BACKGROUND: While there is considerable evidence of income gradients in child and adolescent behaviour problems, evidence relating to children and young people's emotional difficulties is more mixed. Older studies reported no income differentials, while recent reports suggest that adolescents from low-income families are more likely to experience emotional difficulties than their more affluent peers. METHODS: We compared the association between low- versus medium-/high-family income and parent-reported emotional difficulties in 15- and 16-year-olds in three large nationally representative cohorts studied in 1974, 1986 and 1999/2004. We then examined whether increases in the income differential could be accounted for by changes in the association of a range of sociodemographic factors (family type or size, maternal education or housing tenure) with either family income or emotional difficulties. Finally, in the most recent cohorts, we considered whether the effects of these sociodemographic variables were mediated by more proximal family factors (maternal distress, stressful life events or family dysfunction). RESULTS: An increasing income differential in adolescent emotional problems emerged over the period, with standardized coefficients for associations with low income increasing from .07 in 1974 and 1986 to .30 in 1999/2004. This was due partially (~10%) to sociodemographic risk factors for emotional difficulties becoming more strongly associated with low-income families over time, and partially (~40%) to the increasing impact of these risk factors. In the most recent cohorts, about 40% of the effects of sociodemographic risks appear to have been mediated by more proximal family factors. CONCLUSIONS: These findings have implications for our understanding of the health burden of emotional problems, recognition of the health burden associated with inequality and public concern about the consequences of social change.
BACKGROUND: While there is considerable evidence of income gradients in child and adolescent behaviour problems, evidence relating to children and young people's emotional difficulties is more mixed. Older studies reported no income differentials, while recent reports suggest that adolescents from low-income families are more likely to experience emotional difficulties than their more affluent peers. METHODS: We compared the association between low- versus medium-/high-family income and parent-reported emotional difficulties in 15- and 16-year-olds in three large nationally representative cohorts studied in 1974, 1986 and 1999/2004. We then examined whether increases in the income differential could be accounted for by changes in the association of a range of sociodemographic factors (family type or size, maternal education or housing tenure) with either family income or emotional difficulties. Finally, in the most recent cohorts, we considered whether the effects of these sociodemographic variables were mediated by more proximal family factors (maternal distress, stressful life events or family dysfunction). RESULTS: An increasing income differential in adolescent emotional problems emerged over the period, with standardized coefficients for associations with low income increasing from .07 in 1974 and 1986 to .30 in 1999/2004. This was due partially (~10%) to sociodemographic risk factors for emotional difficulties becoming more strongly associated with low-income families over time, and partially (~40%) to the increasing impact of these risk factors. In the most recent cohorts, about 40% of the effects of sociodemographic risks appear to have been mediated by more proximal family factors. CONCLUSIONS: These findings have implications for our understanding of the health burden of emotional problems, recognition of the health burden associated with inequality and public concern about the consequences of social change.