638410

research-article2016
ASMXXX10.1177/1073191116638410AssessmentTolin et al.

Article
Assessment

Psychometric Properties of a Structured

1­–11
© The Author(s) 2016
Reprints and permissions:
Diagnostic Interview for DSM-5 Anxiety, sagepub.com/journalsPermissions.nav
DOI: 10.1177/1073191116638410

Mood, and Obsessive-Compulsive and asm.sagepub.com

Related Disorders

David F. Tolin1,2, Christina Gilliam1, Bethany M. Wootton1,3, William Bowe1,

Laura B. Bragdon1,4, Elizabeth Davis1, Scott E. Hannan1, Shari A. Steinman1,5,6,
Blaise Worden1, and Lauren S. Hallion1

Abstract
Three hundred sixty-two adult patients were administered the Diagnostic Interview for Anxiety, Mood, and OCD and
Related Neuropsychiatric Disorders (DIAMOND). Of these, 121 provided interrater reliability data, and 115 provided
test–retest reliability data. Participants also completed a battery of self-report measures that assess symptoms of anxiety,
mood, and obsessive-compulsive and related disorders. Interrater reliability of DIAMOND anxiety, mood, and obsessive-
compulsive and related diagnoses ranged from very good to excellent. Test–retest reliability of DIAMOND diagnoses
ranged from good to excellent. Convergent validity was established by significant between-group comparisons on applicable
self-report measures for nearly all diagnoses. The results of the present study indicate that the DIAMOND is a promising
semistructured diagnostic interview for DSM-5 disorders.

Keywords
anxiety disorders, mood disorders, obsessive-compulsive and related disorders, interview, diagnosis

Semistructured diagnostic interviews are important for sev- showed acceptable interrater reliability (Lobbestael,
eral applications in psychology and psychiatry. First, when Leurgans, & Arntz, 2011) and test–retest reliability
conducting clinical research trials, researchers must be able (Zanarini et al., 2000); however, to date, no psychometric
to define their sample adequately, and determine whether data have been published for the SCID-5. Additionally,
study participants meet criteria for inclusion or exclusion although the SCID-5 covers a wide variety of disorders, the
diagnoses. Second, in clinical settings, clinicians are often modules are fairly cursory, limiting the extent to which
faced with challenging differential diagnostic cases, and fine-grained (e.g., symptom-specific) analyses can be per-
they must be able to evaluate patients according to clearly formed. The SCID-5 does not, for example, provide detailed
defined diagnostic criteria. Third, training programs fre- guidance about differential diagnoses, allow for examina-
quently use structured diagnostic interviews to teach the tion of the degree of distress and functional impairment
process of diagnostic interviewing and to familiarize train- associated with specific diagnoses, or allow for the system-
ees with diagnostic criteria. atic collection of detailed information about specific symp-
After publication of the DSM-IV (American Psychiatric toms. The Mini International Neuropsychiatric Interview
Association, 1994), several structured diagnostic interviews
were developed, which have become “gold standard” mea-
1
sures for DSM-IV psychiatric disorders. The DSM-5 The Institute of Living, Hartford, CT, USA
2
(American Psychiatric Association, 2013) made several Yale University School of Medicine, New Haven, CT, USA
3
University of Tasmania, Hobart, Tasmania, Australia
important changes from the DSM-IV. There is therefore a 4
Binghamton University, Binghamton, NY, USA
need for structured diagnostic interviews based on the diag- 5
New York State Psychiatric Institute, New York, NY, USA
nostic criteria of the DSM-5, which have now been in place 6
Columbia University Medical Center, New York, NY, USA
for over 2 years. The American Psychiatric Association
Corresponding Author:
recently released their DSM-5 version of the Structured David F. Tolin, The Institute of Living, 200 Retreat Avenue, Hartford,
Clinical Interview for DSM (SCID-5; American Psychiatric CT 06106, USA.
Association, 2015). The SCID-5’s predecessor, the SCID-IV, Email david.tolin@hhchealth.org

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

2 Assessment 

for DSM-5 (MINI-5; Sheehan, 2015), a briefer alternative validity of syndromal models of psychopathology in gen-
to the SCID-5, covers a narrower range of disorders. eral, and of the DSM-5 in particular (e.g., Brown & Barlow,
Importantly for assessment of the anxiety, mood, and obses- 2009; Cuthbert, 2014; Kendell & Jablensky, 2003; Krueger
sive-compulsive and related disorders, the MINI-5 does not & Bezdjian, 2009; Krueger, Markon, Patrick, & Iacono,
assess specific phobia (SpP), separation anxiety disorder 2005). To the extent that the “validity” of a DSM-5 inter-
(SAD), persistent depressive disorder/dysthymia (PDD), view can be examined, such examination is limited at pres-
cyclothymic disorder (CYC), body dysmorphic disorder ent to understanding the interview’s fidelity to the syndromal
(BDD), hoarding disorder (HD), trichotillomania (TTM), or model implied by the DSM-5, rather than the validity of the
excoriation (skin-picking) disorder (EXD). Thus, it is of model itself. Data from the DSM-5 field trials are surpris-
limited utility for clinicians and researchers wishing to ingly sparse, and results for only a handful of diagnoses
assess the range of these conditions. Furthermore, although have been published to date. In these studies, field trial
the MINI-5 is based on the MINI-IV, which shows good diagnoses were obtained using a checklist of DSM-5 diag-
interrater and test–retest reliability, and fair to good conver- nostic criteria (Clarke et al., 2013). Within the mood disor-
gence with other diagnostic interviews (Lecrubier et al., ders, test–retest reliability for bipolar I disorder (BP1) was
1997; Sheehan et al., 1998), to date, no psychometric data good (κ = .56), although the reliability for major depressive
have been published for the MINI-5. disorder (MDD) was questionable (κ = .28). Within the
In addition to the SCID-5 and MINI-5, more circum- anxiety disorders, generalized anxiety disorder (GAD)
scribed structured interviews have been developed. The showed a questionable test–retest reliability of κ = .20
Anxiety and Related Disorders Interview Schedule for (Regier et al., 2013). The investigators were unable to
DSM-5 (ADIS-5; Brown & Barlow, 2013) is closely mod- obtain accurate estimates of κ (defined as a standard error of
eled on the ADIS-IV, which demonstrates fairly strong ≤0.1 and a 95% confidence interval [CI] of ≤0.5) for bipolar
interrater reliability for most disorders assessed (Brown, Di II disorder (BP2) or HD. At present no test–retest data have
Nardo, Lehman, & Campbell, 2001). However, psychomet- been published from the field trials on any other anxiety,
ric data for the ADIS-5 have not yet been published. The mood, or obsessive-compulsive and related disorders.
primary disadvantage of the ADIS-5 may be its long dura- Similarly, no interrater reliability data have been published
tion (administration time of the ADIS-IV is 2 to 4 hours for for any DSM-5 diagnosis. Nevertheless, despite its limita-
the lifetime version in clinical samples (Summerfeldt, tions, the DSM-5 remains the most commonly used diag-
Kloosterman, & Antony, 2010), which may be prohibitive nostic system for research, treatment, and clinical training
in many clinical and research settings). Additionally, the in the United States (Tyrer, 2014), and the categorical model
ADIS-5 does not include modules to assess several obses- of psychopathology remains the dominant scheme in health
sive-compulsive and related disorders (HD, TTM, or EXD); care (First, 2005). Thus, empirically validated semistruc-
schizophrenia spectrum disorders that are fairly common tured interviews that correspond to the DSM-5 are needed.
and often co-occur with anxiety, mood, or obsessive- The Diagnostic Interview for Anxiety, Mood, and OCD
compulsive and related disorders; or other problems such as and Related Neuropsychiatric Disorders (DIAMOND) is a
eating disorders or attention-deficit/hyperactivity disorder new semistructured interview that targets the diagnostic cri-
(ADHD). The Alcohol Use Disorder and Associated teria for a range of DSM-5 disorders, with additional clini-
Disabilities Interview Schedule-DSM-5 Version cal information gathered for the anxiety, mood, and
(AUDADIS-5; Hasin et al., 2015) provides a comprehen- obsessive-compulsive and related disorders. The interview
sive assessment of substance use disorders (SUD) and has was developed in several stages. First, the specific symp-
been validated in the general population. However, there toms for each of the assessed DSM-5 disorders were trans-
are considerable disadvantages that limit its utility in clini- lated into question form, with supplemental behavioral
cal and clinical research settings. First, although the fully observations by the clinician. Second, a panel of expert cli-
structured nature of the interview is advantageous in epide- nicians reviewed the items and suggested typical and atypi-
miological settings, it may be overly restrictive for experi- cal symptom presentations, initial questions, and follow-up
enced clinicians and diagnosticians (e.g., it does not allow questions. Third, the measure was subjected to initial feasi-
for clinical judgment in using follow-up questions). Second, bility testing, with iterative feedback from users at various
interrater reliability is only fair for mood, anxiety, and stages of professional development.
trauma-related disorders. Third, the AUDADIS-5 does not The aim of the present study was to examine the reliabil-
include modules for a variety of common psychological ity and validity (i.e., fidelity to the DSM-5 model and struc-
disorders, including obsessive-compulsive and related dis- ture) of the DIAMOND in a clinical setting. The primary
orders, schizophrenia spectrum disorders, eating disorders, hypotheses were the following: (1) The DIAMOND would
or ADHD. show adequate interrater reliability for the anxiety, mood,
A diagnostic interview can only be as reliable and valid and obsessive-compulsive and related disorders; (2) The
as are the diagnoses themselves. Many have questioned the DIAMOND would show adequate test–retest reliability for

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

Tolin et al. 3

those disorders; and (3) the DIAMOND would show ade- that queries the DSM-5 diagnostic criteria for the anxiety
quate convergent validity for those disorders. In addition, in disorders, bipolar disorders, depressive disorders, obses-
order to determine feasibility of use, we calculated the sive-compulsive and related disorders, trauma- and stressor-
amount of time required to complete the DIAMOND and related disorders, schizophrenia spectrum and other
the amount of time required per diagnosis assigned. We also psychotic disorders, feeding and eating disorders, somatic
examined the performance of the DIAMOND when used by symptom and related disorders, substance-related and
licensed, doctoral-level clinical psychologists versus addictive disorders, and neurodevelopmental disorders. The
trainees. primary focus of the DIAMOND is the anxiety, mood, and
obsessive-compulsive and related disorders, and the diag-
nostic criteria for these disorders are supplemented with
Method
more clinically relevant questions, such as symptom dimen-
Participants sions, as well as information about common differential
diagnoses. The other disorders were included in the DIA-
The DIAMOND was administered as part of a routine MOND because accurate diagnosis of an anxiety, mood, or
intake to consecutive adult (age 18 years and older) English- obsessive-compulsive and related disorder requires the
speaking patients, most of whom were seeking treatment interviewer to carefully rule out alternative diagnoses (e.g.,
(78%) or enrollment in a clinical trial (15%) at a hospital- the diagnosis of BDD may require ascertaining that the
based outpatient clinic for anxiety, mood, and obsessive- patient’s symptoms are not limited to weight or shape con-
compulsive and related disorders. Following the initial cerns secondary to an ED). A suicide screen is also included
interview, participants were then recruited to participate in that queries suicidal ideation, intent, plan, means, behav-
the interrater and test–retest reliability arm of the study. A iors, and protective factors.1
small number of patients (7%) were recruited from other Wording of questions was altered from the DSM-5 crite-
outpatient programs within the hospital (e.g., partial hospi- ria, and the questions were ordered to minimize clinician and
talization programs for serious mental illness) or from other patient burden. Specifically, the following format was used:
sites (e.g., referrals from other clinics, advertisements on
informational web pages). No participants were excluded 1.  Initial questions: Preliminary questions allow the
due to comorbid psychiatric conditions, severity of illness, interviewer to obtain an overview of the presenting
clarity of diagnosis, or complexity of illness. problem prior to inquiring about specific diagnoses.
As shown in Table 1, 362 individuals received the These include the following:
DIAMOND. Of these, 121 provided interrater reliability
data, and 115 provided test–retest reliability data. Mean age a. “Can you describe what kind of problem or
of the samples ranged from 38 to 39 years, and the samples problems you are here to discuss?”
included slightly more women than men. Of the diagnoses b. “How is your physical health? Do you have any
assigned during the initial DIAMOND administration, the significant medical conditions?”
most common mood disorder was MDD. The most com- c. “What medications do you currently take?”
mon anxiety disorders were social phobia (SoP) and GAD. d. “Have you had mental health treatment before?
The most common obsessive-compulsive and related disor- If so, can you describe it? When did it occur?”
ders were obsessive-compulsive disorder (OCD) and HD. e. “Have you ever been hospitalized for psychiat-
Of note, some mood, anxiety, and obsessive-compulsive ric reasons before? If so, can you describe it?
and related disorders were underrepresented in the present Where and when were you hospitalized?”
study, particularly adult SAD, BP2, and CYC. Therefore, f. “Does anyone in your family have a history of
reliability and validity analyses were not conducted on mental health problems? What kind of
these diagnoses. For sample description purposes, Table 1 problems?”
also shows rates of posttraumatic stress disorder, illness g. “Have you been having any thoughts about
anxiety disorder, SUD, ADHD, schizophrenia spectrum, hurting or killing yourself?”
eating disorders, and tic disorders. Of these, only SUD and
ADHD had enough diagnosed participants for further 2.  Symptom questions: The specific DSM-5 symptoms
analysis. were listed and queried (a self-report screening
form, consisting of the symptom questions, was
developed to facilitate this process, although inter-
Measures
viewers were allowed to probe any symptoms men-
Diagnostic Interview for Anxiety, Mood, and Obsessive- tioned during the interview, regardless of the
Compulsive and Related Neuropsychiatric Disorders.  The DIA- participant’s response on the screening form).
MOND (Tolin et al., 2013) is a structured clinical interview Symptoms were queried over the past month, except

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

4 Assessment 

Table 1.  Sample Description. when a longer duration is needed for diagnostic pur-
Test–retest
poses (e.g., a lifetime history of manic episodes for
Total sample Interrater reliability reliability a bipolar disorder diagnosis). For example, in the
(N = 362); subsample subsample case of SoP, the interviewer asked the following
Characteristic n (%) (N = 121); n (%) (N = 115); n (%) questions, required to satisfy Criterion A, “Marked
Demographics   fear or anxiety about one or more social situations in
Age in years, 38.69 (14.79) 37.62 (14.27) 38.28 (14.62) which the individual is exposed to possible scrutiny
M (SD) by others:”
Female 213 (58.2) 72 (59.5) 68 (59.1)
Non-White 30 (9.4) 13 (12.2) 9 (8.6)
a. “In the past month, do you feel very afraid or
Intake diagnoses  
Anxiety   anxious in any social situations, because you
disorders are worried that others will judge you nega-
 SoP 104 (28.7) 36 (29.8) 31 (30.1) tively, or that you will embarrass yourself?”
 PD 60 (16.6) 19 (15.7) 13 (12.6) b. “In the past month, do you feel very afraid or
 AGO 37 (10.2) 14 (11.6) 11 (10.7) anxious in situations where other people might
 GAD 97 (26.8) 27 (22.3) 23 (22.3)
observe you?”
 SpP 32 (8.8) 11 (9.9) 9 (8.7)
 SAD 4 (1.1) 1 (0.8) 1 (1.0)
 Any 225 (62.2) 73 (60.3) 60 (58.3) 3.  Clarifying questions: For additional clarity, when
Bipolar disorders   appropriate, the interviewer asked questions about
 BP1 13 (3.6) 12 (9.9) 12 (11.7) specific symptom dimensions. For example, after
 BP2 4 (1.12) 2 (1.7) 2 (1.9) the aforementioned symptom questions for SoP, the
 CYC 2 (0.6) 2 (1.7) 1 (1.0) interviewer asked, “What kind of situations are you
 Any 19 (5.2) 16 (13.2) 15 (14.6)
afraid of?” A checklist was then provided, allowing
Depressive  
disorders the interviewer to check dimensions, such as public
 PDD 52 (14.4) 12 (9.9) 9 (8.7) speaking; starting or maintaining conversations;
 MDD 123 (34.0) 41 (33.9) 36 (35.0) meeting people you don’t know well; talking to
 PMDD 9 (2.5) 6 (5.0) 6 (5.8) authority figures; asserting yourself; being watched
 Any 177 (48.9) 55 (45.5) 47 (45.6) while working or performing; eating, writing, or
Obsessive-  
performing other activities in public; using public
compulsive
and related restrooms; and others.
disorders 4.  Distress and impairment questions: After the symp-
 OCD 88 (24.3) 14 (11.6) 10 (9.7) toms had been queried, the interviewer asked about
 BDD 24 (6.6) 12 (9.9) 10 (9.7) symptom-related distress and impairment. The dis-
 HD 60 (16.6) 20 (17.4) 20 (19.4)
tress and impairment questions were used at this
 TTM 14 (3.9) 11 (9.1) 10 (9.7)
 EXD 19 (5.2) 9 (7.4) 8 (7.8)
point with the aim of ruling out subclinical symp-
 Any 178 (49.2) 55 (45.5) 49 (47.6) toms as efficiently as possible. For example,
Other disorders  
 PTSD 15 (4.1) 4 (3.3) 3 (2.9) a. “How much does this problem bother or dis-
 IAD 10 (2.8) 4 (3.3) 4 (3.9) tress you?” This was followed by questions
 SUD 50 (13.8) 19 (15.7) 16 (15.2) about the frequency, duration, and intensity of
 ADHD 28 (7.7) 11 (9.1) 9 (8.7)
distress.
  Any SSD 4 (1.1) 3 (2.5) 3 (2.9)
  Any ED 14 (3.9) 5 (4.1) 5 (4.9)
b. “In the past month, does this fear or avoid-
  Any tic 5 (1.4) 2 (1.7) 1 (1.0) ance impair your ability to function, like at
school or work, in your social life, in your
Note. SoP = social phobia (social anxiety disorder); PD = panic disorder; AGO = family, or in your ability to do things that are
agoraphobia; GAD = generalized anxiety disorder; SpP = specific phobia; SAD =
separation anxiety disorder; PDD = persistent depressive disorder (dysthymia); important to you? How?” This was followed
BP1 = bipolar I disorder; BP2 = bipolar II disorder; MDD = major depressive by a checklist of functional impairment
disorder; CYC = cyclothymia; PMDD = premenstrual dysphoric disorder; OCD =
obsessive-compulsive disorder; BDD = body dysmorphic disorder; HD = hoarding
domains: school, work or role functioning,
disorder; TTM = trichotillomania; EXD = excoriation (skin-picking) disorder; social life, family, home responsibilities, lei-
PTSD = posttraumatic stress disorder; IAD = illness anxiety disorder; SUD = sure activities, legal problems, financial
substance use disorder; ADHD = attention-deficit/hyperactivity disorder; OCRD
= obsessive-compulsive and related disorder; SSD = schizophrenia spectrum problems, problems of health or safety, or
disorder; ED = eating disorder. other functional impairment.

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

Tolin et al. 5

5.  Other questions: Other questions that qualified the For depressive and bipolar disorders, we used the depres-
symptoms, but did not necessarily indicate their sion subscale of the DASS (α = .93) and the Mood Disorder
presence or absence, were then asked. For example, Questionnaire (MDQ; Hirschfeld et al., 2000), for which
in the case of SoP, the interviewer then asked ques- we assigned a value of 1 to each “yes” response and calcu-
tions that addressed DSM-5 Criteria E (the fear or lated a sum score for each participant (α = .92).
anxiety is out of proportion to the actual threat posed For obsessive-compulsive and related disorders, we used
by the social situation and to the sociocultural con- the Obsessive-Compulsive Inventory–Revised (OCI-R; Foa
text), F (the fear, anxiety, or avoidance is persistent, et al., 2002), for which we calculated a total score based on
typically lasting for 6 months or more), and H (the the nonhoarding items (α = .91) as well as a total that
fear, anxiety, or avoidance is not attributable to the included all items (α = .90); the Body Dysmorphic Disorder
physiological effects of a substance or another med- Questionnaire (Phillips, 1996), for which we assigned a
ical condition). value of 1 to each “yes” response and calculated a sum
6.  Clinical judgment ratings: Certain items did not score for each participant (α = .80); the Saving Inventory–
require a specific question but rather were to be Revised (Frost, Steketee, & Grisham, 2004), for which we
rated by the interviewer based on all of the available used the total score (α = .97); The MGH Hair-Pulling Scale
information. In the case of SoP, these included (Keuthen et al., 1995; α = .97); and the Skin Picking Scale
Criteria I (the fear, anxiety, or avoidance is not bet- (Keuthen et al., 2001; α = .94).
ter explained by the symptoms of another mental Outside of the Anxiety, Mood, and OCD and Related
disorder) and J (if another medical condition is pres- Disorders, participants also completed the Drug Abuse
ent, the fear, anxiety, or avoidance is clearly unre- Screening Test (Skinner, 1982; α = .84), and the ADHD
lated or is excessive). Symptom Scale (ADHDSS; Barkley & Murphy, 1998), for
7.  Information about differential diagnoses, associated which we used the total score (α = .93).
features, and specifiers: To improve diagnostic
accuracy among both experienced and novice inter-
Procedure
viewers, for each anxiety, mood, and obsessive-
compulsive and related disorder, the DIAMOND Participants (N = 362) were initially assessed using the
provided information about common differential DIAMOND during an intake for clinical treatment or a clini-
diagnoses. For example, in the case of SoP, the cal trial. Interviews were administered by graduate students in
DIAMOND provides information (adapted from the clinical psychology (182 interviews), predoctoral psychology
DSM-5) about distinguishing SoP from normative interns (23 interviews), postdoctoral fellows (59 interviews),
shyness, BDD, agoraphobia (AGO), psychotic dis- or licensed psychologists (98 interviews). All interviewers
orders, panic disorder (PD), autism spectrum disor- received detailed instructions on how to administer the
der, GAD, avoidant personality disorder, SAD, DIAMOND, including watching a senior clinician administer
OCD, SpP, eating disorders, MDD, and other medi- portions of the interview on video. Within 1 week prior to the
cal conditions. Associated features, also adapted interview, participants completed the DASS and provided
from DSM-5, were listed, including inadequate basic demographic information using Research Electronic
assertion; delayed leaving the home; rigid body pos- Data Capture (REDCap) tools (Harris et al., 2009).
ture, poor eye contact, or overly soft voice; self- After the initial interview, participants were invited to
medication with substances; shy, withdrawn, or participate in the second phase of the study. Those agreeing
non–self-disclosing; blushing; seeking jobs or roles to participate signed informed consent for research and
that require little social interaction; and exacerba- were scheduled to be interviewed again within 48 hours
tion of medical issues when anxious. The possible (M = 0.93 days, SD = 1.57) by a second interviewer (inter-
specifier for SoP was performance only. rater reliability), and a third time 1 week later (M = 9.04
days, SD = 3.01) by one of the previous two interviewers
Validity Measures.  Participants also completed a battery of (test–retest reliability). Participants could opt to complete
online self-report measures. For anxiety disorders, we used both the interrater and test–retest interviews, or only one of
the Liebowitz Social Anxiety Scale (Heimberg et al., 1999), these (the large majority completed both interviews; those
for which we used the total score (α = .97); the Panic Disor- who completed only one interview did so because of inabil-
der Severity Scale–Self-Report (Shear et al., 1997; α = .94); ity to schedule the interview within the necessary time
the Mobility Inventory (Chambless, Caputo, Jasin, Gracely, period). They were also asked to complete the battery of
& Williams, 1985), for which we used the total score (α = additional validity measures using REDCap. A total of 121
.96); the Penn State Worry Questionnaire (Meyer, Miller, participants completed the interrater reliability interview,
Metzger, & Borkovec, 1990; α = .95); and the Depression and 115 completed the test–retest reliability interview.
Anxiety Stress Scales (DASS; Lovibond & Lovibond, Participants were reimbursed $20 for each of the reliability
1995), for which we used the anxiety (α = .85) subscale. interviews.

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

6 Assessment 

Data Analytic Plan. Interrater reliability was determined by Table 2.  Interrater Reliability for DIAMOND Diagnoses.
calculating κ coefficients for each diagnosis (present or Diagnosis κ 95% CI t Interpretation
absent) between Rater 1 and Rater 2 at Time 1. Test–retest
a
reliability was determined by calculating κ coefficients for Anxiety disorders SoP .70 0.55-0.84 7.67** Very good
  PD .88 0.78-0.99a 9.80** Excellent
each diagnosis (present or absent) between Time 1 and Time
  AGO .87 0.72-1.01a 9.62** Excellent
2 for the same interviewer. Following recommendations for   GAD .71 0.55-0.86a 7.78** Very good
the DSM-5 field trials (Clarke et al., 2013; Kraemer, Kupfer,   SpP .66 0.43-0.88 7.24** Very good
Clarke, Narrow, & Regier, 2012), κ coefficients of .80 and   Any .73 0.60-0.85a 7.99** Very good
greater are considered “excellent,” from .60 to .79 “very Bipolar disorders BP1 1.00 —a 11.00** Excellent
good,” from .40 to .59 “good,” from .20 to .39 “question-   Any .88 0.75-1.01a 9.78** Excellent
able,” and less than .20 “unacceptable.” We further examined Depressive PDD .65 0.44-0.86 7.23** Very good
disorders
the extent to which each κ coefficient could be considered
  MDD .62 0.47-0.77a 6.84** Very good
statistically precise. When the standard error of κ is high   PMDD .82 0.59-1.06 9.07** Excellent
(resulting in a wide confidence interval), even if the value of   Any .68 0.55-0.81a 7.51** Very good
κ is high, the true κ cannot be estimated with precision. We Obsessive- OCD .62 0.40-0.84 6.84** Very good
therefore used the guidelines from the DSM-5 field trials compulsive and
related disorders
(Clarke et al., 2013) to define a statistically precise estimate
  BDD .95 0.86-1.05a 10.48** Excellent
of κ as those with a standard error of ≤0.1 and a 95% CI ≤0.5.   HD .86 0.75-0.98a 9.53** Excellent
For validity estimates, we used between-groups t tests and   TTM 1.00 —a 11.00** Excellent
between-group effect size estimates (Cohen’s d), using the   EXD .78 0.59-0.99 8.66** Very good
presence and absence of each anxiety, mood, and obsessive-   Any .90 0.82-0.98a 9.93** Excellent
compulsive and related disorder as the independent variables,   SUD .65 0.47-0.82a 7.92** Very good
and scores on the corresponding self-report measures as the   ADHD .60 0.33-0.87 6.71** Very good
dependent variables. To further examine the convergence and Note. DIAMOND = Diagnostic Interview for Anxiety, Mood, and OCD and
divergence of dimensions of psychopathology according to Related Neuropsychiatric Disorders; SoP = social phobia (social anxiety disorder);
DIAMOND diagnoses, we conducted a series of logistic PD = panic disorder; AGO = agoraphobia; GAD = generalized anxiety disorder;
SpP = specific phobia; BP1 = bipolar I disorder; PDD = persistent depressive
regression analyses, initially using a presumed divergent con- disorder (dysthymia); MDD = major depressive disorder; PMDD = premenstrual
struct in the first block and a presumed convergent construct dysphoric disorder; OCD = obsessive-compulsive disorder; BDD = body
dysmorphic disorder; HD = hoarding disorder; TTM = trichotillomania; EXD
in the second and then reversing the order. = excoriation (skin-picking) disorder; SUD = substance use disorder; ADHD =
attention-deficit/hyperactivity disorder.
a
Standard error ≤ 0.1 and 95% CI ≤ 0.5, indicating an acceptable estimate of κ.
Results *p < .05. **p < .001.

Administration Time ranged from very good (κ = .62) to excellent (κ = 1.00),

Mean administration time for the initial DIAMOND was according to interpretive cutoffs used in the DSM-5 field
64.53 minutes (SD = 25.82). The average participant trials (Clarke et al., 2013; Kraemer et al., 2012). For DSM-5
received 2.45 diagnoses (SD = 1.43, range 0-8). Therefore, categories, interrater reliability for any obsessive-compul-
mean administration time (excluding 5 participants who sive and related disorder or any bipolar disorder was excel-
received no diagnosis) was 32.84 minutes (SD = 20.08) per lent. Interrater reliability for any anxiety disorder or any
diagnosis assigned. depressive disorder was very good.
Interviewers for the initial DIAMOND were categorized The κ coefficients for SoP, PD, AGO, GAD, any anxiety
as licensed psychologists or trainees. Psychologists and disorder, BP1, any bipolar disorder, MDD, any depressive
trainees did not differ in terms of the mean time to adminis- disorder, BDD, HD, TTM, any obsessive-compulsive and
ter the DIAMOND (psychologists M = 67.76 minutes, related disorder, and SUD had both a standard error ≤ 0.1
SD = 26.82; trainees M = 63.38 minutes, SD = 25.41, t = and a 95% CI ≤ 0.5, suggesting a precise estimate using the
1.40, p = .16, d = 0.17) or in terms of the time per diagnosis definition from the field trial (Clarke et al., 2013). Precise
assigned (psychologists M = 31.62 minutes, SD = 19.41; estimates of interrater reliability could not be obtained for
trainees M = 33.26 minutes, SD = 20.33, t = 0.66, p = .51, d SpP, PDD, premenstrual dysphoric disorder (PMDD),
= 0.08). OCD, EXD, or ADHD. We compared κ values for interrater
pairings in which (a) both interviewers were licensed psy-
chologists, (b) one interviewer was a psychologist and one
Interrater Reliability a trainee, or (c) both interviewers were trainees. Mean κ
Table 2 shows interrater reliability coefficients for the values across the 14 mood, anxiety, and obsessive-compul-
DIAMOND diagnoses. For all diagnoses, κ coefficients sive and related disorders were .78 (SD = .20) for two

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

Tolin et al. 7

psychologists, .77 (.20) for one psychologist and one Table 3.  Test–Retest Reliability for DIAMOND Diagnoses.
trainee, and .79 (.19) for two trainees.
Diagnosis κ 95% CI t Interpretation
Anxiety  
Test–Retest Reliability disorders
Table 3 shows test–retest reliability coefficients for the  SoP .86 0.76-0.96a 9.21** Excellent
DIAMOND diagnoses (presence vs. absence) as well as the  PD .96 0.89-1.04a 10.33** Excellent
severity ratings for those diagnoses assigned. For all diag-  AGO .90 0.76-1.04a 9.65** Excellent
noses, κ coefficients ranged from good (κ = .59) to excellent  GAD .68 0.52-0.85a 7.35** Very good
 SpP .78 0.57-0.99 8.38** Very good
(κ = 1.00), according to cutoffs used in the DSM-5 field
 Any .75 0.63-0.87a 8.07** Very good
trials (Clarke et al., 2013; Kraemer et al., 2012). For DSM-5
Bipolar disorders  
categories, test–retest reliability for any obsessive-compul-
 BP1 .95 0.86-1.05a 10.22** Excellent
sive and related disorder or any bipolar disorder was excel-
 Any .92 0.81-1.03a 9.88** Excellent
lent. Interrater reliability for any anxiety disorder or any Depressive  
depressive disorder was very good. disorders
The κ coefficients for SoP, PD, AGO, GAD, any anxiety  PDD .59 0.37-0.82 6.46** Good
disorder, BP1, any bipolar disorder, MDD, any depressive  MDD .72 0.59-0.86a 7.78** Very good
disorder, OCD, BDD, HD, TTM, EXD, any obsessive-  PMDD .82 0.58-1.06 8.84** Excellent
compulsive and related disorder, and SUD had both a stan-  Any .76 0.64-0.88a 8.11** Very good
dard error ≤0.1 and a 95% CI ≤0.5, suggesting a precise Obsessive-  
estimate. Precise estimates of test–retest reliability could compulsive and
not be obtained for SpP, PDD, PMDD, or ADHD. related disorders
We compared κ values for test–retest reliability when the  OCD .83 0.69-0.98a 8.92** Excellent
interviewer was (a) a licensed psychologist or (b) a trainee.  BDD 1.00 —a 10.72** Excellent
Mean κ values across the 14 mood, anxiety, and obsessive-  HD .94 0.87-1.02a 10.12** Excellent
compulsive and related disorders were .84 (SD = .16) for  TTM 1.00 —a 10.72** Excellent
 EXD .94 0.83-1.05a 10.12** Excellent
psychologists and .87 (.12) for trainees.
 Any .95 0.89-1.01a 10.17** Excellent
 SUD .76 0.62-0.91a 9.12** Very good
Convergent Validity  ADHD .68 0.42-0.94 7.46** Very good

Table 4 shows self-report measure scores for participants Note. DIAMOND = Diagnostic Interview for Anxiety, Mood, and OCD
with and without the corresponding diagnosis (based on the and Related Neuropsychiatric Disorders; SoP = social phobia (social
anxiety disorder); PD = panic disorder; AGO = agoraphobia; GAD
initial interview). Between-group t tests were significant for
= generalized anxiety disorder; SpP = specific phobia; BP1 = bipolar I
all diagnoses with the exception of participants with and disorder; PDD = persistent depressive disorder (dysthymia); MDD =
without a diagnosis of PMDD, who did not differ on the major depressive disorder; PMDD = premenstrual dysphoric disorder;
depression subscale of the DASS, likely due to the fact that OCD = obsessive-compulsive disorder; BDD = body dysmorphic
disorder; HD = hoarding disorder; TTM = trichotillomania; EXD =
most participants diagnosed with PMDD were not in the excoriation (skin-picking) disorder; SUD = substance use disorder;
week prior to menstruation at the time of the interview. ADHD = attention-deficit/hyperactivity disorder.
Effect size estimates (Cohen’s d) for disorders other than a
Standard error ≤ 0.1 and 95% CI ≤ 0.5, indicating an acceptable estimate
PMDD ranged from moderate (for participants with vs. of κ.
*p < .05. **p < .001.
without MDD on the depression subscale of the DASS) to
very large (HD, TTM, EXD, and BP1).
Table 5 shows the results of logistic regressions to exam- the prediction of a depressive diagnosis over and above
ine convergent and divergent validity. For anxiety disorders, DASS-Anx. In the second regression, DASS-Anx did not
we selected the anxiety subscale of the DASS (DASS-Anx) contribute to the prediction of a depressive diagnosis over
as the convergent measure, and the depression subscale of and above DASS-Dep. For bipolar disorders, we selected
the DASS (DASS-Dep) as the divergent measure. In the first MDQ as the convergent measure and DASS-Anx as the
regression, DASS-Anx contributed to the prediction of an divergent measure. In the first regression, MDQ contributed
anxiety diagnosis over and above DASS-Dep. In the second to the prediction of a bipolar diagnosis over and above
regression, however, DASS-Dep did not contribute to the DASS-Anx. In the second regression, DASS-Anx contrib-
prediction of an anxiety diagnosis over and above DASS- uted to the prediction of a bipolar diagnosis over and above
Anx. For depressive disorders, we selected DASS-Dep as DASS-Dep, though to a lesser degree. For obsessive-
the convergent measure and DASS-Anx as the divergent compulsive and related disorders, we selected OCI-R as the
measure. In the first regression, DASS-Dep contributed to convergent measure and MDQ as the divergent measure. In

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

8 Assessment 

Table 4.  Comparisons on Self-Report Measures for Participants With or Without Specific DIAMOND Diagnoses.

Diagnosis Measure With diagnosis Without diagnosis t d

Anxiety disorders  
 SoP LSAS 63.63 (25.46) 33.07 (24.59) 5.78** 1.22
 PD PDSS 1.39 (0.99) 0.51 (0.73) 4.42** 1.01
 AGO MI 4.38 (1.63) 2.98 (1.26) 3.28* 0.97
 GAD PSWQ 68.91 (8.55) 48.99 (15.94) 5.63** 1.56
 Any DASS-Anx 12.19 (9.70) 7.11 (7.32) 4.42** 0.59
Bipolar disorders  
 BP1 MDQ 11.91 (1.14) 2.73 (3.06) 9.82** 3.97
 Any MDQ 11.67 (1.37) 2.67 (3.01) 10.19** 3.85
Depressive disorders  
 PDD DASS-Dep 22.42 (10.05) 12.56 (11.23) 5.08** 0.93
 MDD DASS-Dep 18.15 (12.64) 12.09 (10.58) 4.04** 0.52
 PMDD DASS-Dep 14.57 (9.29) 13.98 (11.66) 0.13 0.06
 Any DASS-Dep 19.30 (12.13) 9.48 (8.92) 7.52** 0.92
Obsessive-compulsive and related disorders  
 OCD OCI-Ra 24.05 (12.89) 7.26 (6.92) 5.23** 1.62
 BDD BDDQ 1.67 (2.66) 0.41 (1.39) 2.02* 0.59
 HD SI-R 58.76 (15.38) 12.85 (12.37) 14.64** 3.29
 TTM MGH-HPS 16.73 (3.23) 0.82 (3.40) 14.81** 4.80
 EXD SPS 12.89 (7.41) 0.97 (2.24) 11.70** 2.18
 Any OCI-Rb 17.60 (12.67) 7.73 (7.51) 5.02** 0.95
 SUD DAST 4.71 (4.34) 2.03 (0.98) 5.28** 0.85
 ADHD ADHDSS 28.40 (13.43) 11.27 (8.85) 5.53** 1.51

Note. DIAMOND = diagnostic interview for anxiety, mood, and OCD and related neuropsychiatric disorders; SoP = social phobia (social anxiety disorder);
PD = panic disorder; AGO = agoraphobia; GAD = generalized anxiety disorder; BP1 = bipolar I disorder; PDD = persistent depressive disorder
(dysthymia); MDD = major depressive disorder; PMDD = premenstrual dysphoric disorder; OCD = obsessive-compulsive disorder; BDD = body
dysmorphic disorder; HD = hoarding disorder; TTM = trichotillomania; EXD = excoriation (skin-picking) disorder; SUD = substance use disorder;
ADHD = attention-deficit/hyperactivity disorder; LSAS = Liebowitz Social Anxiety Scale; PDSS = Panic Disorder Severity Scale; MI = Mobility Inventory;
PSWQ = Penn State Worry Questionnaire; DASS-Dep = depression subscale of the Depression Anxiety Stress Scales; DASS-Anx = anxiety subscale
of the Depression Anxiety Stress Scales; MDQ = Mood Disorder Questionnaire; OCI-R = Obsessive-Compulsive Inventory–Revised; BDDQ = Body
Dysmorphic Disorder Questionnaire; SI-R = Saving Inventory–Revised; MGH-HPS = Massachusetts General Hospital Hair Pulling Scale; SPS = Skin-Picking
Scale; DAST = Drug Abuse Screening Test; ADHDSS = ADHD Symptom Scale.
a
The three hoarding items were omitted from analysis of the OCI-R. bIncluding the three hoarding items.
*p < .05. **p < .001.

the first regression, OCI-R contributed to the prediction of a criteria from the DSM-5 field trials (Clarke et al., 2013;
depressive diagnosis over and above MDQ. In the second Kraemer et al., 2012). The test–retest reliability of the
regression, MDQ did not contribute to the prediction of a DIAMOND, OCD, BDD, TTM, EXD, SoP, PD, and AGO
depressive diagnosis over and above OCI-R. A table of cor- diagnoses all are very good to excellent. That the same par-
relations between all diagnoses and all measures can be ticipant and interviewer would produce similar interview
found in the supplemental material. results across two time points is perhaps not surprising;
however, we note that test–retest reliability is the only reli-
ability estimate published from the DSM-5 field trials, and
Discussion
those coefficients ranged from questionable to good for the
The aim of this study was to provide a preliminary analysis anxiety and mood disorders (Regier et  al., 2013).
of the psychometric properties of a new semistructured Furthermore, in the present study, very good to excellent
clinical interview for DSM-5 anxiety, mood, and obsessive- interrater reliability, an important statistic not reported in
compulsive and related disorders. Administration time of the field trial, was found with precise estimates for BDD,
the DIAMOND is approximately 1 hour (approximately 30 HD, TTM, SoP, PD, AGO, GAD, BP1, and MDD
minutes for each assigned diagnosis), making it feasible for diagnoses.
use in research, clinical, and training settings. DIAMOND Using the criteria of a standard error of ≤0.1 and a 95%
diagnoses show very good to excellent interrater reliability, CI ≤0.5 for a statistically precise estimate of κ, we were
and good to excellent test–retest reliability using cutoff unable to obtain precise reliability estimates for SpP, PDD,

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

Tolin et al. 9

Table 5.  Logistic Regressions Predicting DIAMOND Diagnoses additional work is needed to more fully understand the reli-
From Convergent and Divergent Measures. ability, validity, and utility of each approach.
DIAMOND diagnostic Model It is worth noting as well that at present, there is no “gold
category Predictors coefficient (χ2) standard” DSM-5 interview with which to compare the
DIAMOND. As discussed in the Introduction, no psycho-
Anxiety disorder Block 1: DASS 4.52* metric data have been published for the SCID-5 (American
depression
Psychiatric Association, 2015), the MINI-5 (Sheehan,
  Block 2: DASS 16.66**
anxiety
2015), the ADIS-5 (Brown & Barlow, 2013), and the
  Block 1: DASS 21.10** AUADADIS-5 (Hasin et al., 2015) does not include mod-
anxiety ules for a variety of common psychological disorders,
  Block 2: DASS 0.09 including obsessive-compulsive and related disorders,
depression schizophrenia spectrum disorders, eating disorders, or
Depressive disorder Block 1: DASS 2.84 ADHD. How the DIAMOND compares to other diagnostic
anxiety interviews, therefore, is unknown.
  Block 2: DASS 9.35* Convergent validity of the DIAMOND diagnoses was
depression verified by higher scores on corresponding self-report mea-
  Block 1: DASS 12.15** sures for participants with specific anxiety, mood, and
depression
obsessive-compulsive and related diagnoses. Regression
  Block 2: DASS 0.03
anxiety
analyses demonstrated that for each diagnostic domain,
Bipolar disorder Block 1: DASS 0.50 convergent measures uniquely predicted DIAMOND diag-
anxiety noses beyond divergent measures. Importantly for training
  Block 2: MDQ 58.18** applications, mean administration time, interrater reliabil-
  Block 1: MDQ 53.33** ity, and test–retest reliability were not affected by trainee
  Block 2: DASS 5.35* status of the interviewers.
anxiety The underrepresentation of certain target disorders is an
Obsessive-compulsive Block 1: MDQ 0.74 important limitation of the present study. In this preliminary
and related disorder analysis, we were unable to obtain a sufficient sample of
  Block 2: OCI-R 8.10* participants with SAD, BP2, or CYC due to very low preva-
  Block 1: OCI-R 8.82* lence in the clinical sites used for the study. As these disor-
  Block 2: MDQ 0.02 ders also did not have precise estimates in the field trials, the
Note. DIAMOND = diagnostic interview for anxiety, mood, and OCD reliability of these diagnoses remains unknown (although
and related neuropsychiatric disorders; DASS = Depression Anxiety the overall category of bipolar disorders showed excellent
Stress Scales; MDQ = Mood Disorder Questionnaire; OCI-R = interrater and test–retest reliability). Another important limi-
Obsessive-Compulsive Inventory–Revised.
tation is the fact that although we assessed the reliability of
*p < .05. **p < .001.
diagnoses (present vs. absent), we did not examine the reli-
ability of the designation of a diagnosis as principal. As
and PMDD. It is likely that the relatively small number of noted in the Results section, the average participant received
participants with each of these disorders hampered our abil- roughly 2.5 psychiatric diagnoses. During development of
ity to obtain precise estimates, as the κ values themselves the ADIS-IV, the investigators examined the reliability of
were in the very good to excellent range. Further research is principal diagnoses, as well as the presence or absence
needed to elucidate the reliability estimates for these of diagnoses overall (Brown et al., 2001). The principal
disorders. diagnoses were identified using clinical severity ratings
As discussed in the Introduction, a diagnostic measure’s assigned to each diagnosis. In the present study, severity rat-
reliability is limited to that of the model of psychopathology ings based on distress and impairment showed questionable
it purports to measure. Categorical assessment models, such reliability, and therefore were deemed insufficient for deter-
as that implied by DSM-5, have been criticized, with many mining whether a given diagnosis was principal or addi-
recommending a dimensional approach to conceptualiza- tional. In future iterations of the DIAMOND, the severity
tion and measurement (e.g., Cuthbert, 2014; Krueger & ratings should be revised. Currently, they are based on a mix
Bezdjian, 2009). Perhaps most prominent among these of distress and impairment variables, the combination of
approaches is the Research Domain Criteria system pro- which may be problematic. It may be that functional impair-
posed by the National Institute of Mental Health (Insel ment is the most reliable and most meaningful criterion with
et al., 2010). A thorough examination of the pros and cons which to determine diagnostic severity, particularly given
of a categorical versus dimensional system is beyond the the fact that functional concerns are a leading reason for
scope of the present article; however, it is clear that individuals to seek treatment (Hunt & McKenna, 1993).

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

10 Assessment 

In constructing a diagnostic interview, there must be a Note

trade-off between thoroughness and efficiency. We deliber- 1. The research version of the DIAMOND also included a
ately added the distress and impairment questions after the severity score for each diagnosis (1-7 scale from “normal” or
symptom questions but before other diagnostic questions, remitted to “extreme”), based on questions about distress and
so that diagnoses would be ruled out more quickly than if functional impairment. However, the average intraclass cor-
the distress and impairment questions were asked at the end relation coefficient for severity ratings was in the question-
of each module. It could be argued that this process might able range (M = 0.54, range 0.03-0.90); therefore, severity
result in type II error, that is, failing to capture an existing ratings will be revised for future editions of the DIAMOND,
diagnosis. In the absence of an alternative “gold standard” and further severity-related analyses are not presented here.
interview, it is difficult to ascertain the false negative rate.
However, several factors alleviate this concern. First, dis- References
tress and impairment are part and parcel of the DSM-5 diag- American Psychiatric Association. (1994). Diagnostic and statis-
noses; therefore, an individual with symptoms that do not tical manual of mental disorders (4th ed.). Washington, DC:
cause distress or impairment by definition cannot be diag- Author.
nosed with a mental disorder (see Bolton, 2013, for discus- American Psychiatric Association. (2013). Diagnostic and statis-
tical manual of mental disorders (5th ed.). Washington, DC:
sion). Second, because the symptoms that would satisfy the
Author.
diagnostic criteria were queried prior to the “rule-out” ques- American Psychiatric Association. (2015). Structured clinical
tions, it seems highly unlikely that the items following the interview for DSM-5 (SCID-5). Retrieved from http://www.
“rule-outs” (e.g., persistence of fears, immediacy of fear appi.org/products/structured-clinical-interview-for-dsm-
response, separation from other mental or medical disor- 5-scid-5
ders) would rule a diagnosis back in. Finally, we note that Barkley, R. A., & Murphy, K. R. (1998). Attention-deficit hyper-
false negatives are generally of less concern for the DSM activity disorder: A clinical workbook (2nd ed.). New York,
than are false positives, as has been noted elsewhere NY: Guilford Press.
(Bolton, 2013; Frances & Widiger, 2012; Wakefield, 2015). Bolton, D. (2013). Overdiagnosis problems in the DSM-IV and the
The reliability and validity of the DIAMOND awaits new DSM-5: Can they be resolved by the distress-impairment
independent replication; however, the interview’s relatively criterion? Canadian Journal of Psychiatry, 58, 612-617.
Brown, T. A., & Barlow, D. H. (2009). A proposal for a dimen-
short administration time and robustness of reliability to
sional classification system based on the shared features of
training status suggest an ease of application that will be of the DSM-IV anxiety and mood disorders: Implications for
benefit in clinical, research, and training programs. The assessment and treatment. Psychological Assessment, 21,
guidance on differential diagnosis may be of particular ben- 256-271. doi:10.1037/a0016608
efit to less experienced interviewers. Overall, the Brown, T. A., & Barlow, D. H. (2013). Anxiety and related dis-
DIAMOND appears to be a promising diagnostic tool, with orders interview schedule for DSM-5, adult and lifetime ver-
good to excellent psychometric properties. sion: Clinician manual. New York, NY: Oxford University
Press.
Acknowledgments Brown, T. A., Di Nardo, P. A., Lehman, C. L., & Campbell, L. A.
(2001). Reliability of DSM-IV anxiety and mood disorders:
The authors thank Kathleen Stanwicks, James Ransom, and
Implications for the classification of emotional disorders.
Amber Billingsley for their administrative assistance, and they
Journal of Abnormal Psychology, 110, 49-58.
thank Jennie Bernstein, Kaylee Burnham, Allison Canfield, Nicole
Chambless, D. L., Caputo, G. C., Jasin, S. E., Gracely, E. J., &
Cappiello, Brian Castelluccio, Jennifer Espinoza, Brett Goshe,
Williams, C. (1985). The mobility inventory for agoraphobia.
Amanda LeTard, Alicia Little-Hodge, Suzanne Meunier, Elizabeth
Behaviour Research and Therapy, 23, 35-44.
Moore, Samantha Moshier, Rebecca Rowland, Christina Toland,
Clarke, D. E., Narrow, W. E., Regier, D. A., Kuramoto, S. J.,
Sara Whiting, and Coleen Williams for conducting DIAMOND
Kupfer, D. J., Kuhl, E. A., . . . Kraemer, H. C. (2013). DSM-5
interviews. field trials in the United States and Canada, Part I: Study
design, sampling strategy, implementation, and analytic
Authors’ Note approaches. American Journal of Psychiatry, 170, 43-58.
Copies of the DIAMOND may be obtained from the first author at doi:10.1176/appi.ajp.2012.12070998
david.tolin@hhchealth.org. Cuthbert, B. N. (2014). The RDoC framework: Facilitating transi-
tion from ICD/DSM to dimensional approaches that integrate
Declaration of Conflicting Interests neuroscience and psychopathology. World Psychiatry, 13,
28-35. doi:10.1002/wps.20087
The authors declared no potential conflicts of interest with respect
First, M. B. (2005). Clinical utility: A prerequisite for the adop-
to the research, authorship, and/or publication of this article.
tion of a dimensional approach in DSM. Journal of Abnormal
Psychology, 114, 560-564. doi:10.1037/0021-843X.114.4.560
Funding Foa, E. B., Huppert, J. D., Leiberg, S., Langner, R., Kichic, R.,
The authors received no financial support for the research, author- Hajcak, G., & Salkovskis, P. M. (2002). The obsessive-
ship, and/or publication of this article. compulsive inventory: Development and validation of a

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016

Tolin et al. 11

short version. Psychological Assessment, 14, 485-496. Krueger, R. F., Markon, K. E., Patrick, C. J., & Iacono, W. G.
doi:10.1037/1040-3590.14.4.485 (2005). Externalizing psychopathology in adulthood: A
Frances, A. J., & Widiger, T. (2012). Psychiatric diagnosis: dimensional-spectrum conceptualization and its implications
Lessons from the DSM-IV past and cautions for the DSM-5 for DSM-V. Journal of Abnormal Psychology, 114, 537-550.
future. Annual Review of Clinical Psychology, 8, 109-130. doi:10.1037/0021-843X.114.4.537
doi:10.1146/annurev-clinpsy-032511-143102 Lecrubier, Y., Sheehan, D. V., Weiller, E., Amorim, P., Bonora,
Frost, R. O., Steketee, G., & Grisham, J. (2004). Measurement of L. I., Sheehan, K. H., . . . Dunbar, G. C. (1997). The MINI
compulsive hoarding: Saving inventory-revised. Behaviour International Neuropsychiatric Interview (M.I.N.I.) A short
Research and Therapy, 42, 1163-1182. doi:10.1016/j. diagnostic structured interview: Reliability and validity
brat.2003.07.006 according to the CIDI. European Psychiatry, 12, 224-231.
Harris, P. A., Taylor, R., Thielke, R., Payne, J., Gonzalez, N., Lobbestael, J., Leurgans, M., & Arntz, A. (2011). Inter-rater reli-
& Conde, J. G. (2009). Research electronic data capture ability of the Structured Clinical Interview for DSM-IV Axis
(REDCap): A metadata-driven methodology and workflow I Disorders (SCID I) and Axis II Disorders (SCID II). Clinical
process for providing translational research informatics Psychology & Psychotherapy, 18, 75-79. doi:10.1002/cpp.693
support. Journal of Biomedical Informatics, 42, 377-381. Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the
doi:10.1016/j.jbi.2008.08.010 Depression Anxiety Stress Scales. Sydney, New South Wales,
Hasin, D. S., Greenstein, E., Aivadyan, C., Stohl, M., Aharonovich, Australia: Psychology Foundation of Australia.
E., Saha, T., . . . Grant, B. F. (2015). The Alcohol Use Meyer, T. J., Miller, M. L., Metzger, R. L., & Borkovec, T. D.
Disorder and Associated Disabilities Interview Schedule–5 (1990). Development and validity of the Penn State Worry
(AUDADIS-5): Procedural validity of substance use disor- Questionnaire. Behaviour Research and Therapy, 28,
ders modules through clinical re-appraisal in a general popu- 487-495. doi:10.1016/00005-7967(90)90135-6
lation sample. Drug and Alcohol Dependence, 148, 40-46. Phillips, K. (1996). The broken mirror: Understanding and
doi:10.1016/j.drugalcdep.2014.12.011 treating body dysmorphic disorder. New York, NY: Oxford
Heimberg, R. G., Horner, K. J., Juster, H. R., Safren, S. A., University Press.
Brown, E. J., Schneier, F. R., & Liebowitz, M. R. (1999). Regier, D. A., Narrow, W. E., Clarke, D. E., Kraemer, H. C.,
Psychometric properties of the Liebowitz Social Anxiety Kuramoto, S. J., Kuhl, E. A., & Kupfer, D. J. (2013). DSM-5
Scale. Psychological Medicine, 29, 199-212. field trials in the United States and Canada, Part II: Test-
Hirschfeld, R. M., Williams, J. B., Spitzer, R. L., Calabrese, J. R., retest reliability of selected categorical diagnoses. American
Flynn, L., Keck, P. E., Jr., . . . Zajecka, J. (2000). Development Journal of Psychiatry, 170, 59-70. doi:10.1176/appi.
and validation of a screening instrument for bipolar spec- ajp.2012.12070999
trum disorder: The Mood Disorder Questionnaire. American Shear, M. K., Brown, T. A., Barlow, D. H., Money, R., Sholomskas,
Journal of Psychiatry, 157, 1873-1875. D. E., Woods, S. W., . . . Papp, L. A. (1997). Multicenter col-
Hunt, S. M., & McKenna, S. P. (1993). Measuring quality of life in laborative panic disorder severity scale. American Journal of
psychiatry. In S. R. Walker & R. M. Rossner (Eds.), Quality of Psychiatry, 154, 1571-1575.
life assessment: Key issues in the 1990s (pp. 343-354). Boston, Sheehan, D. V. (2015). Mini International Neuropsychiatric
MA: Kluwer Academic. Interview 7.0. Jacksonville, FL: Medical Outcomes Systems.
Insel, T., Cuthbert, B., Garvey, M., Heinssen, R., Pine, D. S., Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Amorim, P.,
Quinn, K., . . . Wang, P. (2010). Research domain criteria Janavs, J., Weiller, E., . . . Dunbar, G. C. (1998). The Mini-
(RDoC): Toward a new classification framework for research International Neuropsychiatric Interview (M.I.N.I.): The
on mental disorders. American Journal of Psychiatry, 167, development and validation of a structured diagnostic psychi-
748-751. doi: 10.1176/appi.ajp.2010.09091379 atric interview for DSM-IV and ICD-10. Journal of Clinical
Kendell, R., & Jablensky, A. (2003). Distinguishing between the Psychiatry, 59(Suppl. 20), 22-33;quiz 34-57.
validity and utility of psychiatric diagnoses. American Journal Skinner, H. A. (1982). The drug abuse screening test. Addictive
of Psychiatry, 160, 4-12. doi:10.1176/appi.ajp.160.1.4 Behaviors, 7, 363-371.
Keuthen, N. J., O’Sullivan, R. L., Ricciardi, J. N., Shera, D., Summerfeldt, L. J., Kloosterman, P. H., & Antony, M. M. (2010).
Savage, C. R., Borgmann, A. S., . . . Baer, L. (1995). The Structured and semistructured diagnostic interviews. In M.
Massachusetts General Hospital (MGH) Hairpulling Scale: M. Antony & D. H. Barlow (Eds.), Handbook of assessment
1. Development and factor analyses. Psychotherapy and and treatment planning for psychological disorders (2nd ed.,
Psychosomatics, 64, 141-145. pp. 95-137). New York, NY: Guilford Press.
Keuthen, N. J., Wilhelm, S., Deckersbach, T., Engelhard, I. M., Tyrer, P. (2014). Time to choose: DSM-5, ICD-11 or both?
Forker, A. E., Baer, L., & Jenike, M. A. (2001). The Skin Archives of Psychiatry and Psychotherapy, 3, 5-8.
Picking Scale: Scale construction and psychometric analyses. Wakefield, J. C. (2015). DSM-5, psychiatric epidemiology and
Journal of Psychosomatic Research, 50, 337-341. the false positives problem. Epidemiology and Psychiatric
Kraemer, H. C., Kupfer, D. J., Clarke, D. E., Narrow, W. E., Sciences, 24, 188-196. doi:10.1017/S2045796015000116
& Regier, D. A. (2012). DSM-5: How reliable is reliable Zanarini, M. C., Skodol, A. E., Bender, D., Dolan, R., Sanislow,
enough? American Journal of Psychiatry, 169, 13-15. C., Schaefer, E., . . . Gunderson, J. G. (2000). The collab-
Krueger, R. F., & Bezdjian, S. (2009). Enhancing research and orative longitudinal personality disorders study: Reliability of
treatment of mental disorders with dimensional concepts: axis I and II diagnoses. Journal of Personality Disorders, 14,
Toward DSM-V and ICD-11. World Psychiatry, 8, 3-6. 291-299.

Downloaded from asm.sagepub.com at UNIV OF PITTSBURGH on September 30, 2016