Depression

These guidelines are informational only. They are not intended or designed as a substitute for the reasonable exercise and judgment by practitioners, considering each patient's needs on an individual basis. Guideline recommendations apply to populations of patients. Clinical judgment is necessary to design treatment plans for individual patients.

I. Introduction

These guidelines were developed by a consensus panel of Kaiser Permanente clinicians, researchers, and staff. The panel reviewed evidence-based guidelines and studies containing systematic grading of evidence from Kaiser Permanente, Group Health Cooperative of Puget Sound, The Agency for Health Care Policy and Research, and the Cochrane Collaboration.

Our expert panel also discussed selected evidence from relevant medical literature on identification and treatment of depression in adult primary care practice. Our grading of the evidence represents our panel’s judgments on the conclusions of the evidence based on guidelines we reviewed as well as our expert opinion on additional selected evidence. Evidence is graded as "strong," "moderate," or "expert opinion." A description of the process of producing these guidelines and the evidence grading criteria are included in the Guidelines Appendix, p. 27. A listing of participating clinicians, researchers, and staff can be found at the end of the guidelines.

Scope of Guidelines
These guidelines are intended for use by all Primary Care clinicians to improve detection and treatment of Major Depressive Disorder in adult (age 18 and older) patients in primary care. The CMI Guidelines Panel recognizes that depression is an important problem in other populations, most notably children and adolescents. However, the evidence for making clear recommendations in these populations does not yet meet the criteria established to permit inclusion in these guidelines. Concerns about children and adolescents should be referred to PCPs in child and adolescent health (family practitioners and pediatricians), and to Behavioral Health specialists in child and adolescent evaluation and treatment.

< back to top

II. Detection

Because of the high prevalence of Major Depressive Disorder in the primary care population, it is recommended that PCPs maintain a high index of suspicion for depression in their patients. Best evidence currently available does not demonstrate that general population screening in primary care improves outcomes for depressed patients. CMI’s clinical panel recommends selectively suspecting depression in patients based on known co-morbidities and clinical experience. Because of the high prevalence of depression in certain identifiable populations, we recommend that providers be especially attentive for signs and symptoms in the following groups (expert opinion).

These groups include patients:

• with major medical conditions (e.g., diabetes, CAD, cancer, CHF, COPD, chronic pain syndromes)

• with frequent somatic complaints without organic findings (e.g., fatigue, sleep disturbance, bowel disturbance, malaise)
• with known risk factors for depression (prior episodes of depression, family history of depression, prior suicide attempts, stressful life events, alcohol or other substance abuse)
• who present with other clinical situations in which depression is expected to be highly prevalent (e.g., high utilizers of medical care, recent suspicious or unexplained trauma, domestic abuse)

In some cases providers may elect to use a screening instrument for these patients in order to improve their confidence of detecting depression (see Detection and Diagnosis for screening tools). Regardless of whether depressed patients are identified using a screening instrument or based on clinical impression, the diagnosis should always be confirmed using the DSM-IV criteria.

< back to top

III. Diagnosing Major Depressive Disorder

A. Criteria
The clinical diagnosis is made using DSM-IV criteria for Major Depressive Disorder. These criteria are represented by the "DIG SPACES" (or "SIGECAPS") mnemonic. Five of the nine diagnostic criteria must be present for the same two-week period and must include either depressed mood (dysphoria) or loss of interest or pleasure (anhedonia). Symptoms must cause significant distress and impair functioning.

Depression and mood instability are frequent in patients with alcohol or other substance abuse and in victims of domestic abuse; therefore, inquiries about alcohol and substance abuse as well as domestic abuse are essential (expert opinion) (see information on differential diagnosis in Section C below).

B. Establishing the Diagnosis and Determining Severity
Patient self-report tools, some incorporating severity scales, can be used to make a tentative diagnosis (see Detection and Diagnosis Section for more discussion and tools). If such tools are used, it is still necessary to confirm the patient’s report in clinical interview. To establish the diagnosis, review the criteria outlined above, obtain additional clinical history, evaluate and confirm patient self-report of symptoms and functioning, and determine the severity of the depression. Severity of depression may be determined by counting number of symptoms present, impact on patient functioning and level of distress (see Guidelines Appendix, Table 4, p. 59) Use of a brief tool to judge severity as well as to improve ability to track response to treatment is recommended (expert opinion).

C. Differential Diagnosis and Major Co-Morbidities
Listed below are the major differential diagnoses, co-morbidities, and associated recommendations for assessment and treatment. Initiate treatment for depression if depressive symptoms continue to meet diagnostic criteria after ruling out indicated causes and/or adjusting concurrent treatments. Refer to specialty Behavioral Health Care for indicated conditions.

• Side-effects of medications (e.g., anti-hypertensives and other cardiovascular medications, sedative/hypnotics): Reevaluate and/or adjust medications. (See Guidelines Appendix, p. 61)
• Medical illness (e.g., endocrine disorders, cardiovascular disorders): Treat medical illness.
• Psychiatric disorders (e.g., bipolar disorder, anxiety disorders, dysthymia and sub-syndromal depression): Refer to specialty Behavioral Health Care Services. (See Guidelines Appendix, p. 59)
• Significant alcohol or other substance abuse: Refer to Substance Abuse Treatment. Consider treatment in primary care only after abstinence is established for 4-6 weeks. (See Guidelines Appendix, p. 62)
• Reactions to loss, trauma, and violence (e.g., grief reactions, domestic abuse): Consider referral/consult with specialty Behavioral Health Care Services. (See Guidelines Appendix, p. 62)

D. Assessing for Suicide Potential
Assess suicide risk at diagnosis and regularly thereafter. Direct questioning about history of attempts, suicidal thinking, and impulses is important in all depressed patients. Patients are generally reassured by being asked about suicide and by being educated that such thinking is symptomatic of depression. When the risk is great (a distinct plan to kill oneself or, if in the judgment of the PCP, the patient intends to harm him/herself), consult a Behavioral Health Care Specialist immediately. Clinician judgment is aided by understanding the history and demographics of suicide risk as well as more imminent predictors of suicide. (See Detection and Diagnosis Section for more details).

E. Assessing for Concurrent Depression and Alcohol or Other Substance Abuse
In some patients, depressive symptoms can be caused or aggravated by ongoing substance abuse. Use of the CAGE screening tool is recommended. (See Guidelines Appendix, p. 62 for explanation and description of the tool.) If substance abuse is confirmed, consult/refer for chemical dependency assessment before initiating treatment for Major Depressive Disorder.

If depressive symptoms are still present after 4 to 6 weeks of abstinence, treat the depression (expert opinion).

F. Diagnosing Dysthymia
Dysthymia is a chronic, often debilitating, depressive disorder in which the diagnostic threshold for Major Depressive Disorder is not reached. To confirm a diagnosis of dysthymia, the patient must have at least three of the following symptoms: depressed mood, poor appetite/overeating, insomnia/hypersomnia, fatigue, low self-esteem, poor concentration and hopelessness (one of which must be depressed mood) which have persisted for at least two years (most of the day, more days than not). In addition, the patient must have been free of symptoms for no more than 2 months out of the last 2 years, the symptoms must last nearly all day, the majority of days, and must cause significant distress and/or have a significant impact on psychosocial functioning.

G. Other Depressive Disorders
In the process of evaluating patients for Major Depressive Disorder, PCPs will find many patients with other depressive disorders (e.g., minor depression, adjustment disorders, grief reactions). At this time, evidence does not clearly suggest procedures for diagnosis and treatment of these disorders in primary care practices. Nevertheless, it is important to assure that these patients do not "fall through the cracks" in the care system. It is therefore recommended that patients with these conditions be seen for follow-up by their PCP (follow-up interval determined by the degree of patient distress and dysfunction but no longer than one month). At that time, assuming no significant improvement, consider consultation with or referral to Specialty Behavioral Health Care, referral to Health Education programs, or a trial of anti-depressants.

H. Factors Complicating Treatment in Primary Care Settings: Referrals to, or Consultation with, Specialty Behavioral Health Care
Due to the complicated nature of their illness and/or social situation, certain patients may need specialty evaluation and treatment. Specialty Behavioral Health Care (Note: in some locations, this service line is known as Mental Health and Substance Abuse Services) should be engaged on a referral or consult basis as recommended below (expert opinion).

Arrange referral to specialty Behavioral Health Care Services for patients with:

• Active suicidal or homicidal ideation

• Psychotic symptoms

• Bipolar disorder/manic behavior

• Significant alcohol or other substance abuse

• Lifelong or recurrent depressions

• Domestic abuse

• Other severe psychiatric symptoms which complicate the diagnosis (e.g., very high levels of anxiety or panic)

See Detection and Diagnosis for more information on all of the above.

Schedule referral according to patient need (i.e., potential life threatening emergencies such as suicidality need to be seen immediately; urgent situations that are not life threatening but where the patient is in great distress and/or functioning very poorly need to be seen within 48 hours; other cases can usually be seen within 7 days).

Other situations will be greatly aided by consultation. Consult with specialty Behavioral Health Care for patients with:

• unclear diagnosis

• failure to respond to 2 or more antidepressants

• three months of treatment without desired clinical improvement

• partial response to medication

• a desire for a psychotherapeutic approach without medication or a desire for both medication and psychotherapy

• difficulty adhering to treatment plans

• extreme levels of distress and/or extremely impaired functioning that, in the PCP’s judgment, seem beyond the capabilities of the primary care setting.

< back to top

IV. Treating Major Depressive Disorder

A. Treatment Options
Options include medications, psychotherapy, or both. Treatment options are chosen according to severity of the depression and patient preference. It is important to ask all patients about any concurrent treatments for depression, prescribed or self-selected, in considering depression treatment options.

For mild to moderate Major Depressive Disorder:
Antidepressants and specific types of psychotherapy (e.g., cognitive-behavioral therapy and interpersonal therapy) are equally efficacious for Major Depressive Disorder that is mild to moderate in severity (strong evidence). Start with one or the other for patients with mild to moderate Major Depressive Disorder (expert opinion). Time to onset of benefit for psychotherapy may be longer than for pharmacotherapy. Consider recommending Behavioral Health consultation for patients with mild to moderate depression who desire psychotherapy with or without medication. The recommended psychotherapies may also be added to augment treatment if medication response is sub-optimal.

For moderate to severe Major Depressive Disorder:
Medications are more effective than psychotherapy for the most severely ill patients with Major Depressive Disorder (moderate evidence). The use of antidepressant medication (with or without the recommended forms of psychotherapy) for patients with moderate to severe Major Depressive Disorder is recommended (strong evidence).

For all patients:
Decide on an approach with the patient, explaining the potential risks and benefits of various options. Treatment compliance is aided by patient involvement in the decision-making process (expert opinion). (See the tip sheet, "Your Choices for the Treatment of Depression," in the Tools Section).

B. Supportive Counseling After Diagnosis
Counsel the patient that depression is an illness and not a punishment, moral weakness, or character flaw (in fact, such beliefs are themselves often symptoms of the illness!). Encourage patients taking medications by telling them that medication treatment is highly effective and not addicting, but may take up to 4-6 weeks before optimal improvement is reached (though some may benefit sooner). Encourage resumption of some previously pleasurable activities as soon as possible (expert opinion).

Other important counseling points include:

• Avoid alcohol and illicit drugs, because of their deleterious effects on mood and possible interaction effects with medication.
• Use of caffeine. If significant anxiety or sleep disturbance is present, counsel patients to discontinue caffeine. If no significant anxiety or sleep disturbance is present, caffeine use may be acceptable.
• Encourage regular aerobic exercise as it may have a mild to moderate antidepressant effect.
• If sleep disturbance is significant, encourage good sleep hygiene (See Guidelines Appendix, p. 30 for recommendations).

C. Recommended First-Line Medications
The CMI panel reviewed conclusions of evidence based guidelines and studies, as well as recommendations of many experts from within and outside KP, in formulating medication recommendations. Based on accumulated evidence and prescribing history, the Selective Serotonin Reuptake Inhibitors (SSRIs) and Secondary Amine Tricyclic Antidepressants (TCAs) are the two classes of medication recommended by these guidelines. Literature is clear that efficacy of SSRIs and recommended TCAs is equal provided that medications are taken as prescribed and appropriately titrated to effective doses for an adequate treatment duration (strong evidence). Experts disagree about the conclusions of studies of differential effectiveness of these medications in primary care settings. Considering the above, uniformly recommending either of these medication classes over the other is unwarranted at this time. (See discussion in Guidelines Appendix, p. 31, for more complete information on interpretations of the evidence.)

Choice of medication class should be made on the basis of general indications (see below), PCP experience and expertise in managing the medications, and patient preference (after considering benefits and side effects of alternative treatment choices). Most experts preferentially choose SSRIs based on their relative ease of use and safety profile. The relatively higher side-effect burden of the TCA class of medication and greater toxicity in overdose, coupled with the need to titrate more carefully to effective doses, means that the secondary amine TCAs are a reasonable option only for PCPs comfortable with their indications, selection, titration, and side-effect management.

Choose medications after considering:

• individual patient factors (e.g., medical and psychiatric co-morbidities, prior medication response, family history, patient age)

• medication factors (e.g., side-effect profiles, safety in overdose, ease of prescribing, affordability to the patient, and overall cost-effectiveness)

(See Guidelines Appendix, p. 32 for more complete discussion).

Prior to initiating medication therapy, assess for any over-the-counter or herbal preparations the patient may be using.

In general, SSRIs are recommended if:

• risk of overdose/suicide risk is a deciding factor

• cardiac conditions (especially conduction abnormalities) are present

• other medical conditions are present which could be exacerbated by TCA side-effects (e.g., orthostatic hypotension, benign prostatic hyperplasia (BPH), glaucoma, seizures, chronic constipation)

In general, Secondary Amine TCAs are recommended when:

• cost of the drug to the patient may pose a barrier to compliance

• medical conditions such as chronic pain syndromes, Irritable Bowel Syndrome (IBS), or headaches may be a deciding factor (see Guidelines Appendix, p. 43 for complete discussion and recommendations)

• patients are intolerant of SSRI side effects (after all reasonable adjustments have been made)

In all cases it is recommended that clinicians thoroughly discuss the above factors with patients and involve patients in decision-making about choice of treatment. (See the tip sheet entitled "Your Choices for the Treatment of Depression," in the Tools Section).

Recommended First Line Agents include:

1.	Selective Serotonin Reuptake Inhibitors (SSRIs) • fluoxetine, paroxetine, and sertraline*
2.	Secondary Amine Tricyclic antidepressants (TCAs) • nortriptyline and desipramine

(* Sertraline may be used as formularies permit).

Within class, no clear evidence of differential efficacy exists to recommend any agent over another (strong evidence). Consider secondary effects of medications when selecting a specific agent. See Guidelines Appendix, p. 37 for prescribing information (starting dose, titration to clinically effective dosage, adverse effect profiles), and the pocket card entitled "Treating Depression in Primary Care" for a quick reference tool on titration and managing common side effects.

It is important to remember that improving the system of care for depressed patients, including the key processes of patient identification, diagnosis, treatment, and follow-up, is ultimately more important for better outcomes than selection of class of first line medication (expert opinion).

Monitoring Blood Levels
Therapeutic relationships between plasma level and clinical response have been well established for the recommended TCAs nortriptyline and desipramine. Although routine plasma concentration monitoring is not necessary, it can be useful in certain situations to help with further dosing adjustments. The relationship between plasma concentration and clinical response has not been established for the SSRIs; obtaining drug levels in patients on SSRIs is therefore not recommended (expert opinion).

Additional Considerations
St. John’s wort (hypericum perforatum) is an over-the-counter herbal preparation that limited evidence from European trials suggests is somewhat more effective than placebo for mild to moderate depressive symptoms. It is not known which depressive disorders may respond to treatment with this preparation, whether it is as effective as other antidepressants, if it has fewer side effects than standard antidepressants, or what the optimal dosing regimen should be. Since safety of combining medication has not been evaluated in depression, it is not recommended that St. John’s wort be combined with other antidepressant medications (expert opinion). Also, note that not all herbal preparations are standardized. Formulary-based antidepressants should be considered the preferred medication. Patients who opt to take St. John’s wort should be advised of these issues (expert opinion).

Treating Dysthymia
Antidepressant medications are effective in the treatment of dysthymia (strong evidence). Expert opinion suggests that patients with dysthymia respond to typical pharmacologic or psychotherapeutic treatments used to treat Major Depressive Disorder. Experts disagree, however, about the ease of making the diagnosis in primary care given that patient self-report tools are less reliable than those for Major Depressive Disorder and patient reports about their illness may not be as consistent as for Major Depressive Disorder.

Patients appropriate for medication treatment in primary care are those with moderate to severe symptoms meeting diagnostic criteria and with significant distress or difficulties in functioning. Following are recommendations for PCPs on treating dysthymia:

If the diagnosis can be clearly established:

• begin treatment as for Major Depressive Disorder

If there is doubt about the diagnosis:

• either see the patient again in 2-4 weeks to monitor symptoms, severity, and impairment of functioning, re-evaluate diagnostic information, and, based on your re-assessment begin treatment, OR, refer to specialty Behavioral Health Care for further assessment and treatment.

< back to top

V. Course of Treatment

In general, patients will require at least 7-12 months of therapy, even for an initial episode (expert opinion). In acute phase treatment, the goal is to achieve a minimum period of 3 months of symptom remission. After achieving remission of symptoms, it is recommended that patients be kept on full therapeutic dose of medicine for an additional 4 to 9 months of continuation phase treatment to prevent relapse (expert opinion).

Assess therapeutic response every 2 to 4 weeks (or more frequently if symptoms are severe) for significant evidence of remission of symptoms (see algorithm, entitled "Diagnosis and Treatment of Major Depressive Disorders in Primary Care," in Tools Section).

Carefully assess changes in criterion signs and symptoms. (Patient self-report severity scales are available, see Detection and Diagnosis, but clinical information must confirm results.) Although anxiety and insomnia may improve relatively early in treatment, this does not constitute recovery from depression (which implies improvement over the entire spectrum of depressive symptoms).

Because risks for discontinuing medication treatment are highest in the first six weeks, monitor carefully during this time (expert opinion). Consider weekly phone or office contacts to enhance compliance, monitor for adverse effects, and provide support. During this time, titrate medications to full therapeutic dose as judged by remission of symptoms.

Experts recommend discussing a set of specific messages about treatment with patients which may increase compliance. See pocket card entitled, "Treating Depression in Primary Care" in the Tools Section.

Currently there is no convincing evidence that the onset of therapeutic effect is different among antidepressant medications. Once patients have been prescribed a therapeutic dose of an antidepressant, there is a 2-4 week delay before the clinical benefits are evident. Complete remission of symptoms may not occur for 4-6 weeks or more. Consequently, the duration of a therapeutic trial is considered to be 4-9 weeks on a therapeutic dose of medication.

If substantial remission of symptoms is not achieved at the end of four weeks:

• assess medication adherence and,

• increase dosage of prescribed medication, or

• change to another agent if there are undesirable side effects

If complete remission of symptoms is not present after 3 months:

• assess adherence to medication, and

• increase dosage of the prescribed medication or

• change to another agent if there are undesirable side
  effects or

• add another agent or psychotherapy, and/or

• refer to specialty Behavioral Health Care

No scientific evidence exists to suggest an optimal frequency of visits in any treatment phase. Follow-up is expected to occur at least 3 times in the first 12 weeks of treatment (HEDIS recommendation).

< back to top

VI. Maintenance and Discontinuation

For most patients, after achieving a period of symptom remission of 4 to 9 months (after the first 3 months of treatment) for a first episode, a trial of discontinuing treatment is recommended (expert opinion).

• taper medication over 3-6 weeks (see more detailed information on tapering medications in Guidelines Appendix, p. 39)

• provide patient education on signs and symptoms of relapse

If patient remains symptom-free, reassess periodically. If relapse occurs during or after tapering, resume treatment at full therapeutic dose and continue treatment for another 3-6 months. Then try tapering again.

For patients with a first episode of Major Depressive Disorder, taper medication over 3-6 weeks. If the patient remains stable, assess periodically. If the patient’s symptoms return, resume full therapeutic dose, and treat again for 3-6 months. Once stable, attempt to taper again. Some experts also recommend that patients with two episodes of Major Depressive Disorder and with associated risk factors (see algorithm, entitled "Diagnosis and Treatment of Major Depressive Disorder in Primary Care," in the Tools Section) also receive maintenance treatment, but this is not a consistent finding among the evidence-based guidelines reviewed.

Note: Some patients with two or more episodes of Major Depressive Disorder may be suitable for maintenance therapy; consult with Specialty Behavioral Health Care. For patients with 3 or more episodes of Major Depressive Disorder, maintenance treatment is recommended at full therapeutic dose for several years after remission of symptoms (strong evidence).

< back to top

VII. Considerations for Special Populations

A. The Elderly
Other conditions may masquerade as depression in the elderly. Before concluding that the patient is depressed, consider other diagnoses such as hypothyroidism and early dementia. (See Guidelines Appendix, p. 39 for more complete discussion.) Once confident of the diagnosis of depression, assess carefully for risk of suicide, adherence to current medical regimens, side-effects of concurrent medical illness and medications, alcohol and substance abuse, and for presence and quality of ongoing support systems and services (family, peers, clergy, community involvement, etc.). If medication is indicated, the motto should be "start low, go slow, but KEEP GOING" (expert opinion). Use recommended first-line agents.

Note: Some experts suggest avoiding fluoxetine as the first medication to try in the elderly because of its longer half-life and potential for being associated with greater agitation.

B. Pregnancy and Breastfeeding
When depression accompanies pregnancy or breastfeeding, safety of antidepressant treatment is assessed on a case-by-case basis by weighing the risk of prescribing medication against the risk of withholding or discontinuing medication. Risk to the fetus, perinatal risks for mother and infant, risk associated with treatment during the postpartum and breastfeeding period, and risks to later development of the child must all be considered.

Recommendations for pharmacologic treatment (based on safety profile), if warranted:

• In pregnancy: SSRIs- fluoxetine, paroxetine, sertraline; TCAs- nortriptyline, desipramine
• If breastfeeding: SSRIs- sertraline; TCAs- nortriptyline, desipramine

(See Guidelines Appendix, p. 40-41 for further discussion.)

C. Postpartum
Prevalence of depression within the first 3-6 months after childbirth is 10-15%, which is higher than in non-childbearing controls (5-7%). Women whose maintenance antidepressant treatment was discontinued during pregnancy appear to be particularly at risk for recurrence of depression. In the absence of a contraindication, restart antidepressants after delivery.

Postpartum depression should be treated according to the same principles outlined for other depression. An issue often overlooked is the risk to the infant associated with unresolved maternal depression. Children of mothers with untreated postpartum depression have more developmental problems than children of mothers whose depression has been successfully treated.

It is important to distinguish postpartum depression from the transient 7-10 day depressive condition referred to as "postpartum blues," which does not require medication and is best treated by reassurance. (See Guidelines Appendix, p. 40 for further discussion.)

D. Other Medical Conditions
Controversy exists about the strength of the evidence for recommending one class of antidepressants over another for patients with medical conditions co-morbid with depression. For example, TCAs have been used for many years for the treatment of chronic pain syndromes. Evidence exists to support this practice but experts disagree about the strength of this evidence and few studies have simultaneously examined treatment of concurrent Major Depressive Disorder and chronic pain. Less evidence is available for the use of SSRIs for these conditions. See Guidelines Appendix, p. 43 in the for a more complete discussion of the controversies.

For clinicians who want guidance about prescribing for depression co-morbid with other medical conditions, see "Recommended First-Line Medications" in Guidelines Appendix, p. 19-20, and Table 4 on p. 43 of the Guidelines Appendix, for listing of indications and class of antidepressants some experts would preferentially recommend.

< back to top

Guidelines Development Group

Interregional Expert Panel
 

Neil Baker, MD	Group Health Cooperative of Puget Sound
Joel Feinman, PhD	CMI Clinical Lead;Northeast
Patrick Finley, PharmD	Northern California
Carole Gardner, MD	Georgia
Howard Gould, MD	Georgia
Kirk Hastings, PhD	Southern California
Mark Heine, MD	Northern California
Julia Jacobson, MPH	Southern California
Marguerite Koster	Southern California
Debbie Kubota, PharmD	Southern California
Norman Muilenburg, PharmD	Northwest
David Price, MD	Colorado

< back to top

For more information about this program, contact the CMI Product Information Line at 510-271-6426 or via email at CMIproducts@kp.org.