Uploaded by Loukas Nikolaou

obsessive-compulsive personality disorder evidence for two dimensions

DEPRESSION AND ANXIETY 33:128–135 (2016)
Research Article
Mark A. Riddle, M.D.,1 ∗ Brion S. Maher, M.D.,2 Ying Wang, Ph.D.,1 Marco Grados, M.D.,1
O. Joseph Bienvenu, M.D.,1 Fernando S. Goes, M.D.,1 Bernadette Cullen, M.D.,1 Dennis L. Murphy, M.D.,3
Scott L. Rauch, M.D.,4 Benjamin D. Greenberg, M.D.,5 James A. Knowles, M.D.,6 James T. McCracken, M.D.,7
Anthony Pinto, Ph.D.,8 John Piacentini, Ph.D.,7 David L. Pauls, Ph.D.,9 Steven A. Rasmussen, Ph.D.,5
Yin Yao Shugart, Ph.D.,10 Gerald Nestadt, M.D.,1 and Jack Samuels, Ph.D.1
Background: To determine possible dimensions that underlie obsessive–
compulsive personality disorder (OCPD) and to investigate their clinical correlates, familiality, and genetic linkage. Methods: Participants were selected from
844 adults assessed with the Structured Instrument for the Diagnosis of DSM-IV
Personality Disorders (SIDP) in the OCD Collaborative Genetics Study (OCGS)
that targeted families with obsessive–compulsive disorder (OCD) affected sibling pairs. We conducted an exploratory factor analysis, which included the eight
SIDP-derived DSM-IV OCPD traits and the indecision trait from the DSM-III,
assessed clinical correlates, and estimated sib–sib correlations to evaluate familiality of the factors. Using MERLIN and MINX, we performed genome-wide
quantitative trait locus (QTL) linkage analysis to test for allele sharing among
individuals. Results: Two factors were identified: Factor 1: order/control (perfectionism, excessive devotion to work, overconscientiousness, reluctance to delegate,
and rigidity); and Factor 2: hoarding/indecision (inability to discard and indecisiveness). Factor 1 score was associated with poor insight, whereas Factor 2
score was associated with task incompletion. A significant sib–sib correlation was
found for Factor 2 (rICC = .354, P < .0001) but not Factor 1 (rICC = .129, P
= .084). The linkage findings were different for the two factors. When Factor 2
was analyzed as a quantitative trait, a strong signal was detected on chromosome
1 Department
of Psychiatry and Behavioral Sciences, Johns
Hopkins University School of Medicine, Baltimore, Maryland
2 Department of Mental Health, Bloomberg School of Public
Health, Johns Hopkins University, Baltimore, Maryland
3 Laboratory of Clinical Science, National Institute of Mental
Health, National Institute of Health, Bethesda, Maryland
4 Department of Psychiatry, McLean Hospital, Harvard Medical
School, Boston, Massachusetts
5 Department of Psychiatry and Human Behavior, Brown Medical School, Butler Hospital, Providence, Rhode Island
6 Department of Psychiatry, University of Southern California
School of Medicine, Los Angeles, California
7 Department of Psychiatry and Biobehavioral Sciences,
School of Medicine, University of California Los Angeles, Los
Angeles, California
8 Department of Psychiatry, Columbia University College of
Physicians and Surgeons, New York State Psychiatric Institute, New York City, New York
9 Department of Psychiatry, Psychiatric and Neurodevelopmental Genetics Unit, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
C 2015 Wiley Periodicals, Inc.
10 Unit
of Statistical Genomics, Division of Intramural Research, National Institute of Mental Health, National Institute
of Health, Bethesda, Maryland
Contract grant sponsor: National Institute of Mental Health; Contract grant sponsor: NIH; Contract grant sponsor: NARSAD; Contract grant sponsor: Seaside Therapeutics; Contract grant sponsor:
Roche; Contract grant sponsor: Shire Contract grant sponsor: Dart
Neuroscience; Contract grant numbers: R01MH50214, MH071507,
MH079489, MH079487, MH079488, and MH079494.
∗ Correspondence to: Mark Riddle, Johns Hopkins University School
of Medicine, 550 N. Broadway, Suite 203, Baltimore, MD 21205,
E-mail: [email protected]
Received for publication 22 April 2015; Revised 23 September
2015; Accepted 18 October 2015
DOI 10.1002/da.22452
Published online 23 November 2015 in Wiley Online Library
Research Article: Obsessive–Compulsive Personality Disorder
10 at marker d10s1221: KAC LOD = 2.83, P = .0002; and marker d10s1225:
KAC LOD = 1.35, P = .006. Conclusions: The results indicate two factors of
OCPD, order/control and hoarding/indecision. The hoarding/indecision factor
is familial and shows modest linkage to a region on chromosome 10. Depression
C 2015 Wiley Periodicals, Inc.
and Anxiety 33:128–135, 2016.
Key words: obsessive–compulsive disorder; personality disorder; genetic; assessment and diagnosis; hoarding
bsessive–compulsive personality disorder (OCPD) is
a common and disabling disorder with an estimated lifetime prevalence of 2.1–7.9%.[1–4] In DSM-5, OCPD
is characterized by a “pervasive pattern of preoccupation with orderliness, perfectionism, and mental and interpersonal control, at the expense of flexibility, openness, and efficiency as indicated by four (or more)” of
eight traits.[4] OCPD is one of only a few personality
disorders included in all versions of the American Psychiatric Association’s Diagnostic and Statistical Manual
(DSM; I through 5) over the past 60 years.[4–10] However,
the core traits of OCPD have changed from DSM-II,[6]
when they were first delineated, through DSM5[4] (see
Table 1).
Obsessive–compulsive disorder (OCD), OCPD, and
hoarding disorder (HD) are categorical diagnoses in
DSM-5 that commonly co-occur in clinical settings[11]
and community samples[12] and have overlapping symptoms. For example, in DSM-5, hoarding is a symptom in
all three disorders: as the core feature of HD, as one of
eight core traits of OCPD, and as a symptom of OCD (if
the behavior is related to obsessions).[4] Although categorical diagnoses have pragmatic utility, a deeper understanding of dimensions of behavior that link to biological
substrates is needed to further our understanding of human behavior and psychopathology.
For over 80 years, astute clinical observers, including the eminent British psychiatrist Aubrey Lewis,
have described two types of obsessive–compulsive
personality.[13] Lewis noted, “one [was] obstinate, morose, irritable, the other vacillating, uncertain of himself,
submissive.”[13] Features of the second type were included in DSM-III, with the addition of indecisiveness
and restricted expression of affection.[14] These features
were retained in DSM-III-R, but dropped in DSM-IV.
Indecisiveness appears to have been dropped because
some studies showed low sensitivity and specificity, and
it was relatively common in patients with other personality disorders including paranoid, dependent, avoidant,
and schizotypal.[15] However, a study using stepwise
logistic regression modeling found only modest support
for this decision.[16] It is worth noting that indecision
is listed first in a group of six “other common features”
of the new DSM-5 HD[4] (see Table 1). A review of the
literature revealed a population-based study that found
an association between hoarding and indecision.[17]
Our group previously conducted a factor analysis of
the 79 traits of the 10 personality disorders in DSM-IV in
a community-based sample[18] and the results suggested
five independent factors. The first factor, “compulsive,”
included seven of the eight DSM-IV traits for OCPD,
all with loadings above 0.50. The only trait that did not
load on the compulsive factor or on any of the other four
factors was hoarding. There are many personality disorder factor studies, and all derive a compulsive factor[19, 20]
however, we are not aware of any prior studies of subtypes or factors of OCPD.
The goal of this study was to identify possible OCP dimensions and evaluate their clinical correlates, familiality (sib:sib correlations), and genetic linkage in a convenience sample ascertained through probands with OCD.
Prior research informed our aims.[21–25] Aim 1: Perform
a factor analysis of the eight features of OCPD plus indecision. Aim 2: Using sib:sib pairs, determine if any
factor was familial. Aim 3: Perform linkage analysis on
the factors to identify genetic markers. Aim 4: Determine clinical features that are related to or differentiated
from the factors, including features of OCD, select Axis
1 psychiatric diagnoses, personality disorders, and OC
personality traits.
The OCD Collaborative Genetics Study (OCGS) has been described in detail elsewhere.[26] In brief, the study recruited families
with OCD-affected sibling pairs, and extended these when possible
through affected first- and second-degree relatives. Participants were
recruited from outpatient and inpatient clinics, referrals from clinicians
in the community, web sites, media advertisements, self-help groups,
and the Obsessive Compulsive Foundation (OCF) annual conventions.
To be considered affected, a participant had to meet DSM-IV OCD
criteria at any time in his/her life. Probands were included if, in addition to meeting DSM-IV criteria, their first onset of obsessions and/or
compulsions occurred before the age of 18. Probands with schizophrenia, severe mental retardation, Tourette disorder, or OCD which occurred exclusively in the context of depression (secondary OCD) were
excluded. Subjects had to be at least 7 years old to participate in the
study. Written, informed consent (or assent, for children) to study procedures was obtained prior to the clinical interview. The protocol was
approved by institutional review boards at each site.
Diagnostic assessments were conducted by psychiatrists or PhDlevel clinicians experienced with clinical evaluations.
Depression and Anxiety
Riddle et al.
TABLE 1. Traits of obsessive–compulsive personality
disorder in various DSMs
Inability to relax
Restricted affectivity
Work devotion
Decreased delegation
Detail oriented
Intimacy avoidance
DSM version
III-R IV/V V-Append.
OCD Symptoms. Examiners used the OCGS Assessment Package, modified and developed for the study, as a semistructured format
for the evaluation of psychopathology. The OCD section was adapted
from the Schedule for Affective Disorders and Schizophrenia-Lifetime
version adapted for anxiety disorders (SADS-LA).[27, 28] The Yale–
Brown Obsessive Compulsive Scale (YBOCS) symptom checklist[29]
was refined to include the age of onset and level of severity (i.e., amount
of time and level of distress during the worst period) of each symptom;
a modified version of YBOCS[29] was included, as well as additional
questions on onset, course, and history of treatment for these symptoms. The OCD section also included general questions about whether
the subject had trouble in making decisions, difficulty in completing
tasks, and insight into symptoms. See other Supporting Information for
DSM-IV Diagnoses. The Structured Clinical Interview for
DSM-IV was used for assessing most other major Axis I diagnoses.[30] A
semistructured assessment protocol was used for additional diagnoses
of interest in OCD (pathological nail biting, skin picking, gambling,
trichotillomania, kleptomania, and pyromania).
Personality Disorders. In adults, relevant items from the Structured Instrument for the Diagnosis of DSM-IV Personality Disorders
(SIDP)[31] were used to assess criteria for schizoid, avoidant, dependent, and OCPDs. Each trait was rated 0 (not present), 1 (subthreshold), 2 (present), or 3 (strongly present).
Personality Dimensions. In addition, most subjects selfcompleted the Revised NEO Personality Inventory (NEO-PI-R) questionnaire for the assessment of dimensions of the five-factor model of
personality (neuroticism, extraversion, openness, agreeableness, and
conscientious) and their facets.[32] Personality disorders and dimensions were not evaluated in children or adolescents under age 18 years.
Blood was collected from participants and sent directly to
the NIMH Cell Repository at Rutgers University, where EBVtransformed lymphoblastoid cell lines were created and DNA extracted. DNA for the genome-wide scan was sent to the Center for
Inherited Disease Research (CIDR) for genotyping.
As described previously,[33] genotyping was performed at CIDR
using a modification of the CHLC version 10 marker set (386 microsatellite markers; average spacing of 9 cM; and average heterozygosDepression and Anxiety
ity of .76). Family relationships were analyzed using RELCHECK.[34]
Misspecified relationships were corrected, or removed if there was insufficient information to determine the true relationship. In addition,
PEDCHECK was used for more complicated genotyping error checking for each individual in the dataset.[35] Those genotypes with a posterior probability greater than 75% of being incorrect were removed
from the analysis.[36] In addition, the Merlin program was used to
check for Mendelian inconsistencies and to identify genotypes associated with an excessive number of observed recombinations.[37] Marker
allele frequencies were computed using data from one individual randomly chosen from each pedigree.
Factor Analysis. To identify the nature and number of underlying factors among the OCPD criteria, we performed a principal components analysis, using SPSS statistical software (SPSS 19) with varimax rotation. We used the eight items corresponding to the traits in
OCPD assessed using the SIDP-IV: preoccupation with details, perfectionism, excessive devotion to work, overconscientiousness, hoarding, reluctance to delegate, miserly style, and rigidity. We also included
an additional item: indecisiveness, because of its inclusion in prior versions of the DSM and because a review of the literature revealed a
population-based study that found an association between hoarding
and indecision.[17] A trait was coded (regarded) as positive (present) if
the rating was a 2 (present) or 3 (present and interfering).
We used eigenvalues and the scree plot to estimate the number of factors. We examined the magnitude of the factor loadings
on the nine traits. We included traits with factor loadings greater
than 0.45, and excluded traits that did not discriminate well between
We then calculated factor scales by adding, for each factor, the
number of traits rated “2” or “3” (i.e., “present” or “strongly present”).
We calculated Pearson correlation coefficients between factors and
factor scales with a two-tailed test of significance.
Clinical Correlate Analyses. Contingency table analysis and logistic regression were used to evaluate the association between these
scale scores and categorical variables, whereas the Student t-test,
ANOVA, and linear regression were used to evaluate the association
of the scale scores with continuous variables.
Familiality Analyses. As a formal test of familiality of the phenotypes, we used two approaches to explore familial correlation.
First, we calculated Pearson correlations for OCP for all possible
sibling pairs within the dataset. Second, we used a formalized approach to test univariate and multivariate correlations within familial
data using the FCOR program as implemented in S.A.G.E. v6.3.[38]
The method, described in Mathew et al.[39] allows for the estimation of sibling correlations with valid standard errors and confidence
Linkage Analyses. Linkage analyses were conducted first using
the quantitative factor scores; second using OCD as the outcome measure and stratifying families based on their OCP factor scores. In both
instances MERLIN v1.0.1 was used. In the first instance the QTL option was used for the autosomal genome; this involved a nonparametric
statistic to test for allele sharing among the related individuals. The
MINX version of MERLIN was used for the QTL linkage analysis on
chromosome X,[40] i.e., the nonautosomal genome. Second, to test the
relationship of the factors to OCD, families were stratified into those
having two or more family members with OCP factor score equal or
less than the median and those having zero or one member with OCP
factor scores equal or less than the median. MERLIN v1.0.1, multipoint allele-sharing methods were used to assess evidence for linkage
using OCD as the phenotype on families stratified by factors’ median
Research Article: Obsessive–Compulsive Personality Disorder
In the OCGS, there were 999 participants in 238 families; 277 did not have OCD; 82 had probable OCD; 624
had definite OCD; and 16 had an unknown OCD status. Six hundred fourteen (61.5%) were females, and 385
(38.5%) were males. Of these 999 participants, 844 were
adults (ࣙ18 years old) who were assessed with the SIDP.
The number of participants included in various types of
analyses varied. The factor analysis included 588 adults,
irrespective of OCD status, who had nonmissing scores
on all nine OCP traits (including indecision, which was
added later in the study). For the analyses of the relationships between factor scale scores, OCD features, and
other clinical features, we focused on the 509 adults with
definite OCD; because of missing values for some traits,
for only 473 was a Factor 1 scale score calculated, and for
only 354 was a Factor 2 scale score calculated. For the
sib–sib familiality analysis we included the 360 adult sibling pairs. For the whole genome scan, 999 study samples
were genotyped; of these, 966 participants in 219 families passed genotype quality control and were included
in the linkage analysis.
The data were factor-analyzed for a range of possible solutions. Based upon eigenvalues greater than 1.0
and the shape of the scree plot, as well as the clinical
appreciation of the derived factors, a two-factor solution
appeared to be the most appropriate (See Fig. 1).
The factors were labeled as follows: Factor 1: Order/control (perfectionism, excessive devotion to work,
overconscientiousness, reluctance to delegate, and rigidity); and Factor 2: hoarding/indecision (inability to discard and indecisiveness; see Table 2). The factor loadings
were high, ranging from 0.53 to 0.76 and discriminating for all traits except miserly style (0.176; 0.409) and
preoccupation with details (0.486; 0.496). The two factor solution accounted for 74.3% of the total explained
were associated with poor insight (χ42 = 16.8; P = .002),
whereas higher Factor 2 score were associated with task
incompletion (χ22 = 17.5; P < .001).
Axis I Disorders. Higher Factor 1 scores were associated with higher prevalence rates of many anxiety disorders: panic disorder (χ 2 4 = 11.7; P = .02), agoraphobia
(χ42 = 14.2; P = .007), social phobia (χ42 = 11.2; P = .02),
specific phobia (χ42 = 12.2; P = .02), generalized anxiety
disorder (χ42 = 23.5; P < .001), and posttraumatic stress
disorder (χ42 = 10.7; P = .03). Higher Factor 2 scores
were also related to higher prevalence rates of just two
anxiety disorders, social phobia (χ22 = 8.57; P = .01) and
generalized anxiety disorder (χ22 = 9.11; P = .01), as well
as with pathological skin picking (χ22 = 13.4; P = .001)
and pathological grooming disorders (χ22 = 13.9; P =
Personality Disorders and Personality. Higher
Factor 1 and Factor 2 scores were associated with more
schizotypal (F4;457 = 12.3; P < .001 and F2;346 = 11.1; P
< .001), avoidant (F4;441 = 5.52; P < .001 and F2;330 =
19.5; P = .001), and dependent (F4;441 = 6.12; P < .001
and F2;330 = 48.2; P < .001) personality disorder traits.
Higher neuroticism scores were associated with
higher scores on both Factor 1 (F4;276 = 2.69; P = .03)
and Factor 2 (F2;216 = 5.88; P = .003). However, Factor
1 scores were inversely related to agreeableness (F4;276
= 3.34; P = .01), whereas Factor 2 scores were inversely
related to extraversion (F2;216 = 4.54; P = .01) and conscientiousness (F2;216 = 10.1; P < .001).
A significant sib:sib correlation was found for Factor 2 (rICC = .354, P < .0001) but not for Factor 1
(rICC = .129, P = .084). We further examined correlation within gender-specific sibling-pair groups using
FCOR and found no significant male–male correlations
with 14 pairs for Factor 1 (r = .08, P = .89) or Factor
2 (r = .04, P = .91); however, we did find a significant
correlation among 93 sister–sister pairs for Factor 2 (r =
.36, P = .005) but not Factor 1 (r = .09, P = .48).
OCD Features. We found several differences in the
relationship of the two specific factor scores and OCD
features. Specifically, Factor 1 scores (i.e., the numbers
of Factor 1 traits present) were inversely related to the
age at onset of obsessive–compulsive symptoms (F4;464 =
5.1; P < .001), but Factor 2 scores were not. In addition,
Factor 1 scores were associated with higher prevalence
of symmetry obsessions (χ42 = 31.1; P < .001), ordering
compulsions (χ42 = 27.9; P = <.001), and hoarding compulsions (χ42 = 12.6; P = .013), whereas Factor 2 scores
were associated with higher prevalence of hoarding obsessions (χ22 = 72.0; P < .002) and hoarding compulsions
(χ22 = 69.2; P < .001). Moreover, higher Factor 1 scores
The linkage results differed between the two OCPD
factors. When Factor 1 was analyzed as a quantitative
trait, signals were detected on chromosome 1 (at marker
d1s2845: KAC LOD = 1.88, P = .002; and marker
d1s2660: KAC LOD 1.78, P = .002), chromosome 15
(at marker c15s503: KAC LOD = 1.6, P = .003; and
marker d15s659: KAC LOD = 1.79, P = .002; marker
D15s1507: KAC LOD = 1.4, P = .006), chromosome
18 (at marker d18s862: KAC LOD = 1.42, P = .005),
and chromosome 19 (at marker d19s1034: KAC LOD =
1.32, P = .007).
When Factor 2 was analyzed as a quantitative trait, the
strongest signal was detected on chromosome 10, with a
peak at marker d10s1221: KAC LOD = 2.83, P = .0002;
and marker d10s1225: KAC LOD = 1.35, P = .006.
Depression and Anxiety
Riddle et al.
Figure 1. Scree plot of OCPD criteria.
TABLE 2. Factor loadings of OCPD criteria: two-factor
solution with varimax rotation
Preoccupation with details
Excessive devotion to work
Inability to discard
Reluctance to delegate
Miserly style
Traits that loaded substantially on only one of the two factors are in
There also was a more modest signal on chromosome
7 at marker d7s3056: KAC LOD = 1.3, P = .007. The
results are shown in Tables 3 and 4 and Fig. 2.
The median score for Factor 1 was −0.1376. After
stratification, 80 families have two or more family members with Factor 1 scores equal or less than the median
(low group); 139 families had zero or one member with
a Factor 1 score equal or less than the median (high
group). In the low group, linkage signals were observed
on chromosome 3 (at d3s2398: KAC LOD = 1.82, P =
.002). In the high group, linkage signals were observed
on chromosome 7 (at d7s1818: KAC LOD = 1.77, P
= .002) and chromosome 15 (at d15s642: KAC LOD =
1.99, P = .0012; and marker d15s966: KAC LOD = 1.74,
P = .002).
Depression and Anxiety
TABLE 3. Nonparametric quantitative linkage analysis
for OCPD Factor 1
Chromosome Kosambi distance (cM) KAC LOD
< .01.
The median score for Factor 2 was −0.1776. Similarly, 77 families belonged to the low group, and 142
families belonged to the high group after stratification
on the median. In the low group, we detected linkage
peaks on chromosome 3 (at d3s1262: KAC LOD = 1.99,
P = .0012), chromosome 12 (at marker d12s1064: KAC
LOD = 1.72, P = .002, and marker d12s1300: KAC
LOD = 1.76, P = .002), and chromosome 20 (at marker
TABLE 4. Nonparametric quantitative linkage analysis
for OCPD Factor 2
< .01.
Chromosome Kosambi distance (cM) KAC LOD
Research Article: Obsessive–Compulsive Personality Disorder
Figure 2. (a) Nonparametric quantitative linkage results for OCPD Factor 1 and Factor 2. The dotted line is for Factor 1 and the solid
line for Factor 2. (b) Linkage analysis using OCD as the phenotype in families stratified by OCP factors. The blue line represents –log
P from families with two or more members having OCP factor scores less than or equal to the median. The red line represents –log P
from families with zero or one member having OCP factor scores less than or equal to the median.
d20s851: KAC LOD = 1.77, P = .002). In the high
group, we detected peaks on chromosome 7 (at marker
d7s817: KAC LOD = 1.75, P = .002; marker d7s2846:
KAC LOD = 2.07, P = .001; marker d7s1818: KAC
LOD = 2.74, P = .0002; marker d7s1824: KAC LOD
= 2.25, P = .0006). The results are shown in Fig. 2b.
The results of this study provide evidence for two dimensions of OCPD: one characterized by order and control (perfectionism, excessive devotion to work, overconscientiousness, reluctance to delegate, and rigidity), and
the other by hoarding and indecision (inability to discard and indecisiveness). The evidence includes significant findings from four sources: factor analysis, clinical
correlates, sib–sib correlations and genetic linkage. Also,
these results are congruent with observations by clinical
observers dating back almost a century.[13]
A prototypic individual scoring high on the order and
control dimension is well described in the Alternative
DSM-5 Model of Personality Disorders in the description of the pathological personality trait “rigid perfectionism”: “rigid insistence on everything being flawless,
perfect, and without errors and faults, including one’s
own and other’s performance; sacrificing of timeliness
to ensure correctness in every detail; believing that there
is only one right way to do things; difficulty changing ideas and/or viewpoint; preoccupation with details,
organization, and order.”[4] Such an individual might
also have excessive devotion to work and reluctance to
A prototypic individual scoring high on the hoarding
and indecision dimension would have difficulty discarding objects and difficulty making decisions. In addition,
such an individual might display excessive doubting, be
unsure of his/her self, procrastinate, and have difficulty
completing tasks.
Depression and Anxiety
Riddle et al.
Clinical correlations provide external validation. Both
dimensions are related to certain personality disorders
(schizoid, dependent, avoidant), neuroticism, and certain
anxiety disorders (generalized and social). The first type,
order/control, is significantly related to symmetry obsessions, ordering compulsions, poor insight, disagreeableness, select anxiety disorders (panic disorder, agoraphobia, and specific phobia), anorexia and eating disorders,
and Post-Traumatic Stress Disorder (PTSD). The second type, hoarding/indecision, is significantly associated
with hoarding obsessions, hoarding compulsions, task
incompletion, introversion, low conscientiousness, skin
picking, and other grooming disorders. Although there
are some shared correlates, the two dimensions are quite
different in terms of these associated features.
The significant sib:sib correlation for the hoarding/indecision dimension, but not for the order/control
dimension, suggests that the former may be familial
while the latter may not be. For the hoarding/indecision
factor, linkage to two markers on chromosome 10 suggests a possible genetic vulnerability and locus. The 7.2
Mb region between D10S1221 and D10S1225 on chromosome 10 is interesting. Gene ANK3 (about 4.2 Mb
from D10S1221) in this region has been identified by
several genome-wide association studies of bipolar disorder and schizophrenia, but not OCD or OCPD.[40–42]
Moreover, in addition to the evidence for two dimensions of OCPD and associated clinical correlates,
the linkage signal for OCPD on chromosome 7, marker
d7s1818, was substantially improved when we stratified
for Factor 2 compared to our original (nonstratified)
linkage for OCPD at the same marker, suggesting a possible, more genetically linked, subphenotype.
Factor 1 (order/control) in this study includes five of
the eight symptoms (perfectionism, excessive devotion to
work, overconscientiousness, reluctance to delegate, and
rigidity/stubbornness) in DSM-5’s definition of OCPD.
Factor 2 (hoarding/indecision) includes the core symptom of the new HD (hoarding) plus indecision, which
is listed first in a group of six “other common features.”
Two of the eight symptoms in DSM-5’s OCPD are not
included in either Factor 1 or Factor 2: “preoccupation
with details” loaded in both factors at a level just below
the criterion for inclusion in the factors; “miserly style”
did not load in either factor. These results suggest that in
individuals with OCPD, ascertained through a proband
with OCD, there are two factors that differentiate based
on OCPD features and some other clinical features. One
factor appears to be familial; the other not familial. The
familial factor shows linkage to a genetic marker. Finally, there appears to be some overlap between the two
factors, especially “preoccupation with details.”
A major limitation of this study is that the participants are from a family genetic study of OCD, not from
a representative community sample. Also, other Axis I
Depression and Anxiety
comorbidity and personality features were common in
the sample. So, a major question is: Would this factor
structure of OCPD be derived in other samples? Further
work is needed in independent samples to corroborate
the findings.
Acknowledgments. The authors thank many families participating in the study, the clinicians and coordinators at each site, and the research assistants Erin
Santana, Brittany Gibby, and Rachel Agnello and the
former medical student Delma Postigo at the JHU site.
Conflict of Interest. Drs. Riddle, Maher, Wang,
Grados, Bienvenu, Goes, Cullen, Murphy, Rauch,
Greenberg, Pinto, Piacentini, Pauls, Rasmussen,
Shugart, Nestadt, and Samuels have nothing to disclose.
Dr. Knowles reports grants from NIH and grants from
NARSAD during the conduct of the study.
Dr. McCracken reports grants from NIH during the
conduct of the study; grants from Seaside Therapeutics,
grants and personal fees from Roche, other from Shire,
and personal fees from Dart Neuroscience, outside the
submitted work.
1. Torgersen S, Kringlen E, Cramer V. The prevalence of personality disorders in a community sample. Arch Gen Psychiatry
2. Samuels J, Eaton WW, Bienvenu OJ, et al. Prevalence and correlates of personality disorders in a community sample. Br J Psychiatry 2002;180:536–542.
3. Grant BF, Hasin DS, Stinson FS, et al. Prevalence, correlates,
and disability of personality disorders in the United States: results
from the national epidemiologic survey on alcohol and related
conditions. J Clin Psychiatry 2004;65:948–958.
4. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Washington, DC: American Psychiatric Press; 2013.
5. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 1st ed. Washington, DC: American Psychiatric Press; 1952.
6. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 2nd ed. Washington, DC: American Psychiatric Press; 1968.
7. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. Washington, DC: American Psychiatric Press; 1980.
8. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (Revised). 3rd ed. Washington, DC:
American Psychiatric Press; 1987.
9. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Press; 1994.
10. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed., text revision. Washington, DC:
American Psychiatric Press; 2000.
11. Gordon OM, Salkovskis PM, Oldfield VB, Carter N. The association between obsessive compulsive disorder and obsessive compulsive personality disorder: prevalence and clinical presentation.
Br J Clin Psychol 2013;52:300–315.
Research Article: Obsessive–Compulsive Personality Disorder
12. Samuels JF, Bienvenu OJ, Grados MA, et al. Prevalence and correlates of hoarding behavior in a community-based sample. Behav
Res Ther 2008;46:836–844.
13. Lewis A. Problems of obsessional illness. Proc R Soc Med
14. Hummelen B, Wilberg T, Pedersen G, Karterud S. The quality of
the DSM-IV obsessive-compulsive personality disorder construct
as a prototype category. J Nerv Ment Dis 2008;196:446–455.
15. Pfohl B, Blum N. Obsessive-compulsive personality disorder: a
review of available data and recommendations for DSM-IV. J Personal Disord 1991;5:363–375.
16. Nurnberg GH, Martin GA, Pollack S. An empirical method to
refine personality disorder classification using stepwise logistic regression modeling to develop diagnostic criteria and thresholds.
Compr Psychiatry 1994;35:409–419.
17. Timpano KR, Exner C, Glaesmer H, et al. The epidemiology of
the proposed DSM-5 hoarding disorder: exploration of the acquisition specifier, associated features, and distress. J Clin Psychiatry
18. Nestadt G, Hsu FC, Samuels J, et al. Latent structure of the DSMIV personality disorder criteria. Compr Psychiatry 2006;47:54–62.
19. Livesley WJ, Jackson DJ, Schroeder ML. A study of the factorial
structure of personality pathology. J Personal Disord 1989;3:292–
20. Livesley WJ, Jackson DJ, Schroeder ML. Dimensions of personality pathology. Can J Psychiatry 1991;36:557–562.
21. Nestadt G, Addington A, Samuels J, et al. The identification of
OCD-related subgroups based on comorbidity. Biol Psychiatry
22. Nestadt G, Samuels J, Riddle M, et al. A family study of obsessivecompulsive disorder. Arch Gen Psychiatry 2000;57:358–363.
23. Samuels JF, Bienvenu OJ, 3rd, Pinto A, et al. Hoarding in obsessive-compulsive disorder: results from the OCD
Collaborative Genetics Study. Behav Res Ther 2007;45:673–
24. Samuels J, Bienvenu OJ, 3rd, Riddle MA, et al. Hoarding in obsessive compulsive disorder: results from a case-control study. Behav
Res Ther 2002;40:517–528.
25. Samuels J, Nestadt G, Bienvenu OJ, et al. Personality disorders
and normal personality dimensions in obsessive-compulsive disorder. Br J Psychiatry 2000;177:457–462.
26. Samuels JF, Riddle MA, Greenberg BD, et al. The OCD collaborative genetics study: methods and sample description. Am J Med
Genet B Neuropsychiatr Genet 2006;141:201–207.
27. Mannuzza S, Fyer AJ, Klein DF, Endicott J. Schedule for affective
disorders and schizophrenia—lifetime version modified for the
study of anxiety disorders (SADS-LA): rationale and conceptual
development. J Psychiatr Res 1986;20:317–325.
28. Fyer AJ, Mannuzza S, Gallops MS, et al. Familial transmission of
simple phobias and fears: a preliminary report. Arch Gen Psychiatry 1990;47:252–256.
29. Goodman WK, Price LH, Rasmussen SA, et al. The Yale-Brown
Obsessive Compulsive Scale I. Development, use, and reliability.
Arch Gen Psychiatry 1989;46:1006–1011.
30. Spitzer RL, Williams JB, Gibbon M, First MB. The structured
clinical interview for DSM-III-R (SCID). I: history, rationale, and
description. Arch Gen Psychiatry 1992;49:624–629.
31. Pfohl B, Blum N, Zimmerman M. Structured Interview for DSMIV Personality (SIDP-IV). Iowa City, IA: Department of Psychiatry, University of Iowa; 1995.
32. Costa PT, McCrae RR. Revised NEO Personality Inventory
(NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources; 1992.
33. Shugart YY, Samuels J, Willour L, et al. Genome-wide linkage
scan for obsessive-compulsive disorder: evidence for susceptibility
loci on chromosomes 3q, 7p, 15q, 6q, and 1q. Mol Psychiatry
34. Broman KW, Weber JL. Estimation of pairwise relationships
in the presence of genotyping errors. Am J Hum Genet
35. O’Connell JR, Weeks DE. PedCheck: a program for identification
of genotype incompatibilities in linkage analysis. Am J Hum Genet
36. Douglas JA, Skol AD, Boehnke M. Probability of detection of
genotyping errors and mutations as inheritance inconsistencies in
nuclear-family data. Am J Hum Genet 2002;70:487–495.
37. Abecasis GR, Cherny SS, Cookson WO, Cardon LR. Merlinrapid analysis of dense genetic maps using sparse gene flow trees.
Nat Gen 2002;30:97–101.
38. S.A.G.E. 6.3. Statistical Analysis for Genetic Epidemiology, Release 6.3; 2012. Available at: http://darwin.cwru.edu.
39. Mathew G, Song Y, Elston RC. Interval estimation of familial
correlations from pedigrees. Stat Appl Genet Mol Biol 2011;10:1–
40. Ferreira MA, O’Donovan MC, Meng YA, et al. Collaborative
genome-wide association analysis supports a role for ANK3 and
CACNA1C in bipolar disorder. Nat Gen 2008;40:1056–1058.
41. Athanasiu L, Mattingsdal M, Kähler AK, et al. Gene variants associated with schizophrenia in a Norwegian genome-wide
study are replicated in a large European cohort. J Psychiatr Res
42. Chen DT, Jiang X, Akula N, et al. Genome-wide association study
meta-analysis of European and Asian-ancestry samples identifies
three novel loci associated with bipolar disorder. Mol Psychiatry
Depression and Anxiety