Abstract
Injury is a leading cause of death and hospitalisation in young adults in New Zealand. In the last decade, death or injury due to physical assault and suicidal behaviours have become major public health problems in young adults. Much of the published evidence on predictors for either type of injury is based on cross-sectional study designs, selective samples, small sample size, and retrospective report. Common and unique predictors for assault and/or suicidal behaviours in young adults have not been determined in a cohort study. The present research comprised three prospective cohort studies based on a birth cohort of young adults (N = 1037) enrolled in the Dunedin Multidisciplinary Health and Development Study. The first study (N = 856) determined sex-specific predictors in mid adolescence (age 15) and late adolescence (age 18) for experiencing physical assault in the past year in early adulthood (age 21). The second study (N = 816) determined predictors in mid adolescence for experiencing past-year suicidality (suicidal ideation and/or suicide attempt) in late adolescence and early adulthood. The third study integrated findings from the first two studies and determined sex-specific general (common) and differentiating (unique) predictors in mid adolescence for subsequent experiences of physical assault victimisation and/or suicidality. The predictors included alcohol use, weak attachments, injury, mental ill-health, and antisocial behaviours. The study members have been assessed regularly since their birth on a variety of health, behaviour, and background measures, with additional information from parents and teachers, and official conviction records. Results from the first study, from multivariate Poisson regression analysis showed that perpetration of assault or delinquency in adolescence was the strongest risk factor for physical assault victimisation. Other significant risk factors included a medically treated injury, alcohol use, and school disadvantage for males, and suicidal ideation and negative personality traits for females. Results from the second study, from multivariate logistic regression analysis showed that being female, weak attachments to parents or peers, separation from a parent, depression, and negative feelings about body size in mid adolescence were significant risk factors for subsequent suicidal ideation. There were no significant risk factors in mid adolescence which differentiated young adults who reported subsequent suicidal ideation and suicide attempt from those who reported suicidal ideation only. Results from the third study, from multivariate bivariate logistic regression analysis showed that delinquency, weak attachments to parents, and separation from a parent were significant general predictors in mid adolescence for subsequent assault and/or suicidality. Depression was a significant risk factor for females only, and medically treated injury and delinquency for males only. For males, weak attachment to parents or peers was a significant differentiating or unique predictor for suicidality. The results suggest that antisocial activities, weak attachments to parents or peers, depression, and medically treated injury in mid adolescence are important risk factors for experiencing assault and/or suicidality in early adulthood. Reducing exposure to these risk factors in mid adolescence may be associated with a lower risk of experiencing assault and/or suicidality in early adulthood. These findings need to be replicated with other samples of young adults, and preferably in cohort studies.