for OCSDs; (3) determination of which disorders should be
included in the OCSDs; (4) clarication of subtypes; (5) use of existing
databases; (6) construction of a common endophenotype battery that
includes neurocognition, genotyping, functional brain imaging, symptom scales, structured assessment for comorbidity, and treatment
response; (7) development of self-administered scales for threshold
diagnosis and sensitivity to change; (8) multicenter trials that include
an endophenotyping project; and (9) comparison of the OCSDs to the
other anxiety disorders. Because OCD and especially OCSDs are
underdiagnosed in patients who report a broad symptom of anxiety,
a reclassication of OCD and related disorders into a broader category
would promote better assessment of OC symptoms, more accurate
diagnoses, greater research efforts, and potentially the development of
more effective treatments.