The influence of different types of war experiences on depression and anxiety in a Ugandan cohort of war-affected youth: the WAYS study.

BACKGROUND

War experiences (WE) are frequently associated with mental health problems. Whether different types of WE vary in predicting which problem, or how severe, in former child soldiers (FCS) remains unknown.

METHODS

Using data from the first wave of an on-going longitudinal cohort study (the WAYS study), we investigated relations between types of WE and symptoms of depression/anxiety among FCS in Northern Uganda (N = 539, baseline age = 22.39; SD = 2.03, range 18-25). Using robust Maximum Likelihood estimation in SEM, regression analyses were performed to relate binary indicators of types of WE to a single latent factor capturing symptoms of depression/anxiety.

RESULTS

SEM results showed that "direct personal harm", "witnessing violence", "deaths", "threat to loved ones", "involvement in hostilities", and "sexual abuse" indicators were related to reported symptoms of depression/anxiety irrespective of gender and age. Multivariable models revealed independent associations of "witnessing violence" (β = 0.29, SE = 0.09, p < 0.001) and "deaths" (β = 0.14, SE = 0.05, p < 0.001) with symptoms of depression/anxiety in both sexes. "Sexual abuse" (β = 0.32, SE = 0.16, p < 0.001) independently predicted symptoms of depression/anxiety for female but not male youths whilst "threat to loved ones" (β = 0.13, SE = 0.07, p < 0.05) independently predicted symptoms of depression/anxiety in male but not female youths.

CONCLUSIONS

Dimensions of WE predicted symptoms of depression/anxiety differently, but it is hard to establish their causal status. Our findings suggest that it might be fruitful to consider such exposure variations of WE when designing interventions to mitigate the symptoms of depression/anxiety on male and female FCS.