II. Key clinical features of unipolar depression
MB. Keller and colleagues reviewed important factors predisposing patients to recurrence of depressive symptoms and highlighted several risk factors that should be considered when modelling disease evolution .
Concentrating on prophylactic strategies requires consensus of definitions for specific concepts such as relapse and recurrences of depressive symptoms. In 1988, the MacArthur Foundation Research Network on the Psychobiology of Depression consensus group agreed on the definition of terms required to designate the relevant change points over the course of illness. These definitions have provided a framework for deciding what constitutes "an episode" and have further clarified the concepts of severity and duration.
• Remission is defined as "a relatively brief period during which an improvement of sufficient magnitude is observed so that the individual is asymptomatic, i.e. the patient no longer meets syndromal criteria for the disorder and has no more than minimal symptoms".
• Recovery is defined as "an asymptomatic period that lasts longer than the remission period". This definition is used to designate recovery from the episode, not from the depressive symptoms per se, and implies a sustained remission of symptoms.
• Relapse is defined as "the early return of depressive symptoms following an apparent remission".
• Recurrence is defined as "the appearance of a new episode of major depressive disorder and thus can only occur during a period of recovery".
Figure 1 provides a visual understanding of how distinct phases of depression differentiate relapses from recurrences, and remission from full recovery.
In this paper, a "depressive event" is defined as the occurrence of depressive symptoms. A depressive episode may include several depressive events.
II.2 Important risk factors
The following section highlights key features in terms of risk factors for unipolar major depression. The illustrative data presented hereafter were extracted from published literature [7-10]. Long-term prospective studies of patients with depression are somewhat scarce, therefore this work was mainly based on the National Institute of Mental Health (NIMH) Collaborative Program on the Psychobiology of Depression study [7,9,10]. This study was a prospective, naturalistic long-term follow-up that aimed to describe the episodic course of illness in major depressive disorder. Recruited individuals received either outpatient or inpatient care (outpatients represented 25% of the total sample).
One of the major findings from this long-term follow-up study suggested that the number of previous depressive events a patient experienced significantly influenced their probability of relapse (Figure 3).
In approximately 20% of cases, as duration of depressive symptoms increased, the chances of remission decreased (Figure 4). These findings reinforce the chronic nature of the illness for a substantial number of patients .
The presence of residual depressive symptoms has also been proven to be associated with an increased risk of short-term relapse as well as with a long-term chronic course. Patients' attitude towards treatment has also been widely discussed as a key predictive factor of the long-term course of the disease. Olfson and colleagues  recently showed that approximately 4 out of 10 patients (42.4%) who initiated antidepressant treatment for depression discontinued the antidepressant medication during the first 30 days of treatment, and among those who continued antidepressant therapy for more than 30 days, one-half (52.1%) discontinued the medication during the subsequent 60-day period. Moreover, a 2-year naturalistic study showed superior long-term recovery in patients who were adherent to antidepressant medication compared with non-adherent patients [8,13].
Lastly, sociodemographic characteristics such as age and gender have also been proven to be significant factors to be taken into consideration .
Figure 2 summarizes the key factors for recurrent depression.
Management of patients therefore requires differentiating between their sociodemographic characteristics, disease history, number of prior episodes and compliance to treatment over time.