Background The wellbeing of children and adolescents is a priority area for the Scottish Government. Previous research has shown that mental wellbeing of schoolchildren in Scotland improved between 1994 and 2006, while socio-economic inequalities increased. Between 2006 and 2010 a reduction was seen in the prevalence of adolescent happiness. There are known urban-rural differences in adult mental health, with higher suicide rates in remote rural areas. The aim of this study is to describe urban-rural differences in mental health and wellbeing among 15 year olds living in Scotland in 2010.

Methods Data from the 2010 Scotland Health Behaviour in School-Aged Children survey were modelled using multilevel linear modelling for outcomes: life satisfaction, General Health Questionnaire (GHQ)-12 and number of weekly health complaints, and binary modelling for outcomes: happiness, confidence, subjective health, and feeling left out. Models were adjusted for age, sex, school type, family affluence, area level deprivation and rurality.

Results Preliminary comparisons of means and prevalence showed no difference in life satisfaction between ruralities, but significantly lower mean GHQ-12 score in remote rural areas (19.13 (SE=0.25) compared with 19.84 (0.16) in urban Scotland). Linear regression models found that after adjustment for individual explanatory variables and area level deprivation, adolescents living in remote rural Scotland had a GHQ-12 score 0.71 (SE=0.36) lower than their urban counterparts, and 0.21 (0.11) more weekly health complaints, while life satisfaction was 0.22 (0.10) lower in accessible rural areas. Logistic regression models estimated lower odds of happiness for young people living in accessible rural Scotland (OR=0.69, CI (0.50, 0.95). For all other measures there was no significant difference between ruralities. Categorical family affluence was significant for outcomes: life satisfaction, GHQ-12, weekly health complaints, subjective health, confidence and happiness, with improved health outcomes associated with increasing Family Affluence Scale (FAS). Area-level deprivation (Scottish index of multiple deprivation quintile) was similarly significantly and independently associated with weekly health complaints, GHQ-12 and confidence, independent of FAS and rurality. Variance at the school level was significant for outcome GHQ-12 only. Rurality explained more of the total variance (approximately 2%) than area level deprivation for outcomes weekly health complaints and happiness, while the opposite was true for GHQ-12.

Discussion Young people from rural areas have poorer mental health and wellbeing than those living in urban areas. This suggests that the urban-rural differences in adult mental health may begin in adolescence and should be investigated and addressed at this early life stage.