American College of Physicians: Internal Medicine — Doctors for Adults ®


Special focus: Depression

From the January-February ACP Observer, copyright 2005 by the American College of Physicians.

Depression is a painful, insidious and often chronic disorder—and one that internists are increasingly called upon to diagnose and treat. Recent studies show that primary care physicians may write as many as 60% of all prescriptions for antidepressants.

But making that diagnosis in primary care and providing the necessary follow-up can be frustrating. Unlike patients who seek out mental health professionals for treatment, depressed patients in primary care often present with vague medical complaints, such as headache, stomach ache or gastrointestinal problems. That makes it much harder to reach a diagnosis.

And patients in primary care are also more apt to resist treatment. A study in the December 2003 Journal of General Internal Medicine, for instance, found that patients treated for depression in the primary care setting may be more likely than those cared for in a mental health setting to reject treatment.

They are also more likely to think they'll "get over" the problem on their own, less likely to take medication and more likely to scoff at suggested psychological counseling--an important adjunct for many depressed patients. At the same time, interventions that have proven to work, such as follow-up phone calls from staff members, take time and resources that most office-based physicians don't have.

Despite these hurdles, the stakes for not appropriately treating depression are high. Patients not followed closely during treatment can suffer debilitating side effects, from severe loss of libido to perhaps a higher risk of suicide—the focus of the recent controversy over prescribing antidepressants to adolescents and children.

And many patients suffer needlessly from undertreatment. A study in the June 15, 2004, Annals of Internal Medicine found that only 23% of patients treated for depression achieved remission. And failure to adequately treat depression may be associated with poorer outcomes related to comorbidities.

This edition of ACP Observer Special Focus draws primarily on information from ACP's PIER (Physicians' Information and Education Resource). It is designed to help optimize your ability to treat patients for major depression with recommendations for screening, therapy and thorough follow-up.

Screening and Diagnosis

What to look for

Physicians should consider recurrent screening for patients with a history of depression; multiple unexplained or unrelated somatic complaints; comorbid psychological conditions, such as panic disorder or generalized anxiety; substance abuse; chronic pain; or lack of response to usually effective treatments for comorbid medical conditions.

Other frequent common denominators of depression include being female, having a medical illness, having a family history of depression, suffering sleep disturbances, occupying a low socioeconomic status and experiencing stressful life events. First-degree relatives of patients with mood disorders also have an elevated relative risk for developing major depression.

You can choose from a variety of screening and assessment tools (see "Depression assessment tools") —and you should pay special attention to the elderly. Remember that they could have cognitive impairment problems, visual deficits and comorbid illness. And screen women for postpartum depression four to six weeks after they've given birth, with a special emphasis on women who've had previous psychiatric history or psychosocial stressors.

If you suspect depression, initiate a full diagnostic assessment to determine its severity and any functional impairment, and whether the patient exhibits any suicidal ideation.

Assessment can also help you differentiate major depression from other mood syndromes including bipolar disorder, minor depression, situational adjustment reaction with depressed mood and grief. At the same time, keep in mind that major depression can be a presenting feature of several medical conditions including hypothyroidism and Cushing's disease; substance abuse; or prescription side effects from the use of glucocorticoids, interferon, and reserpine and other centrally-acting antihyptertensives.


Although many now consider the evidence for St. John's wort to be uncertain, you can offer it (at a dose of 0.3% hypericin, 300 mg tid) as a potential treatment for patients with mild to moderate depression or subsyndromal depression.

For patients with depressive disorder, dysthymic disorder or both, initiate single-agent antidepressant drug therapy. Base your choice of agent on differences in side-effect profile, as well as on the following:

  • A history of successful treatment with a particular agent.
  • A patient's history of a first-degree relative successfully treated for depression with a particular agent.
  • Differences in the half-life of different selective serotonin reuptake inhibitors (SSRIs).
  • Differences in the role of cytochrome P450 2D6 metabolism that may influence the risk of drug interactions.
  • Risk of drug and food interactions with monoamine oxidase inhibitors (MAOIs).
  • Potential overdose lethality.

Treat first episodes of depression to achieve complete remission, and expect to treat patients for four to nine months thereafter. You should treat recurrences of major depression with long-term maintenance therapy.

At the same time, you need to counsel patients about antidepressant medication to improve adherence. Stress the importance of taking the medication daily, as well as the fact that antidepressants must be taken for between two and four weeks for patients to feel a noticeable effect. Patients also need to know that they must take their medication even if they start to feel better—and that they must not stop taking their medication without checking with you.

Consider recommending psychotherapy from the outset as a therapeutic possibility, particularly if the patient is reluctant to accept the diagnosis or is hesitant about taking medication. (See "Follow-up for depression" and "When should you refer.")


When taking SSRIs and other antidepressants, patients should, at a minimum, receive three follow-up checkups within 90 days. Here is a breakdown of issues at follow-up:

1-2 weeks

In the initial weeks of treatment, you should assess how patients are accepting medication, reinforce educational messages, address adverse events and reassess for suicidal ideation. Challenges to maintaining treatment during these first weeks include the pessimism and hopelessness intrinsic to depression and the relatively rapid onset of adverse effects before patients notice any benefits.

Your goal is a complete remission of patients' symptoms and return to normal functioning. Keep in mind that one study found that 28% of depressed primary care patients stopped taking their medication during the first month of treatment, while 44% stopped within three months.

6-8 weeks

Because antidepressants take time to be effective, you need to assess patients' specific response goals at between six and eight weeks of treatment. Different assessment tools make that determination easier and more accurate.

Patients are considered to have an adequate response if they exhibit a 50% decrease in symptoms at this point. Keep in mind, however, that treatment-resistant depression is a very common problem among patients in primary care. One study found that between 29% and 46% of patients did not respond fully to an adequate dose of antidepressant treatment at six weeks.

Studies show that partial response, defined as symptomatic improvement that is between 25% and 50%, occurs in 12% to 15% of patients, while nonresponse—defined as less than 25% symptomatic improvement—affects between 19% and 34% of treated patients.

For patients with a less than 50% response, you need to modify their treatment by increasing their initial dose if they have some response, or changing their medication if their response has been minimal or their initial medication can't be tolerated.

You may want to change a patient's class of antidepressant to avoid specific side effects, such as switching from an SSRI to bupropion because of anorgasmia. However, it usually isn't necessary to switch to a non-SSRI medication if the initial medication is an SSRI. Patients who show no response to or can't tolerate one SSRI may successfully use another.

When switching drugs, make sure both you and your patient differentiate between adverse effects from discontinuing the first medication and those that arise from the new therapy.

Note too that, with the exception of changes to or from MAOIs, a washout period is generally not required. However, you can simultaneously titrate the new medication on and the old off. When switching to or from an MAOI, use a minimum two-week washout period—and a six-week washout when changing from fluoxetine to an MAOI.

10-12 weeks

When following up with patients at 10-12 weeks, you again need to use a structured procedure to assess changes in symptom severity.

If the patient is not in remission, you can again increase the dosage, change the antidepressant or add an additional medication. Before adding or switching therapies, maximize the dose of the initial agent as tolerated if the patient experienced a partial response. Alternatively, consider recommending psychotherapy again.

Use these strategies for specific SSRI adverse effects:

  • Sexual dysfunction: switch to bupropion, nefazodone or mirtazapine, or recommend sildenafil citrate.
  • Weight gain: switch to bupropion.
  • Agitation or excessive activation: switch to another SSRI, a tricyclic, nefazodone or mirtazapine, but avoid fluoxetine.

If patients show only partial response, consider combination or augmentation therapies. The advantages of combining or augmenting therapy include a relatively rapid effect compared to withdrawing one medication and then starting another, as well as the potential for synergistic or complementary effect.

Combining or augmenting therapy can also help the patient avoid withdrawal symptoms when you stop the first agent. The disadvantages of both therapeutic routes include the increasing complexity of the regimen and increased opportunities for drug interactions and adverse effects.

Combination therapies include:

  • Bupropion plus SSRI. One trial found that adding bupropion or an SSRI to treat partial responders compared to either one alone showed improved symptom relief in 70%. In another study, 56% of patients on SSRI or venlafaxine monotherapy responded to bupropion added to enhance response or to treat side effects.

  • Tricyclic antidepressant (TCA) plus SSRI. Fluoxetine and paroxetine are P450 (CYP) 2D6 inhibitors and may raise blood levels of TCAs, so carefully monitor TCA levels.

    One study reporting on this combination found that 26 of 30 patients in whom TCAs alone failed responded after the addition of fluoxetine. Eight of 12 patients whose TCA treatment was then stopped relapsed, although all recovered when TCA therapy was reinstituted. Case series support using this combination.

    Although the combined use of MAOIs and TCAs has been reported, this combination is not recommended. When employed by physicians with experience in their combined use, MAOIs should be added to existing TCA therapy—never in the opposite sequence.

  • Mirtazapine plus SSRI. Adding mirtazapine to SSRI treatment is effective in 55% of patients who have not responded to standard treatment alone.

Augmentation therapies include:

  • Lithium. Lithium has been relatively well studied as an augmentation treatment with both TCAs and SSRIs. One study comparing lithium and liothyronine in 50 patients for whom treatment with tricyclics failed showed a greater response with either than with placebo. One randomized controlled trial of 34 patients with incomplete response to TCAs showed a 44% response rate to normal doses of lithium but no difference from placebo (22%) for low doses (18%). An older study of 27 subjects on TCAs showed a significant, rapid improvement compared to placebo.

    One randomized controlled trial studied 24 subjects not responding to an SSRI (citalopram) alone who were comedicated with lithium. Six out of 10 patients on lithium, compared to 2 of 14 on placebo, improved by day 35 of treatment.

    In a study of 62 patients who did not adequately respond to fluoxetine or lofepramine, 50% responded to the addition of lithium compared to 25% of patients on placebo. In a stepped-care single-blind study of 38 refractory subjects, eight responded to fluvoxamine; an additional eight responded after the addition of lithium; and two responded to a combination of fluvoxamine, lithium and perphenazine.

    Keep in mind that elderly patients have more side effects with lithium.

  • Liothyronine. Researchers report mixed results with liothyronine (T3) augmentation. Some studies have shown its efficacy when used to augment TCAs; others have not.

  • Risperidone. One study of eight patients whose treatment with an SSRI failed showed improvement with risperidone.

  • Pindolol. Five controlled studies have shown that pindolol initiated at the beginning of therapy can speed antidepressant response. However, patients' response differences at the end of treatment were less consistent, and the benefit in augmentation therapy in resistant patients has not been shown.


Follow-up for depression

  • Initiation phase: 1-2 weeks: Take history. Assess effectiveness of drug therapy with structured instruments. Ask about symptoms and reassess for suicidal ideation.

  • Titration phase: 6-8 weeks: Same as initiation phase, plus change dosage or drugs; or combine drugs, if necessary.

  • Continuation phase: 10-12 weeks: Same as initiation phase, plus consider augmentation and referral for psychotherapy or drug management by a psychiatrist, if necessary.

  • Maintenance phase: 6 months. Same as initiation phase.

  • Posttreatment phase (after symptom relief): Take history. Assess effectiveness of drug therapy and look for recurrence. Ask about symptoms, if patient is still on medication and about stressful life events; include family in discussion. Do lab testing, depending on chosen therapy.

  • As needed: Initiate nondrug therapy or adjunctive psychotherapy. Increase dose, switch drug or begin combination therapy.


When should you refer?

While primary care physicians are treating a growing number of patients with major depression, internists still need to refer many patients for psychiatric treatment. Situations that warrant referrals include:

  • Drugs familiar to you fail, or different drugs fail.

  • Patients have suboptimal response to therapy and are more severely impaired.

  • Patients have difficult-to-control side effects.

  • Patients have failed multiple, different antidepressants despite appropriate adherence counseling.

  • Patients present with suicidal ideation, and need assessment and management of suicide risk.

  • Patients present with psychotic symptoms or you suspect a primary psychotic disorder. Acute psychosis in the depressed patient, substance abuse or both increases an individual's risk for suicide and should prompt psychiatric referral. According to one review, acute psychosis in a depressed patient is an ominous finding that should lead to hospitalization.

  • Patients have a history of possible or diagnosed bipolar affective disorder or a family history of bipolar disorder and symptoms suggesting mania. Delays in initiating mood-stabilizing drug therapy at illness onset in bipolar disorder, even for relatively mild symptoms, may lead to a higher risk for suicidal behavior, poorer social adjustment and more hospitalizations.

  • Patients are depressed, have complex comorbid mental conditions and substance abuse problems.


Depression assessment tools

  • Beck Depression Inventory. This five minute test assesses how the patient is feeling today.

  • Cornell Scale for Depression in Dementia. This 10-15 minute test examines feelings over the past week.

  • Geriatric Depression Scale (GDS). This scale is also validated for those with mild to moderate degrees of impairment. It tests feelings over the past week.

  • Patient Health Questionnaire (PHQ-9). This two-minute test assesses the presence and severity of the nine symptoms of major depression as established by the Diagnostic and Statistical Manual-IV.

Access to PIER's "Depression" module is online.

Module authors include Steven R. Hahn, MD, professor of clinical medicine, Albert Einstein College of Medicine, Bronx, N.Y.; and Elana R. Sydney, ACP Member, instructor of clinical medicine, Albert Einstein College of Medicine, Bronx, N.Y.

The information included herein should never be used as a substitute for clinical judgment and does not represent an official position of ACP. Source for ACP Observer Special Focus: PIER (Physicians' Information and Education Resource).


[PDF] Acrobat PDF format. Download Acrobat Reader software for free from Adobe. Problems with PDFs?

This is a printer-friendly version of this page

Print this page  |  Close the preview




Internist Archives Quick Links

Not an ACP Member?

Join today and discover the benefits waiting for you.

Not an ACP Member? Join today and discover the benefits waiting for you

ACP offers different categories of membership depending on your career stage and professional status. View options, pricing and benefits.

A New Way to Ace the Boards!

A New Way to Ace the Boards!

Ensure you're board-exam ready with ACP's Board Prep Ace - a multifaceted, self-study program that prepares you to pass the ABIM Certification Exam in internal medicine. Learn more.