New perspectives for the assessment of depression [Elektronische Ressource] : development of an item bank and a screening instrument applying Rasch analysis and structural equation modelling / vorgelegt von Thomas Forkmann

Tforkmann

Sujets

• Gesundheit

Informations

• Publié par: rheinisch-westfalischen_technischen_hochschule_-rwth-_aachen
• Publié le: 01 janvier 2008
• Nombre de lectures: 6
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

Thomas Forkmann – New Perspectives for the Assessment of Depression


New Perspectives for the Assessment of Depression:

Development of an Item Bank and a Screening Instrument
Applying Rasch Analysis and Structural Equation Modelling



Von der Medizinischen Fakultät
der Rheinisch-Westfälischen Technischen Hochschule Aachen
zur Erlangung des akademischen Grades
eines Doktors der Theoretischen Medizin
genehmigte Dissertation


vorgelegt von
Diplom-Psychologe
Thomas Forkmann
aus
Northeim



Berichter: Herr Universitätsprofessor
Dr.phil. Dipl.-Psych. Siegfried Gauggel

Herr Universitätsprofessor
Dr.rer.nat. Klaus Willmes-von Hinckeldey

Tag der mündlichen Prüfung: 16. Dezember 2008

Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online verfügbar.
- 1 - Thomas Forkmann – New Perspectives for the Assessment of Depression
- 2 -
Thomas Forkmann – New Perspectives for the Assessment of Depression
Table of Contents
Table of Contents…………………………………………………………………................3
Abbreviations………………………………………………………………………………….5
1 General Introduction……………………………………………………………………….7
1.1 Continuous Improvement of Depression Diagnostics is Mandatory……………….7
1.2 Advantages of Modern Approaches to Test Development……………………........9
1.2.1 The Rasch Model……………………………………………………………......11
1.3 Studies of the Current Thesis………………………………………………............. 13
1.3.1 Aim of Study One……………………………………………………………….. 13
1.3.2 Aim of Study Two……………………………………………………………….. 13
1.3.3 Aim of Study Three……………………………………………………………... 14
2 Study One: A Core Set of Depressive Symptoms Every Clinical
Expert Would Agree on…………………………………………………………………. 17
2.1 Introduction…………………………………………………………………………….. 17
2.2 Methods………………………………………………………………………………... 19
2.2.1 Sample…………………………………………………………………………… 19
2.2.2 Material…………………………………………………………………………... 20
2.2.3 Data Analysis……………………………………………………………………. 21
2.3 Results…………………………………………………………………………………. 21
2.3.1 Absolute Relevance Ratings………………………………………………………. 21
2.3.2 Inter-Rater Reliability……………………………………………………………….. 24
2.3.3 Psychiatrists vs. Charted Psychotherapists……………………………………… 24
2.4 Discussion……………………………………………………………………………... 26
3 Study Two: Development of an Item Bank for the Assessment of
Depression Using Rasch Analysis……………………………………………………. 31
3.1 Introduction……………………………………………………………………………. 31
3.2 Methods………………………………………………………………………………... 34
3.2.1 Sample…………………………………………………………………………… 34
3.2.2 Material…………………………………………………………………………... 36
3.2.3 Data Analysis……………………………………………………………………. 38
3.3 Results…………………………………………………………………………………. 41
3.3.1 Step (1): Evaluation of Rasch Model Fit……………………………………… 41
- 3 -
Thomas Forkmann – New Perspectives for the Assessment of Depression
3.3.2 Step (2): Evaluation of Differential Item Functioning (DIF)…………………. 42
3.3.3 Step (3): Evaluation of Further Item Bank Characteristics…………………. 47
3.4 Discussion……………………………………………………………………………... 49
3.5 Limitations……………………………………………………………………………… 52
4 Study Three: Development and Validation of the Rasch-based
Screening for Depression (DESC) Using Rasch Analysis and
Structural Equation Modelling………………………………………………………… 53
4.1 Introduction…………………………………………………………………………….. 53
4.2 Methods………………………………………………………………………………... 56
4.2.1 Sample…………………………………………………………………………… 56
4.2.2 Material…………………………………………………………………………... 57
4.2.3 Item Selection and Data Analysis…………………………………………….. 59
4.3 Results…………………………………………………………………………………. 62
4.3.1 Construction of the Rasch-based Screening for Depression
Version 1 (DESC-I) and 2 (DESC-II)…………………………………………. 62
4.3.2 Determination of Sensitive and Specific Cut-off Scores…………………….. 65
4.4 Discussion……………………………………………………………………………... 67
4.5 Limitations……………………………………………………………………………… 70
5 General Conclusion and Future Perspectives……………………………………… 71
6 Zusammenfassung……………………………………………………………………..... 77
7 References……………………………………………………………………………….... 79
8 Appendices……………………………………………………………………………….. 89
Appendix A: Questionnaires for Expert’s Relevance judgments on
Depression Items (Study I)……………………………………………………………….. 91
Appendix B: Aachen Depression Item Pool (Study II)……………………………......103
Appendix C: Rasch-based Screening for Depression I (DESC-I)…………………...113
Appendix D: Rasch-based Screening for Depression II (DESC II)……………….....117
Appendix E: Preliminary English Translation of the Rasch-based Screening
for Depression I and II (DESC-I/ -II)…………………………………………………….121
9 Acknowledgements…………………………………………………………………......125
10 Erklärung § 5 Abs. 1 zur Datenaufbewahrung…………………………………….127
11 Curriculum Vitae………………………………………………………………………..129
- 4 -Thomas Forkmann – New Perspectives for the Assessment of Depression
Abbreviations
ADIB Aachen Depression Item Bank
ADIP Aachen Depression Item Pool
AGFI Adjusted goodness of fit index
AUC Area under the curve
BDI Beck Depression Inventory
BSI Brief Symptom Inventory
CAT Computer adaptive test
CES-D Center for Epidemiologic Studies – Depression Scale
CFA Confirmatory factor analysis
CFI Comparative Fit Index
CI Confidence interval
CP cardiologic patients
CTT Classical test theory
DESC Rasch-based Screening for Depression
DIF Differential item functioning
DP patients treated for a depressive syndrome
DSM-IV Diagnostic and Statistical Manual for Mental Disorders, fourth edition
EM Expectation-maximization
ES Effect size
GDS Geriatric Depression Scale
GFI Goodness of fit index
HADS Hospital Anxiety and Depression Scale
HANDS Harvard Department of Psychiatry/NDSD scale
HDRS Hamilton Depression Rating Scale
ICC Intraclass correlation coefficient
ICD-10 International Classification of Diseases, 10th edition
IDCL International Diagnostic Checklist
IRT Item response theory
LRA Logistic regression analysis
M mean
MLE Maximum likelihood estimation
MnSq Mean square
MRJ Mean relevance judgment
NFI Normed Fit Index
OP Otorhinolaryngologic patiients
PCFAR Principal component factor analysis on the residuals
PCM Partial credit model
RMSEA Root mean square error of approximation
ROC Receiver operating curve
SCL 90-R Symptom-checklist 90-R
SD Standard deviation
SE Standard error
SEoM Standard error of measurement
SEM Structural equation modelling
TIF Test information function
TLI Tucker-Lewis Index
- 5 -
Thomas Forkmann – New Perspectives for the Assessment of Depression
- 6 -Thomas Forkmann – New Perspectives for the Assessment of Depression
1 General Introduction
1.1 Continuous Improvement of Depression Diagnostics is Mandatory
Depressive disorders are among the most prevalent mental disorders with an es-
timated lifetime prevalence of 12% for men and 26% for women (Wittchen & Jacobi,
2005). In certain populations (e.g., patients with chronic somatic illnesses, elderly), the
prevalence can be even higher. Depression represents an important predictor of mortal-
ity and morbidity (e.g. Simon et al., 2006; Rovner, 1993). Studies on quality of life and
psychosocial functioning revealed severe impairments: patients with depression had
functioning scores as low as those with advanced coronary heart disease, scoring lower
than e.g. patients suffering from hypertension or diabetes mellitus (Wells et al., 1989;
Wells & Sherbourne, 1999). Moreover, according to Murray and Lopez (1996), a group
of World Health Organization researchers, unipolar major depression is the leading
cause of disability worldwide. Greenberg and colleagues estimated the economical
costs caused by depression in the United States to be $83.1 billion per year, of which
approximately 70% resulted from premature death and impaired workplace productivity
(Greenberg et al., 2003). In Europe the annual costs of depression was estimated €118
billion and accounts for 1% of the total economy in Europe (Sobocki et al., 2006).
Moreover, depressive disorders often coincide with numerous other mental and
medical illnesses (Ehrt et al., 2006; Porche & Willis, 2006; Starkstein et al., 2005; Valen-
te, 2006; Wallin et al., 2006). In a large scale study by Schulberg et al. (1995), out of
approximately 700 patients suffering from a major depression that could be treated in a
primary care setting nearly 75% had suffered from an additional DSM-IV axis I diagnosis
(especially anxiety disorders and substance use disorders) and 68% suffered from an
axis II diagnosis at some time during their life. Thus, comorbidity in people suffering from
a depressive disorder is very high, which emphasizes the importance of diagnostic tools
that assess depression essentially unidimensionally with very good discriminatory
power.
- 7 -Thomas Forkmann – New Perspectives for the Assessm