Cytokines and Depression

How your immune system causes depression

by Ronald S. Smith

Chapter 6

Depression

After a century of psycho-biomedical research, depression remains an enigma. The cause or causes of depression are a mystery and this is after 100 years of intensive investigation!! The basic pathophysiology (i.e. the disease in terms of chemical-biological dysfunctions) is poorly understood. There are no known physical tests which can be used to positively diagnose depression or to distinguish depression from other diseases. Except for some vague, general notions, there are no proven methods for preventing this mood disorder.

Depression can occur alone or in conjunction with many other diseases. Persons with organic brain pathologies, such as, dementia, multiple sclerosis, Parkinson's disease or stroke have high rates of depression. Other serious maladies, including cancer, heart disease, AIDS, hepatitis, infectious of all types, rheumatoid arthritis, Lyme disease, irritable bowel syndrome, ulcers, schizophrenia, obsessive compulsive disorder and lupus are frequently compounded with depression.

Depression is unique in its commingling with so many human pathologies. In addition, many symptoms of depression are identical to numerous common symptoms of many other maladies. Fatigue, for example, is a symptom of hundreds of physical diseases, yet it is also a pervasive and diagnostically important symptom of depression. Excessive sleep and weight loss are typical signs of many physical ailments and also of depression. Recently, acute phase proteins have been found in the blood of depressed patients. Acute phase proteins are a standard feature in the blood of patients with infectious diseases and other acute ailments. There is a long list of similarities between depression and many, many other diseases.

Diagnosing Depression: The Diagnostic & Statistical Manual (DSM)

From 1930 to 1952 there was very little consistency in diagnosing mental disorders. Diagnosing patients was often based on the psychiatric theories or models which the individual psychiatrist or psychologist subscribed to. The lack of diagnostic uniformity created many barriers to progress in psychiatric research. Often patients classified as depressed in one research paper were very different from patients classified as depressed in another research paper. Thus, experimental results were unreliable because of inconsistent patient diagnoses.

Over 50 years ago the World Health Organization developed guidelines for diagnosing mental disorders. A few years later the American Psychiatric Association set up a committee to establish consistent, uniform criteria for diagnosing depression and other 'mental ailments'. Growing out of these efforts are the two main classification systems used today. The oldest and most widely used outside the United States is the World Health Organization's International Classification of Diseases: Mental Disorders Ninth Revision (ICD-9 ). The standard system for diagnosis within the United States is the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). We will devote our attention to the American Psychiatric Association's DSM system.

The Diagnostic and Statistical Manual of Mental Disorders, 1st Edition, (DSM-I) was first published in 1952. Its classification system was primarily based on Adolf Meyer's reaction theory of mental illness, namely that mental illnesses were reactions of the personality to psychological, social and biological factors. A few years later, the diagnosis and classification of mental illnesses in the United States were modestly changed with the publication of the Second Edition of the DSM(DSM-II).

DSM-II was replaced by DSM-III in 1980. DSM-III completely revised the criteria for classifying mental pathologies. The new manual no longer defined depression and other mental disorders based on theoretical models of psychiatric diseases. Instead, the DSM-III committee used common clusters of symptoms found in patients to define and classify various mental diseases. This is very difficult to do, since there are many overlapping symptoms when comparing patients with differing diagnoses. For example, anxiety disorders and depressive disorders are classified as separate distinct diseases in DSM-III, but many patients diagnosed with depression also suffer from anxiety and vice versa. Consequently, do these patients have one disease (depression), two diseases (depression and anxiety) or some kind of hybrid disease (depressive anxiety)?

For schizophrenia, the dilemma of classification is just as great. In addition to having hallucinations and delusions at some time during their illness, a very high percentage of patients with schizophrenia exhibit intense anxiety and severe depressive symptoms. The key, defining symptoms of schizophrenia are hallucinations and delusions, but how do hallucinations and delusions convert anxiety and depression into completely different diseases? Would it be better to say a patient displaying hallucinations, delusions, anxiety and depression has three diseases at once, namely, schizophrenia, anxiety and depression, rather than just schizophrenia? These questions are impossible to answer because we have no fundamental scientific understanding of anxiety, depression, hallucinations, delusions or schizophrenia. These and other fundamental diagnostic dilemmas still persist and are debated in the professional medical literature.

Another major change from DSM-II to DSM-III was the elimination of the term neurosis as a diagnostic category. For eighty years prior to DSM-III, neuroses and neurotic disorders had been a standard part of psychiatric and clinical psychological practice . Millions of patients had spent billions of dollars for treatment of their neurotic disorders. The DSM-III committee changed all that. They threw out all neurotic disorders. Millions of neurotics suddenly lost their maladies, but alas, they were quickly replaced with new committee designated diseases.

In 1987, DSM-III was made obsolete by the publication of DSM-III-R, a revised edition of DSM-III. DSM-III-R was an expanded and modestly altered version of DSM-III. In 1994, DSM-IV was published and it is the principal diagnostic manual in use today in the United States.

DSM-IV is a massive document, being 886 pages long. As usual, it was a team effort, with over one thousand professionals involved in its preparation. Compared to previous editions, it is significantly expanded, altered and updated. There is extensive scientific documentation underlying DSM-IV, which can be found in the huge five volume set called the DSM-IV Sourcebook. Nevertheless, except for substance-abuse related disorders, there is no suggestion that the cause or causes of psychiatric diseases are known or understood. No claims are made about the basic pathophysiology (i.e. chemical-biological mechanisms) of mental disorders. In DSM-IV, as in previous manuals, diagnosis is based on a collection of symptoms and behaviors. There are no objective physical measurements that can be used to give a definitive diagnosis of any mental disorder. No blood tests, urine tests, biopsies or radiological measurements are used to diagnose these disorders.

DSM-IV has many other limitations. A fundamental one has been expressed by the DSM-IV committee itself. In the Introduction, DSM-IV makes this remarkable disclaimer, "In DSM-IV, there is no assumption that each category of mental disorder is a completely discrete entity with absolute boundaries dividing it from other mental disorders or from no mental disorder. There is also no assumption that all individuals described as having the same mental disorder are alike in all important ways. The clinician using DSM-IV should therefore consider that individuals sharing a diagnosis are likely to be heterogeneous even in regard to the defining features of the diagnosis...." There you have it. The skeptical assessment of DSM-IV's classification system in the very words of the one thousand distinguished psychiatrists who produced DSM-IV!

DSM-IV and Depression Depressive disorders and bipolar disorders (manic-depressive disorders) are put under the general classification of Mood Disorders (see Figure 1). We will not discuss bipolar disorders in this chapter. Major Depressive Disorder, Dysthymic Disorder and Depressive Disorder NOS (Not Otherwise Specified) are the three main subclasses of depressive disorders, with major depressive disorder being the most severe and dysthymic disorder being a chronic, but usually milder condition. Under major depressive disorder are two types, single episode and recurrent episodes. Within the classification of Depressive Disorder NOS are six other depressive disorders which don't meet the criteria for Major Depressive or Dysthymic Disorders. Thus, in all there are nine different depressive disorders classified by the DSM-IV committee under the single heading of Depressive Disorders.

For each of the nine different depressive disorders there are many different subtypes. DSM-IV uses specifiers to indicate the subtypes. For example, under the diagnosis of major depressive disorder, recurrent are a variety of specifiers:

• Mild, moderate or severe without psychotic features
• Severe with psychotic features
• In partial remission
• In full remission
• Chronic
• With catatonic features
• With melancholic features
• With atypical features
• With postpartum onset
• With or without full interepisode recovery
• With seasonal pattern

Thus, there are eleven subtypes for the diagnosis of major depressive disorder, recurrent. A similar pattern of subtypes is used for the other eight depressive disorders, which results in up to 70 different depressive subtypes. In other words, the DSM-IV says that there are 70 different kinds of depression!

There are even more depression linked diagnostic categories. For example, bipolar disorders have depression as part of their diagnostic criteria. According to DSM-IV, the following disorders have conspicuous depressive symptoms:

• Bipolar Disorders
• Mood Disorder due to a General Medical Condition
• Substance-Induced Mood Disorder
• Schizoaffective Disorder
• Schizophrenia
• Delusional Disorder
• Psychotic Disorder Not Otherwise Specified
• Dementia With Depressed Mood
• Adjustment Disorder With Depressed Mood
• Adjustment Disorder With Mixed Anxiety and Depressed Mood

You can see that there is nothing simple (or clear) about DSM-IV's notions on depression. There are a plethora of types and subtypes of depression along with many amalgamated varieties.

DSM-IV's Basic Criteria for Depression

Even though DSM-IV classifies 70 kinds of depression, there are two 'core' types which capture the most basic and general notions on depression. They are Major Depressive Disorder and Dysthymic Disorder.

Major Depressive Disorder In order to be diagnosed with a Major Depressive Disorder, the patient must be experiencing a Major Depressive Episode. The DSM-IV committee decided on nine symptoms to diagnose a major depressive episode. According to the committee's rules, an individual only needs to have five (or more) of the symptoms concurrently for a two week period to qualify for a major depressive episode. Curiously, depressed mood, according to DSM-IV, is not a necessary symptom for this severe form of depression. Lack of interest or pleasure in things can substitute for depressed mood. Furthermore, in children or adolescents, irritability can substitute for depressed mood. Why irritability can replace depressed mood in children and adolescents, but not in adults remains a perplexing committee rule, since irritability (i.e. bad mood, quick temper, quick to anger) is an extremely common trait in depressed persons, whether young or old.

The Nine Symptoms of a Major Depressive Episode (DSM-IV):

1. Depressed mood for most of the day and nearly everyday. In children and adolescents, irritability can be used as a symptom in place of depressed mood.
2. Greatly reduced interest and pleasure in almost all activities throughout the day. Apathy. Boredom.
3. Chronic lack of energy or fatigue.
4. Poor self-esteem and/or excessive or inappropriate feelings of guilt.
5. Chronic indecisiveness or reduced ability to think and concentrate.
6. Recurrent thoughts of death or suicide.
7. Pronounced weight change, either gain or loss, without going on special diets. Or significant change in appetite, either increase or decrease, nearly everyday.
8. Either insomnia or excessive sleep (hypersomnia) nearly every day.
9. Psychomotor agitation (mental anxiety accompanied by increased movement, pacing, restlessness) or psychomotor retardation (noticeably slowed thinking and movement).

Only five (or more) of the nine symptoms are necessary for the diagnosis of major depressive episode, but the symptoms must include #1 or #2. Let's use three hypothetical patients to illustrate some peculiar aspects for the criteria of a major depressive episode.

Let's compare the Patient A seems like a classic example of serious depression, with depressed mood, fatigue, poor self-esteem and thoughts of suicide. But according to DSM-IV, patient A does not qualify for major depressive episode because he/she only exhibits four symptoms and not five. In contrast, Patient B, by exhibiting mental anxiety, increased movement, insomnia, weight loss and reduced pleasure in things, appears more anxious than seriously depressed, nevertheless Patient B fits all the DSM-IV criteria for major depressive episode.

Now, let's compare Patient B to Patient C. They exhibit distinctively opposite symptoms, with Patient B losing weight, sleepless and having psychomotor agitation, while in sharp contrast, Patient C is gaining weight, sleeping excessively and has psychomotor retardation, which is the opposite of psychomotor agitation. The two patients are remarkably dissimilar, yet according to DSM-IV, both of them qualify for a diagnosis of major depressive episode. Could psychiatry really be talking about the same disease here?

DSM-IV puts two important restrictions on the diagnosis of major depressive episode. The first restriction is: Do not include symptoms that are clearly due to a general medical condition. At first sight, this restriction may seem reasonable, but on closer examination it is patently absurd. In actual practice the statement means: If a physician believes he can determine the cause of a depressive symptom, then it is not a depressive symptom. But if a physician doesn't believe he can determine the cause of a depressive symptom, then it is a depressive symptom. Thus, by definition, major depression is a collection of symptoms of unknown cause. Once the cause is believed to be discovered, you no longer have major depression!! Clearly this is an absurd way to define a disease.

DSM-IV's other important but very similar restriction is this: The symptoms are not due to a general medical condition or to the direct physiological effects of a substance, such as a drug of abuse or a medication. This restriction is absurd also. In practice it means almost the same as the previous restriction, namely, if a physician believes he can determine the cause of the depressive symptoms, then the patient does not have depression. But, if a physician cannot determine the cause of the depressive symptoms, then the patient does have depression. Therefore, DSM-IV defines major depression as a disease of unknown cause, which is an absurd criteria for the definition of a disease.

Dysthymic Disorder The key feature of dysthymic disorder is a chronically depressed mood lasting at least two years (one year for children and adolescents). As in a major depressive episode, irritable mood can substitute for depressed mood in children and adolescents. Fewer symptoms are needed for the diagnosis of dysthymic disorder compared to major depressive disorder.

The Seven Symptoms of Dysthymic Disorder (DSM-IV)

1. Depressed mood during most of the day, on the majority of days, for at least two years. For children and adolescents, the duration is one year and irritable mood can substitute for depressed mood. Symptom #1 must be present for the diagnosis of dysthymic disorder.

In addition, two of the following symptoms must be present:

2. Poor appetite or over eating.
3. Insomnia or excessive sleep.
4. Low energy or fatigue.
5. Low self-esteem.
6. Poor concentration or indecisiveness.
7. Feelings of hopelessness.

As in major depressive episode, DSM-IV imposes the restriction that the symptoms are not due to a general medical condition or to the direct physiological effects of a substance, such as a drug of abuse or a medication. Thus, dysthymic disorder, by definition, is also a disease of unknown cause.

DSM-IV's substitution of irritability for depressed mood in children and adolescents is a vexing problem. For example, an irritable adolescent who, for one year has poor appetite and insomnia, could easily be diagnosed with dysthymic disorder. Yet if those same symptoms continued until he was 21, then according to the rules of DSM-IV, the dysthymic disorder has disappeared, even though the exact same symptoms continued! This is obviously absurd.

Associations With Depression

Emotional and Interpersonal Pathology In addition to official lists of depressive symptoms, DSM-IV lists numerous other symptoms, signs and behaviors which are closely associated with depression. Many of the associated symptoms are emotional and interpersonal. For example depression very frequently occurs with "tearfulness, irritability, anger, brooding, obsessive rumination, anxiety, phobias, panic attacks, excessive worry over physical health, difficulty in intimate relationships, less satisfying social interactions, social withdrawal and sexual dysfunction." (DSM, pg. 323 & 346)

Psychiatric Disorders Depression very often occurs with other psychiatric disorders. One common finding is so-called double depression, that is, having dysthymic disorder and major depressive disorder at the same time! Other frequent associations with depression are: substance related disorders, panic disorder, obsessive-compulsive disorder, anorexia nervosa, bulimia nervosa and borderline personality disorder.

Somatic Complaints Depressed individuals usually have a plethora of somatic (body) complaints. Headache is probably be the most common somatic problem. Headache and depression occur together so often that some headache specialists have called headache a 'mask' for depression. A number of studies have reported that over 70% of the patients consulting physicians for headache, upon further examination, have met the criteria for a depressive disorder. Other painful conditions, such as, back pain, neck pain and joint pain are extremely common with depression also. Gastrointestinal disorders are very prevalent in depressed persons. Irritable bowel syndrome, colitis, celiac disease, diarrhea, constipation, stomach ache and ulcers are some of the diseases typically correlated with depression.

Psychosocial Disease Violence, crime, poor school performance, alcoholism, homelessness, drug addictions, family violence, child abuse, chronic unemployment and mental illness are often grouped together under the broad class of diseases called psychosocial disease. Depression is a common component of most psychosocial diseases.

Physical Illness Frank physical illnesses like cancer, stroke, heart disease, atherosclerosis, kidney disease, diabetes, pneumonia, infections, brain disease, dementia and many others have extremely high rates of depression. In a study of patients hospitalized for a physical illness, 45% of them were found to be mildly to severely depressed. Over 80% of Parkinson's disease patients are depressed, as are 42% of cancer patients, 45% of stroke patients, 50% of heart attack patients and 75% of multiple sclerosis patients. Often the depressive symptoms in physically ill patients are ignored or go unrecognized by the examining physician.

Depression in physically ill persons is not merely unhappiness over their physical condition. It is a profound mood, attitude and feeling change. If depression accompanies a physical illness, then there is much greater morbidity and mortality during the next few years. Many studies have shown that physically ill patients with depression have much higher death rates than physically ill patients without depression. Even without a physical illness, depressed persons have a four fold higher rate of death than non-depressed individuals.

There are several other important findings on physical illness and depression. First, persons originally diagnosed with depression have very high rates of previously undetected physical illnesses. Second, patients with depression visit doctors and stay in hospitals much more frequently than patients without depression. Third, sometimes depression is one of the first indicators of a serious physical illness. For example, with cancer or multiple sclerosis, depression often occurs before any other symptoms or signs appear.

What Is Depression?

DSM-IV's criteria for the diagnosis of depression has not answered the question, what is depression? The criteria for the diagnosis of depression is arbitrary and provides absolutely no insight into the causes or means of prevention of depression. DSM-IV doesn't even venture a guess as to the pathophysiology of depression. It describes no objective measurements or physical tests for the diagnosis. Furthermore, depression is not just one unknown disease, but rather a committee defined collection of over 80 different types, subtypes and amalgams.

At first glance, DSM IV's criteria for diagnosing depression appears to incorporate three objectively measurable changes, namely, body weight, body movement and sleep. But let's look closer. DSM IV's rules for depression allow contrary values for each of the measurements to be used for the diagnosis depression! Concerning weight: both weight gain and weight loss are signs of depression. Concerning sleep: both excessive sleep and insomnia are signs of depression. On body movement: both psychomotor agitation (continual rapid body movement) and psychomotor retardation (slow, infrequent body movement) are signs of depression.

There are no physical diseases where opposing values of objectively measurable changes can be used to diagnose the same illness. This state of affairs never occurs with scientific definitions. A scientific definition would not allow opposing values to specify the same disease. The fact that the diagnosis of depression allows opposing values suggests that DSM IV's definition of depression is pre-scientific.

After a century of depression research on 80 types, subtypes and amalgams of depression, our scientific understanding of this moody enigma is more tentative than ever. The plethora of pathologies, signs and symptoms associated with depression have made it almost impossible to characterize the fundamental nature of depression. The four editions of the DSM since 1952 have attempted to pin down this slippery disease by promulgating rules and definitions for diagnosis. These dogmas have not solved the problem. To the contrary, they may have provided us with intellectual blinders making it appear that we know something, instead of admitting we don't.

Depression As A Multifaceted Sign The universal association of depression with almost all other diseases, whether physical, mental or psychosocial, is quite unique. There are no other diseases known to medical science that commingle with such a plethora of other pathologies. It is one of the most distinguishing and baffling characteristics of depression. Since this characteristic is not found with other diseases, it suggests that depression, with its 80 types, subtypes and amalgams, isn't really a disease after all.

If depression isn't a disease, then what is it? The answer has eluded medical science for over 100 years, but a massive amount of recent immunological evidence offers a completely new understanding of depression. The new immunological evidence supports the concept of depression as a multifaceted sign of disease, rather than a disease itself. More specifically, the model we propose in this book is that depression is a multifaceted sign of chronic immune system activation. This concept, if correct, is a fundamental revolution in our understanding of depression.

What is a sign and how does it differ from a disease? A sign is an objective indicator of disease or internal pathology, but it is not the disease itself. There are thousands of specific signs that have been observed by physicians which facilitate the diagnosis of thousands of diseases. Most of the specific signs are named after the physicians who first observed them. For example, Kehr's sign is severe pain in the left shoulder, which may indicate a ruptured spleen. The shoulder pain, obviously, is not the disease, but it is a signal or sign that the patient may have a ruptured spleen. Kehr's sign does not prove the presence of a ruptured spleen, but when combined with a variety of other indicators of ruptured spleen, a diagnosis can be made with a high degree of confidence.

There are a number of universal signs, such as fever, rash, congestion, tenderness, nausea and inflammation which are found with a tremendous number diseases. These well known universal signs are invariably caused by an activated immune system. Consequently, universal signs are actually indicators of an activated immune system, rather than indicators of a specific disease or pathology. Fever is our most widely measured universal sign of immune system activation.

Fever as a Disease Many years ago, before the age of modern medicine, fever was classified as a disease, just as depression is presently classified as a disease. Fever disease occurred with a confusing variety of other diseases just as, at the present time, "depression disease" appears with an astonishing variety of other diseases.

Over 200 years ago, in 1771, the First Edition of Encyclopaedia Britannica was published. The lengthy article on Medicine in the First Edition discussed fever diseases extensively. The 225 year old article on fever has fundamental similarities with the modern DSM IV article on depression. The most uncanny similarity is the classification of various fever subtypes, depending on associated symptoms and signs. Also, the 225 year old article in Encyclopaedia Britannica discussed "fever disease" with no understanding of fundamental nature of fever, just as depression is presented with no understanding in DSM IV.

Listed below are the five different types and 37 different subtypes of fever diseases reviewed 225 years ago in Encyclopaedia Britannica. Fever is classified and subclassified into 37 different subtypes, based on its temporal characteristics and its association with other signs, symptoms and diseases. This is remarkably similar to DSM-IV's approach for classifying and sub classifying depression.

Eventually it was understood that fever is not one disease nor 37 different kinds of fever diseases, but rather it is a trustworthy universal sign of acute immune system activation. Fever is a sign of acute immune system activation, regardless of any other signs, symptoms or diseases that it may be associated with. After this realization, fever was no longer a bewildering and complex disease, but instead, a simple, direct and easily understood signal of acute immune system activation.

In like manner depression should be understood as a multifaceted sign of chronic immune system activation instead of a collection of 80 different diseases. Using this new understanding, the key question for a patient exhibiting depression would be, "what is causing your chronic immune system activation?" This question, of course, can be extremely difficult to answer. Nevertheless, the rewards would be great because by finding the cause of a patients chronic immune system activation you would be finding the cause of the patients depression. Once the real cause is discovered, then prevention, better treatment and cure wouldn't be far behind.

Briefly, some implications of depression as a multifaceted sign. The concept of depression as a sign of chronic immune system activation has many implications for the practice of medicine, psychiatry, psychology and most importantly, for individuals suffering with depression.

First, any patient presenting with symptoms of depression should be thoroughly examined to find the cause of the chronic immune system activation. This includes searching for all possible physical diseases. Infections of all kinds should be looked for. Many infections, when they become chronic, such as brucellosis or Lyme disease, are extraordinarily difficult to detect. Yeast infections are very neglected, although several authors attribute neuropsychiatric symptoms to yeast infections. Searching for damaged, dying or abscessed tissue or malfunctioning organs should be routine. Cancer is always a possibility, since depression is often a leading sign of cancer. All of this of course would lead to earlier and better diagnosis and treatment of serious physical diseases.

Second, depression accompanying a diagnosed physical illness, such as heart disease or cancer, is merely a sign the patient is much sicker, just as the degree of fever can indicate the severity of an illness. Furthermore, questions like, did the heart disease cause the depression or did the depression cause the heart disease would cease. This would eliminate the unfair, guilt laden suggestion that a patient's mood may have caused his cancer, heart disease or other physical pathology.

Third, since the gastrointestinal tract is the largest immune organ in the body and it is chronically exposed to more pathogens and antigens than any other site, the immune model of depression directly implicates the gut as an important site to look for the cause of some depressions. Almost a century of published clinical evidence has claimed food can cause depression and associated behavioral pathology. These reports, and there are a great many of them, have been universally scoffed at and ignored by the biomedical establishment, especially psychiatry and psychology. Two common reasons for their rejection are: 1). they are too good to be true and 2). there is no known mechanism to explain how food could be causing depression. The immune model of depression provides a straight forward mechanism to account for the effects of food on mood and behavior. This will be covered in chapter 7.

Fourth, the immunological insights on depression suggest that polyunsaturated fats may have long term effects on the incidence and severity of depression. In chapter 7, evidence will be presented which is consistent with polyunsaturated vegetable oils as promoters of depression and polyunsaturated fish oils as attenuators of depression.

Fifth, the new insights remove the stigma of the term 'mental illness' from depression, since chronic immune system activation is a physical, biological phenomena. Instead of a mental illness, a depressed patient has a chronically activated immune system, just as a patient with rheumatoid arthritis has a chronically activated immune system..

Sixth, the blaming game will be eliminated. No longer will depression be blamed on mothers, fathers, families, teachers, poverty, repressed memories, socio-economic factors, traumatic experiences or stressful jobs. Neither will the patient be blamed for his 'bad' thoughts, bad mood, irritability, fatigue and sense of hopelessness. Depressed patients shouldn't be held accountable for their depression just as feverish patients aren't criticized for having a fever. It's time we blame the real culprit, the chronically activated immune system.

Next chapter: Immunological Evidence Supporting The Immune-Cytokine Model of Depression