An investigation of family SES-based inequalities in depressive symptoms from early adolescence to emerging adulthood

https://doi.org/10.1016/j.alcr.2010.04.001Get rights and content

Abstract

Using the life course cumulative advantage/disadvantage (CAD) perspective, this study examines the influence of early family SES on trajectories of depressive symptoms spanning from early adolescence to early adulthood, as well as variations in SES-based inequality in depressive symptoms trajectories over this period. This study looks at direct influences of family SES and SES-age interactions (exposure-dependent CAD mechanisms), as well as indirect influences through SES-linked youth experiences (path-dependent CAD mechanisms) to explain variations in SES-based inequality. Data was derived from the Add Health study- a national longitudinal survey of 14,000 adolescents. Results showed large and significant effects of early family SES and associated factors on depressive symptoms in early adolescence, but diminishing effects in middle and late adolescence, supporting the hypothesis of equalization in adolescent health across levels of SES. Disparities in depression reemerged as adolescents entered adulthood, supporting the view that SES-based health equalization is only a temporary process. These findings also strengthen the concept of life course CAD processes, stemming from family characteristics, coming into play later in life. Early family SES was directly and indirectly related to a set of transition-related risks and challenges during emerging adulthood, to which young adults from families of higher SES responded more effectively than those of lower SES. This paper discusses theoretical and methodological implications of the findings.

Introduction

Social and epidemiological research has documented a strong association between socioeconomic status (SES) and physical and mental health problems (Adler et al., 1994, Chen et al., 2002, Link and Phelan, 1995). Individuals of lower SES possess fewer resources and often experience more stressful life events than people of higher SES, contributing to health disparities (Pearlin, 1989, Turner and Noh, 1983, Turner et al., 1995). More importantly, a growing amount of life course and developmental research has confirmed that childhood family SES exerts lifelong influences, increasing health disparities in later stages of the life course (Robinson, 2001, Ross and Wu, 1996, Walesman et al., 2009, Wickrama et al., 2005a).

Life course research has increasingly focused on how disadvantages experienced during childhood may regulate health inequalities over the life course (Hatch, 2005, Hayward and Gorman, 2004, O’Rand, 1996, O’Rand and Hamil-Luker, 2005, Wilson et al., 2007). Early childhood socioeconomic conditions may initiate a lifelong cumulative advantage/disadvantage (CAD) process (Dannefer, 2003, Merton, 1988) with individuals of lower family SES experiencing more rapid decline or less rapid improvement in health over the life course compared to individuals of higher family SES, leading to widening inequalities and diverging trajectories with age (Hatch, 2005, House et al., 2005, Jackson, 2006, Lynch, 2003, O’Rand and Hamil-Luker, 2005, Prus, 2007, Ross and Wu, 1996, Wilson et al., 2007). Such SES-based health inequalities are evident in both physical and mental health domains (Turner et al., 1995, Wickrama et al., 2005b).

Most life-course research on mental health disparities has often focused on the SES-depression association in adulthood (Walesman et al., 2009, Wheaton and Clarke, 2003). Although a large volume of research has focused on adolescent mental health (Conger et al., 1994, Elder et al., 1996, Gore and Aseltine, 2003), only a few studies has specifically investigated the SES-depression association during the period of adolescence (e.g., Meich & Shanahan, 2000). Adolescence, a sensitive developmental period of rapid changes, adjustments, and life transitions, consists of developmental phases (early, middle, and late adolescence) that are distinguished by physiological markers and core socialization tasks. Developmental studies suggest that emotional trajectories, their association with family SES factors, and consequently, SES-based inequalities vary during different phases of adolescence, as well as during the transition to adulthood (Arnett, 2000, Ge et al., 1994, Kerckhoff, 2002, Larson et al., 2002). However, more research is needed to help gain insight about the variation in SES-based inequalities in depressive symptoms over the early life course.

The life course perspective suggests that SES-based health inequality increases with developmental stages through a cumulative advantage/disadvantage (CAD) process producing divergent health trajectories. Life course research has documented several mechanisms involved in CAD process (Wilson et al., 2007). Drawing from life course literature, we have identified several potential mechanisms applicable to the present study.

First, CAD processes may operate through various direct and indirect mechanisms (Wilson et al., 2007). Direct exposure to early family SES and SES-linked family resources (e.g., parenting) creates early SES health-gaps across SES groups (exposure-dependent mechanism). If the influences (the rate of return) of these SES factors remains unchanged over time, the early health-gaps may also maintain over the life course, thus producing parallel SES-group health trajectories (without a divergence or convergence). If the influences of early family SES factors on depressive symptoms increase with age, early health-gaps may widen, producing divergent SES-group health trajectories. This is the exposure by age mechanism of CAD process: higher SES groups gain increasingly larger health advantage or/and lower SES groups experience increasingly larger disadvantage. Conversely, if the direct influence of early family SES factors on depressive symptoms diminishes with age, it may produce convergent trajectories (a reversed CAD process: an equalization).

Second, CAD processes may operate through indirect influences of early family SES and associated resources whereby earlier advantages/disadvantages set in motion a series of cascading age-graded life experiences (path-dependent mechanism) (O’Rand & Hamil-Luker, 2005). That is, earlier developmental successes or failures are linked to the quality of life experiences in a successively contingent manner, creating a social chain of risks and resources from early adolescence to adulthood (Hayward and Gorman, 2004, O’Rand and Hamil-Luker, 2005). Previous research has well documented linkages between age-grade experiences and resources, and family SES characteristics (e.g., Conger et al., 1994, Whitbeck et al., 1997). As in the case of family SES, the influences (rate of return) of SES-linked age-grade life experiences (e.g., school attachment) and SES-linked core life resources (e.g., self-esteem) may change. For example, if the influence of early school attachment on depressive symptoms increases with age, early SES gaps may widen, thus producing divergent depressive symptom trajectories. This is the path by age mechanism of CAD process. Conversely, if the influence of school attachment on depressive symptoms decreases with age (a reversed CAD process), it may produce convergent trajectories (an equalization). We expect deviations from typical CAD processes when family SES factors influence depressive symptom trajectories during adolescence.

Using longitudinal data, this study examines depressive symptom trajectories to understand the ways in which family SES contributes to health inequalities over a period spanning from early adolescence to emerging adulthood. Our study has two specific objectives.

First, we will investigate changing influences of family SES on depressive symptoms over adolescence and emerging adulthood. In order to investigate the overall influence of family SES, we will examine the influences family SES factors corresponding to both ‘exposure-dependent’ and ‘path-dependent’ mechanisms (Wilson et al., 2007) including family SES characteristics, SES-linked family resources (e.g., effective parenting), SES-linked core life resources (e.g., self-esteem and mastery), and SES-linked age-graded experiences (e.g., school attachment and disruptive events). This objective requires an analyses of individual depressive symptoms trajectories across three developmental segments: early/middle adolescence (12/13–15/16 years), middle/late adolescence (15/16–19/20 years), and emerging adulthood (19/20–25 years).

Second, we will examine the varying magnitude of inequalities across SES-group depressive symptom trajectories over adolescence and emerging adulthood, and the age-related pattern of depressive symptom trajectories (inequalities across SES-group trajectories may vary, while preserving their average curvilinear pattern). To accomplish this objective, we will examine mean trajectories of SES-based groups. In order to assess the magnitude of SES-based inequality, we will also estimate between-SES-group inequality in depressive symptom levels at several age-points. We will discuss analytical strategies used to fulfill objectives in the methods section.

Consistent with the exposure-dependent mechanism (Wilson et al., 2007), early family SES characteristics such as parental education, poverty, and race/ethnicity and SES-linked resources largely determine depressive symptom levels in childhood and early adolescence, thus, creating initial SES-depression inequalities (Brody et al., 2006, Conger et al., 1994, Ge et al., 1994; Grunden & Ernst, 2007; Matjasko et al., 2007, Miech et al., 2005, Whitbeck et al., 1997). Particularly, parents with higher education have greater access to social-psychological resources, parenting skills, and information, all of which protect children from the adverse influence of stressful family circumstances and mental health consequences (Whitbeck et al., 1997, Wickrama et al., 2003, Wickrama and Noh, 2009).

Consistent with the family investment model (Meich & Shanahan, 2000), family socioeconomic disadvantage (lower SES) limits resources and opportunities for youth, including reduced access to quality housing, proper sanitation, vehicles, household equipment, food, and health insurance, resulting in poor physical and mental health. In addition, children in families facing economic hardship often live in disadvantaged communities with poor health, educational, recreational, and social services (Wheaton and Clarke, 2003, Wickrama and Bryant, 2003, Williams, 2005).

Consistent with the family stress process model (Conger et al., 1994), stressful daily life experiences associated family socioeconomic disadvantage (lower SES) have psychological consequences for parents (Conger et al., 1994, Whitbeck et al., 1997, Wickrama and Noh, 2009); distressed parents are more likely to be irritable, authoritarian, and rejecting, often resulting in poor parenting practices (Conger et al., 1994). Negative parental affect (e.g., hostility and rejection) and poor child management (e.g., lack of involvement or supervision and ineffective discipline) are two aspects of poor parenting that often influence adolescent psychopathology (Sroufe, Duggal, Weinfield, & Carlson, 2000).

Previous research has also documented that family SES characteristics and resources including family economic problems, parental education, and parental practices indirectly influence the level of depressive symptoms of adolescents through children/early adolescents’ age-graded experiences such as school attachment and GPA and core life resources, such as early psychological competence (e.g., self-esteem, mastery) (Conger et al., 1994, Conger et al., 2009, Koivusilta et al., 2006, Whitbeck et al., 1991). We hypothesize that family SES factors, including family characteristics, family resources, adolescents’ early age-graded experiences, and core life resources, will influence the initial level of depressive symptom trajectories during early adolescence (Hypothesis 1; H1). We expect these influences to produce initial SES inequalities in the levels of depressive symptoms of adolescents.

There are reasons to expect that the salience of family SES, SES-linked early experiences, and core life resources for adolescent depressive symptoms begin to diminish with age during adolescence resulting in a decrease in SES-based health inequalities (SES equalization – a reversed CAD process). As several studies have suggested, several countervailing factors including individual autonomy and peer affiliation during adolescence may decrease the salience of family SES factors (Blane et al., 1994, Clausen, 1993, Shanahan, 2000). Particularly, adolescent experiences in school often enable social mixing across socioeconomic lines (Koivusilta et al., 2006). Thus, schools offer adolescents the opportunity to gain life experiences and associated benefits regardless of family SES, which can retard CAD process (West, 1997). In addition, youth experience physical changes during puberty and stress-inducing transformations, diminishing the salience of family SES factors that may contribute to a decrease in SES-based health inequalities (Brooks-Gunn et al., 1994, Ge et al., 1994). Therefore, during this period it is plausible that adolescents from families of higher SES may not have the expected relative health advantages over adolescents from families of lower SES.

However, other studies suggest that the influence of extra-familial experiences may be limited, especially during early and middle adolescence because high levels of social and racial segregation and homogeneous peer groupings can constrain opportunities for social mixing in school and community settings (Hallinan & Teixeira, 1987). A lack of extracurricular activities in schools may also reinforce racial and social segregation among students (Moody, 2001). Moreover, adolescents’ level of autonomy, peer affiliation, and social mixing may also be constrained by structural socioeconomic characteristics of the family. While such constraints are possible, some degree of social mixing across SES classes is expected to occur during early and middle adolescence.

We hypothesize that the influences of family SES factors on depressive symptoms will decrease during early to middle adolescence (Hypothesis 2A; H2A). As a result, we expect a decrease in SES inequality (an equalization) in depressive symptoms during this period.

The developmental and life course studies also indicate that the effects of early family disadvantage can continue throughout middle and late adolescents by limiting opportunities for educational and occupational success, as well as promoting psychosocial failures such as premature termination of education (Bernhardt et al., 2005, Sampson and Laub, 1993, Wickrama et al., 2003).

Youth who are unsuccessful in school and have poor motivation and social skills are more likely to experience disruptive life events, such as dropping out of school, unplanned pregnancy, and early cohabitation or marriage. Previous research has shown that adverse family circumstances may push adolescents prematurely into stressful adult roles (Sampson and Laub, 1993, Wickrama and Noh, 2009). These disruptive events and circumstances may be associated with adolescents’ depressive symptoms, and hence, partly replace the direct impact of family SES on depressive symptoms during middle and late adolescence (Scaramella et al., 1998, Wickrama et al., 2005a, Wickrama et al., 2005b).

However, during the late adolescence, their own level of psychosocial maturity as well as school and other institutional support may operate as countervailing factors and may decrease the causal influences of SES-linked age-graded experiences such as disruptive events on depressive symptom levels. As adolescents become older and more independent, some youth may avoid adverse consequences of disruptive life events by making effective choices (Clausen, 1993, Elder et al., 1996, Shanahan, 2000). This may contribute to a reversed CAD process during middle-late adolescence. Thus, we hypothesize that the influences of family SES factors on depressive symptoms decreases with age during middle to late adolescence (Hypothesis 2B; H2B). As a result, we expect a decrease in SES inequality (an equalization) form depressive symptoms during this period.

Meich and Shanahan (2000) found a strong association between family SES characteristics and depressive symptoms in early adulthood, reflecting increased SES-based health inequalities. Beyond adolescence, the CAD process regains momentum. Consistent with the path-dependent mechanism, experiences of successes and failures persist into adulthood and precipitate as key transition failures such as failures to pursue higher education or to gain a full-time employment (Elder, 1998, Hatch, 2005, O’Rand and Hamil-Luker, 2005). Although some youth escape the consequences of early family adversities and failures by making well-planned and effective choices (Clausen, 1993, Elder et al., 1996, Shanahan, 2000), most youth are unable to escape those consequences and likely to experience key transition failures during the transition to adulthood.

After graduating from, or dropping out of high school, young adults usually either pursue higher education or attempt to gain full-time employment (Gore & Aseltine, 2003). Individuals without higher education often face low-paying and unstable work (Kerckhoff, 2002). Key transitional experiences, such as higher education and full-time employment, may exert strong influences on young adults’ depressed mood. More importantly, the detrimental influence of failures in these key life transitional experiences may increase over time contributing to a differentiation trend or reemergence of SES health inequality.

In addition, educational and occupational attainments are directly influenced by parental SES and SES-related social capital during transition to adulthood independent of life experiences during adolescence (Gore & Aseltine, 2003). Continued familial assistance during early adulthood has increased substantially during the last decade due to longer school attendance and delayed first marriage (Sandefur et al., 2005, Schoeni and Ross, 2005). Family wealth also enhances access to the information and technology required to develop competitive technological skills that can increase employment prospects (Mortimer & Larson, 2002). Conversely, lower aspirations and limited encouragement from lower SES parents can interfere with both educational and occupational aspirations and attainments (Steinberg, Lamborn, Darling, Mounts, & Dornbusch, 1994). These family influences during the transition to adulthood may contribute to a reemergence of SES-based health inequality in depressive symptoms. As Osgood, Ruth, Eccles, Jacobs, and Barber (2005, p. 345) stated, “The path one takes typically reflects the social-class values and resources of one's natal family.”

Although we expect to find an SES equalization during adolescence, SES inequalities may reemerge in transition to adulthood. Moreover, consistent with path by age mechanisms of the CAD process, SES inequalities may increase with age. That is, we contend that equalization during adolescence is temporary and is attributable to the decreased impact of family SES factors during adolescence (Larson et al., 2002). We hypothesize that the influences of family SES characteristics and SES-linked key transitions on depressive symptoms will increase during the transition to adulthood (Hypothesis 3; H3). As a result, we expect a reemergence of SES inequality during this period.

Inequalities across youth depressive symptoms trajectories may vary, while preserving the trajectory pattern (shape) over adolescence and emerging adulthood. Previous research has shown that a general curvilinear (convex) pattern in depressive symptoms is apparent in individual, group average, and overall average trajectories, and is largely related to changes in stressful circumstances associated with physical changes and puberty during adolescence (Brooks-Gunn et al., 1994, Ge et al., 1994). We expect to replicate a typical curvilinear trajectory in depressive symptoms suggested in previous developmental research (Ge et al., 1994, Larson et al., 2002, Moneta et al., 2001).

It is important to ensure, empirically, that the effects of family SES and associated resources on youth depressive symptoms are estimated net of individual racial/ethnic minority status. Previous research shows that racial/ethnic minority status exhibits influences on health, beginning in childhood and continuing across the life course, regardless of income, education, and family socioeconomic characteristics (Spencer, 2001, Walesman et al., 2009). We anticipate direct, main effects of race/ethnic minority status on youth depressive symptoms. Hence, the regression coefficients of family SES factors and associated resources and risks on depressive symptom trajectories will be assessed while controlling for the influences of race/ethnic minority status.

Section snippets

Sample

Data for this study came from a nationally representative sample of adolescents participating in the National Longitudinal Study of Adolescent Health (Add Health: see title page for a listing of participating organizations). The primary purpose of this multi-agency funded panel study was to investigate adolescent health and health behavior outcomes. Adolescents and parents participating in the Add Health study provided family socioeconomic information during in-home interviews. The data were

Investigating objective 1: changing influences of family SES factors on depressive symptoms (H1, H2A, H2B, H3)

Our first objective was to examine changing influences of family SES factors with age over adolescence and emerging adulthood. As described in the analysis section, our preliminary analyses focused on identifying distinct potential slope segments exist in the data.

Preliminary analyses: exploring slope segments in depressive symptoms trajectories

We estimated a continuous growth curve of depressive symptoms for 13–23 year age-period with three distinct slope segments corresponding to early/middle adolescence (ages 13–16), middle/late adolescence (ages 16–20), and emerging

Discussion

Life course theorists contend that childhood advantages related to the family of origin initiate a cumulative advantage/disadvantage (CAD) process that produces increasing SES-based health inequality over the life course. However, there has been a noticeable gap in the existing empirical literature regarding variation in SES-based mental health inequality, particularly across the early life course. Thus, this study focused on SES-based inequality in depressive symptom trajectories during the

Acknowledgments

This research is based on data from the Add Health project, a program project designed by J. Richard Udry (PI) and Peter Bearman, and funded by grant P01-HD31921 from the National Institute of Child Health and Human Development to the Carolina Population Center, University of North Carolina at Chapel Hill, with cooperative funding participation by the National Cancer Institute; the National Institute of Alcohol Abuse and Alcoholism; the National Institute on Deafness and Other Communication

References (81)

  • D. Blane et al.

    Social patterning of medical mortality in youth and early adulthood

    Social Science and Medicine

    (1994)
  • P. West

    Health inequalities in the early years: Is there equalization in youth?

    Social Science and Medicine

    (1997)
  • N.E. Adler et al.

    Socioeconomic status and health: The challenge of the gradient

    American Psychologist

    (1994)
  • P. Allison

    Missing data

    (2001)
  • J.J. Arnett

    Emerging adulthood: A theory of development from the late teens through the twenties

    American Psychologist

    (2000)
  • E. Bernhardt et al.

    The impact of childhood family structure and conflict on routes out of the parental home in Sweden

    Acta Sociologica

    (2005)
  • P. Bourdieu

    The logic of practice

    (1990)
  • G.H. Brody et al.

    Perceived discrimination and the adjustment of African American youths: A five-year longitudinal analysis with contextual moderation effects

    Child Development

    (2006)
  • J. Brooks-Gunn et al.

    Studying links between hormones and negative affect: Models and measures

    Journal of Research on Adolescence

    (1994)
  • Centers for Disease Control and Prevention
    (1996)
  • K. Chantala et al.

    Non-response in wave III of the Add Health Study

    (2004)
  • E. Chen et al.

    Socioeconomic differences in children's health: How and why do these relationships change with age?

    Psychological Bulletin

    (2002)
  • K.J. Conger et al.

    Development of mastery during adolescence: The role of family problem solving

    Journal of Health and Social Behavior

    (2009)
  • R.D. Conger et al.

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

    Child Development

    (1994)
  • J. Clausen

    American lives: Looking back at the children of the great depression

    (1993)
  • D. Dannefer

    Cumulative advantage/disadvantage and the life course: Cross-fertilizing age and social science theory

    Journals of Gerontology

    (2003)
  • L.R. Derogatis

    SCL-90-R: Administration, scoring and procedures manual II

    (1983)
  • A. Donner

    A review of inference procedures for the intraclass correlation coefficient in the one-way random effects model

    International Statistical Review

    (1986)
  • T.E. Duncan et al.

    An introduction to latent variable growth curve modeling: Concepts, issues, and applications

    (1999)
  • G.H. Elder

    The life course as developmental theory

    Child Development

    (1998)
  • G.H. Elder et al.

    Psychological stress over the life course

  • X. Ge et al.

    Trajectories of stressful life events and depressive symptoms during adolescence

    Developmental Psychology

    (1994)
  • S. Gore et al.

    Race and ethnic differences in depressed mood following the transition from high school

    Journal of Health and Social Behavior

    (2003)
  • M.T. Hallinan et al.

    Students’ interracial friendships: Individual characteristics, structural effects, and racial differences

    American Journal of Education

    (1987)
  • S. Harper et al.

    Methods for measuring cancer disparities: A review using data relevant to Healthy People 2010 cancer-related objectives

    National Cancer Institute

    (2006)
  • S.L. Hatch

    Conceptualizing and identifying cumulative adversity and protective resources: Implications for understanding health inequalities

    Journals of Gerontology

    (2005)
  • M.D. Hayward et al.

    The long-arm of childhood: The influence of early-life social conditions on men's mortality

    Demography

    (2004)
  • J.S. House et al.

    Continuity and change in the social stratification of aging and health over the life course: Evidence from a nationally representative longitudinal study from 1986 to 2001/2002 (Americans Changing Lives Study)

    Journals of Gerontology

    (2005)
  • P.B. Jackson

    Health inequalities among minority populations

    Journals of Gerontology

    (2006)
  • A.C. Kerckhoff

    The transition from school to work

  • M.S. Kreiter

    Grown up at 21? No way

    (2003)
  • L.K. Koivusilta et al.

    Health inequality in adolescence. Does stratification occur by familial social background, family affluence, or personal social position?

    BMC Public Health

    (2006)
  • R.W. Larson et al.

    Continuity, stability, and change in daily emotional experience across adolescence

    Child Development

    (2002)
  • A.E. Lewis

    Race in the schoolyard: Negotiating the color line in classrooms and communities

    (2003)
  • B.G. Link et al.

    Social conditions as fundamental causes of disease

    Journal of Health and Social Behavior

    (1995)
  • S.M. Lynch

    Cohort and life-course patterns in the relationship between education and health: A hierarchical approach

    Demography

    (2003)
  • M.G. Marmot

    General approaches to migrant studies: The relation between disease, social class and ethnic origin

  • J.L. Matjasko et al.

    Structural and dynamic process family risk factors: Consequences for holistic adolescent functioning

    Journal of Marriage and Family

    (2007)
  • T.J. Mathews et al.

    Mean age of mother 1970–2000

    Centers for Disease Control, National Vital Statistics Reports

    (2002)
  • R.K. Merton

    The Matthew effect in science. II: Cumulative advantage and the symbolism of intellectual property

    Isis

    (1988)
  • Cited by (0)

    View full text