Is the Children's Depression Inventory Short version a valid screening tool in pediatric care? A comparison to its full-length version

https://doi.org/10.1016/j.jpsychores.2012.08.016Get rights and content

Abstract

Objective

This is the first study to validate and to compare the Children's Depression Inventory (CDI) and its short version (CDI:S) as screening tools for medically ill children.

Methods

A sample of 406 pediatric hospital patients, aged 9 to 12 years (56.2% male, 77.1% inpatients), completed the German CDI. Criterion validity of the 26-item CDI and the 10-item CDI:S was calculated by receiver operating characteristic (ROC) curves. DSM-IV diagnoses of depression based on the structured diagnostic interview for mental disorders in children and adolescents (Kinder-DIPS) served as the reference standard. Areas under the ROC curves as well as sensitivities and specificities for the optimal cutoffs were compared for both versions.

Results

Diagnoses of major or minor depression were established for 7.4% of the children. Areas under the curve for the 26-item CDI (87.7%) and the 10-item CDI:S (88.2%) were comparable. For the CDI, the cutoff ≥ 12 yielded the best balance between sensitivity (83.3%) and specificity (82.7%). At the optimal cutoff ≥ 3, the CDI:S resulted in a high sensitivity of 93.3% and a specificity of 70.7%. Thus, the CDI:S proved to be as sensitive as the CDI, but was less specific than the full-length version.

Conclusion

Both the CDI and the CDI:S are valid screening instruments for depression in medically ill children. The sensitive and brief CDI:S is a promising tool in time-pressed settings such as pediatric care, but has to be followed by a thorough diagnostic assessment to rule out false positive cases.

Introduction

Depression is a common mental disorder even in childhood: A meta-analysis of epidemiological studies reported a point-prevalence of 2.8% for prepubertal children under the age of 13 [1]. Prevalence rates are even higher for children suffering from medical illnesses. Substantially elevated rates of major and minor depression ranging between 11% and 29% were found for various somatic conditions, such as migraine and epilepsy [2], congenital heart disease [3], inflammatory bowel disease and cystic fibrosis [4], cancer [5], burn injuries [6] and orthopedic procedures [7].

Since early-onset depression negatively affects the course of somatic diseases [8] and tends to persist into adolescence [9], early recognition is essential. However, pediatricians often feel uncomfortable diagnosing psychiatric disorders in their patients [10], [11], and indeed, only one in five mentally ill children or adolescents is being identified [12]. Specifically for depression, an even lower detection rate of 12.5% was found [13]. Self-report questionnaires to screen for depression can be one helpful tool for pediatricians to improve low recognition rates [14], [15].

The most established self-report measure of depressive symptoms for child samples is the 27-item Children's Depression Inventory (CDI, [16], [17]). The CDI has been translated into several languages. In the present study, a slightly modified German version was used [18].

Research on the criterion validity of the CDI recommends various cutoffs to distinguish between depressed and non-depressed patients in samples differing in psychosocial impairment.

A high cutoff score of 19 was recommended to differentiate between severely affected psychiatric inpatients and a non-symptomatic school sample [19].

For screening in moderately affected clinically referred samples, cutoffs between 12 and 17 have been found. Timbremont et al. [20] reported a cutoff of 16 in patients aged 8 to 18 who were referred from in- and outpatient facilities to centers of childhood psychopathology. A lower cutoff of 12 was found by Lobovits and Handal [21] in a sample of 8- to 12-year old children referred to outpatient psychology service centers. Using this cutoff, 88% of the 50 children were correctly classified as depressed or non-depressed with regard to the DSM-III criteria for major affective disorder.

So far, only one research group explored cutoff scores in mildly affected medically ill children who had a pediatricians' referral to evaluate their emotional state [22]. In this sample of 81 8 to 19 year olds mainly suffering from chronic medical conditions, a cutoff of 11 was found [22]. For validation of the CDI a best estimate procedure was applied based on a structured diagnostic interview and medical records. The number of correctly identified cases (sensitivity) was 80%, whereas 70% were correctly classified as non-depressed (specificity).

For screening purposes in time-limited pediatric settings, Kovacs [17] recommends the 10-item CDI Short version (CDI:S). So far, criterion validity of the CDI:S in terms of diagnostic accuracy for screening for depression has not been examined, even though there are a number of studies applying the CDI:S. Yet, they focused on different aspects of validity, such as concurrent validity [e.g., [23], [24], [25]].

Apart from the CDI:S, there are two other established screening instruments for prepubertal children, the Short Mood and Feelings Questionnaire (MFQ-SF, [26]) and the Child Behavior Checklist (CBCL, [27]). In a study that particularly addressed medically ill children using the MFQ-SF as opposed to the structured Diagnostic Interview Schedule for Children (DISC, [28]) an overall accuracy of 84%, a sensitivity of 80% and a specificity of 81% were reported [29]. The sample consisted of children aged 11 to 17 with and without asthma who were enrolled in a health maintenance organization. For the CBCL, a diagnostic accuracy of 77% was found in clinically referred 6 to 18 year olds [30].

This is the first study to assess the validity of the Children's Depression Inventory Short version (CDI:S) as opposed to the CDI for screening among prepubertal medically ill patients. It was hypothesized that the CDI:S would be equally valid as the CDI.

Section snippets

Procedure

Recruitment took place in three pediatric hospitals and three pediatric surgery hospitals and was arranged sequentially in time units of three months, each between September 2009 and November 2010. Monday through Friday, all newly admitted in- and outpatients between 9 and 12 years and their parents were invited to participate in the study. Children had to meet the following inclusion criteria as documented by the medical staff to assure they could complete the screening instrument autonomously:

Sample description

Participating patients (n = 406) and excluded patients (n = 227) did not differ significantly with respect to age (t(631) =  1.80, p = .073) or sex (χ2(1, 633) = 3.71; p = .054). The mean age of participants was 10.5 years (SD = 1.10) including 43.8% females and 56.2% males. Thirty participants (18 boys, 12 girls) fulfilled the diagnostic criteria of a depressive disorder according to the DSM-IV-TR as represented in the Kinder-DIPS interview. Point prevalence was 7.4%, 3.2% accounting for minor depression and

Discussion

The aim of the present study was to investigate and to compare the criterion validity of the Children's Depression Inventory (CDI) and its short version (CDI:S) in a sample of medically ill children.

The CDI showed good diagnostic accuracy as well as balanced values for sensitivity and specificity (both above 80%) at an optimal cutoff ≥ 12 for the German 26-item version. An identical cutoff of 12 for the original 27-item version was reported by Lobovits and Handal [21] for clinically referred

Conflict of interest

All authors have completed the Unified Competing Interest form and declare that the Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy received funding by the Bavarian State Ministry of the Environment and Public Health in the scope of the initiative Gesund.Leben.Bayern. Apart from that, the authors have no competing interests to report.

Acknowledgments

This study was funded by the Bavarian State Ministry of the Environment and Public Health in the scope of the initiative Gesund.Leben.Bayern. We thank the Departments of Pediatrics and Pediatric Surgery of the Hospital “Dritter Orden”, the Departments of Pediatric and Pediatric Surgery of the University Hospital “Schwabing”, the Department of Pediatric Surgery of the University Hospital “Dr. Haunersches Kinderspital” and the Department of Pediatrics of the Hospital “Harlaching” for their

References (43)

  • E.J. Costello et al.

    Is there an epidemic of child or adolescent depression?

    J Child Psychol Psychiatry

    (2006)
  • M. Pinquart et al.

    Depressive symptoms in children and adolescents with chronic physical illness: an updated meta-analysis

    J Pediatr Psychol

    (2011)
  • P.A. Karsdorp et al.

    Psychological and cognitive functioning in children and adolescents with congenital heart disease: a meta-analysis

    J Pediatr Psychol

    (2007)
  • L. von Essen et al.

    Self-esteem, depression and anxiety among Swedish children and adolescents on and off cancer treatment

    Acta Paediatr

    (2000)
  • F.J. Stoddard et al.

    A diagnostic outcome study of children and adolescents with severe burns

    J Trauma

    (1989)
  • J.H. Kashani et al.

    Depression in children admitted to hospital for orthopaedic procedures

    Br J Psychiatry

    (1981)
  • M.C. Delmas et al.

    Asthma and major depressive episode in adolescents in France

    J Asthma

    (2011)
  • W.P. Fremont et al.

    Comfort level of pediatricians and family medicine physicians diagnosing and treating child and adolescent psychiatric disorders

    Int J Methods Psychiatr Res

    (2008)
  • J. Williams et al.

    Diagnosis and treatment of behavioral health disorders in pediatric practice

    Pediatrics

    (2004)
  • A.K. Allgaier et al.

    Screening for depression in adolescents: validity of the patient health questionnaire in pediatric care

    Depress Anxiety

    (2012)
  • U.S. Preventive

    Services Task Force. Screening and treatment for major depressive disorder in children and adolescents: US Preventive Services Task Force Recommendation Statement

    Pediatrics

    (2009)
  • Cited by (141)

    View all citing articles on Scopus
    View full text