Time trends in mental well-being: the polarisation of young people's psychological distress.

PURPOSE: Previous research on time trends of young people's mental health in Britain has produced conflicting findings: evidence for deterioration in mental health during the late 20th century followed by stability and slight improvement during the early 21st century is contrasted with evidence showing continued deterioration. The present study adds to the evidence base by assessing time trends in means, variances, and both low and high psychological distress scores covering a similar period.

METHODS: GHQ-12 (Likert scale) was regressed on time (adjusting for age) using a sample of young people aged 16-24 between 1991 and 2008 from the British Household Panel Study. Change in variance was assessed using Levene's homogeneity of variance test across 9-year intervals. Polarisation was assessed by a comparison of the prevalence of scores ≥1 standard deviation and ≥1.5 standard deviations above and below the pooled mean.

RESULTS: There was a small but significant increase in mean GHQ-12 among young women (b 0.048; 95% CI 0.016, 0.080) only. Variance increased significantly (p < 0.05) across 9-year intervals in seven out of nine comparisons for women and in six out of nine comparisons for men. There were significant increases in low (OR: 1.19; 95% CI 1.05, 1.35), high (OR: 1.27; 95% CI 1.13, 1.42), and very high scores (OR: 1.42; 95% CI 1.23, 1.64) for young women, and increases in low (OR: 1.39; 95% CI 1.21, 1.59) and very low (OR: 1.53; 95% CI 1.21, 1.92) scores for young men.

CONCLUSIONS: The evidence suggests a polarisation of the psychological distress of young women in Britain between 1991 and 2008.