Uneven progress in reducing exposure to violence at home for New Zealand adolescents 2001-2012: a nationally representative cross-sectional survey series.

Objective: To explore trends, and identify risk factors, that may explain changes in adolescent exposure to family violence over time.

Methods: Data for this study was drawn from the Youth 2000 series of cross-sectional surveys, carried out with New Zealand high school students in 2001, 2007 and 2012. Latent class analysis was used to understand different patterns of exposure to multiple risks for witnessing violence at home among adolescents.

Results: Across all time periods, there was no change in witnessing emotional violence and a slight decline in witnessing physical violence at home. However, significant differences were noted between 2001 and 2007, and 2007 and 2012, in the proportion of adolescents who reported witnessing emotional and physical violence. Four latent classes were identified in the study sample; these were characterised by respondents' ethnicity, concerns about family relationships, food security and alcohol consumption. For two groups (characterised by food security, positive relationships and lower exposure to physical violence), there was a reduction in the proportion of respondents who witnessed physical violence but an increase in the proportion who witnessed emotional violence between 2001 and 2012. For the two groups characterised by poorer food security and higher exposure to physical violence, there were no changes in witnessing of physical violence in the home. Implications for public health: In addition to strategies directly aimed at violence, policies are needed to address key predictors of violence exposure such as social disparities, financial stress and alcohol use. These social determinants of health cannot be ignored.