Elsevier

Social Science & Medicine

Volume 197, January 2018, Pages 153-160
Social Science & Medicine

Adolescent mental health and behavioural problems, and intergenerational social mobility: A decomposition of health selection effects

https://doi.org/10.1016/j.socscimed.2017.11.026Get rights and content

Highlights

  • Adolescent mental health and behavioural problems affect socioeconomic attainment.

  • To what extent does the childhood environment account for this association?

  • Half the effect of conduct problems is accounted for by earlier risk factors.

  • No such attenuation for emotional symptoms or hyperactivity.

  • Smaller effects on mobility of emotional symptoms than conduct or hyperactivity.

Abstract

According to the health selection hypothesis, poor mental health and behavioural problems in adolescence limit socioeconomic attainment. But to what extent is health selection driven by prior social causation? This paper quantifies the extent to which health selection – here, restricted or downward intergenerational social class mobility due to poor mental health or behavioural problems – can be attributed to the influence of modifiable childhood risk factors. The UK National Child Development Study provides measures of socioeconomic deprivation and childhood adversities, as well as multiple-informant ratings of adolescents' mental health and behavioural problems, for which confirmatory factor analysis supports a three-factor model. Decomposition analysis is used to robustly assess the extent of attenuation of selection effects when conditioning on the childhood environment. Conduct problems, hyperactivity, and to a lesser extent emotional symptoms at age 16 are associated with individuals' chances of achieving (un)desirable mobility outcomes. When prior childhood risk factors are taken into account, the association of conduct problems with mobility is attenuated by around 50%, indicating a substantial role for confounding and earlier processes of social causation in the generation of this health selection effect. Further analyses indicate that this attenuation is greater for those from the most disadvantaged backgrounds and is mostly driven by the inclusion of indicators of generalised disadvantage such as crowded housing and low income. On the other hand, the effects of emotional symptoms and hyperactivity on mobility outcomes are not significantly accounted for by childhood risk factors. This study adds to the health inequalities literature by interrogating the empirical validity of the usual interpretation of health selection effects as indicating the causal priority of the onset of poor mental health.

Introduction

Inequalities in health outcomes are ubiquitous, and this is true of mental as well as physical health. Explanations usually focus on the mechanisms of health selection and social causation (Dohrenwend et al., 1992, Miech et al., 1999, Power et al., 2002). The social causation hypothesis holds that low socioeconomic status entails greater exposure to factors which negatively influence mental health. In short, SES affects health. According to the health selection hypothesis, mental health contributes to the sorting of individuals into SES positions, whether by influencing educational attainment or labour market outcomes. In this case, health affects SES. A recent formulation states that ‘[the health selection] hypothesis gives causal priority to the onset of mental illness as a factor preceding the disadvantaged placement of individuals into socioeconomic positions or social classes’ (Muntaner et al., 2013: 219). Another is that ‘men and women with pre-existing illness drift down the social scale’ (Power et al., 2002: 1989).

To what extent is health selection driven by prior social causation? This paper attempts to quantify the extent to which health selection – here, restricted or downward intergenerational social class mobility due to poor mental health or behavioural problems – can be attributed to the influence of modifiable (that is, at least partially amenable to social policy) childhood risk factors. These risk factors for poor mental health and restricted mobility – here, socioeconomic deprivation and adversities – may account for an apparent health selection effect in two ways (Fig. 1). The first is that adolescent mental health (AMH) may mediate the effect of childhood factors on SES in young adulthood. This reflects social causation occurring during childhood, and is therefore referred to as mediated social causation. The second is that childhood factors may confound the relationship between adolescent mental health and SES in young adulthood. Despite similarly attributing causal priority to childhood environmental factors in accounting for the social gradient in mental health, this is referred to in the literature as the ‘indirect health selection’ hypothesis (DiStefano and Morgan, 2014, Wilkinson, 1986). This paper evaluates the joint contribution of these processes, but does not disentangle the two.

Health selection may occur both between and within generations. This paper focuses on intergenerational health selection. That is, when poor mental health restricts the intergenerational social mobility of an individual: her mental health prevents her from achieving the social class position relative to her parents' which she otherwise would have attained. There is much reason to expect such selection: most mental disorders begin in adolescence and young adulthood, with half of all lifetime cases having onset by age 14 (de Girolamo et al., 2012, Kessler et al., 2005), and poor mental health in adolescence interferes with education in particular and the status attainment process in general (Miech et al., 1999, Power et al., 2002, Schoon et al., 2003).

In sum, this paper examines the following two questions. First, are health selection effects upon intergenerational social mobility evident? Second, if health selection effects are evident, to what extent is health selection operating in the sense that poor AMH is independent of any measured risk factors, and to what extent, alternatively, does it simply represent mediation of, or confounding by, these risk factors?

Section snippets

SES and health: the importance of the transition to adulthood

The reality of the health selection and social causation theories of health inequality is likely that they are best viewed as intertwined, with reciprocal causation occurring between SES and health over the life course (Kröger et al., 2015, Mulatu and Schooler, 2002). The observation that social causation may precede and underlie health selection may therefore appear trivial, since this social causation may itself be driven by earlier health selection. However, though this picture of reciprocal

Previous work

The childhood environment is predictive of mental health in childhood and adolescence, though associations are stronger and more consistent for externalising problems, such as conduct problems and hyperactivity (Amone-P'Olak et al., 2011, Bolger et al., 1995). This body of research includes the natural experiment upon which Costello et al. (2003) capitalise, finding that an exogenous increase in income was associated with a substantial reduction in conduct and oppositional defiant disorders,

Data

The National Child Development Study (NCDS) follows a British cohort made up of all births in England, Scotland and Wales in a single week in 1958, collecting data in a variety of domains. Data collected at birth and ages 7, 11, 16, 23 and 33 are used here. The NCDS includes a rich set of measures of the childhood environment, and measures of adolescent mental health from multiple informants. At age 33, 11,468 cohort members were interviewed (70.7% of the eligible sample). These participants

Analytical strategy

Logistic regressions are used to model the probability both of upward (versus stable or downward), and of downward (versus stable or upward) mobility. Cases in the highest and lowest origin class are excluded from the analysis of upward and downward mobility, respectively, since one cannot be upwardly mobile from the highest origin class and vice versa. In each case, I use the KHB method to compare a reduced model with a full model, whilst adjusting for the rescaling of the coefficients due to

Results

Descriptive statistics for the 3321 complete cases are shown in Table 1. Due to standardisation of the latent variables in the CFA, the AMH factor scores each approximate a mean-centred normal distribution. The count of childhood socioeconomic deprivation items ranges from 0 to 21. The values at the 25th, 50th and 75th percentiles are 1, 3, and 6 respectively. As expected, childhood adversities are comparatively rare.

Table 2 is a mobility table for the complete-case sample. There is substantial

Discussion

Growing up in deprivation and experiencing adversities such as bullying could partially account for the health selection effect of poor adolescent mental health and behavioural problems on socioeconomic attainment in adulthood. Previous research has raised this point in passing but not examined it in detail. This paper uses the NCDS to examine, first, whether selection effects are evident, and if so, second, to what extent they are accounted for by modifiable factors that place children at risk

References (44)

  • R.D. Conger et al.

    Economic stress, coercive family process, and developmental problems of adolescents

    Child. Dev.

    (1994)
  • E.J. Costello et al.

    Relationships between poverty and psychopathology: a natural experiment

    JAMA

    (2003)
  • G. de Girolamo et al.

    Age of onset of mental disorders and use of mental health services: needs, opportunities and obstacles

    Epidemiol. Psychiatr. Sci.

    (2012)
  • C. DiStefano et al.

    A comparison of diagonal weighted least squares robust estimation techniques for ordinal data

    Struct. Equ. Model. A Multidiscip. J.

    (2014)
  • B.P. Dohrenwend et al.

    Socioeconomic status and psychiatric disorders: the causation-selection issue

    Science

    (1992)
  • J. Elander et al.

    An update on the status of the rutter parents' and teachers' scales

    Child Psychol. Psychiatry Rev.

    (1996)
  • J. Elander et al.

    Use and development of the rutter parents' and teachers' scales

    Int. J. Methods Psychiatr. Res.

    (1996)
  • E. Ferri

    Life at 33: the Fifth Follow-up of the National Child Development Study

    (1993)
  • A. Goodman et al.

    The long shadow cast by childhood physical and mental problems on adult life

    Proc. Natl. Acad. Sci. U. S. A.

    (2011)
  • R. Goodman et al.

    Using the strengths and difficulties questionnaire (SDQ) to screen for child psychiatric disorders in a community sample

    Br. J. Psychiatr. J. Ment. Sci.

    (2000)
  • K.L. Gould et al.

    Gene-environment interactions in ADHD: the roles of SES and chaos

    J. Abnorm. Child Psychol.

    (2017)
  • J.J. Hudziak et al.

    The new genetics in child psychiatry

  • Cited by (25)

    • Does perceived social mobility affect health? Evidence from a fixed effects approach

      2022, Social Science and Medicine
      Citation Excerpt :

      Some recent studies started to investigate how changes in SSP across individuals' lives, i.e., subjective intra-generational mobility, are associated with health (Euteneuer et al., 2021; Mendoza et al., 2018; Yan et al., 2018). While changes in SSP over time are an important aspect of health inequalities, another fundamental dimension of equality of opportunity is the intergenerational transmission of (dis)advantages and the health consequences of this transmission process (Anderson, 2018; Boyle et al., 2009; Bulczak et al., 2021; Gugushvili and Kaiser, 2020; Präg et al., 2022; Präg and Richards, 2019; Venkataramani et al., 2020). Recently, scholars began to study the health and wellbeing consequences of subjective perceptions of social mobility (Berger and Engzell, 2020; Gugushvili, 2021a; Präg and Gugushvili, 2021; Vaquera and Aranda, 2017).

    • Cumulative social disadvantage and risk of attention deficit hyperactivity disorder: Results from a nationwide cohort study

      2020, SSM - Population Health
      Citation Excerpt :

      Other studies identify a similar inverse relationship between parental educational attainment and symptom scores of hyperactivity/inattention (Bøe et al., 2012; Davis et al., 2010) or risk of ADHD diagnosis in children (Miech et al., 1999; Rowland et al., 2018; Russell et al., 2014). A few studies identify significant associations between childhood ADHD and other SES factors such as parent unemployment (Davis et al., 2010), occupational status (Miech et al., 1999), or receipt of social assistance (Anderson, 2018; Hjern et al., 2010). Although previous research identify significant negative associations between measures of SES and childhood ADHD, several questions and issues remain to be addressed.

    • Social causation or social selection? The longitudinal interrelationship between poverty and depressive symptoms in China

      2020, Social Science and Medicine
      Citation Excerpt :

      There is convincing evidence that depression can globally be considered a consequence of poverty (Darin-Mattsson et al., 2018; Flisher et al., 2007; Lund et al., 2010a,b; Patel and Kleinman, 2003; Wickham et al., 2017), and often leads to income loss (Dohrenwend et al., 1992; Huddleston-Casas et al., 2009; Lund et al., 2013; Saraceno et al., 2005), implying that depression and poverty could interact in a vicious cycle (Haushofer and Fehr, 2014; Lund and Cois, 2018; Patel et al., 2010; Saraceno and Barbui, 1997). In spite of this information, little is still known about the interrelationship between poverty and depression (Anderson, 2018; Elwell-Sutton et al., 2019; Lund et al., 2018; Simmons et al., 2008), due to the fact that longitudinal studies remain sparse and causal mechanisms are difficult to identify (Cooper et al., 2012). Moreover, to our knowledge, no studies have investigated the complex pathways between poverty and depression to identify potential mechanisms and targets for interventions that address both issues.

    • ‘Falling from grace’ and ‘rising from rags’: Intergenerational educational mobility and depressive symptoms

      2019, Social Science and Medicine
      Citation Excerpt :

      This is chiefly because we use cross-sectional data that do not allow us to rule out the possibility of reverse causation as predicted by social selection theory. Although studies using high quality longitudinal data largely find support for social causation rather than social selection in the risk of developing depressive symptoms (Anderson, 2018; Hudson, 2005; Ritsher et al., 2001), it is possible that, for example, the downwardly mobile had more depressive symptoms than their immobile counterparts before they completed their educational careers, i.e., before we can establish whether or not they experienced any educational mobility. However, we believe that the first step in any social research is that of ‘establishing the phenomena’ through a sophisticated description.

    View all citing articles on Scopus
    View full text