Depression is a serious, yet common illness that affects people of all age, ethnic and socioeconomic groups. It is estimated that 10 percent of the population experiences depression annually, and 17 percent of the population becomes depressed at some point within a lifetime (Kessler, 1994). Depression is often chronic (recurring) in nature and is associated with high rates of relapse and recurrence, psychosocial and physical impairment and high morbidity/mortality. For example, people who experience one episode of depression have a 50 percent chance that the illness will recur while those who experience two episodes have a 70 percent chance. Those who experience three episodes of depression have a 90 percent chance that they will be depressed in the future. Furthermore, patients who do not achieve full remission from their depression and continue to experience symptoms of depression after being treated for it are at an increased risk for relapse.

In the 1990s, chronic depression had a reported prevalence rate of 3 to 5 percent among the general population in the United States, and these cases accounted for 30 to 35 percent of all United States cases of depression (Keller, 1995). Depression noticeably impairs a person's functioning in both social and occupational roles, possibly impacting both familial well-being and economic status.

These statistics point to the necessity of proper diagnosis and adequate treatment of depression, particularly treatment to the point of full remission of depressive symptoms. Yet, despite the existence of numerous effective antidepressant medications, depressed patients are often under-diagnosed and under-treated by both psychiatrists and primary care physicians. Lack of effective diagnosis and treatment results in numerous patients experiencing residual symptoms of depression, which continue to impair their psychosocial functioning and increase their risk of experiencing another depressive episode.

Course of Illness for Depression
The treatment of depression can be divided into three phases:

1. acute, which aims at symptom remission,
2. continuation, which continues the effective medication and aims at preventing a return of the most recent episode (relapse); and
3. maintenance, which aims at preventing the onset of a new recurrence.

Some people use the terms relapse and recurrence interchangeably, but they are separately defined, and must not be confused. The definition of a relapse is the return of symptoms during the first few months following a response to treatment or spontaneous remission. It is important to realize that the period of risk for relapse is highest in the 4-6 months following full remission. After this time, the relapse rates decline considerably (Belsher, 1988), suggesting a natural break between the events of relapse and recurrence.

Recurrence is a new episode of depression that occurs after a sustained period of remission. Risk factors of recurrence include a previous history of depressive episodes, lack of full recovery between episodes, being at an older age at onset, experiencing a long duration of the first episode, documented family history of depression, experiencing stressful life events, possessing poor social support and co-morbid anxiety. Maintenance phase therapy can prevent recurrence and increase the likelihood of sustained recovery. Continued efficacy in long-term trials suggest that maintenance treatment is required for at least 12 months, probably longer for those at risk for recurrent depression. Given the statistics for relapse, all patients experiencing their third depressive episode, and some patients having their second, should be considered for maintenance treatment.

Goals of Treatment
The aim of treatment should be to induce remission of the depressive disorder (full relief of the signs and symptoms) with full restoration of psychosocial function. Furthermore, each patient should be guided to establish a long-term state of wellness to prevent relapse or recurrence of the depressive state.

Of studies comparing antidepressant therapies, the majority use response as the standard criterion by which to compare efficacy, the ability of a drug to be proved effective. Response usually indicates that the patient has improved by at least 50 percent on a standardized clinician-rated assessment measure, such as the Hamilton Rating Scale for Depression (HRS-D), compared to the patient's first visit. Yet, depending on the severity of the patient's depression on his or her first visit, a 50 percent improvement does not ensure that the patient is symptom-free. For example, a patient with severe depression may score a 28 on the HRS-D at baseline and a 14 at the end of a clinical trial. By this definition, the particular patient has "responded" to the medication, but still shows significant depressive symptoms. Remission, which can be defined by a HAM-D score of less than 8, may be more relevant to clinical practice because it indicates that the patient is well and not experiencing significant residual symptoms of depression. Though most randomized, controlled clinical trials do not document complete symptomatic remission and restoration of functioning, evidence from long-term efficacy studies suggest that patients who fare best during continuation-phase treatment are those who have attained the most complete remission of their symptoms during the acute phase (AHCPR, 1993). Achievement of remission, rather than response would provide a more realistic measure of antidepressant efficacy. There are many reasons why full remission is important, so instead of positive or high response, remission should more often be the criterion by which to measure the effectiveness of treatments for depression.

Rationale for the Goal of Full Remission
Unremitting depression has numerous consequences for the individual, family and society. Patients who respond only partially have a higher percentage of relapse (Judd et al, 1998) and decreased functioning (Miller et al, 1998) compared to those who remit fully. Patients with residual depressive symptoms also have increased health service utilization, general medical complications, decreased productivity and shortened life span.

Outcome Assessment in Depression
Since the goal of acute phase treatment is to induce full remission of symptoms and restoration of functioning, measures of symptom severity and functional outcome can be very useful in determining whether full remission, rather than a simple response has been attained. In addition, measurements of symptom reduction and functional improvement can yield invaluable information at critical points in decision making related to clinical care. Outcome measurements encourage clinicians to create a treatment plan that provides a basis for making tactical decisions (e.g., changing dosages, extending the trial) for their patients. Outcome assessment can also serve as necessary documentation to supply a clear basis for treatment-related decision. Healthcare providers and other administrators are increasingly demanding accountability, which ongoing outcome assessment can provide.

Clinically relevant outcomes that need to be assessed include symptoms of depression and function, which are interrelated. Depressive symptoms include mood, interest, sleep, appetite, concentration and motivation, among others. Function is gauged in the context of work, marriage, family and friends. Clinician-rated symptom measurements, while time consuming, have been shown to be more sensitive than patient-reported symptom measurements in detecting improvement with treatment, thus providing a firm basis for making treatment decisions. There are numerous rating scales for depression, which vary in their coverage of symptom domains and format. The Hamilton Rating Scale for Depression (HRS-D) is considered the "gold standard" in depression research, though many different versions exist. The Inventory for Depressive Symptomatology (IDS) covers the full range of depressive symptoms and is available in both clinician-rated and self-report formats, which facilitates the comparison of these two perspectives. A widely used measure of social functioning is the Social Adjustment Scale Self Report (SAS-SR). It is the only scale that assesses a patient's functioning in family, marital and parental roles in addition to work.

The routine use of clinical rating scales can provide useful information to assist the physician in determining when to adjust a patient's dose, or change his or her medication. Measurement of symptom outcomes can also help clinicians inform managed care professionals and other administrators when continuation treatment is indicated.

Strategies for Achieving Remission
No pretreatment clinical predictors have been identified to consistently predict whether an individual patient is likely to attain a full symptomatic remission. In fact, evidence from studies of double depression suggests that full remission can be attained with careful treatment, even in those patients with chronic depression. However, patients who achieve full symptomatic remission during the acute phase of treatment need no revision in their medication management. Continuation therapy for these patients should entail continuing the effective medication and monitoring their psychosocial functioning.

For patients who did not respond fully, but show a partial response, careful monitoring of response with possible dosage adjustments may be sufficient. For patients with no response at six weeks or those with only a partial response by 12 weeks, a revision in treatment strategies is indicated. The clinician has several choices:

1. make no change in treatment tactics and wait for several more weeks,
2. increase the dose of the current medication,
3. augment the partially effective medication with a second medication to attain full remission,
4. switch from the partially successful medication to a new medication, typically of a different class, or
5. add psychotherapy.

There is some evidence that indicates, especially for those with more chronic depression, a prolonged acute-phase treatment period may be required to attain a full therapeutic response (Keller et al, 1998); however, after an additional few weeks with little improvement, a change is warranted. If an increase in dose reveals a less-than-satisfactory response, switching or augmentation should be considered.

Switching to a different class of antidepressants may be the most logical choice if patients have failed to respond to another medication in this class because longer term treatment would then involve, if successful, only a single medication. Monotherapy has the advantage of minimizing the likelihood of drug interactions, reducing pharmaceutical expenses and increasing compliance because fewer medications will be need to be taken less often. Switching medications is also preferred if the patient experiences numerous side effects. On the other hand, especially for patients who have already failed to respond to several medication classes, yet have attained a partial response to acute-phase treatment, augmentation may be the better choice. Augmentation has been associated with a more rapid response (Thase, Howland, Frieman, 1998). Augmentation may ultimately be more effective for patients who have not achieved a satisfactory response on different types of medications separately due to the combination of mechanisms of action (Nelson, 1998). Choosing one antidepressant with more than dual mechanisms of action, such as Venlafaxine XR or mirtazapine may offer more robust efficacy than medications that have a single mechanism of action. Finally, another approach undergoing evaluation in clinical trials is the use of cognitive-behavioral therapy (CBT) or interpersonal psychotherapy (IPT). CBT and IPT have been shown to be equally effective to pharmaceuticals in the treatment of depressive disorders.

Conclusion
The original standard of attaining 50 percent improvement in the depressive patient is insufficient, since residual symptoms of depression are associated with continued dysfunction increases in the risk of developing further depressive episodes. Furthermore, adherence to this rationale may promote future chronic depression in some patients who are not in remission. For these, continued aggressive treatment and evaluation can bring the clients to the point of true remission. Full remission and long-term recovery, rather than short-term response are the desired outcomes from antidepressant therapy. Once a response is seen, further tactical (e.g. dosage adjustment or augmentation) or strategic (e.g., addition of psychotherapy or rehabilitative services) options should be considered before accepting a response that is anything short of remission.t

Dr. Madhukar H. Trivedi is Director, Depression Module Texas Medication Algorithm Project; Director, Depression and Anxiety Program and Associate Professor, Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas, Texas. He has had a distinguished career with experience in radiology, psychiatry and other related fields at highly respected institutions. His honors and awards include 1997-1999 Scientific Content Expert, Newer Pharmacotherapies for Depression from the San Antonio Evidence Based Practice Center and 1992-1994 National Alliance for Research in Schizophrenia and Depression (NARSAD) Young Investigator Award. Dr. Trivedi has published extensively in renowned journals. He can be contacted by e-mail at: madhukar.trivedi@utsouthwestern.edu.

Tracie Voegtle is a senior research associate in the Depression and Anxiety Disorder Program of the Department of Psychiatry, University of Texas Southwestern Medical Center, Dallas, Texas.