Socio-economic inequalities in adolescent mental health in the UK: multiple socio-economic indicators and reporter effects

Abstract

There are socio-economic inequalities in the experience of mental ill-health. However, less is known about the extent of inequalities by different indicators of socio-economic position (SEP). This is relevant for insights into the mechanisms by which these inequalities arise. For young people's mental health there is an additional layer of complexity provided by the widespread use of proxy reporters. Using data from the UK Millennium Cohort Study (N=10,969), we investigated the extent to which five objective SEP indicators (parent education, household income, household wealth, parent occupational status, and relative neighbourhood deprivation) predict adolescent internalising mental health and how this varies as a function of reporter. Both parent report and adolescent self-report were considered. Regression models demonstrated that whilst all five SEP indicators were associated with parent-reported adolescent mental health (regression coefficients for the most disadvantaged groups and adolescent mental health: parent education β=0.53 [0.44;0.62], household income β=0.56 [0.50;0.62], household wealth β=0.18 [0.10;0.27], parent occupational status β=0.40 [0.35;0.46], and relative neighbourhood deprivation β=0.41 [0.33;0.49]), only income (β=0.11 [0.04;0.17]), wealth (β=0.12 [0.02;0.21]), and occupational status (β=0.08 [0.03;0.13]) were associated with self-reported mental health. The magnitude of these effects was greater for parent-reported than self-reported adolescent internalising symptoms: SEP indicators jointly predicted 5.2% of the variance in parent-reported compared to 1.4% of the variance in self-reported internalising mental health. Income predicted the most variance in both parent (4.2% variance) and self-reported internalising symptoms (0.5% variance). Interestingly, the gradient of parent-reported adolescent mental health across SEP indicators mirrors that of parent's own mental health (for example, income predicted 7.3% variance). Our findings highlight that the relevance of different SEP indicators to adolescent internalising mental health differs between parent and adolescent reports. Therefore, it is important to consider the various perspectives of mental health inequalities gained from different types of reporters.