Anxiety and depression are highly prevalent in children and young people (learn more in my last Mental Elf blog) and early identification can lead to better long-term outcomes (Parry, 1992). However, most studies do not consider comorbid anxiety and depression as a distinct group. This is surprising given that 25-50% of youth with depression report comorbid anxiety (Axelson & Birmaher, 2001) and therefore require different treatment and support.
Using data from the Avon Longitudinal Study of Parents and Children (ALSPAC; a UK birth cohort study beginning in 1999, which has collected data from expectant mothers and their children over the last 25 years) Isabel Morales-Muñoz and colleagues (2023) from the University of Birmingham investigated associations between persistent levels of anxiety and/or depression (defined as high levels of symptoms over time) in childhood, and a range of adverse physical health, mental health and lifestyle outcomes in adulthood. Specifically, they wanted to know:
- How does anxiety, depression and comorbid anxiety/depression develop between the ages of 8–13?
- Does having anxiety, depression or comorbid anxiety/depression in childhood impact the likelihood of adverse outcomes at age 24?
- Which has the greatest impact on adverse outcomes at age 24 – childhood anxiety, depression or comorbid anxiety/depression?
8,122 mothers reported levels of anxiety and/or depression for their children at 8, 10 and 13 years old using the Development and Well-Being Assessment scale. This information was used in a latent class growth analysis to identify 3 groups:
- Children with anxiety
- Children with depression
- Children with comorbid anxiety/depression.
From this, models were used to further divide the groups into several unique classes, allowing for a more nuanced analysis when using logistic regressions.
Owing to the nature of cohort studies, attrition (drop out from the research study) was high. By the time children were 24 years old and completing the questionnaires themselves, participant numbers had reduced to 3,882.
1. How does anxiety, depression and comorbid anxiety/depression develop between the ages of 8–13?
Latent class growth modelling analyses were carried out to identify subgroups of different developmental trajectories, resulting in participants being split into the following groups:
|Participants with childhood anxiety|
|Class 1||n = 6,331, 72.9%||Persistent and increasing low levels|
|Class 2||n = 1,882, 21.7%||Persistent and decreasing intermediate levels|
|Class 3||n = 469, 5.4%||Persistent and decreasing high levels|
|Participants with childhood depression|
|Class 1||n = 695, 7.4%||Persistent and increasing intermediate levels|
|Class 2||n = 8,324, 88.2%||Persistent and decreasing low levels|
|Class 3||n = 421, 4.5%||Persistent and increasing high levels|
|Participants with childhood comorbid anxiety/depression|
|Class 1||n = 11,154, 91.2%||Decreasing low levels|
|Class 2||n = 703, 5.8%||Increasing intermediate levels|
|Class 3||n = 369, 3.0%||Increasing high levels|
Analysis found that a 3-class model best explained these differences.
2. Does having anxiety, depression or comorbid anxiety/depression in childhood impact the likelihood of adverse outcomes at age 24?
The authors ran logistic regressions on class 3 of each group as these had the highest severity of symptoms and therefore the highest risk of negative future outcomes.
Results found that class 3 participants did have an increased likelihood of adverse outcomes in adulthood, but in different ways:
- Participants with persistent anxiety in childhood were more likely to develop adverse outcomes in adulthood (p < 0.001, OR = 2.09, 95% CI [1.63 to 2.69]), especially panic disorder (p = 0.001).
- Participants with persistent depression in childhood were also more like to develop adverse outcomes in adulthood (p < 0.001, OR = 2.07, 95% CI [1.50 to 2.87]), most associated with developing Generalised Anxiety Disorder (p < 0.001).
- Participants with persistent comorbid anxiety/depression in childhood were more likely to develop adverse outcomes in adulthood (p < 0.001, OR = 1.99, 95% CI [1.49 to 2.65]), especially psychotic disorder and severe depression (both p < 0.001).
3. Which has the greatest impact on adverse outcomes at age 24 – anxiety, depression or comorbid anxiety/depression in childhood?
All 3 groups had an increased likelihood of developing physical health problems at age 24, but only participants in the class 3 depression (p = 0.002, OR = 1.27, 95% CI [1.09 to 1.48]) and class 3 comorbid anxiety/depression (p < 0.001, OR = 1.40, 95% CI [1.21 to 1.62]) groups were significantly more likely to develop asthma or arthritis.
Interestingly, only those in the class 3 comorbid anxiety/depression group were significantly associated with substance misuse at age 24 (p < 0.001, OR = 1.57, 95% CI [1.15 to 2.15]).
All 3 groups (especially the comorbid anxiety/depression group) were significantly associated with having employment or education problems at age 24 (anxiety: p = 0.001, OR = 1.56, 95% CI [1.20 to 2.03]; depression: p = 0.047, OR = 1.38, 95% CI [1.00 to 1.91]; comorbid: p = 0.018, OR = 1.48, 95% CI [1.07 to 2.05]).
This is the first longitudinal study to investigate the association between persistent anxiety and/or depression in childhood and different adverse outcomes in adulthood. The authors addressed this knowledge gap by establishing the different developmental trajectories of anxiety and depression, and subsequently demonstrated that persistent levels do seem to “limit the psychological, academic and social functioning in individuals” over time.
Aligning with previous studies (e.g., Belzer et al., 2004; Kalin, 2020), all three groups were significantly associated with adverse outcomes in adulthood, with the comorbid anxiety/depression group being the most at risk of developing mental health, physical health and lifestyle problems.
The authors conclude that knowing how these disorders develop through childhood allows for early targeted identification and intervention.
Strengths and limitations
A longitudinal study is an appropriate method when understanding changes in populations over time. Existing, well-validated scales were used and the authors attempted to control for confounding variables (e.g., gender, ethnicity) within the analysis or noted it as a limitation.
However, there are limitations to this study, some of which the authors acknowledge:
- Selection bias: By using an existing dataset, the authors had little control over the sample. As expectant mothers living in the Avon region were the target for recruitment, selection bias was introduced, meaning that the results may not be generalisable to other parts of the UK. The focus on mothers also meant that the perspectives of fathers were not considered.
- Attrition: The sample size reduced from 8,122 to 3,882 when the children (now adults) were self-reporting at the age of 24, resulting in a huge attrition rate of 47% which begs the question of just how reliable the findings are and whether further selection bias was introduced. Did only participants with certain characteristics continue to respond, therefore skewing the results?
- Use of self-reports: The use of self-reports always adds a risk of recall bias, as well as social desirability effect – participants may respond in ways that they think they should as opposed to being completely truthful. Arguably, one of the biggest drawbacks of using data from ALSPAC is that the responses change from mothers’ self-reports when the child is 8, 10 and 13 years old to the child self-reporting at 24 years old. This introduces potential reliability issues, namely two different samples who may respond in different ways being combined into one.
- Calculations used in logistic regression: The authors chose to only analyse the class 3 participants in each of the anxiety, depression and comorbid anxiety/depression groups. In some ways this makes sense – they have the most pervasive and increasing symptoms in childhood. However, this means the analysis was completed on only a small number of total participants, making the subsequent results hard to generalise and possibly painting a worse case scenario considering that, when looking at the total sample, 88% reported having no mental health problems, 86% no physical health problems and 63-88% no lifestyle problems at age 24.
Overall, results should be interpreted with caution.
Implications for practice
The implications of this study are very relevant to front-line psychological practice. If we have a better understanding of how anxiety and depression may develop over the course of childhood and adolescence, and who may be at greatest risk of adverse outcomes in adulthood, this may inform targeted early identification and prevention strategies. The authors suggest that this could beneficially impact policy and practice at certain touchpoints, such as the transition from children’s to adult’s mental health services, which has been highlighted by previous studies as an “international concern” due to the consequences of lack of continuity of care (Hendrickx et al., 2020 p. 163).
The authors discuss potential reasons as to why persistent anxiety, depression or comorbid anxiety/depression may be associated with adverse outcomes in adulthood, including:
- Biological causes (chronic elevation of stress hormones or dysregulation of the automatic nervous system).
- Lifestyle causes (having a mental health condition in adulthood is associated with increased smoking and poor diet, which in itself can elevate the risk of mental or physical health conditions).
- Potential interactions between the two (experiencing anxiety or depression is likely to impact the child’s social, cognitive or academic development, which in turn could result in adverse outcomes in adulthood).
This could feel quite deterministic and paints the picture that experiencing anxiety or depression in childhood is a one-way road to adverse outcomes for the rest of your life. But, there were groups within ALSPAC that show this is not the case. There are many respondents who, despite reporting anxiety and/or depression symptoms and high levels of family adversity indicators (e.g., financial difficulties, poor parental mental health or substance use) in childhood, continued to report no adverse outcomes in adulthood. It would good to see future research investigating protective factors associated with this sub-group in order to promote them within policy and practice.
Finally, the authors acknowledge that they did not look into all possible confounding variables and encourage future research to explore “other potential contributing factors such as cognition, social interactions, lifestyle and family factors and/or obstetric complications”. Adopting a social psychology standpoint and using qualitative methods could also add valuable insight by considering what impact having a ‘label’ of anxiety and/or depression at an early age has on children, and whether this impacts the child’s view of themself and how much agency they feel they have in shaping their own future.
Statement of interest
Morales-Muñoz, I., Mallikarjun, P. K., Chandan, J. S., Thayakaran, R., Upthegrove, R., & Marwaha, S. (2023). Impact of anxiety and depression across childhood and adolescence on adverse outcomes in young adulthood: a UK birth cohort study. The British Journal of Psychiatry, 222(5), 212-220.
Axelson, D. A., & Birmaher, B. (2001). Relation between anxiety and depressive disorders in childhood and adolescence. Depression and Anxiety, 14(2), 67-78.
Belzer, K., & Schneier, F. R. (2004). Comorbidity of anxiety and depressive disorders: issues in conceptualization, assessment, and treatment. Journal of Psychiatric Practice, 10(5), 296-306.
Hankey, L. (2023). Online support more helpful for youth anxiety than depression, according to recent review. The Mental Elf.
Hendrickx, G., De Roeck, V., Maras, A., Dieleman, G., Gerritsen, S., Purper-Ouakil, D., … & Tremmery, S. (2020). Challenges during the transition from child and adolescent mental health services to adult mental health services. BJPsych Bulletin, 44(4), 163-168.
Kalin, N. H. (2020). The critical relationship between anxiety and depression. American Journal of Psychiatry, 177(5), 365-367.
Parry, T. S. (1992). The effectiveness of early intervention: a critical review. Journal of Paediatrics and Child Health, 28(5), 343-346.