The pandemic has brought links between unemployment and mental health to the fore. With joblessness having risen across the globe, new research looking at the longer-term effects is particularly timely. Liam Wright and colleagues from UCL’s Department of Epidemiology and Public health describe new research which could motivate efforts to target vulnerable groups and use resources efficiently.
We have known for some time that unemployment has a detrimental effect on mental health. And we know, too, that these effects can last for many years. A large body of research tells us those who have a spell out of work when young are more likely to suffer from poor wellbeing, depression and anxiety even decades later.
It’s particularly useful to study these effects through the lens of youth unemployment, for two main reasons. First, unemployment rates are higher for those under 25 and recessions have a disproportionate effect on them. Second, unemployment at this formative stage of life may have a greater impact than unemployment later on: it can affect the way young people see themselves and can set off ‘chains of risk.’
Negative experiences during youth can have a measurable impact on our responses to stress, and that this can have a lifelong effect on our physical and mental health.
But until now we don’t know much about the reasons behind these links, or about whether these impacts were experienced differently by different groups of people. By learning more about these things, we should be able to direct resources more effectively to those who are likely to need extra help.
We decided to look at whether unemployment had a stronger association with later mental health for some individuals than others using a statistical technique called quantile regression. We also looked at whether the association was stronger for those with longer unemployment spells, was larger in men or women, and whether later employment success (which is thought to explain the association) was associated with relatively better mental health.
We used data from Next Steps, formerly the Longitudinal Study of Young People in England, which followed a cohort of English school children who were aged 13 and 14 in 2003-4. The group were followed up to the age of 25– by that stage, there were 7,700 in our sample.
The mental health of the participants was measured at age 25 using the General Health Questionnaire or GHQ-12, a screening tool which can pick up mood and anxiety disorders and which scores them on a 36-point scale of seriousness – with 36 the most serious. We also took account of whether the respondents had a disability, their mental health during adolescence, as well as how they saw their general physical health.
We were able to compare these health scores with the employment status of the young people, focusing on those who had been unemployed for six months or more around the ages of 18 to 20 – this took place between 2008-10 and coincided with the global financial crisis, after which youth unemployment rose significantly.
We took account of a range of other factors such as gender, ethnicity, neighbourhood deprivation, educational attainment at age 25 and risk-taking behaviours such as drug-taking, alcohol, smoking and anti-social behaviour.
The results supported our key hypothesis that the association between youth unemployment and later mental health was driven by a relatively small proportion of formerly unemployed individuals who had very poor levels of mental health. Our model suggested that among a set of hypothetical individuals with average characteristics, more than 30 per cent of those who had been unemployed more than six months would have GHQ scores over 15 on the 36-point GHQ scale; 10 percentage points more than those who had not.
These effects could be seen even among those who were employed by age 25, and there was some evidence that the association was greater for men than for women.
Who’s at risk?
Our findings support and extend our existing knowledge, and they also pose questions: who are the individuals most at risk? We know men are more vulnerable in this respect than women, though this may be in part due to the greater likelihood that they are seeking work as opposed to looking after children, for example. But do men suffer more in economic, as opposed to mental health, terms?
We might also look at whether certain personality traits can help or hinder the wellbeing of those who find themselves unemployed while young. For example, does it help to feel that one has control over one’s own destiny, rather than taking a more fatalistic approach?
And we might look, too, at the mechanisms through which scarring takes place. Could unemployment while young affect people’s neuro-behavioural development? Or should we focus more on the ways in which an early spell of unemployment can cause problems later in the jobs market?
The answers to these questions could help us to identify vulnerable groups more accurately, and to point towards policy solutions which could potentially reduce these scarring effects in the future.
Heterogeneity in the Association Between Youth Unemployment and Mental Health Later in Life: A Quantile Regression Analysis of Longitudinal Data from English Schoolchildren, is research by Liam Wright, Jenny Head and Stephen Jivraj of the Department of Epidemiology and Public Health, University College London, and is published in BMJ Open (http://dx.doi.org/10.1136/bmjopen-2020-047997).