Essay: Depression in adolescents

Adolescents have a heightened vulnerability to depression. Specifically, prevalence rates of major depressive disorder (MDD) have been shown to increase immensely from approximately 2% in early adolescence to 15% in middle adolescence (Hankin, Abramson, Moffitt, et al., 1998). The implications of adolescent depression include poorer educational and psychosocial outcomes and long-term psychiatric problems, leading to individual as well as societal costs (Birmaher, Ryan, Williamson et al., 1996; Fergusson & Woodward, 2002; Knapp, McCrone, Fombonne et al., 2002). Therefore, it is of great importance to identify effective treatment- and prevention strategies. In order to realize this, more knowledge about the etiology of depression at this sensitive developmental time is required.
Adolescent depression frequently emerges in response to psychosocial stress, like social disappointment or a break-up (Lewinsohn, Allen, Seeley & Gotlib, 1999; Monroe, Rohde, Seeley, Lewinsohn, 1999; Thapar, Collishaw, Pine & Thapar, 2012). Adolescents spend an increased amount of time with their peers and accordingly, peer rejection is a particularly strong source of psychosocial stress (Platt, Cohen Kadosh & Lau, 2013). Previous studies have shown that adolescents diagnosed with or at risk for depression display more negative expectations and affect in reaction to peer rejection compared to their healthy peers (Platt, Cohen Kadosh & Lau, 2013). Vulnerability factors such as gender (female), genetics and cognitive bias (negative attributional style and negative self-schemas) are widely assumed to play a mediating role in this relationship. However, the exact underlying mechanisms causing the negative vulnerability of depressed adolescents to social rejection remain largely unknown.
Previous studies using functional magnetic resonance imaging (fMRI) have offered more insight into how the brain mediates the impact of peer rejection on adolescents in general and into what mechanisms possibly underlie adolescent depression. Various studies have found that distress following peer rejection in healthy adolescents was positively correlated to neural activation of affective brain regions, such as the insula, anterior cingulate cortex and precuneus and negatively correlated to regulatory regions, such as the VLPFC and DMPFC (Platt, Cohen Kadosh & Lau, 2013). With regard to depression, various fMRI studies found that amygdala hyperactivity in response to negative emotional stimuli was linked to the pathogenesis of depression (Abler, Erk, Herwig & Walter, 2007; Davey, Allen, Harrison & Y??cel, 2011; Siegle, Thompson, Carter, Steinhauer & Thase, 2007; Siegle, Steinhauer, Thase, Stenger et al., 2002). However, to date, no study regarding depression-linked differences in the neural substrates of peer rejection has been conducted. The current study therefore aims to address this research gap in order to gain more insight into why depressed adolescents respond more negatively to peer rejection compared to healthy adolescents.
We will examine the following hypotheses: (1) conform earlier findings as described by Platt et al (2013), depressed adolescents will display more negative expectations and affect in reaction to peer rejection compared to their healthy peers (they have a ‘pessimism-bias’), while healthy participants do not show this bias (2) Based on previous fMRI studies and on our first hypothesis, we expect that neural activation of affective brain regions will be positively correlated to negative distress in reaction to peer rejection in all participants, but even more so in depressed adolescents. (3) We expect that neural activation of regulatory regions will be negatively correlated to negative distress in reaction to peer rejection in all participants (4) Based on previous fMRI studies we expect to find amygdala hyperactivity in response to peer rejection in depressed adolescents, but less so in the healthy controls.
APPROACH
Participants
A total of 100 adolescents (50 participants with major depressive disorder as determined by the Structured Clinical Interview (SCI) for DSM-IV and the Beck Depression Inventory (First, Spitzer, Gibbon & Williams, 1997) 50 healthy controls), aged between 15 and 24 years, will be recruited. The age range extends from early to late adolescence and is compatible with our current knowledge of the continuities in brain and social development through this developmental period ( Davey, Y??cel & Allen, 2008). Exclusion criteria are (comorbid) psychiatric disorders, intellectual disability, neurological disorders, extensive drug or alcohol usage and brain trauma (screened by an online questionnaire provided to the participants beforehand). Antidepressant medication will not be an exclusion criterion for the sake of ethicality, but will be controlled for during the analysis by comparing BDI scores. Depressed participants will be matched with the healthy controls on age, gender, and full-scale IQ. Depressed participants will be recruited through mental healthcare facilities and healthy controls through advertisements in daily newspapers. Control participants will be screened on history of mental illness by the SCI. Ethically approved informed consent forms will be provided to the participants (or their parents in case they were under 18 years of age). Excessive head movements during scanning will lead to exclusion of the participant.
Research Design and Analysis
A between-subject design with 2 groups (depressed vs. control) as well as a within-subject design (all participants) will be used in order to answer the hypotheses (see figure 1. for a schematic overview of the hypotheses and methods used to test them). A modified version of the social judgment paradigm will be implemented (see Davey et al., 2010; Somerville et al., 2006; Gunther Moor et al., 2010 for a detailed description of this paradigm). This widely used paradigm simulates peer rejection and enables us to examine social evaluation (and the hypothesized pessimism bias), which is necessary for the current study. Moreover, this task can be used during fMRI recording (Davey et al., 2010). Participants will be given a cover story, stating they would participate in an investigation about first impressions. Afterwards there will be a check to ensure participants believed the cover story. The experiment consists of the completion of the social judgment paradigm and another cognitive control task that will be used to control for cognitive performance during analysis (order of presentation will be counterbalanced between participants), while lying in an A 3T Siemens Magnetom Trio magnetic resonance scanner to acquire whole-brain functional T2*-weighted echo-planar images (further specifics about image preprocessing can be obtained from Davey et al., 2010). Participants will be asked to rate (on a Likert scale) how they felt on receiving feedback (as a measure of distress after rejection) and how difficult they found maintaining their attention throughout the task (control). A bias score will be derived from the number of acceptance judgments by the number of total judgments in order to examine whether participants had either an optimism bias (>50%) or a pessimism bias (<50%) (Van der Veen et al., 2013). Between-group analysis will be conducted on the images. Group differences for the peer rejection (negative feedback) versus control-feedback contrast will be explored in the regions of interest (ROIs) using a two-factor ANOVA (group x feedback).