What are the effects of treatments in mild to moderate or severe depression?

BENEFICIAL

Prescription Antidepressant Drugs (Tricyclic Antidepressants [Including Low Doses], Selective Serotonin Reuptake Inhibitors [SSRIs], Monoamine Oxidase Inhibitors [MAOIs], Reboxetine, or Venlafaxine) Improved Symptoms Compared with Placebo in Mild to Moderate and Severe Depression)

Systematic reviews and subsequent randomized controlled trials (RCTs) including persons 18 years or older in primary and secondary care showed that prescription antidepressant drugs were effective treatments for all grades of depressive disorders compared with placebo. However, the most robust available evidence on the effectiveness of antidepressant drug treatment is in the pharmacologic management of moderate and severe depression.

One systematic review and two subsequent RCTs including persons 55 years or older with all grades of depressive disorders showed that tricyclic antidepressants, SSRIs, or MAOIs reduced the proportion of persons who did not recover over 26 to 49 days compared with placebo. The reviews included little information on severe adverse effects of antidepressant drugs compared with placebo. There is evidence of publication bias in trials of SSRIs. Current evidence indicates no clear relationship between SSRIs and increased risk of suicide in adults, but SSRIs and tricyclic antidepressants may induce or worsen suicidal ideation and behavior during the early phases of treatment. This may be because of increased agitation and activation. SSRIs generally increase adverse events in children and adolescents compared with placebo, and regulatory authorities in several countries are reviewing the safety of these drugs.

Tricyclic Antidepressants vs. Other Prescription Antidepressant Drugs

Three systematic reviews showed no significant difference in outcomes with different kinds of antidepressant drugs. One systematic review and one subsequent RCT showed no significant difference between tricyclic antidepressants and venlafaxine in the proportion of persons who responded over one to 12 months. Another systematic review suggested that tricyclic antidepressants were more effective than MAOIs in persons with severe depressive disorders, but that it may be less effective in atypical depressive disorders with biologic features (e.g., increased sleep and appetite). A third systematic review showed that tricyclic antidepressants were associated with higher rates of adverse effects than SSRIs, but the difference was small. One systematic review and two case–control studies showed no significant difference in risk of suicide between SSRIs and tricyclic antidepressants.

Two RCTs, primarily including persons with severe depression, showed no significant difference in symptoms between desipramine or imipramine and reboxetine at four weeks, but results were sensitive to outcome scales used. One systematic review showed no significant difference between milnacipran and imipramine in persons with major depressive disorder, although it showed that imipramine had an overall higher incidence of adverse events. One systematic review showed no significant difference between low-dose and standard-dose tricyclic antidepressants in the proportion of responders at six to eight weeks. One systematic review showed that combined antidepressants (primarily tricyclic antidepressants) with benzodiazepines increased response rates within one week compared with antidepressants alone, although this difference was not significant at six weeks.

SSRIs and Related Drugs vs. Other Prescription Antidepressant Drugs

Three systematic reviews showed no significant difference in outcomes with different kinds of antidepressant drugs, although one systematic review showed that SSRIs were less effective than venlafaxine in increasing the proportion of persons who responded. RCTs including persons with major depression showed similar response rates for fluoxetine and reboxetine at six weeks, but they showed that reboxetine may be slightly more effective in improving social functioning. One RCT showed that mirtazapine was significantly more effective than sertraline during the first two weeks of treatment, although this benefit did not last longer than two weeks. Another RCT showed no significant difference between paroxetine and mirtazapine after 24 weeks.

One systematic review showed that SSRIs were associated with fewer adverse effects than tricyclic antidepressants, but the difference was small. Two systematic reviews, one retrospective cohort study, and two case-control studies showed contradictory evidence about the risk of suicide in adults treated with SSRIs. One of the systematic reviews and the case-control studies reported no significant difference in risk of suicide between SSRIs and tricyclic antidepressants. One RCT and observational data suggested that abrupt withdrawal of SSRIs was associated with adverse symptoms (e.g., dizziness, rhinitis) and that these symptoms were more likely to occur with drugs with a short half-life (e.g., paroxetine). There is evidence of publication bias in trials of SSRIs.

Current evidence indicates no clear relationship between SSRIs and increased suicide in adults; however, SSRIs and tricyclic antidepressants may induce or worsen suicidal ideation and behavior during the early phases of treatment. This may be because of increased agitation and activation. SSRIs generally increase adverse events in children and adolescents compared with placebo, and regulatory authorities in several countries are reviewing the safety of these drugs.

MAOIs vs. Other Prescription Antidepressant Drugs

Three systematic reviews showed no significant difference in outcomes with different kinds of antidepressant drugs. One systematic review showed that MAOIs were less effective than tricyclic antidepressants in persons with severe depressive disorders. However, the review showed that MAOIs may be more effective for atypical depressive disorders with biologic features (e.g., increased sleep and appetite).

Venlafaxine vs. Other Prescription Antidepressant Drugs

One systematic review and one RCT showed no significant difference between venlafaxine and tricyclic antidepressants in the proportion of persons who responded to treatment over one to 12 months. The review showed that venlafaxine increased the proportion of persons who responded compared with SSRIs, although one subsequent RCT showed no significant difference between extended-release venlafaxine and escitalopram. One RCT showed no significant difference between venlafaxine and mirtazapine in the proportion of persons who responded, but both treatments improved baseline depressive symptoms.

Electroconvulsive Therapy (in Severe Depression)

One systematic review including persons, mostly inpatients, with moderate to severe depressive disorder showed that electroconvulsive therapy improved symptoms over one to six weeks compared with simulated electroconvulsive therapy or antidepressant drugs. The review showed that bilateral electroconvulsive therapy improved symptoms compared with unilateral therapy and that high-dose electroconvulsive therapy was more effective than low-dose therapy. The degree of reported short-term cognitive impairment seemed to be inversely related to treatment effectiveness.

Another systematic review provided insufficient evidence to assess electro-convulsive therapy in older adults. Electro-convulsive therapy is a short-term treatment and may be inappropriate for some patients; therefore, the consensus is that the therapy normally should be reserved for persons who cannot tolerate or have not responded to antidepressants. Electroconvulsive therapy may be useful when a rapid response is required.

Cognitive Therapy (in Mild to Moderate Depression)

One systematic review including younger and older adults showed that cognitive therapy improved symptoms compared with no treatment. Three systematic reviews including younger and older adults with mild to moderate depression showed that psychological therapies (mainly interpersonal psychotherapy and cognitive therapy) increased the proportion of persons whose depression was in remission over 10 to 34 weeks compared with control (i.e., usual care, usual care plus a placebo pill, or supportive therapy). These reviews did not report results for cognitive therapy alone compared with control.

One systematic review showed no significant difference between cognitive therapy and interpersonal therapy, although it showed that cognitive therapy was more successful than combined interpersonal therapy, brief psychodynamic therapy, or supportive therapy. One systematic review showed no significant difference between short-term psychodynamic psychotherapy and cognitive therapy in remission rates or improvement of depressive symptoms. One systematic review of poor-quality RCTs including persons 55 years or older with mild to moderate depression showed no significant difference in symptom improvement between psychological treatments (e.g., cognitive therapy, cognitive behavior therapy) and no treatment.

Interpersonal Psychotherapy (in Mild to Moderate Depression)

Two systematic reviews including adults with mild to moderate depression showed that psychological therapies increased the proportion of persons whose depression was in remission over 10 to 34 weeks compared with control. These reviews did not report results for interpersonal therapy alone compared with control. One RCT identified by a third review showed that interpersonal therapy increased response rates compared with usual care, although the review showed no difference between cognitive therapy and interpersonal therapy. One RCT showed that group interpersonal psychotherapy improved depressive symptoms in adults in Uganda compared with no treatment. One systematic review including older adults showed no significant difference among interpersonal therapy, psychodynamic psychotherapy, and no treatment.

LIKELY TO BE BENEFICIAL

Reboxetine vs. Other Antidepressant Drugs (in Mild to Moderate or Severe Depression)

Two RCTs that primarily included persons with severe depression showed no significant difference in symptoms at four weeks among reboxetine, desipramine, and imipramine, but results were sensitive to the outcome scales used. RCTs including persons with major depression showed similar response rates at six weeks between reboxetine and fluoxetine, although reboxetine may have slightly improved social functioning.

St. John's Wort (More Effective Than Placebo, as Effective as Other Antidepressants in Mild to Moderate Depression)

One systematic review showed that St. John's wort ( Hypericum perforatum) improved depressive symptoms over four to 12 weeks compared with placebo. The review showed no significant difference between St. John's wort and tricyclic antidepressants or SSRIs. The results of the RCTs should be interpreted cautiously, because many patients did not use standardized preparations of St. John's wort and dosages of antidepressants varied. Questions have been raised regarding the methodologic quality of available studies, which have examined heterogenous patient populations and have used standardized symptom rating instruments inconsistently. St. John's wort likely interacts with many more drugs than previously reported.

Nondirective Counseling (in Mild to Moderate Depression)

One systematic review including persons 18 years or older with recent-onset psychological problems, including depression, showed that brief, nondirective counseling in the primary care setting reduced short-term (less than six months) symptom scores in persons with mild to moderate depression compared with usual care. However, the review showed no significant difference in long-term (more than six months) scores. One RCT showed that eight counseling sessions significantly improved anxiety and depression symptoms compared with no treatment.

Combining Prescription Antidepressant Drugs and Psychological Treatments (in Mild to Moderate and Severe Depression)

Two systematic reviews showed that a combination of pharmacotherapy and psychotherapy improved depressive symptoms compared with either treatment alone. One subsequent poor-quality RCT showed no significant difference in response rate between a combination of interpersonal therapy and sertraline and either therapy alone. A subsequent RCT showed no significant difference between short-term psychodynamic psychotherapy plus antidepressants and antidepressants alone, although subgroup analysis showed that combination therapy was more effective in persons with depression and a personality disorder.

UNKNOWN EFFECTIVENESS

Problem-Solving Therapy (in Mild to Moderate Depression)

One systematic review including younger and older adults presenting to primary care physicians with mild to moderate depression showed that psychological therapies (including problem-solving therapy) improved outcomes compared with usual care. The review did not report results of problem-solving therapy alone in persons with moderate depression. It showed no significant difference between problem-solving therapy and usual care in symptoms at six to 11 weeks in persons with mild depression or dysthymia. One large subsequent RCT showed that problem-solving therapy increased the proportion of persons who were not depressed at six months compared with usual care. However, it showed no significant difference at one year. A smaller RCT showed no significant difference between problem-solving therapy and usual care in symptoms at eight or 26 weeks in persons, most of whom were depressed, presenting to primary care physicians with emotional disorders. RCTs showed insufficient evidence to assess the relative effectiveness of drug and nondrug treatments of severe depression.

Befriending (in Mild to Moderate Depression)

One small RCT provided insufficient evidence to assess befriending in persons with mild to moderate depression.

Exercise (in Mild to Moderate Depression)

One systematic review including adults showed limited evidence from poor-quality RCTs that exercise may improve symptoms compared with no treatment, and that it may be as effective as cognitive therapy. One poor-quality RCT including older adults identified by the review showed limited evidence that exercise may be as effective as antidepressant drugs in improving symptoms and may reduce relapse over 10 months.

Lithium Augmentation

RCTs provided insufficient evidence to assess augmentation of prescription antidepressants with lithium in younger and older adults with treatment-resistant depression.

Pindolol Augmentation

RCTs provided insufficient evidence to assess augmentation of prescription antidepressants with pindolol in younger and older adults with treatment-resistant depression.

Which interventions reduce relapse rates?

BENEFICIAL

Continuing Prescription Antidepressant Drugs (Reduced Risk of Relapse After Recovery in Persons with Mild to Moderate Depression)

One systematic review showed that continuing prescription antidepressant drug treatment after recovery reduced the proportion of persons who experienced a relapse over one to three years compared with placebo. The effect of continuing antidepressant therapy was independent of the underlying risk of relapse, the duration of treatment before randomization, or the duration of previous antidepressant treatment.

Five subsequent RCTs also showed that continuing antidepressant treatment reduced absolute risk of recurrence by up to one half compared with treatment discontinuation. Two RCTs including persons 60 years or older showed that continued treatment after recovery with dosulepin (dothiepin) or citalopram reduced the risk of relapse over one to two years compared with placebo, but it may increase the risk of ischemic heart disease. One RCT showed no benefit from continuing treatment with sertraline for another two years compared with placebo.

UNKNOWN EFFECTIVENESS

Cognitive Therapy (Weak Evidence That May Reduce Relapse Over One to Two Years After Stopping Treatment in Persons with Mild to Moderate Depression Compared with Antidepressant Drugs)

One systematic review, including younger and older adults with mild to moderate depression, that combined relapse rates across RCTs showed limited evidence that cognitive therapy reduced the risk of relapse over one to two years after stopping treatment compared with antidepressant drugs. Four subsequent RCTs showed limited evidence that cognitive therapy reduced relapse rates up to one year after remission. We found no systematic review or RCT that specifically included older adults.

Relapse Prevention Program (Improved Symptoms Over One Year After Recovery in Persons with Mild to Moderate Depression but No Significant Difference in Relapse Rates)

One RCT identified by a systematic review showed that a relapse prevention program improved depressive symptoms over one year compared with usual care in persons who had recovered after eight weeks of antidepressant treatment. However, there were no significant differences in relapse rates.

What are the effects of interventions to improve delivery of treatments?

LIKELY TO BE BENEFICIAL

Care Pathways (in Mild to Moderate Depression)

Two systematic reviews and eight subsequent RCTs showed that care pathways improved the effectiveness of treatment for depression (i.e., antidepressant drugs or cognitive behavior therapy) compared with usual care in persons older than 18 years with mild to moderate depression. These pathways included collaboration between primary care physicians and psychiatrists, intensive patient education, case management, and telephone support. The RCTs also showed that clinical practice guidelines and educational strategies without other organizational processes did not improve detection or outcome of depression compared with usual care. It was not possible to determine from the results which components of care pathways were most effective.

One RCT including older persons with depression showed that initial care by a community mental health team did not improve depressive symptoms compared with usual care from a primary physician. Another RCT showed that management of depression in older persons by a community nurse improved symptoms and increased recovery after three months compared with usual primary care. One RCT including persons older than 60 years with major depression, dysthymic depression, or both who were treated in a variety of primary care clinics showed that collaborative care was more effective than usual care in reducing depressive symptoms. One RCT showed that treatment recommendations had no significant effect on depressive symptoms in older persons compared with usual primary care. However, diagnosis of depression and anti-depressant prescribing were more prevalent in the intervention group.

Definition

Depressive disorders are characterized by persistent low mood, loss of interest and enjoyment, and reduced energy. They often impair day-to-day functioning. Most of the RCTs assessed in this chapter classify depression using the Diagnostic and Statistical Manual of Mental Disorders, 4th ed (DSM-IV)¹ or The ICD-10 Classification of Mental and Behavioral Disorders.² DSM-IV divides depression into major depressive disorder or dysthymic disorder.

Major depressive disorder is characterized by one or more major depressive episodes (i.e., at least two weeks of depressed mood or loss of interest accompanied by at least four additional symptoms of depression). Dysthymic disorder is characterized by at least two years of depressed mood on more days than not accompanied by additional symptoms that do not meet criteria for major depressive disorder.¹

ICD-10 divides depression into mild to moderate or severe depressive episodes.² Mild to moderate depression is characterized by depressive symptoms and some functional impairment. Severe depression is characterized by additional agitation or psychomotor retardation with marked somatic symptoms.² Treatment-resistant depression is defined as an absence of clinical response to treatment with a tricyclic antidepressant at a minimum 150-mg dose of imipramine (or an equivalent drug) daily for four to six weeks.³ In this chapter, DSM-IV and ICD-10 classifications are used, but treatments are considered to have been assessed as severe depression if the RCT included inpatients.

Older adults generally are defined as 65 years or older. However, some of the RCTs of older persons in this chapter included persons 55 years or older. Older persons may have an atypical presentation of depression: low mood may be masked, and anxiety or memory impairment may be the principal presenting symptom. Dementia should be considered in the differential diagnosis of depression in older adults.⁴ This chapter does not cover intervention in women with postnatal depression or in persons with seasonal affective disorder.

Incidence

Depressive disorders are common, with major depression affecting 5 to 10 percent of persons seen in primary care settings.⁵ Two to three times as many persons may have depressive symptoms but do not meet DSM-IV criteria for major depression. Women are affected twice as often as men. Depressive disorders are the fourth most important cause of disability worldwide and are expected to become the second most important cause by 2020.^6,7 Ten to 15 percent of older persons have depressive symptoms, although major depression is relatively rare in this population.⁸

Etiology

The causes of depression are unclear but are thought to include childhood events and current psychosocial adversity. Recent studies suggest that genetic factors also may be important, indicating that several chromosomal regions may be involved. Phenotypes, however, do not seem to exhibit classic mendelian inheritance. Psychiatric research also has focused on the role of psychosocial factors (e.g., social context, personality dimensions) in depression. Many theories emphasize the importance of temperament (i.e., differences in adaptive systems), which can increase vulnerability to mood disturbances. Impairment in social relationships, gender, socioeconomic status, and dysfunctional cognition also may play a role. It seems that integrative models, which take into account the interaction of biologic and social variables, offer the most reliable approach to the complex etiology of depression.

Prognosis

About one half of persons presenting with their first episode of major depressive disorder experience further symptoms within the next 10 years.⁹ One systematic review (search date 1996, 12 prospective cohort studies, 1,268 persons, mean age 60 years) showed that the prognosis may be especially poor in older persons with chronic or relapsing depression.¹⁰ Another systematic review (search date 1999, 23 prospective cohort studies including persons 65 years or older, five identified by the first review) showed that depression in older persons was associated with increased mortality (15 studies; pooled odds ratio, 1.73; 95% confidence interval, 1.53 to 1.95).¹¹