Original article
Psychometric Evaluation of the Rasch-Based Depression Screening in Patients With Neurologic Disorders

Presented to the German Society for Neuropsychology, October 16, 2009, Munich, Germany.
https://doi.org/10.1016/j.apmr.2010.04.021Get rights and content

Abstract

Forkmann T, Norra C, Wirtz M, Gauggel S, Boecker M. Psychometric evaluation of the Rasch-based depression screening in patients with neurologic disorders.

Objectives

To provide a first evaluation of the Rasch-Based Depression Screening (DESC) in consecutive patients with neurologic disorders.

Design

Cross-sectional study.

Setting

Hospital specializing in neurologic disorders.

Participants

Eligible patients (N=323; mean age ± SD, 53.4±17.2y; 49.3% women) age 18 years and older.

Interventions

Not applicable.

Main outcome Measures

DESC with 2 parallel versions.

Results

Both versions of the DESC showed good item separation (DESC-I=4.96; DESC-II=4.94) and person separation (DESC-I=2.01; DESC-II=2.14) as well as item separation reliability (DESC-I=.96; DESC-II=.96) and person separation reliability (DESC-I=.80; DESC-II=.82). All items fit the Rasch model (infit and outfit mean squares ≤1.3). There were no signs of violation of unidimensionality. One item of DESC-I showed differential item functioning because of age and 1 item because of sex. Parallel test reliability of the 2 DESC forms was r equal to .92.

Conclusions

The DESC shows good psychometric characteristics that suggest high psychometric quality of the instrument in patients with neurologic disorders. Differential item functioning on single items needs further examination but should not substantially interfere with the valid interpretation of DESC sum scores.

Section snippets

Methods

The DESC was completed by 323 consecutive inpatients at admission to the neurologic department of a German university hospital. All participants were white. The mean age ± SD of the participants was 53.4±17.2 years; 49.3% were women. The mean length of stay in the hospital ± SD was 9.5±5.3 days (table 1).

Participants took part voluntarily without payment and signed an informed consent prior to testing. Test administration was conducted by trained personnel. The study was approved by the local

Separation and Reliability

Both versions of the DESC showed good item separation (DESC-I=4.96; DESC-II=4.94) and person separation (DESC-I=2.01; DESC-II=2.14) as well as item reliability (DESC-I=.96; DESC-II=.96) and person reliability (DESC-I=.80; DESC-II=.82).

Rasch Model Fit

All items of both DESC versions fit the Rasch model according to both infit and outfit mean squares statistics (see table 2). The spread of item category thresholds ranged from –3.48 to 2.35 logits for DESC-I and from –3.52 to 2.73 logits for DESC-II.

Unidimensionality and Local Independence

The principal

Discussion

This study aimed at providing a first evaluation of the DESC in consecutive inpatients with neurologic disorders. Overall, we found good Rasch model fit according to infit and outfit mean squares and indicators for unidimensionality and local independence of both parallel rating scales. Parallel test reliability was very good, and classifications of patients into depressed versus nondepressed were concordant between the 2 forms of the DESC. This indicates that the 2 versions of the instrument

Conclusions

Although future studies are needed to validate the DESC further in patients with neurologic disorders, the promising results of the present study suggest high psychometric quality of the instrument in this population.

Acknowledgments

We thank the staff of the neurologic hospital of the RWTH Aachen University for their support in conducting the present study.

References (38)

  • Diagnostic and statistical manual of mental disorders, fourth edition. DSM-IV

    (1994)
  • The ICD-10 classification of mental and behavioral disorders: clinical descriptions and diagnostic guidelines

    (1992)
  • A.D. Sadovnick et al.

    Depression and multiple sclerosis

    Neurology

    (1996)
  • M.T. Wallin et al.

    Depression and multiple sclerosis: review of a lethal combination

    J Rehabil Res Dev

    (2006)
  • R.W. Licht et al.

    Validation of the Bech-Rafaelsen Melancholia Scale and the Hamilton Depresson Scale in patients with major depression; is the total score a valid measure of illness severity?

    Acta Psychiatr Scand

    (2005)
  • E. Svensson

    Guidelines to statistical evaluation of data from rating scales and questionnaires

    J Rehabil Med

    (2001)
  • H.J. Moeller

    Methodological aspects in the asessment of severity of depression by the Hamilton Depression Scale

    Eur Arch Psychiatry Clin Neurosci

    (2001)
  • T. Forkmann et al.

    Development of an item bank for the assessment of depression in persons with mental illnesses and physical diseases using Rasch analysis

    Rehabil Psychol

    (2009)
  • K.A. Choquette et al.

    Effects of retesting with the Beck and Zung depression scales in alcoholics

    Alcohol Alcohol

    (1987)
  • Cited by (10)

    • The role of childhood abuse for suicidality in the context of the interpersonal theory of suicide: An investigation in German psychiatric inpatients with depression

      2019, Journal of Affective Disorders
      Citation Excerpt :

      We calculated the sum score for further analysis. The internal consistency (α = .92) of the DESC-I could be proved as excellent (α = .92) in a representative German sample (Forkmann et al., 2010) and as good in the present data (α = .85). All statistical analyses were conducted using SPSS version 24 (IBM, 2016).

    • Adaptive screening for depression - Recalibration of an item bank for the assessment of depression in persons with mental and somatic diseases and evaluation in a simulated computer-adaptive test environment

      2013, Journal of Psychosomatic Research
      Citation Excerpt :

      Using the software WINSTEPS 3.60.1 Forkmann et al. [18] showed that the ADIB is essentially unidimensional, fits the Rasch model and captures a wide range of the latent continuum. The ADIB proved to be useful for the derivation of high quality static short scales for the assessment of depression supporting its general psychometric quality [19–23]. However, Forkmann et al. [18] further reported that there were small signs of a potential secondary dimension constituted by items about suicidal ideation and behavior.

    • Cross-sectional validation of the Rasch-based Depression Screening (DESC) in a mixed sample of patients with mental and somatic diseases

      2013, Comprehensive Psychiatry
      Citation Excerpt :

      So results generally give confidence that the scales are truly Rasch scales. Across different studies that evaluated the psychometric characteristics of the DESC the “sad” item either fitted well [17,18], or showed only marginal misfit (infit MNSQ between 1.2 and 1.3 such as in Forkmann et al. [13,14] and the present study). Therefore, future research should further examine the fit of the “sad” item, and may aim at identifying potential sources for marginal misfit, such as individual worry or sorrow about a medical condition or coping styles with the disease.

    • Evaluation of the "Consultation and Relational Empathy" (CARE) measure by means of Rasch-analysis at the example of cancer patients

      2011, Patient Education and Counseling
      Citation Excerpt :

      Especially due to the correction of the skewed distribution and straightening differences in the parameter distribution's extreme ranges, enhanced discriminability [60] and sensitivity for changes [61] can be expected. Furthermore, all psychometrical advantages and evaluation strategies (Appendix; [25–30]) which hold only for Rasch-homogeneous scales can now be applied for the PE construct by using the CARER-measure or the corrected item pool from the original CARE-measure, respectively. Evaluation of moderating effects on the scale structure showed that general scale misfit for specific patient characteristics as well as item biases exist.

    View all citing articles on Scopus

    Supported by the START-program of the Faculty of Medicine (grant no. 690812), RWTH Aachen, and the German Research Foundation (grant no. DFG, WI3210/2-1).

    No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.

    View full text