Peer victimisation during adolescence and its impact on depression in early adulthood
A prospective cohort study in the United Kingdom
“Peer victimisation during adolescence and its impact on depression in early adulthood: prospective cohort study in the United Kingdom by Lucy Bowes and colleagues (BMJ 2015;350:h2469).
Objective—To investigate the strength of the association between victimisation by peers at age 13 years and depression at 18 years.
Design—Longitudinal observational study.
Setting—Avon Longitudinal Study of Parents and Children, a UK community based birth cohort.
Participants—6719 participants who reported on peer victimisation at age 13 years.
Main outcome measures—Depression defined according to international classification of diseases, 10th revision (ICD-10) criteria, assessed using the clinical interview schedule-revised during clinic assessments with participants when they were aged 18 years. 3898 participants had data on both victimisation by peers at age 13 years and depression at age 18 years.
Results—Of the 683 participants who reported frequent victimisation at age 13 years, 101 (14.8%) were depressed according to ICD-10 criteria at 18 years; of the 1446 participants reporting some victimisation at age 13 years, 103 (7.1%) were depressed at age 18 years; and of the 1769 participants reporting no victimisation at age 13 years, 98 (5.5%) were depressed at age 18 years. Compared with children who were not victimised those who were frequently victimised by peers had over a twofold increase in odds of depression (odds ratio 2.96, 95% confidence interval 2.21 to 3.97, P<0.001). This association was slightly reduced when adjusting for confounders (2.32, 1.49 to 3.63, P<0.001). The population attributable fraction suggested that 29.2% (95% confidence interval 10.9% to 43.7%) of depression at age 18 years could be explained by peer victimisation if this were a causal relation.
Conclusion—When using observational data it is impossible to be certain that associations are causal. However, our results are consistent with the hypothesis that victimisation by peers in adolescence is associated with an increase in the risk of developing depression as an adult
Why do the study?
Depression is common and will affect between 4% and 10% of people throughout their lifetime worldwide. It is diagnosed when a patient has a persistent low mood and an absence of positive affect, accompanied by a range of symptoms.
Symptoms are most commonly diagnosed using either the World Health Organization’s international classification of diseases, 10th revision (ICD-10) criteria, or the American Psychiatric Association’s 5th edition of the Diagnostic and Statistic Manual of Mental Disorders (DSM-5).
Over recent decades, clinical diagnosis and treatment of child and adolescent depression have increased worldwide.Rates of youth suicide have seen a corresponding increase across a similar time frame in the World Health Organization European Region.If depression is becoming more common among adolescents, what can we do about it? Targeting schoolchildren with school based cognitive behaviour programmes is one solution, but it has shown little efficacy.
Meta-analyses of longitudinal studies have shown that bullying is a significant risk factor for depression, with bullied teenagers being about twice as likely as non-bullied ones to be depressed up to seven years after the bullying stopped. Studying whether peer victimisation during adolescence leads to depression in later life is therefore of interest.
If an association exists, interventions might be targeted to schoolchildren who are bullied, to help them cope better and perhaps prevent depressive symptoms from worsening when they are older.
What did the authors do?
The authors used data from the Avon Longitudinal Study of Parents and Children (ALSPAC). The data were initially collected between 1 April 1991 and 31 December 1992, recruiting 14 541 pregnant women, their children arising from pregnancy, and their partners in Bristol, United Kingdom, and the surrounding area.
The ALSPAC study followed this cohort of women and their children for more than 20 years, using questionnaires, interviews, and linked data to measure whether there was a causative link between bullying and depression at 18 years old. Obtaining data both prospectively and retrospectively helped the authors to see if there is an association between various exposures and outcomes.
When analysing the data, the authors identified whether participants had depressive symptoms at the age of 18 using the ICD-10 to classify mental disorders, including depression.
Next, the authors sought to identify whether the children were bullied or victimised at the ages of 8, 10, and 13 by using a bullying and friendship interview. This helped researchers to see if time plays a part in developing depression—that is, are adolescents who are bullied longer more likely to develop depression. Participants were asked about overt and relational victimisation, including exclusion by peers, being beaten up, and being threatened.
Children were given a score between 0 and 25, where 0 represented never being bullied. To investigate a possible dose-response relation between victimisation and depression, children were further coded on a three level ordinal variable (for victimisation) scale. Code 0 represented never being bullied (score 0), code 1 represented being occasionally bullied (score 1-3), and code 2 represented being bullied often (score 4 or more).
A confounder is a risk factor that is associated with both the exposure (victimisation) and the outcome (depression).
The authors collected data on two broad categories of confounders: individual characteristics of the adolescent and family characteristics, via various questionnaires and interviews of adolescents and mothers. Individual characteristics included were being male, having emotional problems, conduct problems, concurrent depressive symptoms, or concurrent bullying perpetration (at 13 years).
In relation to family characteristics, having a lower parental social class, lower level of maternal education attainment, maternal depression, or maltreatment were all identified as confounders.
More than 14 000 mothers initially enrolled in the ALSPAC study, but complete data on exposure, outcome, and confounders existed for only 2668 adolescents in the published study (see figure). 1 Data on peer victimisations at 13 years were collected for 6719 participants, and on both victimisation at age 13 years and depression at age 18 years for 3898 participants.
The authors acknowledged that missing data was a limitation of the study and adjusted for it in their analysis with multivariate imputation by chained equations (MICE). MICE is a statistical technique that replaces missing values with substituted values for data that are considered “missing at random. MICE was conducted first on missing data within confounders to give a sample size of 3898, then on outcome—that is, depression—to give a sample size of 6472.
What did they find?
Evidence exists to support an association between peer victimisation in adolescence and depression at age 18 years, which meets the ICD-10 criteria, including dose-response relation. This remained true after adjustment for individual and family confounders.
Association with depression
Participants who were victimised were more likely to develop depression by age 18 years compared with those who were not victimised.
Of teenagers who were not victimised (n=3090), 5.5% developed depression, compared with 7.1% of teenagers in the occasionally victimised group (n=2430) and 14.8% in the often victimised group (n=1199).
|Victimisation status||No (%) depressed||Unadjusted odds ratio (95% CI)||Odds ratio (95% CI)|
|All available data (n=3898)||Complete cases (n=2668)||Adjusted (n=2668)||Unadjusted using imputed dataset* (n=6472)||Adjusted using imputed dataset† (n=3898)||Adjusted using imputed dataset* (n=6472)|
|Occasional||1446 (7.1)||1.31 (0.98 to 1.74)||1.34 (0.93 to 1.93)||1.08 (0.74 to 1.59)||1.35 (1.01 to 1.81)||1.08 (0.80 to 1.46)||1.13 (1.02 to 1.12)|
|Frequent||683 (14.8)||2.96 (2.21 to 3.97)||3.33 (2.32 to 4.78)||2.32 (1.49 to 3.63)||2.82 (2.05 to 3.87)||2.00 (1.39 to 2.87)||1.87 (1.29 to 2.72)|
|Linear trend||—||1·13 (1·09 to 1·17)||1.14 (1.09 to 1.18)||1.08 (1.02 to 1.14)||1.12 (1.08 to 1.16)||1.07 (1.03 to 1.12)||1.07 (1.02 to 1.12)|
Compared with the never victimised group, those who were occasionally victimised were 1.31 times more likely to develop depression (95% confidence interval, 0.98 to 1.74), and those who were often victimised were almost three times more likely to develop depression (odds ratio 2.96 (95% confidence interval 2.21 to 3.97)). After confounding was adjusted for, both odds ratio decreased slightly to (1.08 (0.74 to 1.59) and 2.32 (1.49 to 3.63)) respectively.
After adjusting for confounders, those who were continuously victimised were 1.08 times (95% confidence interval, 1.02 to 1.14) more likely to develop depression compared with those who were never victimised. A linear trend was observed, confirming a dose-response relation.
To assess how much of the total risk of depression at age 18 could be explained by peer victimisation in adolescence, the authors calculated the population attributable fraction—contribution of a risk factor to a disease or a death—that is, depression. The population attributable fraction was 29.2% (10.9% to 43.7%).
Strengths and limitations
However, can we establish causation for this study question? It would be unethical and impractical to answer this question using the gold standard of a randomised controlled trial because children could not be allocated and exposed randomly to peer victimisation. We must therefore rely on risk factors research, such as this study, to help inform us on associations between exposures and outcomes of this nature.
The follow-up of this cohort remains of interest. To have a relatively extended follow-up on a cohort of this size is unique, particularly with the level of detail on risk factors, outcome, and confounders. However, is looking at the effect of bullying by the age of 18 enough? It is likely that some young adults will not be diagnosed with depression until later on in their life, a proportion of which will be related to peer victimisation. The drop-out rate of respondents for this cohort cannot be ignored. Less than 20% of the original cohort remained, and we must ask whether this led to selection bias—that is, were people who dropped out of the study more likely to have depression. The authors defend this possibility by carrying out MICE, sensitivity, and missing data analyses, which confirmed preliminary and final study results. However, selection bias remains a possibility.
Another notable limitation of the data is the lack of analysis of the effects of cyberbullying. This type of bullying is increasing but was considerably less common when initial data were collected in 2003-5. About 90% of cyberbullying victims are also “traditionally” bullied, making these results clinically relevant.
What does the study mean?
Perhaps unsurprisingly, the results suggest that peer victimisation during adolescence can lead to depression. In this data rich cohort, particularly with adjustment of known confounders, the authors call for more robust antibullying measures in schools. How should this take place?
In a linked BMJ editorial, Ttofi argues that we must better educate children and adolescents to endorse antibullying attitudes, ensuring they inform parents and school staff and are careful not to internalise victimisation. This may affect how they progress to adolescence and then young adulthood.
As many of the study participants did not report being bullied to teachers or their mothers, better counselling based support systems between home and school might be a good start.Neil Chanchlani, foundation year 2 doctor
1Colchester General Hospital
Correspondence to: firstname.lastname@example.org
Competing interests: None declared.
Provenance and peer review: Commissioned; not externally peer reviewed.
- Waraich P, Goldner EM, Somers JM, Hsu L. Prevalence and incidence studies of mood disorders: a systematic review of the literature. Can J Psych 2004;49:124-38.
- Collishaw S. Annual research review: Secular trends in child and adolescent mental health. Child Psychol Psychiatry 2015;56:370-93.
- Thapar A, Collishaw S, Pine DS, Thapar AK. Depression in adolescence. Lancet 2012;379:1056-67.
- Stallard P, Sayal K, Phillips R, Taylor JA, et al. A classroom based cognitive behavioural therapy in reducing symptoms of depression in high risk adolescents: pragmatic cluster randomised controlled trial. BMJ 2012;345.
- Ttofi MM, Farrington DP, Lösel F, et al. Do the victims of school bullies tend to become depressed later in life? A systematic review and meta-analysis of longitudinal studies. J Aggress Confl Peace Res 2011;3:63-73.
- Graham JW. Missing data analysis: making it work in the real world. Ann Rev Psych 2009;60:549-76.
- Hawker DS, Boulton MJ. Twenty years’ research on peer victimization and psychosocial maladjustment: a meta-analytic review of cross-sectional studies. J Child Psychol Psych 2000;41:441-55.
- Kaltiala-Heino R, Rimpela M, Marttunen M, et al. Bullying, depression, and suicidal ideation in Finnish adolescents: school survey. BMJ 1999;319:348-51.
- Olweus D. Cyberbullying: An overrated phenomenon? Eur J Dev Psychol 2012;9:520-38.
- Ttofi, Maria M. Adolescent bullying linked to depression in early adulthood. BMJ 2015;350:h2694.
Cite this as: Student BMJ 2015;23:h4215