Children with a marked increase in internalizing behaviour as tweens: determinants and mental health at 16 years.
Professor Vivienne Moore
The University of Adelaide
Anxiety and mood disorders are among the most common mental health problems experienced by young people. Not only do they have acute impacts on learning and social relationships, they are often followed by long-term impacts on mental health and life opportunities. Prevention and early intervention are thus recognized as essential. As well as society-wide initiatives, targeted efforts are needed in order to make the best use of finite resources. Thus identifying at-risk families and at-risk children is important, as is knowledge of what could be modified to improve mental health trajectories.
Vivienne Moore, Lynne Giles, Michael Davies, Michael Sawyer, Melissa Whitrow
The project concerns the development of anxiety disorders over childhood, culminating in late adolescence. It builds on work with an established cohort of some 550 children who were born in 1998-2000 and have had their competencies as well as behavioural and emotional problems assessed at 2, 3, 5, 9-10 and 12-13 years. Within this cohort, children were categorized according to their trajectory of internalizing behaviour (i.e. withdrawn, anxious/depressed, somatic symptoms) up to age 12-13 years. Distinct groups included children whose anxiety level was stable – either persistently high, moderate, low or very low. There was also a small group of children (3.5% of the total) who had a rising trajectory of anxiety, which suggested an influence of the wider environment, rather than an inherent disposition. A constellation of factors in early life were associated with anxiety trajectories. The degree of conflict between parents was progressively related to anxiety group. For those with a rising trajectory, the combination of maternal anxiety and inadequate resources was key. Work on the project is ongoing.
The project concerns the development of anxiety disorders over childhood, culminating in late adolescence. It builds on work with an established cohort of some 550 children who were born in 1998-2000 and have had their competencies as well as behavioural and emotional problems assessed at 2, 3, 5, 9-10 and 12-13 years.
Within this cohort, children were categorized according to their trajectory of internalizing behaviour (i.e. withdrawn, anxious/depressed, somatic symptoms) up to age 12-13 years. Distinct groups included children whose anxiety level was stable – either persistently high, moderate, low or very low. For this project, the focus was a small group of children (3.5% of the total) who had a rising trajectory of anxiety (RTA) which suggested an influence of the wider environment, rather than an inherent disposition.
The aims of this study were to:
(1) investigate historical and proximal determinants of RTA;
(2) assess the mental health of these children at 16-17 years;
(3) make comparisons with mental health of peers in the cohort.
In relation to the first aim, a variety of statistical analyses were used to investigate the circumstances in early life that were linked to the type of anxiety trajectory that children exhibited. When considered separately, a number of variables pertaining to early childhood (≤ 3 years) appeared to be differentiate the RTA group, including maternal age at child’s birth (on average, older), mother’s educational attainment (least likely to have left high school early) and family type (relatively more likely for mother to become a sole parent). Anxiety among mothers of the RTA group was elevated (close to that of mothers of the group of children with persistently high anxiety), as was maternal depression, and the extent to which family resources were reported to be inadequate. Parental conflict tended to be greater than reported for families of children with persistently low or very low anxiety, but not as severe as in families where children had persistently high or very high anxiety. Graphs and calculations suggested that these factors did not occur or operate in isolation, but formed a constellation. For example, anxiety scores in mothers were strongly correlated with depression scores (r = 0.75) and were moderately correlated with constrained family resources (r = 0.38).
When considered in an integrated manner, the factors that were most clearly associated with anxiety trajectory were conflict between parents (as per Parent Problems Checklist, graded progressive relationship with anxiety trajectories, p<0.01) and the combination (statistically significant interaction, p=0.02) of maternal anxiety (Spielberger A-state) and inadequate family resources (Family Resources Scale). We are now drafting a paper to convey this complexity.
We also planned to invite adolescents to complete a set of online questionnaires to profile current mental health. We modified the original plan to use the Centre for Epidemiologic Studies Depression Scale, Spielberger A-state and selected modules from the Diagnostic Interview Schedule for Children (DISC-IV). Instead we decided to use the Depression, Anxiety and Stress Scale (21 item version) and three other scales (comprising 8 items) because this set of questionnaires was used by Venning et al. (2013) to profile a large sample of South Australian adolescents (thus providing normative data for our study) and it also captures the full spectrum of mental health from flourishing through to overt problems.
The aim was obtain information from at least 418 of the cohort members (75%). We have not yet achieved this, but we plan to undertake a further round of invitations that should address this. Thus we have not yet completed the second and third aim, although we have made substantial progress.
The impediment that we did not foresee was that the Human Research Ethics Committee did not want us to approach young people aged 16-17 years directly (by letter or email). We were asked to write to the parent/guardian and seek their permission for the young person to be involved. We think that both the cumbersome nature of the process and the idea that this was a parent-sanctioned task contributed to a reduced response. However, we are now able to approach young people who have not yet responded directly.