The MADRS is a clinician-rated depression severity scale that Montgomery and Åsberg built in 1979 by keeping only the CPRS items most sensitive to change during treatment, and it remains a frequent outcome measure in antidepressant clinical trials, particularly European ones.
Its psychometric record is strong: high internal consistency, high inter-rater agreement even when psychiatrists are paired with general practitioners or nurses, strong convergence with the Hamilton scale, and roughly twice the Hamilton's measurement precision at average severity.
A trained clinician rates symptoms from a semi-structured interview covering the past week; totals are interpreted against published severity bands, and anchor-based thresholds define minimal and substantial clinically meaningful improvement. A self-rating companion, the MADRS-S, agrees closely with the interview version.
The scale is optimized for detecting treatment change rather than for diagnosis, it requires rater training for reliable use, and its suicidal-thoughts item obliges researchers and clinics to have an explicit safety protocol.
15-20 minutes, semi-structured clinical interview by a trained clinician
Response format
Each item rated 0-6 (7-point scale) with descriptive anchors; ratings cover the past week; total 0-60
Scores
Single total score (0-60); research also supports four correlated symptom factors (Quilty et al., 2013)
Validated populations
Adults, 18 and older (development sample ages 18-69; later validation samples up to 75)
License
Copyright held by the publisher of the British Journal of Psychiatry (Royal College of Psychiatrists). Free for individual research and clinical use with attribution; written permission required for commercial use and large-scale reproduction
Original citation
Montgomery & Åsberg (1979), British Journal of Psychiatry
Introduction
The Montgomery-Åsberg Depression Rating Scale (MADRS) is a ten-item clinician-rated measure of depression severity developed in 1979 by the British psychiatrist Stuart Montgomery and the Swedish psychiatrist Marie Åsberg. Its ten items were selected from the 65-item Comprehensive Psychopathological Rating Scale (CPRS) as those most sensitive to change with treatment, and the scale was designed from the outset to detect treatment-induced change in depression rather than to sample the syndrome exhaustively (Montgomery & Åsberg, 1979).
A clinician rates each of the ten items from 0 to 6 on the basis of a semi-structured interview covering the previous week, and the item ratings are summed to a total between 0 and 60. Item selection was empirical: ratings from combined English (n = 54) and Swedish (n = 52) patient samples identified the CPRS items that changed most during antidepressant treatment, a bilingual derivation intended to reduce cultural bias (Montgomery & Åsberg, 1979). In the development study, the resulting ten-item scale differentiated treatment responders from non-responders better than the Hamilton Rating Scale (point-biserial r = 0.70 vs 0.59; Montgomery & Åsberg, 1979).
Theoretical Foundation
The MADRS reflects a measurement philosophy rather than a theory of depression: the criterion for including an item was demonstrated sensitivity to change, not diagnostic coverage. Compared with the Hamilton Depression Rating Scale, the resulting item set places less weight on somatic symptoms and applies one uniform 0-6 response scale to every item (Carmody et al., 2006). Later psychometric work bears on how coherent this item set is: item-response-theory analyses found the scale unifactorial in two outpatient samples (Carmody et al., 2006), while confirmatory factor analysis supported four correlated symptom factors under a higher-order general depression factor (Quilty et al., 2013).
📏 Key insight: The MADRS inverted the usual logic of depression scale construction; instead of sampling the syndrome broadly, it kept only the CPRS items that moved most when patients improved, making sensitivity to change the criterion of item selection (Montgomery & Åsberg, 1979).
Key Features
Assessment Characteristics
10 clinician-rated items selected from the CPRS for sensitivity to treatment change (Montgomery & Åsberg, 1979)
Uniform 0-6 rating per item with descriptive anchors; total score 0-60
15-20 minute semi-structured interview covering symptoms over the past week
Adult populations (18 and older)
Free for individual research and clinical use with attribution; permission required for commercial reproduction
Dimensions Assessed
Apparent sadness – observable signs of depression during the interview
Reported sadness – the patient’s subjective experience of depressed mood
Inner tension – feelings of ill-defined discomfort, edginess and turmoil
Reduced sleep – shortened or shallow sleep relative to the patient’s usual pattern
Reduced appetite – loss of appetite relative to when well
Concentration difficulties – difficulty collecting one’s thoughts and sustaining attention
Lassitude – difficulty getting started and slowness in initiating activities
Inability to feel – reduced interest and blunted emotional response (anhedonia)
Pessimistic thoughts – guilt, self-reproach and negative expectations
Suicidal thoughts – from feeling life is not worth living to explicit plans and preparations
Versions & Adaptations
Original clinician-rated MADRS, 10 items (Montgomery & Åsberg, 1979)
MADRS-S, the self-rating version derived from the CPRS self-rating program, scored 0-3 in half-steps rather than the clinician scale’s 0-6 (Svanborg & Åsberg, 1994)
SIGMA, a structured interview guide for administering the clinician MADRS (Williams & Kobak, 2008)
No validated child or adolescent form was identified during verification; the validation samples reviewed here were adults
Research Applications
Antidepressant trials – used frequently in European registration and other clinical trials as an outcome measure (Carmody et al., 2006)
Treatment monitoring – tracking symptom change over the course of therapy, the purpose the scale was built for
Measurement research – the standard comparator in psychometric studies of the Hamilton Depression Rating Scale (Carmody et al., 2006)
Patient-reported monitoring – longitudinal follow-up with the MADRS-S self-rating version (Svanborg & Åsberg, 1994)
Administer the ten-item clinician-rated depression assessment used frequently in antidepressant clinical trials.
Scoring and Interpretation
Response Format
Each item is rated by the clinician on a 7-point scale (0-6) on the basis of a semi-structured clinical interview. Ratings reflect symptom severity over the previous week, and descriptive anchors are provided at alternate scale points (0, 2, 4, 6) to support consistent scoring across raters (Montgomery & Åsberg, 1979).
Illustrative Items
The full scale, with anchor definitions for all ten items, is printed in the appendix of Montgomery and Åsberg (1979). Three items illustrate the anchor structure:
Item 1: Apparent Sadness (0-6 scale)
0: No sadness
2: Looks dispirited but does brighten up without difficulty
4: Appears sad and unhappy most of the time
6: Looks miserable all the time, extremely despondent
Item 7: Lassitude (0-6 scale)
0: Hardly any difficulty in getting started, no sluggishness
2: Difficulties in starting activities
4: Difficulties in starting simple routine activities which are carried out with effort
6: Complete lassitude, unable to do anything without help
Item 10: Suicidal Thoughts (0-6 scale)
0: Enjoys life or takes it as it comes
2: Weary of life, only fleeting suicidal thoughts
4: Probably better off dead, suicidal thoughts are common, suicide considered as a possible solution but without specific plans or intention
6: Explicit plans for suicide when there is an opportunity, active preparations for suicide
Scoring Procedure
Rate each of the 10 items from 0 to 6 on the basis of the interview, using the published anchor definitions.
Sum the item ratings to obtain a total score between 0 and 60.
Interpret the total against the published severity bands below, and evaluate change against the response and remission conventions used in treatment research.
Severity Interpretation
Total Score
Severity Band
Source
0-6
Absent / normal
Snaith et al. (1986)
7-19
Mild depression
Snaith et al. (1986)
20-34
Moderate depression
Snaith et al. (1986)
35-60
Severe depression
Snaith et al. (1986)
No normative means and standard deviations for general-population samples were verified for this page; interpretation rests on the published severity bands above and the treatment-change criteria below, not on population norms.
Treatment Response Criteria
Response: ≥50% reduction from the baseline score, the standard convention in antidepressant trials
Remission: total score ≤10, the conventional trial cut-off (some studies use ≤8)
Clinically meaningful change: a 6-point improvement, anchored to a 1-point improvement on the Clinical Global Impression-Severity scale (Turkoz et al., 2021)
Clinically substantial change: a 12-point improvement, anchored to a 2-point CGI-S improvement (Turkoz et al., 2021)
Suicide risk: any score above 0 on Item 10 warrants immediate clinical attention
Research Evidence and Psychometric Properties
Reliability Evidence
Internal consistency: Cronbach’s α = 0.90 (Study 1, n = 233) and 0.92 (Study 2, n = 985) in nonpsychotic major-depression outpatients, the highest of the measures compared with the Hamilton scale (Carmody et al., 2006)
Item-total correlations: median 0.75 (Study 1) and 0.78 (Study 2), with most items correlating ≥0.60 with the total score (Carmody et al., 2006)
Inter-rater reliability: correlations of 0.89-0.97 for total and difference scores across rater pairs, including a psychiatrist paired with a general practitioner or a nurse (Montgomery & Åsberg, 1979)
Structured administration: the SIGMA interview guide was developed to standardize administration, with good to excellent inter-rater reliability for most items (Williams & Kobak, 2008)
Validity Evidence
Convergent validity: MADRS and HRSD-17 total scores correlated r = 0.88 (Study 1) and 0.92 (Study 2) at study exit (Carmody et al., 2006)
Severity estimation: scores correlated significantly with the Hamilton Rating Scale and with global severity ratings in the development study, supporting the scale as a general severity estimate (Montgomery & Åsberg, 1979)
Response discrimination: the ten-item scale differentiated treatment responders from non-responders better than the Hamilton scale (point-biserial r = 0.70 vs 0.59; Montgomery & Åsberg, 1979)
Content specificity: less somatic emphasis than the Hamilton scale, and unifactorial in both study samples, consistent with a purer depression severity measure (Carmody et al., 2006)
Factor Structure
The evidence on dimensionality is mixed, and both findings are compatible with using the total score:
Unifactorial in IRT analyses: principal-components and item-response-theory analyses found the MADRS to load on a single depression factor in two outpatient samples (Carmody et al., 2006)
Four-factor hierarchical model: confirmatory factor analysis supported four correlated factors – Sadness (items 1, 2), Negative Thoughts (items 9, 10), Detachment (items 6, 7, 8) and Neurovegetative symptoms (items 3, 4, 5) – loading on a higher-order general depression factor (Quilty et al., 2013)
Invariance: the four-factor structure showed configural and metric invariance across time and full (strict) invariance across gender, supporting use of both the total score and the subscales (Quilty et al., 2013)
Treatment Sensitivity
Superior change measurement: the development study found the scale differentiated responders from non-responders better than the Hamilton scale, and the IRT comparison concluded the MADRS “would be superior to the HRSD17 in the conduct of clinical trials” (Montgomery & Åsberg, 1979; Carmody et al., 2006)
Greater precision: about twice the measurement precision of the HRSD-17 at average levels of depression severity (Carmody et al., 2006)
Anchored change thresholds: 6-point and 12-point improvements correspond to 1-point and 2-point improvements on the CGI-S, giving trialists empirically anchored definitions of meaningful and substantial change (Turkoz et al., 2021)
Self-Report Version (MADRS-S)
Agreement with the clinician scale: self-rated and interview-based MADRS scores correlated r = 0.90 (first occasion) and 0.93 (second occasion) in depressed patients, with correlations of 0.80-0.94 across depression and anxiety groups (Svanborg & Åsberg, 1994)
Usage Guidelines and Applications
Primary Applications
Outcome measurement in antidepressant efficacy trials, where the scale is used frequently in European registration and other clinical trials (Carmody et al., 2006)
Quantifying symptom change during treatment in research and clinical settings
Patient-reported symptom monitoring between clinician assessments with the MADRS-S (Svanborg & Åsberg, 1994)
Research Design Considerations
Entry thresholds: a MADRS total of at least 20 (moderate-to-severe depression) is commonly required for clinical-trial entry (Quilty et al., 2013)
Define endpoints in advance: report the response (≥50% reduction) and remission (≤10) definitions used, since conventions vary across studies
Rater training: plan clinician training and inter-rater calibration; the SIGMA structured interview guide supports consistent administration (Williams & Kobak, 2008)
Clinician and self-report forms differ: the MADRS-S uses a different response format (0-3 in half-steps) and should be reported as a distinct instrument, not as an interchangeable substitute (Svanborg & Åsberg, 1994)
Cultural Considerations
Items were selected from combined English and Swedish patient samples specifically to reduce cultural bias in item content (Montgomery & Åsberg, 1979)
The four-factor structure was invariant across male and female depressed outpatients (Quilty et al., 2013); broader cross-language and cross-cultural validation evidence was not reviewed for this page, so check local validation work before use in translation
Limitations and Cautions
Clinician-administered: reliable use requires a trained rater conducting a clinical interview; it is not a self-administered questionnaire in its original form
Time investment: the 15-20 minute interview may limit use in high-volume settings
Narrow focus: optimized for detecting treatment change rather than comprehensive symptom assessment or diagnosis
Safety protocol required: the suicidal-thoughts item can elicit risk-relevant information, and studies and clinics need an explicit procedure for responding to elevated scores
Copyright and Usage Responsibility: Check that you have the proper rights and permissions to use this assessment tool in your research. This may include purchasing appropriate licenses, obtaining permissions from authors/copyright holders, or ensuring your usage falls within fair use guidelines.
The Montgomery-Åsberg Depression Rating Scale was originally published in 1979 in the British Journal of Psychiatry, a journal of the Royal College of Psychiatrists, which holds the publication copyright. Individual researchers and clinicians may use the scale for their own research and clinical practice; written permission from the Royal College of Psychiatrists is required for commercial use, large-scale distribution, and reproduction beyond individual use.
Proper Attribution: When using or referencing this scale, cite the original development:
Montgomery, S. A., & Åsberg, M. (1979). A new depression scale designed to be sensitive to change. British Journal of Psychiatry, 134(4), 382-389. https://doi.org/10.1192/bjp.134.4.382
Montgomery, S. A., & Åsberg, M. (1979). A new depression scale designed to be sensitive to change. British Journal of Psychiatry, 134(4), 382-389. https://doi.org/10.1192/bjp.134.4.382
Structured Interview Guide (SIGMA):
Williams, J. B. W., & Kobak, K. A. (2008). Development and reliability of a structured interview guide for the Montgomery-Åsberg Depression Rating Scale (SIGMA). British Journal of Psychiatry, 192(1), 52-58. https://doi.org/10.1192/bjp.bp.106.032532
Psychometric Comparison with the Hamilton Scale:
Carmody, T. J., Rush, A. J., Bernstein, I., Warden, D., Brannan, S., Burnham, D., Woo, A., & Trivedi, M. H. (2006). The Montgomery-Åsberg and the Hamilton ratings of depression: A comparison of measures. European Neuropsychopharmacology, 16(8), 601-611. https://doi.org/10.1016/j.euroneuro.2006.04.008
Factor Structure:
Quilty, L. C., Robinson, J. J., Rolland, J.-P., De Fruyt, F., Rouillon, F., & Bagby, R. M. (2013). The structure of the Montgomery-Åsberg depression rating scale over the course of treatment for depression. International Journal of Methods in Psychiatric Research, 22(3), 175-184. https://doi.org/10.1002/mpr.1388
Severity Interpretation:
Snaith, R. P., Harrop, F. M., Newby, D. A., & Teale, C. (1986). Grade scores of the Montgomery-Åsberg Depression and the Clinical Anxiety Scales. British Journal of Psychiatry, 148, 599-601.
Self-Report Version (MADRS-S):
Svanborg, P., & Åsberg, M. (1994). A new self-rating scale for depression and anxiety states based on the Comprehensive Psychopathological Rating Scale. Acta Psychiatrica Scandinavica, 89(1), 21-28.
Treatment Sensitivity and Change Thresholds:
Turkoz, I., Alphs, L., Singh, J., Jamieson, C., Daly, E., Shawi, M., Sheehan, J. J., Trivedi, M. H., & Rush, A. J. (2021). Clinically meaningful changes on depressive symptom measures and patient-reported outcomes in patients with treatment-resistant depression. Acta Psychiatrica Scandinavica, 143(3), 253-263. https://doi.org/10.1111/acps.13260
Related Assessments: HDRS: Hamilton Depression Rating Scale and BDI: Beck Depression Inventory (links added when their pages go live)
A burdened donkey trudging through fog — symbolizing heaviness, exhaustion, and depressive symptoms measured by the MADRS (Montgomery-Åsberg Depression Rating Scale)
Frequently Asked Questions
What does the MADRS measure?
Depression severity, with a design emphasis on detecting treatment-induced change. Its 10 items cover apparent sadness, reported sadness, inner tension, reduced sleep, reduced appetite, concentration difficulties, lassitude, inability to feel, pessimistic thoughts, and suicidal thoughts. The items were selected from the Comprehensive Psychopathological Rating Scale as those most sensitive to change during antidepressant treatment (Montgomery & Åsberg, 1979).
Who administers the MADRS and how long does it take?
A trained clinician rates the 10 items from a semi-structured interview lasting about 15-20 minutes, covering symptoms over the previous week. In the development study, agreement remained high even when a psychiatrist was paired with a general practitioner or a nurse (Montgomery & Åsberg, 1979).
Is the MADRS free to use?
Individual researchers and clinicians may use it for their own research and clinical practice with attribution to Montgomery and Åsberg (1979). Written permission from the Royal College of Psychiatrists is required for commercial use, large-scale distribution, and reproduction beyond individual use.
How is the MADRS scored and interpreted?
Each item is rated 0-6 and the ratings are summed to a total of 0-60. Published severity bands are 0-6 (absent/normal), 7-19 (mild), 20-34 (moderate), and 35-60 (severe) (Snaith et al., 1986). In treatment research, response is conventionally a reduction of at least 50% from baseline and remission a total of 10 or below.
What counts as a clinically meaningful change on the MADRS?
An anchor-based study in treatment-resistant depression found that a 6-point improvement corresponds to a 1-point improvement on the Clinical Global Impression-Severity scale (clinically meaningful change), and a 12-point improvement to a 2-point CGI-S improvement (substantial change) (Turkoz et al., 2021).
How does the MADRS differ from the Hamilton Depression Rating Scale?
It is shorter (10 items vs 17), uses one uniform 0-6 scale for every item, and places less weight on somatic symptoms. In a direct comparison it showed higher internal consistency (α = 0.90-0.92, the highest of the measures compared), correlated r = 0.88-0.92 with the Hamilton total, and was about twice as precise at average severity levels (Carmody et al., 2006).
Is there a self-report version?
Yes. The MADRS-S is a self-rating version scored 0-3 in half-steps rather than the clinician scale's 0-6. Self-rated and interview-based scores correlated r = 0.90 and 0.93 across two occasions in depressed patients (Svanborg & Åsberg, 1994).
Frequently Asked Questions
What does the MADRS measure?
Depression severity, with a design emphasis on detecting treatment-induced change. Its 10 items cover apparent sadness, reported sadness, inner tension, reduced sleep, reduced appetite, concentration difficulties, lassitude, inability to feel, pessimistic thoughts, and suicidal thoughts. The items were selected from the Comprehensive Psychopathological Rating Scale as those most sensitive to change during antidepressant treatment (Montgomery & Åsberg, 1979).
Who administers the MADRS and how long does it take?
A trained clinician rates the 10 items from a semi-structured interview lasting about 15-20 minutes, covering symptoms over the previous week. In the development study, agreement remained high even when a psychiatrist was paired with a general practitioner or a nurse (Montgomery & Åsberg, 1979).
Is the MADRS free to use?
Individual researchers and clinicians may use it for their own research and clinical practice with attribution to Montgomery and Åsberg (1979). Written permission from the Royal College of Psychiatrists is required for commercial use, large-scale distribution, and reproduction beyond individual use.
How is the MADRS scored and interpreted?
Each item is rated 0-6 and the ratings are summed to a total of 0-60. Published severity bands are 0-6 (absent/normal), 7-19 (mild), 20-34 (moderate), and 35-60 (severe) (Snaith et al., 1986). In treatment research, response is conventionally a reduction of at least 50% from baseline and remission a total of 10 or below.
What counts as a clinically meaningful change on the MADRS?
An anchor-based study in treatment-resistant depression found that a 6-point improvement corresponds to a 1-point improvement on the Clinical Global Impression-Severity scale (clinically meaningful change), and a 12-point improvement to a 2-point CGI-S improvement (substantial change) (Turkoz et al., 2021).
How does the MADRS differ from the Hamilton Depression Rating Scale?
It is shorter (10 items vs 17), uses one uniform 0-6 scale for every item, and places less weight on somatic symptoms. In a direct comparison it showed higher internal consistency (α = 0.90-0.92, the highest of the measures compared), correlated r = 0.88-0.92 with the Hamilton total, and was about twice as precise at average severity levels (Carmody et al., 2006).
Is there a self-report version?
Yes. The MADRS-S is a self-rating version scored 0-3 in half-steps rather than the clinician scale's 0-6. Self-rated and interview-based scores correlated r = 0.90 and 0.93 across two occasions in depressed patients (Svanborg & Åsberg, 1994).
