Background
Mental disorders are a leading cause of loss of health to disease in middle and high-income countries, and the resulting economic costs can be huge [
1]. Mental health is multidimensional, including mental disorders but also positive mental functioning. According to Keyes, there is wide variation in levels of mental health in the general population, with some people ‘flourishing’ (enthusiastic about life and actively engaged with other people), others ‘languishing’ (‘a life of quiet despair’) [
2], and the remainder ‘moderately mentally healthy’ [
3]. Keyes’ view that mental health should be regarded not just as the absence of mental illness but as a state of complete emotional, psychological, and social wellbeing is part of a growing international interest in what has come to be called positive mental health, often referred to as mental wellbeing [
3,
4].
Adolescent mental health problems are risk factors for future mental distress and psychopathology. Adolescent conduct problems are strongly associated with risk of psychiatric disorders, such as depression and substance abuse, in later life [
5‐
8]. Many studies have also reported that adolescent depression is associated with a strong, specific, and direct risk of recurrence in adulthood [
9‐
13]. Recent reports from birth cohort studies have indicated that not only conduct and emotional problems during adolescence, but also low self-organisation (defined in terms of ‘effortful regulation of the self by the self’) [
14], may be a significant developmental precursor predicting future mental health problems including substance dependence, depression, and hallucinations [
13,
15,
16]. While such empirical evidence leaves no doubt that adolescent mental health problems are associated with poor mental health in later life, to our knowledge no study has investigated adolescent mental health problems in relation to mental wellbeing in later life.
Childhood factors, particularly childhood cognitive ability and early-life socioeconomic position may affect adolescent mental health [
17], and educational attainment by early adulthood may be a possible mediator between adolescent mental health problems and future mental wellbeing [
6,
9]. Analyses that consider the effects of these variables are needed when examining association between adolescent mental health problems and mental wellbeing in old age. In addition, an analysis that considers mental ill health such as symptoms of anxiety/depression in later life as a confounder is needed to assess whether association between adolescent mental health problems and mental wellbeing in later life are fully explained by mental ill health in later life or whether adolescent mental health problems also have implication for future mental wellbeing.
The current study aims to investigate associations between adolescent mental health problems (lower self-organisation, conduct and emotional problems) and mental wellbeing and life satisfaction in early old age considering the effects of childhood cognitive ability, early-life socioeconomic position, educational attainment and mental ill-health in early old age, using longitudinal data from The Medical Research Council National Survey of Health and Development (NSHD, the 1946 British birth cohort study) which is one of the longest continuously running studies of human development and aging in the world [
18].
Discussion
Using longitudinal data from a national birth cohort, we found that emotional problems in adolescence predicted mental wellbeing and life satisfaction in early old age, after controlling for gender, social class of origin, childhood intelligence, and educational attainment. The association with life satisfaction may operate through symptoms of anxiety/depression in early old age. We did not find any significant associations between adolescent self-organisation or conduct problems and future mental wellbeing or life satisfaction.
Empirical evidence leaves little doubt that adolescent emotional problems are associated with risk of future mental ill-health such as depression [
9,
30]. Our current study revealed that adolescent emotional problems may also affect mental wellbeing (positive mental health based on an instrument that captures positive mental functioning and satisfying personal relationships as well as positive affect and the cognitive-evaluative dimension of wellbeing) in early old age even after controlling the effect of concurrent symptoms of anxiety/depression, suggesting that a pathway which was not fully explained by symptoms of anxiety/depression may be operating. On the other hand, in our study, a significant association between adolescent emotional problems and life satisfaction measured by SWLS in early old age was fully attenuated after adding concurrent symptoms of anxiety/depression. Life satisfaction has been defined as an overall assessment of individual’s quality of life according to his/her chosen criteria [
21]. It may be reasoned that mental health problems such as depression may play important role in the cognitive-evaluative process of life satisfaction. In contrast, data from the 1970 British Cohort Study show that greater conduct problems in childhood and adolescence were associated with lower life satisfaction at age 34 [
31]. Explanation for this discrepancy may possibly be due to the ages at which outcomes were assessed, the different life satisfaction or conduct instruments employed, or different long-term implications of conduct problems during the adult lives of those born more recently.
Some birth cohort studies have reported that poor self-organisation in adolescence might be a significant developmental precursor predicting a broad range of adverse outcomes in later life including educational and occupational underachievement, poor physical health, and trouble with the criminal justice system [
13,
15]. However, these studies have not considered the effects of both conduct and emotional problems in adolescence as confounders. Our previous study revealed that poor self-organisation in adolescence was a risk factor for future psychopathology, such as hallucination, after controlling for the effects of adolescent conduct and emotional problems [
16]. On the other hand, our current results suggest that self-organisation in adolescence was not associated with mental wellbeing and life satisfaction after controlling for the effects of adolescent conduct and emotional problems. Together, these pieces of evidence suggest that lack of self-organisation in adolescence might be a driver for later mental illness, but that self-organisation is not related to later positive mental wellbeing once we consider the effects of past conduct and emotional problems.
Our current results indicate that conduct problems in adolescence were not associated with mental wellbeing or life satisfaction in later life (after controlling for the effects of emotional problems and self-organisation in adolescence). Among teenagers, rates of comorbidity between depression and conduct problems are high [
32]. Conduct problems are strongly associated with early-onset depression, the most common pattern being that conduct problems precede the expression of affective symptomatology [
9]. Therefore, it is necessary to consider the mutual effects of these two factors when investigating the associations between conduct and emotional problems in adolescence and future outcomes.
It is also worth noting that childhood cognitive ability was associated with poorer life satisfaction in early old age after adjustment for social class in origin, educational attainment, and symptoms of anxiety/depression in later life. Analysis of a younger birth cohort to age 34 showed a positive correlation between life satisfaction and childhood and adolescent cognitive ability, though this was smaller in magnitude than the correlations between life satisfaction and either emotional or conduct problems in childhood and adolescence [
31]. In contrast, the Lothian birth cohort 1921 study, which consisted of 550 older adults, showed that correlations between IQ at age 11 and the life satisfaction measured by SWLS at age 79 were null (
r = 0.00) [
33], although the study did not control for adolescent mental health, educational attainment, or depression in later life. In order to explain the absence of this relation, the authors refer to the fact that higher ability is equally likely to lead to positive and negative outcomes (e.g. increasing one’s resources through entry to better employment, and an awareness of alternative lifestyles or striving for greater achievement, respectively), which may be involved in measuring subjective wellbeing [
34]. Our results suggest that higher childhood cognitive ability is more likely to lead to negative outcomes when considering the additional effects of adolescent mental health, educational attainment, and symptoms of anxiety/depression in later life.
Strengths of this study include the use of a national population-based sample, the availability of independent teacher ratings for adolescent mental health, and a comprehensive range of potential explanatory variables, including cognitive ability measured during childhood.
Against these strengths we acknowledge some limitations. A limitation of this study is attrition of survey members over the 60 year follow-up period. Typically for longitudinal health-related studies, there was disproportional loss to follow-up in the NSHD for those who were relatively less socially advantaged, had poorer adolescent mental health, and who had lower childhood cognitive ability. However, while this might have led to some degree of underestimation of association strengths, we have no reason to believe that this would have influenced the
pattern of these associations. As there were no data on mental wellbeing and life-satisfaction in adolescence in NSHD, the autoregressive effect of them could not be controlled in our analysis. It therefore remains unclear whether adolescent emotional problems predict mental wellbeing once pre-existing level of mental wellbeing is controlled. The NSHD data contain only teachers’ assessments of the children’s emotional adjustment and behaviours, with no information from the parents or the children themselves. However, teachers can make an important contribution to the identification of adolescent behavioural problems, which is often missed when children or parents report on behaviour [
35]. We used data of teachers’ assessments of the children’s emotional adjustment and behaviours collected almost 50 years ago to make a recently-identified construct, self-organisation. In our study, adolescent self-organisation was defined through secondary analysis by a relatively small number of items, rather than by an instrument specifically designed to capture this construct; however, the psychometric techniques used to distinguish it from adolescent emotional and conduct problems were rigorous, and the resulting measure was found to have discriminant properties [
24].