Introduction
The term depression can be used to qualify isolated depressive symptoms, which may be present in normal (as in euthymia) or associated with systemic medical conditions (such as hypothyroidism). The term depression can characterize a clinical syndrome when found several signs and symptoms of depression and mourning in depressions associated with systemic medical disorders, other psychiatric disorders or disorders arising from the use of psychoactive substances. It may also be a primary manifestation of mood, such as the depressive episode with recurrent depression, dysthymia and bipolar depression.
Although there are no physiological or biological parameters to assess the clinical manifestations of depression, rating scales are used to measure and characterize the phenomenon, that is, reflect the clinical phenomenon of objective and quantitative information. The information provided by the rating scales can be used to assist in diagnosis, to document the clinical status of depressed at any given time or to supplement patient information that has undergone a prior clinical evaluation. In general, the scales for evaluation of depressive states aim to describe the samples used patients showing symptoms present or absent in the clinical picture and evaluate changes taking place in the course of treatment. There are several rating scales, which can be self-assessment or evaluation of the observer (Table 1). In this review we will discuss some features of the scales for depression Hamilton (Hamilton, 1960) and Montgomery & Åsberg (Montgomery & Åsberg, 1979). Both have been used in research to establish clinical criteria for inclusion, development measures or recovery from a depressive episode (Snaith, 1993).
TABLE 1
Scales for Depression
Reviewer instrument Author
Rating Scale Hamilton Depression (HAM-D) Professional Hamilton, 1960
Rating Scale for Montgomery & Åsberg Depression (MADRS) Professional Montgomery & Åsberg, 1979
Melancholy Assessment Scale-deBech Rafaelsen Professional Bech and Rafaelsen 1980
Clinical Interview for Depression Professional Paykel, 1985
Self-Rating Scale for Zung Depression Self-Assessment Zung, 1965
Inventory Beck Depression Self-Assessment Beck et al, 1961
Adapted from Marder 1995
SCALE OF HAMILTON DEPRESSION RATING (HAM-D)
The Rating Scale for Depression Hamilton has been developed for over 40 years, but has kept its scale position administered by the researcher most widely used worldwide. It serves as a comparative scale standard for other developed more recently. Hedlung and Vieweg (1979) reviewed 23 studies comparing the total scores of the HAM-D scores of the scales of Beck and Zung (both self-assessment) and also with the Clinical Global Assessment Scale and concluded that the scale HAM-D reflects consistently the modifications made by the patient during treatment. The HAM-D was developed for evaluation and quantification of depressive symptoms in patients with mood disorders. Anxiety symptoms were included, despite the negative correlation between depression and anxiety in version 1960 (Hamilton, 1960). It is recommended to another type of patient or for normal people, and does not constitute a diagnostic tool for identifying depression.
The HAM-D emphasizes somatic symptoms, which makes it particularly sensitive to changes experienced by severely depressed patients, and contributes to the spread of its use in clinical trials with antidepressants.
The HAM-D scale has 17-21 items depending on the version, or 24 items (three additional topics are helplessness, hopelessness and worthlessness). The items are evaluated in accordance with the intensity and frequency within a given period of days. The evaluation is based on interviews and information from other sources, such as medical records, nursing information and / or family members (Appendix 1). It does not have any standardized interview, but depends on the interviewer's ability to collect information and make decisions about the scores. Therefore, it should only be used by professionals with clinical experience, as it requires knowledge of how to interview depressed patients and evaluate symptoms. The author recommended that the interview lasted 30 minutes and that the evaluation was generally carried out by two researchers, one conducting the interview and the other making supplementary questions at the end. recently it designed a Structured Manual for HAM-D Interview (Williams, 1988) in order to standardize the questions of the interviewer. This manual has the advantage of reliability of individual items without significantly increasing the time to manage the scale (annex 2).
Originally, Hamilton described 21 items and subsequently reduced the scale for a version with 17 items. The four remaining daytime variation of humor, derealization, paranoid symptoms and obsessive symptoms were excluded because they did not measure depression or its intensity or because they occur with low frequency, no longer useful (Hamilton, 1960). In general, the work refers to the total score, which is the sum of all items. This total score is often used as a criterion for inclusion in research in clinical trials (Del Porto, 1989). Hamilton (1960) never quoted a score to differentiate morbidity normal, but the researchers defined their samples of depressed patients, based only on a score on the scale, despite the instruction not be used as a diagnostic tool. In general, scores below 10 were considered too light for inclusion in studies of patients with medication. The author proposed a cutoff standard, but in practice it is accepted that scores above 25 points identify severely depressed patients; between 18 and 24 points, moderately depressed patients, and between 7 and 17, slightly depressed patients.
As a result of its popularity, the Hamilton rating scale for depression has been the instrument most widely studied depression. Its validity has been demonstrated in several studies on their scores are compared between groups of patients with different severity disease and studies in which scores are compared with the overall severity of disease. Hedlund and Vieweg (1979) summarized the reliability information obtained from 9 studies. Internal consistency coefficients ranged from 0.83 to 0.94. The reliability of the scale of evaluators has been consistent over the numerous studies. Generally it is above 0.85. Although usually get a good agreement between evaluators after proper training, it is appropriate in all the studies is to evaluate the reliability among evaluators (Del Porto, 1989).
A recent assessment by Snaith (1996), about the current use of the HAM-D found that the instructions to fill it out are often not followed. The author reviewed 114 articles of 5 major psychiatric journals. It was found that 71 of them were used in some scale for depression, HAM-D and applied to 66% of the studies. It was found that the scale was not used only in cases diagnosed primary depressive disorder, as it should, but for diagnostic purposes and / or criteria for inclusion in 47% of trials. There were also some arbitrariness in determining scores to assess severity and recovery. Other limitations reviewed by Snaith (1993) refer to the fact scale lacks internal consistency and that some items are compromised by age or gender of the subject. The presence of several somatic symptoms can lead to misleading conclusions when the scale is used in patients with physical disorders.
RATING SCALE FOR DEPRESSION OF MONTGOMERY & Asberg (MADRS)
It is also a scale widely used in trials with antidepressant drugs, which has been specially developed to measure clinical changes in the course of treatment. It was drawn from the administration of psychopathological Rating Scale Summary (CPRS – "Comprehensive Psychopathological Rating Scale") in 106 patients in England and Sweden, with a diagnosis of primary depressive illness, according to the criteria of Feighner (Montgomery & Åsberg 1979). The authors selected the 17 most common items and their sum, a measure of the severity of the disease, had a high correlation with the scores of the Hamilton scale and global clinical evaluation. From these 17 items were selected the 10 that showed a higher correlation with treatment. These 10 items are the final scale (Annex 3).
The MADRS differs from the HAM-D because it does not include somatic or psychomotor symptoms. However, evaluates some of the main symptoms of depressive disorder, such as sadness, decreased sleep, lassitude, pessimism, and suicidal thoughts. Your items include biological, cognitive, affective and behavioral. It contains all the essential symptoms of depression is highly apparent validity (validity face). Its validity has been demonstrated by their high correlation with the HAM-D (Kaplan and Sadock, 1995) and several studies show that the scale is valid and reliable when applied to clinical interviewer (Craighead and Evans, 1996). It has also been shown to be valid and reliable when applied by trained nurses and based on observations and interactions with patients admitted to psychiatric units. Their ability to differentiate changes between patients who respond to antidepressant treatments were those who do not respond better than the HAM-D.
Craighead and Evans (1996) reported that, because it is a scale designed to measure the effect of treatment and the degree and speed of change may vary between symptoms, the use of scores from a subscale of the MADRS is necessary. Also because some items (for example, internal tension, pessimistic and suicidal thoughts) are not reliable when compared individually and when you want to compare different treatment modalities, it may be useful to determine if there are specific differences in the scores and symptom type. To this, the authors studied 340 adult patients admitted to a Affective Disorders Program and carried out the factorial analysis of the MADRS. Identified four factors labeled as pessimistic cognitive, affective, congnitivo anxious and vegetative. The first reflects the patient's account of sadness, pessimism, guilt thoughts and the various levels of suicidal thinking and planning. The second factor, the cohesion of the load behavior of lack of affection. The patient is assessed by looking sad, discouraged and unhappy, the loss of interest in things around him and the difficulty in initiating activities, even routine. Cognitive anxious factor reflects the combination of concentration problems and tension that characterize anxious depression. The vegetative symptoms, sleep disturbances and appetite, are the fourth factor.
COMPARISON BETWEEN SCALES
Depression severity can be assessed by the patient for scales developed for this purpose. The Visual Analog Mood Scale (VAMS) (Folstein and Luria, 1973), for example, is a self-rating scale, in which the patient seeks to locate your mood between the poles "best I've ever felt" and "worse than ever I felt. "
Inasmuch as the three scales (HAM-D and MADRS VAMS) have the same purpose – to evaluate the intensity of depression – Dratcu et al. (1985) found that the assessment made by one of them finds correspondence in another, or if the change in one is reflected in the variation observed in the other. also analyzed the clinical utility of the three scales checking if the result of evaluating the intensity of symptoms for these instruments – or severity of depression – is related to the option for certain medical management. Reassessing patients diagnosed as suffering from depression for several criteria, they found 28 (60%) who met criteria for Major Depressive Disorder, established by the "Research Diagnostic Criteria" (RDC) (Spitzer et al., 1980). They applied the three scales (HAM-D, MADRS and VAMS) and analysis of the results showed correlation between them. Patients admitted in hospitalization regime had significantly higher average points in HAM-D and MADRS scales than the average of the other groups (admitted to partial hospitalization or outpatient). Thus, it is once again demostrada the usefulness of these scales to evaluate the severity of depression by reflecting the judgment and clinical management.
The significant positive correlation between the ratings by the HAM-D and MADRS scale indicated that both measure the intensity of depressive symptoms in the same direction, that is, the higher the number of points greater severity. Both, therefore, are also useful tools for assessing the severity of depressive disorders. Some differences between them, however, seem important. HAM-D consists of 17 items, while the MADRS has only 10, which makes it appear easier. Furthermore, each of the ten items MADRS allows a maximum of six points, with intermediate points among those that describe the manifestation of symptoms; the interviewer thus has greater flexibility to evaluate and can observe the intensity of symptoms that do not correspond exactly to the descriptions. Another important difference lies in the distinction that provides MADRS between the reported symptoms and observed, avoiding questions that may damage assessment. Finally, the MADRS has no contradictory symptoms, as some found in the HAM-D (p. Ex., Retardation and agitation) and make it virtually impossible for the maximum score of 52 points. The labor Dratcu et al. (1985), reviewed the maximum reached by the HAM-D point was 41 (79% of the maximum possible), while the maximum reached by the MADRS rating was 55 (92% of the maximum possible). This suggests that the MADRS characteristics may render it suitable for a reflection intensity of the symptoms in cases of depression. One can also assume that the MADRS be more sensitive to subtle changes in symptomatology and early, which become able to evaluate the response to treatment with greater accuracy and advance, especially in severe cases. The authors conclude that the comparative study between the two scales suggests that the MADRS seems to be more comprehensive and sensitive to the HAM-D, assess the severity of depressive symptoms.
THE EXPERIENCE OF GRUDA
In psychopharmacology trials carried out in Affective Diseases Group (GRUDA) of the Institute of Psychiatry, Hospital of the Faculty of Clinical Medicine, University of São Paulo used the HAM-D and MADRS scales in projects to evaluate efficacy and tolerability of antidepressants. In a study evaluating efficacy and tolerability of nortriptyline compared with amitriptyline was used HAM-D Scale 21 items as inclusion criteria (score less than 18, that is, moderate to severe depression) in patients previously diagnosed according to the criteria of DSM III-R. The scores of the HAM-D scale 21 items showed that both treatment groups had the same performance (F = 0.108 – ns), significantly acting (F = 56.557 – p <0.001) after the first week of drug therapy (Moreno, 1998 ).
In the comparative study, double-blind, randomized, venlafaxine versus amitriptyline in the treatment of outpatients with major depression according to the criteria of DSM IV, the HAM-D scale 21 items was used as an inclusion criterion for the minimum score of 20 and both HAM-D and MADRS as measure of efficacy (Moreno, 1998). In another ongoing study to compare the efficacy of mirtazapine and fluoxetine in severely depressed patients for 8 weeks of treatment, it is using the HAM-D 17 items (minimum score of 25, that is, severe depression) as inclusion criteria and measuring effectiveness and MADRS only as measure of efficacy (Moreno, 1998) in depressed previously diagnosed by DSM-IV. Patients who experience decrease in HAM-D scale score after the washout period (a week) are excluded from the study because they were considered responsive to placebo.
For each study was performed after discussion of each item of the scale, an inter-rater reliability training, with the presence of researchers from all the centers involved. We recommend that this practice be followed whenever you use the scales in studies involving various researchers and especially when there are different centers researchers, in order to homogenize possible conceptual differences of psychopathology. Items less goals such as depressed mood and feelings of guilt of the HAM-D, tend to differences in scores. It is also important to standardize the criteria used for the quantification of symptoms present.
A practice widely used in psychopharmacology studies of antidepressants is to consider effective a substance that reduces by at least 50% of the initial score of the scales. Thinking hypothetically in severe depressed with initial score between 25 and 30 by the HAM-D, 50% reduction in score means only partially effective and no remission or recovery of the depressive episode from a clinical point of view. This fact should be considered when interpreting the results.