Background: In Northern Ireland (N.I) the overall prevalence rates of mental ill health
are approximately 25% higher than other parts of the U.K (Bamford., 2006). Research
suggests approximately 45,000 adolescents in N.I suffer with mental ill health at any one
time (Betts & Thompson., 2017). Despite this, the current available evidence on the
prevalence of mental ill health in adolescent in N.I and in the Republic of Ireland (R.O.I)
is scarce (Betts & Thompson., 2017; Dooley, Fitzgerald & Giollabhui., 2015; Harley et
al., 2015). The lack of existing prevalence data within N.I for mental health disorders in
adolescence is particularly concerning due to the rising rates of self-reported poor
emotional wellbeing, higher rates of antidepressant prescribing among young people and
increasing rates of self-harm (SH) and completed suicide among young people in N.I
(PHA., 2014; O'Neill, Corry, McFeeters, Murphy & Bunting., 2015). Unfortunately, the
lack of existing prevalence data for mental health disorders in adolescence means that our
understanding of mental health problems in N.I remains largely elusive.
Anxiety and mood disorders are the most prevalent disorders among young people in
today’s society and as such are considered a major public health concern (Stockings et
al., 2015). Developing either of these disorders in the younger years is associated with
increased risk of suicide, future psychopathology and a range of psychosocial issues.
(Balazs et al., 2013; Yap et al., 2018). Furthermore, it is well documented that a
substantial degree of comorbidity exists between anxiety and mood disorders in
adolescence. In fact a solid empirical evidence base suggests that when it comes to
depression and AD’s, comorbidity is the rule rather than the exception (Essau et al.,
2017). Research has suggested that the clinical outcomes of co-occurring anxiety and
depression are significantly more impairing than for either disorder alone (Garber &
Weersing., 2010; Zhou et al., 2017). Moreover, recent research has also demonstrated the
presence of co-occurring symptoms of depression and anxiety in subclinical forms among
adolescents within the general population (van Lang, Ferdinand, Ormel & Verhulst.,
2006; Wadsworth, Hudziak, Heath & Achenbach., 2001). Despite the fact that these sub
clinical symptoms of depression and anxiety do not meet diagnostic thresholds, they have
been found to contribute to high levels of distress for those who experience them and
greatly increase the risk of future psychopathology and suicidality (Balázs et al., 2013;
Jinnin et al., 2017; Løvaas et al., 2018; Mullarkey et al., 2019). However, despite the clear clinical relevance of sub threshold symptoms of anxiety and depression in
adolescence, it remains a largely understudied area (Jinnin et al., 2017). This, coupled
with lack of consensus regarding how best to distinguish anxiety and depression during
childhood and adolescence, with some researchers suggesting there is little to separate
anxiety and depression in the younger years and therefore may take the form of an
unidimensional construct (Garber & Weersing., 2010; McElroy, Fearon, Belsky, Fonagy
& Patalay., 2018; Rouquette et al., 2018), creates a key gap within the current literature
to understand how depression and anxiety symptoms are expressed among adolescents
within the general population. Overall, issues such as high rates of comorbidity,
heterogeneity of symptoms and the limited evidence supporting depression and anxiety
disorders as discrete diagnostic entities (Beard et al., 2016; Fried., 2015), has challenged
researchers to begin to study the complexity of these constructs in a unique way.
Specifically Fried and colleagues suggested that “this calls for more transdiagnostic
work” (p. 6, Fried et al., 2017). With this in mind, exploring a factor demonstrated to play
a transdiagnostic role in the onset and maintenance of both anxiety and depression
symptoms would therefore be a fruitful avenue of exploration. One such factor identified
in the literature is emotional regulation (ER; Schäfer et al., 2017). Therefore, this thesis
aims to explore self-reported symptoms of anxiety and depression, specifically
generalised anxiety disorder (GAD) and major depressive disorder (MDD) based
symptoms among adolescent females within the general population through the lens of
ER.
Aims: This thesis is concerned with three key lines of enquiry; (1) how common
symptoms of GAD and MDD are among school aged females in the general population,
(2) how are these symptoms expressed among school aged females, (3) what role does
ER ability play in relation to GAD and MDD symptom interplay and expression among
school aged females.
Methods: This study utilised data from a cross sectional school-based study. A total of
615 adolescent females (M = 13.32; SD = 2.02) were recruited from two post primary
schools in the north-west region of N.I. Participants completed measures assessing
demographic information, deficits in ER, major depressive disorder (MDD) and
generalised anxiety disorder (GAD) symptomatology, self-harm (SH) and thoughts of SH
and peer victimisation. A series of advanced statistical models including factor analysis,
latent variable modelling and network analysis (NA) were employed to test the research questions. Firstly, confirmatory factor analysis (CFA) and latent class analysis (LCA)
were utilised to (i) identify the dimensional structure of MDD and GAD in a school based
sample of adolescent females and (ii) explore the potential heterogeneity of MDD and
GAD symptoms among adolescent females. Next, exploratory factor analysis (EFA),
CFA and latent profile analysis (LPA) were employed to (i) identify the dimensional
structure of ER ability in a school-based sample of adolescent females and (ii) explore
the potential heterogeneity of ER ability among adolescent females. A series of regression
based analyse were then used to explore the differences between the ER classes using a
range of risk variables and mental health outcomes. Finally, network analysis (NA) was
used to examine the structure of MDD and GAD based symptoms in the sample and
examine the distinctions that emerge in the network structure of anxiety-depression
between adolescents with varying degrees of ER ability.
Results: In this sample 10.9% met the criteria for self-reported MDD symptoms and
16.9% met the criteria for self-reported GAD symptoms. Lifetime prevalence of SH and
thoughts of SH across the entire sample was 11.7% and 10.2% respectively. The structure
of GAD and MDD was successfully modelled as i) four homogenous sub-populations
characterised by different patterns of comorbidity, no distinct groups with ‘pure’ GAD or
‘pure’ MDD occurred, and ii) a network of causally associated symptoms, with no
evidence of modularity regarding the distinct domains of GAD and MDD. Regarding ER,
CFA confirmed a six-factor solution was the most parsimonious model. LPA revealed the
presence of three distinct profiles, each with varying degrees of deficits in ER ability. The
dysregulated class (characterised by higher levels across all six factors), exhibited greater
risk and were significantly more likely to experience several mental health outcomes. The
structure of three separate GAD-MDD networks was compared using the three ER ability
profiles generated from the LPA. The ‘low ER ability’ network as predicted, contained
the most positive associations, with a network structure consistent with previous similar
studies which have used psychiatric adult samples. However, no evidence of modularity
was demonstrated. Symptom connectivity for the ‘low ER’ group was significantly
greater compared to the ’high ER’ group. The findings tentatively suggest the significance
of the ‘transdiagnostic quality’ of ER in GAD and MDD symptoms in adolescence from
both a common cause and casual systems perspective, highlighting the potential
implications when ER processes are compromised. Conclusions: The current thesis is novel in nature for two key reasons i) contributing to
and updating the empirical evidence base surrounding the prevalence of common mental
health problems in the younger years in N.I. and, ii) offering a unique contribution to the
examination of the role of protective factors i.e. ER in the study of symptoms of
adolescent MDD and GAD from both a common cause and network perspective.