Abstract:
Many adolescents at some point in their lives experience bullying victimisation at
school. This has been amplified by its co-occurrence with depression and suicide that
has also been linked to bullying. In recent years the need for empirically proven
effective programs targeted at bullying and has increased significantly globally and
locally. The study investigated psychosocial dynamics impacting the mental health of
adolescents in public secondary schools in Nairobi County, Kenya, focusing on
bullying victimisation, depression, and suicidal behaviour. It also assessed the impact
of a Teacher-led anti-bullying psychoeducation program on these dynamics within the
educational context. A concurrent parallel mixed methods design incorporating a two
group post-test only control group design in the quantitative arm was used. A total of
539 students across 5 schools formed the sample for the study. The study population
comprised form 1 students who had attended the selected secondary schools for at least
one month. The study instruments included a questionnaire and guides for in-depth
interviews and focus group discussions. For quantitative data, predictors of depression
were assessed using a generalized linear model (GLM), using a backward stepwise
Poisson distribution with a log-link function, was used to estimate both the unadjusted
prevalence ratios (uPR) and the adjusted prevalence ratios (aPR). Variables were
included into the multivariable model based on a relaxed p-value of 0.2 in the
univariable analysis. The qualitative data was transcribed and analysed thematically.
The prevalence of depression was 14.5% among adolescents joining secondary school
students in Nairobi County, with a mean PHQ-A score of 6.16 (SD=3.16). Among
males, the prevalence of depression was 15.9% (n=34) compared to females who had
a prevalence of 13.5% (n=44). Suicide risk among adolescents was found to be 20.0%
(n=108) with a mean SBQ-R score of 4.88(SD=2.90). Majority (85.7%) of the students
had experienced bullying victimisation. Depression was the only statistically
significant predictor of bullying victimisation in multivariable analysis (aPR=1.33;
95%CI=1.05-1.68, p=.026). In the depression multivariable model, the risk of suicidal
behaviour [aPR=3.07, CI (1.94-4.88); p<.001] and lifetime alcohol use [aPR=2.24, CI
(1.36-3.68); p=.001] remained as statistically significant predictors. Bullying
victimisation was retained in the multivariable model, but it was not statistically
significant [aPR=2.88, CI (0.90-9.24); p=.075]. From the qualitative findings, students
had experienced bullying victimisation and depression and witnessed suicidal
attempts. Additionally, school-related gender-based violence was an emergent form of
bullying victimisation in mixed secondary schools. Ultimately, the psychoeducation
program was not effective in reducing bullying victimisation, depression, and suicide
risk. It is possible that these issues are deep-seated and perhaps needful of a longer-
term continuing intervention. It is recommended that interventions targeting
adolescent alcohol use should be commenced early in the secondary school period
given a likelihood of earlier age of onset. School-based cognitive behavioural therapy
is an emerging intervention that needs to be explored further to reduce and prevent
bullying victimisation, depression, and suicidality in secondary schools. The National
School Health Policy should be reviewed to create a framework for reporting and
addressing adolescent bullying, depression, and suicidal behaviour in the school
setting.