Youth Mental Health Decline Looks More Generational Than Period-Driven

Key Takeaways

• Among adults aged 15 to 85 in nationally representative Australian and UK household panels, mental health differed more by birth cohort than by period.
• Younger cohorts had worse mental health than older cohorts in both countries, with Gen Z clearly entering adulthood from a lower baseline; Millennials also showed significant disadvantages relative to older cohorts.
• Australian Gen Z showed a U-shaped recovery after age 20, but absolute mental health levels remained significantly below those of older cohorts.
• Explanatory patterns diverged across countries, with the Gen Z cohort penalty in UK models attenuating markedly after perceived social isolation was added, while measured Australian economic, family, and perceived support variables did not remove it.

Worsening distress in younger people has raised a persistent question in psychiatry and public mental health: does it reflect a difficult but expected life stage, or a structurally different generational experience that persists into adulthood? Australia and the United Kingdom (UK) are comparable high-income settings that have faced different recent social pressures. Investigators explored that question in two long-running national household panels, without implying direct applicability beyond those countries.

In the comparative Australia-UK mental health analysis, investigators analyzed the Household, Income and Labour Dynamics in Australia survey from 2001-2023 and Understanding Society in the UK from 2009-2023, restricting the sample to adults aged 15 to 85.

The final dataset included 239,244 Australian observations and 416,453 UK observations. Australia used the Mental Health Inventory-5 (MHI-5), whereas the UK used the 12-item Short Form Mental Component Summary (SF-12 MCS), so the countries were analyzed separately rather than treated as directly comparable on a single scale. Generalized additive models (GAMs) were used to visualize trajectories, and hierarchical age-period-cohort (HAPC) cross-classified random-effects models tested economic factors, life-course social benefits, and social networking after adjustment for age, age squared, gender, migrant status, region, and baseline mental health.

The youngest age group had the lowest descriptive mean mental health scores in both countries, at 4.21 in Australia and 3.81 in the UK. In Australia Model 1, the cohort random-effect SD was 0.0330 versus 0.0090 for the period random effect, and the same directional pattern held across models and in both countries. Younger cohorts had worse mental health than older cohorts in both countries, with Gen Z clearly entering adulthood from a lower baseline. Millennials also showed disadvantages relative to older cohorts.

In Australia, deterioration accelerated after 2009, with slight recovery in the youngest adults after 2021 and a U-shaped Gen Z pattern after age 20, but absolute mental health levels remained significantly below those of older cohorts. UK Gen Z showed no comparable recovery, and the adverse trajectory extended further into midlife.

In the UK models, the Gen Z cohort penalty attenuated markedly after perceived social isolation was added, whereas the Millennial deficit persisted after economic and social adjustment. In Australia, measured economic, family, and perceived social support variables did not eliminate the generational gap.

Direct cross-country point-estimate comparison remained limited because the outcome instruments differed, and the UK perceived social isolation construct was not equivalent to the Australian perceived social support construct. The analysis deleted observations missing the dependent variable and used listwise deletion for time-varying covariates, which the authors noted could interact with non-random attrition and selective mortality in longitudinal age-period-cohort work. The attenuation patterns therefore show how cohort gaps changed after adjustment rather than proving causal mechanisms. The findings identify country-specific patterns rather than a pooled cross-national effect size.

Youth mental health decline tracked more closely with birth cohort than with shared period shocks in both countries, but the apparent explanatory pattern differed between the UK and Australia. The authors interpreted the UK findings as more consistent with social disconnection among Gen Z, whereas the Australian gap persisted despite adjustment for the measured economic, family, and support variables.

Clinician Questions

Why were Australia and UK mental health scores interpreted separately rather than pooled in this cohort analysis?

The investigators used different mental health instruments in each country—MHI-5 in Australia and SF-12 MCS in the UK—so one-unit changes were not treated as metrically equivalent across countries. The comparison therefore centered on within-country patterns and on how cohort penalties changed after adjustment, rather than on direct cross-country coefficient comparisons.

How was social connectedness measured differently in the UK and Australian mental health models?

The UK models used a perceived social isolation construct based on subjective disconnection, whereas the Australian models used a perceived social support construct that leaned more toward functional and subjective support. The authors treated these as related appraisals of social integration rather than interchangeable measures of the same construct, which affects how the UK attenuation finding and the Australian persistence of the gap are interpreted.

How far into adulthood did the adverse mental health trajectory extend in the UK compared with Australia?

In the UK, the adverse trajectory extended up to age 55, whereas in Australia the decline was described as typically beginning below age 45. That contrast indicates a broader midlife extension of the UK pattern rather than a treatment implication.