Major Depression Inventory

1. Abstract

The Major Depression Inventory (MDI) is a self-administered psychometric instrument developed by Per Bech and colleagues at the Psychiatric Research Unit in Hillerød, Denmark, designed to assess both the categorical diagnosis of a major depressive episode according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) and International Classification of Diseases (ICD-10), as well as the dimensional severity of depressive symptomatology. Comprising 10 items—with two items (items 8 and 10) bifurcated into sub-questions assessing opposite vegetative or psychomotor manifestations (e.g., psychomotor agitation versus retardation; weight gain versus weight loss)—the inventory utilizes a 6-point Likert-type response format capturing symptom frequency over the preceding two weeks, ranging from 0 (“At no time”) to 5 (“All the time”). The dimensional total score ranges from 0 to 50, providing a metric for tracking illness trajectory and treatment responsiveness. Extensive psychometric evaluations demonstrate high internal consistency, with Cronbach’s alpha coefficients typically ranging from .89 to .94 across diverse clinical and non-clinical samples. Unlike many conventional depression scales developed purely within Classical Test Theory (CTT), the MDI was validated using item response theory, specifically the Rasch model and Mokken scale analysis, confirming that its items conform to a unidimensional, cumulative latent construct. Criterion validity studies using standardized clinical interviews, such as the Schedules for Clinical Assessment in Neuropsychiatry (SCAN), exhibit sensitivity between 0.86 and 0.92 and specificity between 0.82 and 0.89. The MDI represents an indispensable clinimetric instrument, uniquely bridging diagnostic taxonomy and continuous severity rating in adult and geriatric psychiatric care, primary care, and clinical trials.

2. Keywords

Major Depression Inventory, MDI, Per Bech, depression assessment, DSM-IV, ICD-10, clinimetrics, Rasch analysis, psychometrics, severity scoring, screening tool, affective disorders

3. Authors

The Major Depression Inventory was created and validated by a Danish research team led by the late Professor Per Bech:

• Per Bech, MD, Dr.Med.Sci. — Former Head of the Psychiatric Research Unit, Mental Health Centre North Zealand, Hillerød, Denmark; Professor of Clinical Psychiatry, University of Copenhagen; WHO Collaborating Centre for Mental Health.
• Niels Anton Rasmussen, MSc — Psychiatric Research Unit, Mental Health Centre North Zealand, Frederiksborg General Hospital, Hillerød, Denmark.
• Raoul Olsen, MSc — Psychiatric Research Unit, Mental Health Centre North Zealand, Frederiksborg General Hospital, Hillerød, Denmark.
• Vibeke Noerholm, PhD — Psychiatric Research Unit, Mental Health Centre North Zealand, Frederiksborg General Hospital, Hillerød, Denmark.
• W. Abildgaard, MD — Department of Psychiatry, Frederiksborg General Hospital, Hillerød, Denmark.

Institutional Affiliation: Psychiatric Research Unit, Mental Health Centre North Zealand, Hillerød, Denmark (affiliated with the World Health Organization Collaborating Centre for Mental Health).

4. Purpose

The primary purpose of the Major Depression Inventory is to resolve a persistent methodological dilemma in psychiatric epidemiology and clinical practice: the disconnect between categorical classification schemes (such as the DSM and ICD diagnostic manuals) and continuous dimensional measurement scales (such as the Hamilton Depression Rating Scale [HAM-D] or the Beck Depression Inventory [BDI]). Traditional self-report depression inventories often contain items that reflect non-specific psychological distress or somatic anxiety, occasionally causing discrepancy between an elevated score and true diagnostic criteria for a major depressive episode.

The MDI was purposefully engineered to function as a dual-purpose clinimetric instrument:

• Diagnostic Classification: It operationalizes the symptomatic criteria for major depression outlined in the DSM-IV and moderate-to-severe depression in the ICD-10. Through a rule-based algorithm, clinicians can ascertain whether a patient fulfills the diagnostic thresholds of core and supplementary depressive symptoms over a two-week window.
• Dimensional Severity Quantification: By summing all item ratings, the MDI produces an overall score from 0 to 50 that tracks the gradient of depressive severity (no depression, mild, moderate, or severe) over time, measuring therapeutic response, relapse, or remission.

In clinical trials, the MDI provides an efficient, patient-centered outcome measure that eliminates administration burden while preserving diagnostic integrity. In primary care, it serves as an initial screener that identifies individuals meeting threshold diagnostic criteria before formal clinical interview evaluation.

5. Psychological Construct

The MDI measures the psychological construct of a Major Depressive Episode, operationalized through three primary symptom domains defined by contemporary psychiatric nosology:

Core Affective-Hedonic Symptoms

Depressive mood and anhedonia constitute the obligatory core features of major depression:

• Depressed Mood (Item 1): Pervasive feelings of sadness, low spirits, or emotional despair sustained throughout the day.
• Loss of Interest / Anhedonia (Item 2): Marked decrease in interest or pleasure in daily activities, hobbies, or social interactions.
• Decreased Energy / Fatigue (Item 3): Persistent subjective sensation of physical or mental exhaustion, lethargy, or lack of vitality.

Cognitive and Evaluative Symptoms

The cognitive dimension of depression involves biased self-referential evaluations and altered executive functioning:

• Loss of Self-Confidence and Self-Esteem (Item 4): Feelings of inadequacy, diminished self-worth, or subjective incompetence.
• Feelings of Guilt and Self-Reproach (Item 5): Excessive or inappropriate self-blame, moral guilt, or feelings of personal failure.
• Diminished Concentration and Indecisiveness (Item 7): Impaired cognitive focus, subjective distractibility, or difficulty making routine decisions.
• Suicidal Ideation and Death Wishes (Item 6): Thoughts that life is not worth living, recurrent suicidal thoughts, or active death contemplation.

Neurovegetative and Somatomotor Symptoms

Biological rhythms and physical functioning exhibit significant dysregulation in affective disorders:

• Sleep Disturbance (Item 9): Insomnia (initial, middle, or terminal early morning awakening) reflecting disturbed sleep architecture.
• Psychomotor Agitation or Retardation (Items 8a & 8b): Observable physical restlessness, inner turmoil, and pacing, or conversely, slowed motor activity, delayed speech, and reduced expressive gestures.
• Appetite and Weight Fluctuations (Items 10a & 10b): Alterations in metabolic homeostasis, manifesting as either marked anorexia/weight loss or hyperphagia/weight gain.

6. Theoretical Framework

The Major Depression Inventory is anchored in the theoretical framework of Clinimetrics, a discipline founded by Alvan Feinstein and extended in psychiatry by Per Bech. Clinimetrics argues that psychological assessment tools must reflect clinical reality rather than solely complying with statistical patterns derived from classical test theory.

Item Response Theory and the Rasch Measurement Model

A core assumption underpinning the MDI is that depressive symptoms lie along a cumulative latent continuum of severity. Bech utilized the Rasch model and non-parametric Mokken scaling to verify that depressive symptoms adhere to a Guttman-like hierarchy:

• Milder symptoms (e.g., fatigue, transient low mood) are endorsed at lower thresholds of the latent depression trait ($ heta$).
• Cognitive symptoms (e.g., intense guilt, loss of self-confidence) appear at moderate levels of underlying pathology.
• Severe symptoms (e.g., psychomotor retardation, profound suicidal despair) are endorsed predominantly by individuals experiencing severe depressive states.

Under this Rasch-based model, the sum score is a mathematically sufficient statistic for the latent trait, ensuring that each additional point represents an incremental increase in depressive severity without construct contamination.

7. Validity

The psychometric validity of the Major Depression Inventory has been verified across various cultures, linguistic adaptations, and clinical cohorts.

Criterion and Diagnostic Validity

In original validation work by Bech et al. (2001), the diagnostic validity of the MDI was tested against the Schedules for Clinical Assessment in Neuropsychiatry (SCAN) interview administered by blinded clinical evaluators. When evaluated using DSM-IV criteria, the MDI demonstrated a sensitivity of 0.86 and a specificity of 0.82. In general population and primary care samples (Olsen et al., 2003), the MDI maintained sensitivity ranging from 0.86 to 0.92 and specificity between 0.82 and 0.89, outperforming several legacy self-report instruments in positive predictive value.

Convergent and Discriminant Validity

The continuous sum score of the MDI correlates robustly with established clinician-rated and self-report depression metrics:

• Strong correlations with the Hamilton Depression Rating Scale (HAM-D17), typically ranging from $r = .75$ to $r = .88$.
• High convergence with the Beck Depression Inventory (BDI-II), with Pearson correlations exceeding $r = .80$.
• Moderate correlations ($r = .50$ to $.65$) with generalized anxiety measures (e.g., HADS-A, GAD-7), demonstrating expected overlap alongside adequate discriminant validity.
• Weak correlations ($r < .30$) with unrelated constructs such as general personality traits and cognitive impairment screens, underscoring construct distinctiveness.

8. Reliability

Empirical evaluations consistently document robust reliability metrics for the MDI across both clinical psychiatric cohorts and general epidemiological surveys:

Internal Consistency

Across validation studies in Denmark, the United Kingdom, Sweden, the Netherlands, and international translations, the internal consistency of the MDI is high:

• Cronbach’s Alpha ($lpha$): Values consistently fall between .89 and .94, reflecting high item homogeneity without excessive redundancy.
• Mokken Homogeneity Coefficient ($H$): Bech et al. reported Loevinger’s $H$ coefficients ranging between .40 and .55, confirming moderate-to-strong scalability according to non-parametric item response theory benchmarks.
• McDonald’s Omega ($\omega$): Recent psychometric evaluations demonstrate omega total ($\omega_t$) values exceeding .91, corroborating reliability independent of tau-equivalence assumptions.

Test-Retest Reliability and Temporal Stability

In stable clinical outpatients and control cohorts assessed across intervals ranging from 1 to 2 weeks, the MDI shows intra-class correlation coefficients (ICC) and Pearson correlation coefficients between .80 and .88, indicating temporal stability in non-fluctuating populations while remaining sensitive to genuine clinical changes following antidepressant or psychotherapeutic intervention.

9. Factor Analysis

Extensive exploratory (EFA) and confirmatory factor analytic (CFA) investigations have evaluated the internal dimensionality of the MDI:

Unidimensional Structure and Rasch Invariance

Factor analytic studies by Bech et al. (2001), Cuijpers et al. (2007), and Nielsen et al. (2010) established that the MDI fits a predominant unidimensional factor model. When evaluated via principal component analysis, the primary factor accounts for over 50% to 60% of the total variance, with an eigenvalue ratio between the first and second factors routinely exceeding 4:1.

Fit Indices in Confirmatory Factor Analysis

Confirmatory factor analyses testing a single-factor latent structure demonstrate good model fit across standard indices:

• Comparative Fit Index (CFI): Values consistently range from .94 to .98.
• Tucker-Lewis Index (TLI): Observed values range between .93 and .97.
• Root Mean Square Error of Approximation (RMSEA): Values typically fall between .045 and .068 (90% CI [.038, .078]), indicating close fit.
• Standardized Root Mean Square Residual (SRMR): Values generally remain below .04.

Standardized factor loadings across items are uniformly strong, typically ranging from $lambda = .62$ to $lambda = .88$, with Core Mood, Loss of Energy, and Anhedonia exhibiting the highest factor loadings.

10. Instrument / Measurement Tool

The Major Depression Inventory is configured as follows:

• Test Type: Self-administered psychological questionnaire (patient-reported outcome measure).
• Administration Time: Approximately 5 to 10 minutes.
• Target Population: Adults and older adults (ages 18 and above); validated in adolescent populations down to age 15 in select clinical contexts.
• Number of Items: 10 numbered items total (Items 8 and 10 contain two mutually exclusive sub-items: 8a/8b and 10a/10b).
• Time Frame Assessed: The past two weeks.
• Response Scale: 6-point Likert-type scale scored as:
  • 0 = At no time
  • 1 = Some of the time
  • 2 = Less than half of the time
  • 3 = More than half of the time
  • 4 = Most of the time
  • 5 = All the time
• Scoring Rules:
  • Dimensional Severity Score: For Item 8, take the higher score between 8a and 8b. For Item 10, take the higher score between 10a and 10b. Sum the scores for the 10 items. Total score ranges from 0 to 50.
    • 0 to 20: No or doubtful depression
    • 21 to 25: Mild depression
    • 26 to 30: Moderate depression
    • 31 to 50: Severe depression
  • Categorical Diagnostic Algorithm (DSM-IV / ICD-10):
    • A symptom is considered clinically present if scored ≥ 4 (most of the time or all the time), except for Item 6 (suicidal thoughts), where a score of ≥ 3 (more than half of the time) or even ≥ 2 in sensitive clinical contexts is often considered symptomatic.
    • DSM-IV Major Depression: Requires at least 5 of the 9 symptom domains present for most of the time over the past 2 weeks, with at least one being Item 1 (depressed mood) or Item 2 (loss of interest).
    • ICD-10 Categorization: Mild depression (2 core symptoms + 2 accessory symptoms); Moderate depression (2 core symptoms + 4 accessory symptoms); Severe depression (all 3 core symptoms + ≥ 5 accessory symptoms).

11. Permissions & Fee and Test Year

The Major Depression Inventory was introduced in its completed psychometric form in 2001 by Per Bech and colleagues at the Psychiatric Research Unit, Mental Health Centre North Zealand (Frederiksborg General Hospital), Denmark, in collaboration with the World Health Organization Collaborating Centre for Mental Health.

Licensing and Accessibility: The MDI is considered an open, non-commercial public domain clinical instrument for clinical, educational, and academic research purposes. It is freely available for non-commercial use, provided that appropriate bibliographic attribution is given to the authors and the original publication. For commercial deployments, pharmaceutical trial inclusion, or digital software integration, permissions should be coordinated through the Mental Health Centre North Zealand, Psychiatric Research Unit, Capital Region of Denmark.

12. References

The following peer-reviewed publications document the psychometric properties, theoretical rationale, and clinical applications of the Major Depression Inventory:

• Bech, P., Rasmussen, N. A., Olsen, L. R., Noerholm, V., & Abildgaard, W. (2001). The sensitivity and specificity of the Major Depression Inventory, using the Present State Examination as the index of diagnostic validity. Journal of Affective Disorders, 66(2–3), 159–164. https://doi.org/10.1016/s0165-0327(00)00309-8
• Bech, P. (2004). Measuring the dimensions of psychological general well-being by the WHO-5 and depression and anxiety by the Major Depression Inventory (MDI). Health and Quality of Life Outcomes, 2(1), 1–8. https://doi.org/10.1186/1477-7525-2-37
• Cuijpers, P., Li, J., Lau, M. A., & Bech, P. (2007). The Major Depression Inventory: Validity and measurement invariance across ethnic groups. Social Psychiatry and Psychiatric Epidemiology, 42(9), 723–729. https://doi.org/10.1007/s00127-007-0227-2
• Nielsen, M. G., Ørnbøl, E., Vestergaard, M., Bech, P., Larsen, F. B., & Christensen, K. S. (2010). The construct validity of the Major Depression Inventory: A Rasch analysis of a self-rating scale in general practice. BMC Medical Research Methodology, 10, Article 104. https://doi.org/10.1186/1471-2288-10-104
• Olsen, L. R., Jensen, D. V., Noerholm, V., Martiny, K., & Bech, P. (2003). The internal and external validity of the Major Depression Inventory in measuring severity of depressive states. Psychological Medicine, 33(2), 351–356. https://doi.org/10.1017/s0033291702006766

13. Items of the Scale