What are the Risk Factors for Childhood and Adolescent Depression?

AS and A Level Psychology

Jackie Tilston – Extended Essay – 2001

What are the Risk Factors for

Childhood and Adolescent Depression?

PS51010a

Extended Essay

Name: Jackie Tilston

Date: 30 March 2001

Popular psychology tends to blame childhood and adolescent depression on factors such as experience of divorce. In reality however the risk factors for depression across these ages are much more complicated. There are many perspectives that have been used to address childhood and adolescent depression, including developmental, psychobiological, and psychodynamic approaches. There are therefore a number of different levels of explanation and many specific explanations within each level. Although not all explanations can be discussed here, it is the aim of this essay to extract as many of the risk factors as possible that overlap at least one of these levels. To arrive at as full a description as possible it is necessary to ask why it is useful to analyse what causes depression in children and adolescents, both theoretically and practically. It is then necessary to, firstly, examine what is meant by depression generally and, secondly, to briefly analyse what the differences are between adult depression and depression in young people, before providing a short description of the parameters of childhood and adolescence. The risk factors or causes of depression within these parameters will then be described, including gender, hormones, genetic contributions, stress, attachment issues and maternal depression. It should be noted that space does not permit an exhaustive analysis of all risk factors, rather the most salient causes have been extracted. Finally, an attempt will be made to answer the question of whether there are any implications of risk factors for treatment.

One theoretical reason why it is useful to analyse what causes depression is because up until recently it was not uncommon for papers on childhood and adolescent depression to come from only one perspective of psychology and for this to have detrimental implications. For example, the orthodox psychoanalytic perspective was taken and then that theory applied to the likelihood and characteristics of depression in children of different ages (Cicchetti & Schneider-Rosen, 1986). This particular perspective has led to such statements as made by Rie (1966; in Cicchetti & Schneider-Rosen, 1986): prepubertal children are unable to be depressed because they do not have a fully developed ego. This view has been expressed in other ways too. Malmquist’s claims (1971, 1975; in Cicchetti & Schneider-Rosen, 1986, p. 74) have been described thus: “… children’s lack of ego differentiation limits their range of potential affective expressions”. Anthony (1975; in Cicchetti & Schneider-Rosen, 1986) supported this view by proposing that as children were more easily influenced by their environment their depressive reactions were transient and brief. In summarising and further supporting these claims to a certain extent, Cicchetti & Schneider-Rosen (1986) propose that any model of childhood depression should take account cognitive development, thus implying that depression cannot occur in children whose cognitions are not sufficiently complex.

It should be noted that clearly these points of view have important practical implications for treatment – would a young child fitting a depressive diagnosis not be treated appropriately if they were considered to be too young to actually be depressed? This view, though dominant for some time, has now fallen out of favour (Greenberg et al., 1990 in Graham & Easterbrooks, 2000). Furthermore, studies have now found that children can indeed suffer depression, for example the so-called Isle of Wight Studies (Rutter, 1979; Rutter et al. 1970; in Rutter, 1986). This question is just one justification for meticulous study of the etiology of depression in children and adolescence. More recently it has been found that childhood onset depression has a 60 to 70 % risk of continuing into adulthood and 20 to 40 % of developing into bipolar disorder within five years (Weller & Weller, 2000). Both these reasons have implications for treatment and as etiology may affect accurate diagnosis and treatment efficacy these reasons provide further support for an integrative study of the risk factors of depression in children and adolescents.

Before looking at what might cause depression in young people it is worthwhile to briefly look at what depression is – its symptoms and special points relating to children and adolescents. In a general discussion of depression Hammen (1997) describes depression under four categories: affective, cognitive, behavioural and physical symptoms. These symptoms might include listlessness and apathy; irritability (especially in children); feelings of guilt, thoughts of a bleak future, difficulty in concentrating; behavioural symptoms include an inability to get out of bed, psychomotor changes; and physical symptoms could include changes in appetite, sleep and energy. Weinberg et al. (1978; in Poznanski, 1979) set the following as criteria for the diagnosis of childhood depression. He required: 1) dysphoric mood, and 2) self-deprecatory ideation. Two or more of the following were also required: 1) aggressive behaviour; 2) sleep disturbance; 3) change in school performance; 4) diminished socialisation; 5) change in attitude toward school; 6) somatic complaints; 7) loss of usual energy, and 8) unusual change in appetite or weight. Additionally, these symptoms needed to remain for at least a month. In their review of patterns of emotion in depression, Izard & Schwartz (1986) describe studies that have analysed depressive emotions such as anger, sadness and detachment in infants and young children. Interestingly, emotions of fear, shame and guilt – common in adolescent and adult depression – have not been detected in children (Scarr & Salapatek, 1970; in Izard & Schwartz, 1986) and it has been suggested that this is because children’s cognitive development is not high enough to have developed a stable enough self-image, on which these emotions might be based (Groh, 1980; in Izard & Schwartz, 1986). It has only been relatively recently (since the 1970s) that it has been widely accepted that depression exists more than rarely in children (Rutter, 1986). While some researchers have maintained that little difference exists between adult and youth depression, it has been asserted that there are in fact differences in the form and frequency of depression between these two groups (Rutter, 1986).

This is a preview of the whole essay

Emde et al. (1986) chose to define childhood as encompassing four periods: infancy, preschool, school age and adolescence. These periods coincide with (Emde et al., 1986, p. 141): “(1) traditional divisions with respect to society’s role expectations, (2) theoretical divisions based on major transformation in cognitive development as demarcated by Piaget (Piaget & Inhelder, 1969), and (3) theoretical divisions of psychosexual development as demarcated by Freud (1905/1953) and Erikson (1950).” Within each of these groups, prevalence has been found to be as follows. In children younger than 12, the incidence of depression is said to be about 2 to 3 % of the population (Angold & Costello, 1993 in Hammen, 1997), although some have put the figure as high as 10 to 12 % (Nolen-Hoeksema et al., 1992 in Graham & Easterbrooks, 2000); in adolescents (adolescence has been said to exist from 12 to 20 [Allchin, 1975]) the figure rises and in preschool-aged children the figure lowers to less than 1% (Kashani & Carlson, 1987 in Hammen, 1997). Clearly depression during formative years has the potential to interfere with development. The effects on development of depression in children and adolescents (Coie et al., 1992 in Graham & Easterbrooks, 2000; Rutter, 2000) provide another reason for continued research into its etiology.

As there are differences between adult and youth depression, and childhood and adolescent depression, there are also gender differences in depression, with much empirical support. Before the onset of adolescence boys are more likely to be depressed than girls. However, from adolescence and throughout their childbearing years females are more likely to suffer from depression (e.g. Angold et al., 1998; in Angold et al., 1999). For example, a study (Li et al., 2001) examining the risk factors for depression in children in group homes in South Carolina found that out of a number of risk factors the only statistically significant risk factor was gender, with female adolescents more likely to be depressed and consequently more likely to attempt suicide. One study (Joyner & Udry, 2000, p. 369) has also found a connection between adolescent romance and depression: their results suggest “females’ greater vulnerability to romantic involvement explains a large part of the emerging sex difference in depression during adolescence”.

Why should girls be more likely to suffer from depression in adolescence? Angold et al. (1999) proposed that morphological changes associated with puberty and hormonal changes might be strongly associated with depression in adolescence and set out to test which factor may be more greatly associated with depression in teenage girls. Morphological changes were suspected because an analysis of onset of unipolar depression increased significantly in girls reaching Tanner stage III of puberty – indicating that body changes might be to blame for depression rather than simply age. Interestingly, it was found that body morphology and the resultant psychosocial effects on social interactions were eliminated by changes in androgen and oestrogen levels added to the model. This does not mean that rises in these hormones lead to depression, rather that girls whose levels rise quickly will be more susceptible to other risk factors such as life stressors or cognitive style. Cyranowsi et al. (2000) have proposed extending the risk factor of changes in hormone levels to include the following as risk factors in adolescent girls. Cyranowski et al. cite gender socialisation and intensification (which can be supported by neurobiological evidence) as a potential cause for mood change in adolescent females if this intensification comes at the wrong time. Gender intensification refers to the convergence of males and females to their respective stereotypical behaviours, for example, females tend to spend even more time together whereas males will become more solitary during adolescence. This is an example of a gender difference of affiliative behaviour – others are insecure parental attachment, anxious or inhibited temperament, and low instrumental coping skills. Any of these problems, possibly coupled with a negative life event, can create an at-risk adolescent female, and thus this model might help to explain elevated levels of depression in this demographic.

Although hormones play a role in the gender difference in depression, they have also been analysed more generally, across the sexes. A rise in dihydroepiandrosterone (DHEA) and cortisol secretion (hormones) has been found to predict onset of first-episode major depression in adolescents as studied by Goodyer et al. (2000). This study tested this level of hormone in adolescents from two groups – at risk and not at risk. Although neither hormone was associated initially with cognitive dysfunction, mood or risk status on initial assessment, when participants were tested a year later those who had subsequently experienced an onset of major depression had in fact had higher levels of DHEA during the initial assessment. Although there has been much work on adult cortisol secretion a consensus is still to be arrived at for studies on children and adolescents (Puig-Antich, 1986). The general pattern of adult hormone hypersecretion does however appear to be generally to mirrored in adolescents (e.g. Robins et al., 1982; in Puig-Antich, 1986) as the Goodyer et al. (2000) study replicates. Prepubertal children on the other hand show very low levels of cortisol hypersecretion related to depression (Puig-Antich et al., unpublished data, 1979; in Puig-Antich, 1986). Although there is not enough evidence to suggest that hypersecretion of hormones actually causes depression, the evidence does suggest that the cortisol secretion may predict onset of major depression in adolescents.

Neurobiology is another perspective from which studies looking at childhood depression are beginning to emanate from. It has been suggested that stress in early childhood might produce changes in the brain (Kaufman et al., 2000). Kaufman et al. (2000) have reviewed the connections between stress and changes in neurobiology, finding that preclinical studies have shown that early stress can alter the development of the hypothalamic-pituitary-adrenal axis, hypothalamic and extrahypothalamic corticotrophin releasing hormone, monoaminergic, and gamma-aminobutyric acid / benzodiazepine systems (these systems relate to the previous point made about hormone hypersecretion in relation to childhood depression). Forms of stress will be discussed below, however one source of stress, child abuse, is associated with higher rates of depression – this association seems like soon having a concrete biological basis. As Kaufman et al. (2000; p. 778) suggests: “More research is needed to understand how inherent factors interact with experiences of abuse and other psychosocial factors to confer vulnerability to develop depression”.

Methods of genetic research into the etiology of depression include family studies, twin studies, adoption studies and more recently molecular genetics. Family studies involve studying a proband and his or her family to identify heritability of a disorder. One such study found depression in first-degree relatives of between 7 and 30% - much higher than in the general population (Gherson, 1990; in Hammen, 1997). Another study, consisting of 900 patients and controls, found major depression in 10.4% of first-degree relatives of depressed patients, compared with 4.9% for controls (Winokur et al., 1995; in Hammen, 1997). Interestingly, when all forms of depression were counted, that is, both major depression and mild, no significant difference was found between patients and controls’ family rates. A finding, which is less clear as to its pure genetic element, is that 50 to 80% of children of depressed parents show some sort of diagnosable disorder, with about half of those diagnosable as depressed (e.g. Gotlib & Lee, 1990; Hammen, 1991a; Klein, et al., 1988; in Hammen, 1997). This statistic is of course clouded by insufficient parenting amongst other social factors (Hammen, 1997).

Twin studies have provided most of the accumulated knowledge about the heritibility of depression (Reiss & Neiderhiser, 2000). Theoretically, the study of monozygotic twins (sharing 100% of their genes) should provide a far more accurate picture of the heritability of depression than both family studies and also dizygotic twins (sharing an average of 50% of their genes). McGuffin et al. (1996; in Hammen, 1996) studied 214 twin pairs through probands who had been inpatients being treated for depression. They found that concordance for lifetime depression for monozygotic twins was 46%; 20% for dizygotic twins – both figures being much higher than in the general population. Kendler et al. (1992; in Hammen, 1996) studied 1033 female twin pairs and found concordance between monozygotic twins to be 48% compared to a 42% concordance rate in dizygotic twins – both groups were rated as having depression based on DSM-III-R criteria. While the McGuffin et al. (1996; in Hammen, 1996, p. 61) study concluded that “genetic factors play a moderate role in family patterns of depression”, the Kendler et al. study (1992; in Hammen, 1996, p. 62) concluded (via a complex statistical model) that genetic factors play “a substantial, but not overwhelming, role in the cause of depression”.

It should be noted, however, that there have been criticisms of the underlying assumptions of twin studies in genetics research. One of the most common criticisms is the equal environments assumption – the basic tenet on which twin studies are based (Reiss & Neiderhiser, 2000). The assumption, first stated by Galton, is that the environmental influences on any trait under study are equally correlated in monozygotic twins and dizygotic twins. While Kendler (e.g. 1993; in Hammen, 1996) has asserted that this assumption is indeed firm, Reiss & Neiderhiser (2000, p. 360) suggest that “it needs to be subjected to continuing criticism as we learn more about the nuances of sibling relationship and how they evolve over time. It may well be that subtle differences in the social worlds of identical twins in comparison to fraternal twins are still masquerading as genetic influence.” Reiss & Neiderhiser do not suggest ignoring the findings of twin studies. Instead they suggest integration with observations about the social environment in developmental theories. It has also been suggested that genetics research be combined with neurodevelopmental theories. With the recent advances in genetics it perhaps should seem possible now to map the genetic path of individual brain differences (Skuse, 2000). However, especially in relation to child psychopathology, studies linking genetics and neurodevelopmental problems have been hard to replicate (Rutter et al., 1999b; in Skuse, 2000).

In his review of the attempt at bringing together genetics and neuroscience, Skuse (2000) notes that while general medicine become more and more scientifically based in the last half of the twentieth century, psychiatry and child psychiatry more specifically remained rooted in psychoanalysis. Although it would appear that this has now changed to a large extent, theories which have stemmed from the psychoanalytic approach still feature in etiological theories of depression (e.g. Poznanski, 1979; Rutter & Sroufe, 2000) in children and adolescents. An example of this continuing paradigm is the following quote which in fact is very similar to a statement first made by Freud in 1917 (in Lewis & Lewis, 1979): “internalisation of maladaptive beliefs about the self and relationships, such as a diminished sense of self-worth” (Rudolph et al., 2000, p. 217). Modern psychological theories play a very important role in models of depression in children and some of their components will be outlined below.

Stress and coping models of developmental psychopathology have begun to assume a central position in the search for risk factors for depression in children (Rudolph et al., 2000). One study (Rudolph et al., 2000) in particular has proposed a model of depression which is specifically based on children’s interactions in the context of stress. The study made two distinctions: between interpersonal stress and non-interpersonal stress, and between episodic and chronic stress. Interpersonal stress may lead to internalisation of maladaptive beliefs about the self and relationships (as mentioned above); decreased perceptions of competence and control; negative attributions; and a tendency to focus on negative aspects of interpersonal situations. In turn this stress may also lead to children’s resources being overwhelmed and thus a sense of helplessness and problematic coping styles may set in. This chain of events may put a child at risk for depression. Non-interpersonal stressors, such as poor academic performance have not been found to correlate with depression as highly as interpersonal stressors. Another point made by the study is that depressive symptoms and associated impairment may in fact precipitate stress and this in turn may perpetuate or trigger future depression.

While the study just described found a strong correlation between interpersonal stress and depression in children, the paper did not delve very deeply into the different types of interpersonal stress that cause stress and in turn depression. To elaborate on stress as a risk factor for depression some issues will be discussed in further detail as they are addressed in many studies and can be seen to cross domains. More specifically, attachment problems will be addressed and maternal depression and socio-economic status will also be briefly described as risk factors for depression in children and adolescents.

The first interpersonal stressor to be described is also perhaps the most universally studied and described – a child’s relationship with his or her parents. It has been proposed that it is through a child’s relationship with his or her parents that he or she forms initial working models of the self in relation to the world (Graham & Easterbrooks, 2000). Beck (1976; in Graham & Easterbrooks, 2000) has suggested that negative parent-child attachment might lead to depression. Attachment theory has tended to concentrate on the dyadic relationship between a child and its mother. Although it is now known that the developmental process, especially in relation to psychopathology, is much more complicated, this relationship is still seen to be very important in predicting mental health (Rutter & Sroufe, 2000). Avoidant attachment in infancy is associated with later conduct problems, it is also associated with low self-esteem, negative attributional biases and rejection by peers and teachers (Rutter & Sroufe, 2000).

Related to attachment patterns as a risk factor for depression is the fact that maternal depression can contribute if not cause depression in children (Graham & Easterbrooks, 2000) (of course it is necessary to be aware of the interaction between genetics and psychology here). In a study of inner city mental health needs in London Davis et al. (2000) found that the most common risk factor for depression in children and adolescents was maternal mental health problems. This is because children may be deprived of an empathetic relationship and mothers may be extremely sad, irritable, and even hostile (Graham & Easterbrooks, 2000). As the mother is unlikely to be particularly responsive children may be unable to form secure attachments, hence the connection with attachment as a factor of depression (e.g. Egeland & Sroufe, 1981; in Graham & Easterbrooks, 2000).

Why should socio-economic status contribute to depression in children? The inability to meet basic needs including housing, clothing and food, exposure to situations such as unemployment and violence, and stress within the family may all lead to inadequate parenting involving inadequate supervision and discipline, a lack of support for early development and lower levels of affection (e.g. Hashima & Amoto, 1994; in Graham & Easterbrooks, 2000). Graham & Easterbrooks (2000) have found a correlation (-0.45) between depressive symptomatology in school-aged children and economic risk. In another recent study Davis et al. (2000) also found that an increasing number of risk factors were present with decreasing income. Although it might be thought that economic risk would be correlated with maternal depression, these variables have not been found to be correlated (Graham & Easterbrooks, 2000). Explanations for this finding include the fact that parents may have more people to rely on for emotional support and also that school-aged children compare themselves with each other thus exacerbating emotional problems arising from these comparisons (Graham & Easterbrooks, 2000). Although low socio-economic status cannot be said to cause depression it could serve as a sort of ‘umbrella’ risk factor because it covers a number of individual risk factors.

Wallace et al. (in Davis et al., 2000, p. 171) stated that in practice, many services contribute to child mental health “by direct support or by alleviating risk factors”. This is a clear statement regarding the connection between risk factors and treatment: clearly risk factors need to be identified before any form of prevention (for example, child and parent befriending schemes) can be undertaken. It is easy to see how stressors, as described above, could be identified. Puig-Antich (1986, p.347) takes this connection further to suggest: “… it is conceivable that onset in childhood rather than in adult life of depressive illness may have serious implications for etiology, pathogenesis, course, and treatment”. Throughout the literature on childhood and adolescent depression connections have been made between etiology, identification of symptoms, diagnosis, and treatment (e.g. the Weinberg et al. [1978; in Poznanski, 1979] study – diagnostic criteria for identifying childhood depression – was in fact used to describe the efficacy of treatment). It is imperative that an interdisciplinary study is done with the specific aim of establishing links between risk factors and treatment for depression in children and adolescents.

One of the main themes throughout this essay is well-summed up by the following quote: “Some of the most interesting connections between early risk factors and later psychological adjustment, connections of prime importance to prevention, cross domains” (Reiss & Neiderhiser, 2000, p. 357). Perhaps the most important conclusion to be made from a study of the risk factors for depression in children and adolescents is that these risk factors interact and it would be a rare occurrence where a cause could be identified as a sole, discrete reason for a young person becoming depressed. Interactions exist, for example, between gender and social class and between genetics and maternal depression. Despite an overlapping etiology for depression, some extent of identification of cause should be helpful for deciding on which prevention measures or treatments or combinations of treatment would best benefit the at-risk child or teenager, and for developing more appropriate diagnostic tools for children and adolescents specifically.

(3926 words)

References

Allchin, B. (1975). Some Observations on Depression and Suicide in Adolescents. In Meyerson, S. (Ed.) Adolescence and Breakdown: A Study of Adolescence by Members of the Tavistock Clinic and other British Experts. London: George Allen & Unwin Ltd.

Cicchetti, D. & Schneider-Rosen, K. (1986). An Organizational Approach to Childhood Depression. In Rutter, M. et al. (eds). Depression in Young People: Developmental and Clinical Perspectives. New York: The Guilford Press.

Cyranowski, J.M.; Frank, E.; Young, E. & Shear, M.K. (2000). Adolescent Onset of the Gender Difference in Lifetime Rates of Major Depression: A Theoretical Model. Archives of General Psychiatry, 57: 21-27.

Davis, H.; Day, C.; Cox, A. & Cutler, L. (2000). Child and Adolescent Mental Health Needs Assessment and Service Implications in an Inner City Area. Clinical Child Psychology and Psychiatry, 5 (2): 169-188.

Emde, R.N.; Harmon, R.J. & Good, W.V. (1986). Depressive Feelings in Children: A Transactional Model for Research. In Rutter, M. et al. (eds). Depression in Young People: Developmental and Clinical Perspectives. New York: The Guilford Press.

Goodyer, I.M.; Herbert, J.; Tamplin, A. & Altham, P.M. (2000). First-episode major depression in adolescents: Affective, cognitive and endocrine characteristics of risk status and predictors of onset. British Journal of Psychiatry, 176 (1): 142-9.

Graham, C. A. & Easterbrooks, M. A. (2000). School-aged children’s vulnerability to depressive symptomatology: The role of attachment security, maternal depressive symptomatology, and economic risk. Development and Psychopathology, 12, 201-213.

Hammen, C. (1997). Depression. Hove: Psychology Press Ltd.

Izard, C. E. & Schwartz, G. M. (1986). Patterns of Emotion in Depression. In Rutter, M. et al. (eds). Depression in Young People: Developmental and Clinical Perspectives. New York: The Guilford Press.

Joyner, K. & Udry, J.R. (2000). You don’t bring me anything but down: adolescent romance and depression. Journal of Health & Social Behaviour, 41 (4): 369-91.

Kaufman, J.; Plotsky, P.M.; Nemeroff, C.B. & Charney, D. S. (2000). Effects of early adverse experiences on brain structure and function: clinical implications. Biological Psychiatry, 48 (8): 778-90.

Lewis, M. & Lewis, D.O. (1979). A Psychobiological View of Depression in Childhood. In French, A. & Berlin, I. Depression in Children and Adolescents. New York: Human Sciences Press.

Poznanski, E.O. (1979). Childhood Depression: A Psychodynamic Approach to the Etiology and Treatment of Depression in Children. In French, A. & Berlin, I. Depression in Children and Adolescents. New York: Human Sciences Press.

Puig-Antich, J. (1986). Psychobiological Markers: Effects of Age and Puberty. In Rutter, M. et al. (eds). Depression in Young People: Developmental and Clinical Perspectives. New York: The Guilford Press.

Reiss, D. & Neiderhiser, J.M. (2000). The interplay of genetic influences and social processes in developmental theory: Specific mechanisms are coming into view. Development and Psychopathology, 12: 357-74.

Rudolph, K.D.; Hammen, C.; Burge, D.; Lindberg, N.; Herzberg, D. & Daley, S.E. (2000). Toward an interpersonal life-stress model of depression: The developmental context of stress generation. Development and Psychopathology, 12: 215-234.

Rutter, M. (1986). Some Developmental Issues. In Rutter, M. et al. (eds). Depression in Young People: Developmental and Clinical Perspectives. New York: The Guilford Press.

Rutter, M. & Sroufe, A. (2000). Developmental psychopathology: Concepts and challenges. Developmental and Psychopathology, 12, 265-296.

Skuse, D. H. (2000). Behavioural Neuroscience and Child Psychopathology: Insights from Model Systems. Journal of Child Psychology and Psychiatry, 41, 1, 3-31.

Weller, E.B. & Weller, R. A. (2000). Depression in adolescents: growing pains or true morbidity? Journal of Affective Disorders, 61, Suppl 1 (1): 9-13.