Clinical features related to age at onset in bipolar disorder

https://doi.org/10.1016/j.jad.2003.10.002Get rights and content

Abstract

Background: Early age at illness onset has been associated with poor functional and syndromal outcome in bipolar disorder, although debate remains about the likely robustness of this variable, especially while controlling for other illness parameters. Method: Fifty-six consecutive bipolar outpatients underwent semistructured interviews to assess psychopathology and outcome. We hypothesized that early age at onset would be linked with more prevalent rapid cycling, psychosis, comorbid substance abuse, and suicide attempts. Results: Illness onset before age 19 arose in 46% of subjects. Separate logistic regression analyses revealed that onset before age 19 was associated with developing comorbid substance abuse/dependence (OR=7.714, 95% CI=1.863–31.944, Wald χ2=7.949, df=1, P=0.005), as well as the eventual development of rapid cycling (OR=6.000, 95% CI=1.250–25.893, Wald χ2=5.348, df=1, P=0.021). No significant or near-significant associations were observed between age at onset and lifetime suicide attempts or lifetime psychosis. After an initial manic or mixed episode, delaying the introduction of antidepressants tended to be protective against the eventual development of rapid cycling (OR=0.927, 95% CI=0.856–1.004, Wald χ2=3.440, df=1, P=0.064). Limitations: The sample size may have been too small to detect group differences of small magnitude. The use of retrospective life chart assessments to ascertain age at onset and lifetime illness course may impose limitations on generalizability in the absence of prospective data. Conclusions: Early onset of bipolar illness appears related to the development of rapid cycling and of comorbid substance abuse/dependence. The findings raise developmental implications for the pathogenesis of illness complexity and poor outcome states.

Introduction

Early age at illness onset has been widely reported as a negative prognostic variable in major Axis I disorders such as schizophrenia Ho et al., 2000, Lieberman et al., 1994, Meltzer et al., 1997. Among individuals with bipolar disorder, early onset has been linked with greater illness severity (Schulze et al., 2002), psychosis Ballenger et al., 1982, Carlson et al., 2000, greater Axis I comorbidity (McElroy et al., 2001), more depressive episodes (Benazzi, 2001), slower recovery times (Tohen et al., 2000), poorer outcome Black et al., 1988, Carlson et al., 2002, and suicidality Lopez et al., 2001, Tsai et al., 1999. However, others have linked early age at onset with comparable or even better outcomes than later onset presentations McGlashan, 1988, Tsai et al., 2001, while still others dispute the predictive utility of age at onset Carlson et al., 1977, Coryell et al., 1998. Such conflicting findings may partly reflect failures to control for possible intervening variables that may covary with age at onset and outcome states. Elucidation of these inter-relationships is, therefore, of both theoretical and clinical importance for the management of bipolar disorder.

Previously we reported that lengthier delays from initial symptom onset to the initiation of a first mood stabilizer pharmacotherapy were associated with both early age at onset as well as a number of negative clinical outcome domains (Goldberg and Ernst, 2002). The present study extends these findings by further clarifying the relationship between early age at onset relative to four major aspects of clinical psychopathology: rapid cycling, psychosis, comorbid substance misuse, and lifetime suicide attempts. We hypothesized that early age at onset would be associated with an increased likelihood for each of these phenomena, even while controlling for a number of demographic or symptom-based correlates of poor outcome. In addition, based on recent observations that rapid cycling may arise more often for some bipolar patients given antidepressants prior to first lifetime manias or hypomanias (Yildiz and Sachs, 2003), we hypothesized that rapid cycling would be more likely to arise in those with early onset depressions than manias.

Section snippets

Study subjects

The study group consisted of 56 patients with bipolar disorder type I (n=46), II (n=7) or not otherwise specified (NOS; n=3). Diagnoses of bipolar disorder, as well as of comorbid substance abuse or dependence, were made using the Structured Clinical Interview for the DSM-IV (SCID; First et al., 1995). SCID assessments were conducted by trained and experienced research assistants at the time of entry into a comprehensive clinical research program for bipolar disorder based at the Payne Whitney

Study group characteristics

At the time of assessment, subjects had a mean (S.D.) age of 40.8 (12.8) years. Forty-five percent were female and 77% were Caucasian. Eighty-four percent of the study group had completed at least some college, and 39% had never been married. Their mean (S.D.) HAM-D31 and YMRS scores at the time of assessment were 17.8 (13.0) and 3.7 (5.2), respectively. The group had a mean (S.D.) age at onset for first affective illness of 21.7 (9.6) years. Twenty-six of the 56 subjects (46%) had their

Discussion

The present findings suggest that early age at onset in bipolar disorder is more often insidious than acute and may be linked with the eventual development of both rapid cycling and comorbid substance misuse. Stratification by outcome states preliminarily suggests that rapid cycling is more likely to arise in early onset bipolar patients when substance abuse/dependence is present rather than absent. No significant associations were found between early onset and suicidality or psychosis.

The

Acknowledgements

Supported in part by NIMH Career Development Award Grant K23 01936 (J.F.G.), a NARSAD Young Investigator Award (J.F.G.) and by resources from a fund established in the New York Community Trust by DeWitt Wallace.

References (29)

  • G.A. Carlson et al.

    A comparison of outcome in adolescent- and later- onset bipolar manic-depressive illness

    Am. J. Psychaitry

    (1977)
  • G.A. Carlson et al.

    Phenomenology and outcome of subjects with early- and adult-onset psychotic mania

    Am. J. Psychiatry

    (2000)
  • G.A. Carlson et al.

    Age at onset, childhood psychopathology, and 2-year outcome in psychotic bipolar disorder

    Am. J. Psychiatry

    (2002)
  • M.B. First et al.

    Structured Clinical Interview for DSM-IV

    (1995)
  • Cited by (78)

    • Older Age Bipolar Disorder

      2020, Clinics in Geriatric Medicine
    • Bipolar disorders in older adults

      2020, Handbook of Mental Health and Aging
    • Comparisons of the clinical outcomes between early- and adult-onset bipolar disorders: A prospective cohort analysis

      2020, Journal of Affective Disorders
      Citation Excerpt :

      Early-onset bipolar disorders (EOBD) has been considered as a distinct subtype of bipolar disorders with a particularly ominous outcome (Joslyn et al., 2016). Evidence from previous cross-sectional studies and one recent meta-analysis suggests that EOBD patients exhibit increased psychiatric comorbidities, higher suicidality, increased number of mood episodes, and more severe long-term functional impairment than those with adult-onset bipolar disorder (AOBD) (Azorin et al., 2013; Baldessarini et al., 2012; Bellivier et al., 2001; Benazzi, 2009; Biffin et al., 2009; Birmaher et al., 2006; Carter et al., 2003; Ernst and Goldberg, 2004; Etain et al., 2012; Goldstein and Levitt, 2006; Grunebaum et al., 2006; Holtzman et al., 2015; Javaid et al., 2011; Kennedy et al., 2005; Leverich et al., 2007; Lin et al., 2006; Moor et al., 2012; Perlis et al., 2004; Sala et al., 2013; Tozzi et al., 2011). In addition, family studies found a higher rate of BD among the relatives of EOBD patients than those of AOBD patients (Lin et al., 2006).

    • Predictors of longitudinal psychosocial functioning in bipolar youth transitioning to adults

      2019, Journal of Affective Disorders
      Citation Excerpt :

      While the Diagnostic and Statistical Manual of Mental Disorders-IV and 5 (DSM-IV/5) criteria for depression, mania, and hypomania episodes require that symptoms be associated with marked psychosocial impairment or change in psychosocial functioning (Association, 2004), little is known about how psychosocial functioning may evolve longitudinally from youth through young adulthood in Bipolar Disorder (BD) youth. Studies that have examined psychosocial functioning during this transitional period have been limited by their recruitment of adults reporting retrospectively on past functioning, and the use of a cross-sectional design (Ernst and Goldberg, 2004; Leverich et al., 2007; Suominen et al., 2007; Tasha et al., 2003). The Course and Outcome of Bipolar Youth (COBY) study, a multi-site longitudinal research study of youth with BD followed into adulthood, provides a unique opportunity to examine longitudinal psychosocial functioning outcomes.

    View all citing articles on Scopus
    View full text