Factors that differentiate early vs. later onset of major depression disorder

Abstract

This report explores the relationship between age of first onset of major depression and other demographic and clinical features in the first 1500 patients entering the Sequenced Treatment Alternative to Relieving Depression (STAR*D) study. Outpatients, 18–75 years of age, with nonpsychotic major depressive disorder (MDD) from either primary care or psychiatric practices constitute the population. Age of onset was defined at study intake by asking patients to estimate the age at which they experienced the onset of their first major depressive episode. This report divides the population in terms of pre-adult (before age 18) onset and adult (age 18 or later) onset. The results suggest that MDD that begins before age 18 has a distinct set of demographic (female gender) and clinical correlates (longer duration of illness; longer current episodes; more episodes; more suicidality; greater symptom severity; more psychiatric symptoms associated with Axis I comorbidity; and more sadness, irritability, agitation and atypical symptom features), and it appears associated with significant psychosocial consequences (lower educational attainment and marriage rates). Thus, pre-adulthood onset MDD is a particularly severe and chronic condition.

Introduction

The average age of first onset of major depressive disorder (MDD) is in the mid-20s, although first onset can occur from childhood through old age. Epidemiologic studies in lifetime onset of MDD reveal trends towards increasing overall rates, as well as an earlier age of onset, with each successive birth cohort (Lavori et al., 1993). While recent research has focused more intently on MDD beginning in adolescence or childhood, there remain limited systematically gathered data on the uniqueness of early-onset MDD. For example, is earlier onset associated with unique presenting symptoms, severity of illness, comorbidity, course of illness or treatment requirements? This report takes advantage of systematically gathered information from the largest depression treatment study ever attempted, Sequenced Treatment Alternative to Relieve Depression (STAR*D) (Fava et al., 2003, Rush et al., 2004), to comprehensively explore factors associated with early as opposed to later age of onset of MDD.

Diagnostic criteria for major depressive episodes (MDE) are similar throughout the life cycle. One difference for children and adolescents is that irritability is felt to be particularly common in pre-adulthood and an irritable or cranky mood may replace a sad or dejected mood as the so-called A.1 Criterion (American Psychiatric Association, 2000, p. 349). Whether this distinction is maintained into adulthood in those individuals who remain depressed or develop recurrent MDE as adults compared with individuals whose first onset of MDD occurs in adulthood is unknown. Similarly, some investigators have posited more mood lability and a greater tendency towards atypical features (i.e., mood reactivity, oversensitivity to interpersonal rejection, overeating, oversleeping and leaden paralysis) (Ryan et al., 1987, Carlson and Kashani, 1988, Benazzi, 2000, Klein et al., 1998) in adolescents compared with adults with MDD, but again, this may be a distinction more related to patient age than to age of onset.

Some studies have found different risk factors for childhood/adolescent onset than for adult-onset MDD. An early age of onset has been associated with increased familial loading for depression (Weissman et al., 1988, Klein et al., 2001, Neuman et al., 1997), particularly when the affected parent has had an early age of onset (Wickramaratne and Weissman, 1998). Some (Kornstein et al., 2000) but not other (Weissman et al., 1993) studies have reported that females have an earlier age of onset of MDD than males. Kessler and Magee (1993) reported that family violence, parental psychopathology and the early death of a parent increased the risk for early (by age 20) but not later onset of depressive symptoms.

Following a representative birth cohort prospectively from birth to age 26 years, Jaffee et al. (2002) found that pre-adulthood-onset MDD may be a distinct subtype associated with both genetic and early childhood risk factors. They identified four distinct subgroups: (1) those diagnosed with MDD in childhood but not adulthood (n=21), (2) those diagnosed with MDD in adulthood but not childhood (n=314), (3) those diagnosed in childhood and adulthood (n=34) and (4) never-depressed individuals (n=629). Compared with adult-onset MDD, those with childhood or adolescent onset experienced more perinatal insults and motor skill deficits, caretaker instability, criminality and psychopathology in their family of origin and behavioral and socioemotional problems. The adult-onset group's risk profile was similar to the never-depressed group with the exception of elevated childhood sexual abuse.

Other clinical characteristics of childhood and adolescent MDD have been well delineated. For example, several investigators have noted high rates of comorbidity in child and adolescent MDD (Biederman et al., 1995, Birmaher et al., 1996). The most consistent finding has been high rates of anxiety disorders in children and adolescents with MDD. In childhood, Separation Anxiety Disorder appears particularly prevalent, whereas Social Anxiety Disorder appears more characteristic of depressed adolescents (Ryan et al., 1987, Geller et al., 1985, Puig-Antich and Rabinovitch, 1986, Rohde et al., 1991). High rates of comorbid Conduct Disorders and Attention Deficit Hyperactivity have been noted by several investigators (Biederman et al., 1991, Ryan et al., 1987).

In terms of course of illness, adolescent MDD has been associated with elevated rates of subsequent MDEs in early adulthood (Harrington et al., 1990, Weissman et al., 1999). In contrast, childhood depression may be a risk for bipolar disorder (Geller et al., 1994, Kovacs and Goldston, 1991, Weissman et al., 1999), future conduct disorder and substance abuse (Geller et al., 1985, Harrington et al., 1991, Kovacs et al., 1989, Ryan et al., 1987) but not necessarily for recurrent MDE in adulthood (Weissman et al., 1999). Other studies have suggested elevated rates of suicidality (Kovacs et al., 1993, Rao et al., 1993), medical and psychiatric hospitalizations (Klein et al., 1999, Harrington et al., 1990) and work, family and social impairment (Rao et al., 1995) in adolescents with MDE. Thus, there appears to be heterogeneity among early-onset MDD, with differences between prepubertal and adolescent-onset depressions in comorbidity patterns and courses of illness.

While much is known about the clinical characteristics and course of MDD in children and adolescents, most studies to date have not simultaneously compared adults with early-onset MDD compared with those with adult-onset MDD. Thus, it is difficult to draw accurate inferences about the precise role of age of onset, as opposed to present age, on other clinical features. Studies that have examined large adult populations classified based on the age of onset of the first MDE help separate the relative contributions of age and age of onset to the clinical characteristics of MDD. Such investigations have generally reported that early-onset MDD may be a more severe variant, with increased chronicity and disability, increased levels of anxiety or rates of anxiety disorders, alcohol and other drug dependence, and behavioral and personality difficulties (Parker et al., 2003, Ramklint and Ekselius, 2003, Klein et al., 1999). In the largest of such studies, 404 adults with a current diagnosis of MDE who were consecutively enrolled in treatment studies at the Massachusetts General Hospital Clinical Psychopharmacology Unit were divided into early-onset (<age 18) and adult-onset groups. As predicted, higher rates of several personality disorders (avoidant, passive–aggressive, histrionic, narcissistic, borderline and antisocial), social and simple phobias and alcohol abuse/dependence were found in the early-onset (age<18 years) than in the adult-onset group (Fava et al., 1996, Alpert et al., 1999).

This report explores the relationship between age of onset (before age 18 years or thereafter) and other demographic and clinical features of MDD in a large, consecutive series of the first 1500 patients from an expected series of 4000 patients from primary care and psychiatric practices entering the STAR*D depression treatment protocol (http:www.star-d.org). This preliminary report looks at factors that might affect risk for early- vs. late-onset MDD, as well as factors that might be affected by age of onset. We specifically evaluated whether increased family loading, severity, impairment, atypicality and current comorbid symptomatology differentiated early-onset from later onset MDD.