Research report
Adolescent risk factors for episodic and persistent depression in adulthood. A 16-year prospective follow-up study of adolescents

https://doi.org/10.1016/j.jad.2007.06.001Get rights and content

Abstract

Background

We examined mid-adolescent psychosocial problems as risk factors for subsequent depression up to adulthood proper, and differences in these for episodic and persistent depression.

Methods

In a 16-year follow-up of an urban Finnish community cohort (547 males and 714 females) from age 16 years risk factors for subsequent depression (S-BDI) were studied. Data were collected with a classroom questionnaire at 16 years and a postal questionnaire at 22 and 32 years. Differences in predictors for episodic depression (only at age of 22 or 32 y) and persistent depression (both at 22 and 32 y) were studied using logistic and multinomial regression analyses.

Results

Mid-adolescent depressive symptoms predicted persistent and female sex episodic depression. Low self-esteem, dissatisfaction with academic achievement, problems with the law, having no dating experiences, and parental divorce all predicted both episodic and persistent depression.

Limitations

We had two assessment points in adulthood, but no information about depression between these.

Conclusions

The associations between mid-adolescent psychosocial problems and subsequent depression extended up to adulthood proper, somewhat differently for episodic and persistent depression. Preventive efforts should be focused towards young people at risk.

Introduction

Depression has an increasing prevalence in adolescence, high recurrence rates, and continuity to adulthood (Hankin et al., 1998, Kaufman et al., 2001, Roza et al., 2003). Depression in adolescence seriously impairs subjective well-being and social development (Fergusson et al., 2005, Kaufman et al., 2001, Kessler et al., 2001). The course of depression may, however, be episodic or chronic (Kaufman et al., 2001), and not all adolescents with significant psychopathology continue to have serious emotional problems in adulthood (Kessler et al., 2001, Lewinsohn et al., 2000, Reinherz et al., 2003).

Longitudinal studies show that mid-adolescent depression, depressive symptoms (Aalto-Setälä et al., 2002, Hankin et al., 1998, Reinherz et al., 2003), sub-clinical symptoms of depression (Fergusson et al., 2005, Pine et al., 1999) and multiple family stressors and psychological distress associate with later depression (Reinherz et al., 2003). Thereby young people prone to depression commonly have suffered from generalized psychosocial impairment and mood fluctuations in adolescence, assessed previously as being “normal”. Adolescence is thus a critical developmental phase for studying vulnerability to depression (Hankin et al., 1998), and research on risk factors besides depressive symptoms or disorders for subsequent depression may reveal additional risk factors and thereby help to identify youth at risk.

Several longitudinal studies have covered the period from adolescence to young adulthood (Hankin et al., 1998, Hofstra et al., 2001, Steinhausen et al., 2006), but less is known about the predictive impact of multiple psychosocial problems in mid-adolescence on well-being in adulthood proper.

We have previously reported that mid-adolescent (at 16 years) psychological distress symptoms, behaviour problems, poor family atmosphere, and interpersonal problems associated with subsequent depression in young adulthood (at 22 years) in a representative population based sample of adolescents (Pelkonen et al., 2003). In the present longitudinal study the same cohort was followed up to adulthood proper over a 16-year period. We studied psychosocial problems in mid-adolescence as risk factors for subsequent depression in adulthood and further, whether there were differences in these for subsequent episodic and persistent depression.

Section snippets

Subjects

The original study population included all ninth-grade pupils (N = 2269, participation rate 96.7%) attending secondary school in the spring of 1983 in Tampere, an industrial and university town in southern Finland with 166,000 inhabitants. These youngsters were studied at 16, 22 and 32 years of age (Fig. 1). The final sample size of the follow-up was 1262. The study protocol was approved by the Ethics Committee of the National Public Health Institute.

Predictor variables in adolescence

At age 16, the measure of depression was

Depression at T.2 and T.3

There was depression at 22 years only (T.2.) in 7.7% of the females and in 5.1% of the males, depression only at age 32 years (T.3.) in 12.0% of the females and in 8.8% of the males, and persistent depression (present at T.2. and T.3.) in 5.0% of the females and in 4.0% of the males. Gender specific proportions of subjects by categories of depression, and the gender rations are in Table 1.

To rule out the possibility that the incidence of depression at follow-up was a mere function of depressive

Discussion

In this follow-up study of a sizable and representative population based cohort of adolescents from mid-adolescence up to adulthood proper, the predictive value of mid-adolescent problems in psychosocial well-being for subsequent depression was found to extend over sixteen years. Most mid-adolescent risk factors associated both with episodic and persistent depression.

Our findings on the importance of adolescent well-being for adult depression accord with previous research (Hofstra et al., 2001,

Role of the funding source

The Yrjö Jahnsson Foundation and the Academy of Finland had no further role in study design, in the collection, analysis and interpretation of data, in the writing of the report and in the decision to submit the paper for publication.

Conflict of interest

No conflict declared.

Acknowledgements

The study was financially supported by the Yrjö Jahnsson Foundation and the Academy of Finland, grant number 80414. Jaakko Kaprio is supported by the Academy of Finland Center of Excellence in Complex Disease Genetics.

First page preview

First page preview
Click to open first page preview

References (41)

  • Aalto-SetäläT. et al.

    Depressive symptoms in adolescence as predictors of early adulthood depressive disorders and maladjustment

    Am. J. Psychiatry

    (2002)
  • AroH.

    Life stress and psychosomatic symptoms among 14 to 16-year old Finnish adolescents

    Psychol. Med.

    (1987)
  • AroH.

    Parental discord, divorce and adolescent development

    Eur. Arch. Psychiatr. Neurol. Sci.

    (1988)
  • BeckA.T. et al.

    Screening depressed patients in family practice. A rapid technic

    Postgrad. Med.

    (1972)
  • BeckA.T. et al.

    Short form of depression inventory: cross-validation

    Psychol. Rep.

    (1974)
  • CherlinA.J. et al.

    Effects of parental divorce on mental health throughout the life course

    Am. Sociol. Rev.

    (1998)
  • CyranowskyJ.M. et al.

    Adolescent onset of the gender difference in lifetime rates of major depression

    Arch. Gen. Psychiatry

    (2000)
  • EerolaM. et al.

    The problem of attrition in a Finnish longitudinal survey on depression

    Eur. J. Epidemiol.

    (2005)
  • FergussonD.M. et al.

    Subthreshold depression in adolescence and mental health outcomes in adulthood

    Arch. Gen. Psychiatry

    (2005)
  • FombonneE. et al.

    The Maudsley long-term follow-up of child and adolescent depression. 1. Psychiatric outcomes in adulthood

    Br. J. Psychiatry

    (2001)
  • Cited by (63)

    • Depression in Children and Adolescents

      2022, Comprehensive Clinical Psychology, Second Edition
    • Long-term influences of parental divorce on offspring affective disorders: A systematic review and meta-analysis

      2017, Journal of Affective Disorders
      Citation Excerpt :

      Nine case-control studies presented a range of STROBE scores from 63% (Gabardi and Rosen, 1991) to 81% (Nomura et al., 2002; Summers at al., 2002) (mean =74%). Ten perspective studies that were assessed for STROBE presented a range of STROBE scores from 68% (Gilman et al., 2003) to 94% (Pelkonen et al., 2008) (mean =84%). A one-way ANOVA revealed significant differences in quality among designs (p=0.04).

    View all citing articles on Scopus
    View full text