Downloaded from ebmh.bmj.com on February 22, 2010 - Published by group.bmj.
com
The clinician rated CY-BOCS scale is more
sensitive than the self-reported LOI-CV in children
and adolescents with obsessive compulsive
disorder
BMJ Publishing Group Ltd, Royal College of Psychiatrists and British
Psychological Society
Evid Based Mental Health 2006 9: 67
doi: 10.1136/ebmh.9.3.67
Updated information and services can be found at:
http://ebmh.bmj.com/content/9/3/67.full.html
These include:
References This article cites 6 articles
http://ebmh.bmj.com/content/9/3/67.full.html#ref-list-1
Email alerting Receive free email alerts when new articles cite this article. Sign up in the box at
service the top right corner of the online article.
Topic collections Articles on similar topics can be found in the following collections
Epidemiologic studies (23648 articles)
Obsessive-compulsive disorder (48 articles)
Schizophrenia spectrum (288 articles)
Epidemiology (4706 articles)
Screening (epidemiology) (6087 articles)
Notes
To order reprints of this article go to:
http://ebmh.bmj.com/cgi/reprintform
To subscribe to Evidence Based Mental Health go to:
http://ebmh.bmj.com/subscriptions
Downloaded from ebmh.bmj.com on February 22, 2010 - Published by group.bmj.com
DIAGNOSIS 67
The clinician rated CY-BOCS scale is more sensitive than the self-
reported LOI-CV in children and adolescents with obsessive
compulsive disorder
Stewart SE, Ceranoglu TA, O’Hanley T, et al. Performance of clinician versus self-report measures to identify obsessive-compulsive
disorder in children and adolescents. J Child Adolesc Psychopharmacol 2006;15:956–63.
...............................................................................................................................
Q How sensitive are clinician administered and self-reported surveys in the diagnosis of childhood obsessive compulsive
disorder (OCD)?
METHODS NOTES
This study could not provide data on specificity of these scales as all
participants had a diagnosis of OCD. The results of this study may not
Design: Cohort study. be generalisable to a mixed population including children and
adolescents with and without OCD.
Setting: Massachusetts General Hospital, Boston, Massachusetts, Commentary
USA; timeframe not stated.
T
he prevalence and severity of paediatric obsessive compulsive
disorder (OCD) have resulted in increased attention from research-
Patients: 81 children and adolescents with DSM-IV OCD (mean
ers and clinicians alike in the realms of assessment and treatment.1
age 11.5 years, mean age of OCD onset 8.3 years).
Poor detection abilities may contribute to underdiagnosing of paediatric
OCD and inaccurate estimates of prevalence. A number of measures
Test: Leyton Obsessional Inventory—Child Version (LOI-CV) have been developed to assess the presence and severity of paediatric
survey form self-report: participants completed the 44 item OCD symptoms. The primary clinician rated instrument is the Children’s
children and adolescent version of the Leyton Obsessional Yale-Brown Obsessive Compulsive Scale (CY-BOCS),2 and the Leyton
Inventory for current OCD symptoms prior to clinical assessment, Obsessional Inventory—Child Version (LOI-CV) is a commonly used
without assistance. The LOI-CV items fall into three subgroups: child-report questionnaire.3 Although strong psychometric properties
symptoms (each item scored 0 or 1), interference (each item exist for the CY-BOCS,2 4 less consistent support has been found for the
scored 0–3) and resistance (each item scored 0–3). The shorter LOI-CV. However, child-report measures such as the LOI-CV have been
form of the LOI-CV, the LOI-CV survey form, uses 20 of the used in epidemiological studies to estimate prevalence rates.5
symptom and interference items. Children’s Yale-Brown With this in mind, Stewart et al studied the agreement between the CY-
Obsessive Compulsive Scale (CY-BOCS): a clinician rated scale
BOCS and LOI-CV in a sample of 81 youth with OCD to examine the
of 10 items, each rated from 0 (no symptoms) to 4 (extreme
adequacy of the LOI-CV survey form as a screening instrument. Although
symptoms), scores are summed to give a maximum score of 40.
the CY-BOCS correlated strongly and positively with the Global
Diagnostic standard: DSM-IV diagnosis using the Kiddie— Assessment of Functioning rating, results indicated weak associations
Schedule for Affective Disorders and Schizophrenia— between the CY-BOCS and LOI-CV. Further, the LOI-CV demonstrated
Epidemiological Version (K-SADS-E), Fifth Revision: a semi- poor sensitivity, whereas the CY-BOCS showed generally good ability to
structured DSM-IV-based psychiatric diagnostic interview of both detect OCD cases.
parents and child by trained interviewers. This paper has significant implications. First, earlier estimates of
paediatric OCD prevalence that were based on the LOI-CV5 likely
Outcomes: Sensitivity.
underrepresent true rates. Using the same cutoff as Flament et al,5 only
12.5% of this sample would have been identified as meeting criteria.
Second, this paper speaks to the limited utility of child-report measures in
paediatric OCD assessment. The notion that children are ‘‘better’’
reporters of internalising symptoms may not hold true for some youth with
MAIN RESULTS
OCD, particularly those with comorbid externalising behaviours or
Using the ‘‘high interference’’ cutoff score (>25) on the LOI-CV limited insight.6 Finally, these results suggest that the LOI-CV may best be
survey form for diagnosis of OCD, as used in a previous study, yielded used as an adjunct to clinician rated measures, rather than as a primary
a sensitivity of 12.5%. To achieve more than 95% sensitivity a cutoff outcome.
score of 6 was needed on the LOI-CV 44 item form, or a cutoff total
score of 2 on the LOI-CV survey form. The standard CY-BOCS cutoff Eric A Storch, PhD
score used to diagnose mild OCD (>8) yielded a sensitivity of 100%, Departments of Psychiatry and Pediatrics, University of Florida,
and a cutoff score of 12 still yielded over 95% sensitivity. Gainesville, FL, USA
1 Lewin A, Storch EA, Adkins J, et al. Current thinking about pediatric
obsessive-compulsive disorder. Pediatr Ann 2005;34:128–37.
CONCLUSIONS 2 Scahill L, Riddle MA, McSwiggin-Hardin M, et al. Children’s Yale-Brown
The clinician rated CY-BOCS scale is more sensitive than the self- Obsessive Compulsive Scale: Reliability and validity. J Am Acad Child
report LOI-CV in identifying childhood and adolescent OCD. Studies Adolesc Psychiatry 1997;36:844–52.
that have used the LOI-CV to assess OCD in children and adolescents 3 Berg CZ, Rapoport JL, Flament M. The Leyton Obsessional Inventory-Child
may have underestimated the prevalence of this disorder. Version. J Am Acad Child Adolesc Psychiatry 1986;25:84–91.
4 Storch EA, Murphy TK, Geffken GR, et al. Psychometric evaluation of the
Children’s Yale-Brown Obsessive Compulsive Scale. Psychiatry Res
2004;129:91–8.
5 Flament MF, Whitaker A, Rapoport JL, et al. Obsessive compulsive disorder
............................................................. in adolescence: An epidemiological study. J Am Acad Child Adolesc
For correspondence: Daniel Geller, Pediatric Psychopharmacology Research Psychiatry 1988;27:764–71.
Program, 185 Alewife Brook Parkway, Suite 2000, Cambridge, MA 02138, 6 Merlo LJ, Storch EA, Murphy TK, et al. Assessment of pediatric OCD: a
USA; dageller@partners.org critical review of current methodology. Child Psychiatry Hum Dev
2005;36:195–214.
Source of funding: National Institutes of Mental Health.
EBMH Volume 9 August 2006 www.ebmentalhealth.com
