|dc.description.abstract||Injuries are an important health problem for young adults (15-24 years of age), especially males. In New Zealand, the leading causes of young adult injuries that result in death or hospital admission are motor vehicle traffic crashes and intentional self-inflicted injury. The causes of less serious injury are more diverse, and these injuries occur mainly during sport, work and domestic activities.
With the exception of traffic-related injury, little research has addressed the young adult injury problem. Most studies have suffered from methodological shortcomings (e.g. crosssectional study design, lack of multivariate analyses). A few recent studies, however, have addressed these shortcomings. Of particular relevance was a prospective study where lifestyle factors (based on Problem Behaviour theory) were shown to be predictors of traffic crashes among young adults. The present research aimed to extend this earlier study to examine the role of lifestyle factors as predictors of injury crashes and non-injury crashes, and also other serious injury, not traffic related, among young adults.
This research was part the Dunedin Multidisciplinary Health and Development Study (DMHDS) which is a longitudinal study of the health, development and behaviour of a cohort of young New Zealanders. The cohort (n=1037) has been assessed at regular intervals from birth through to age 21 years. The present research used data collected at ages 15 and 18 as predictors of injuries and traffic crashes that occurred between 18 and 21 years (recalled at the age 21 assessment). The predictor variables covered a wide range of factors related to background (personal and family), general behaviour (e.g. mental health, tobacco use, alcohol use); road traffic behaviour (e.g. driver's licensure, drinking and driving, exposure to risk); and personality. Four separate outcomes measures were examined: any crash, injury crash, non-injury crash, and serious injury (not traffic related). The main results were obtained using multivariate logistic regression analysis. The variables for the final models were selected using a bootstrap procedure.
Overall, relatively few lifestyle factors were shown to be important predictors of crashes or injury among young adults. For males, across all crash outcomes, the strongest predictors at age 18 were: having a motorcycle licence, low involvement with family, low constraint, high substance dependence, and high involvement in spare-time activities. For the females, across all crash outcomes, the strongest predictors at age 18 were: not being a poor reader, having a motorcycle driver's licence, high substance dependence, having at least one court conviction, and high involvement in physical activity. Serious injury was predicted by high involvement in physical activity (males and females) and high conduct disorder (males only). On the whole, the age 15 measures did not contribute meaningfully to the results.
It was concluded that focusing injury prevention efforts on changing the lifestyles of young adults may do little to reduce the crash or serious injury problem. Greater gains may be made by targeting specific safety issues, such as motorcycle use and substance dependence, and developing and implementing initiatives targeted at these.||