Mental health benefits of retirement not evenly shared, study says

Mental health benefits of retirement not evenly shared, study says

The impact of retirement on mental health depends on income level, the nature of the job left behind, and the age at which people leave the workforce, a study suggests.

Researchers found that although retirees’ wellbeing generally improved once they finished work, some groups were at risk of dips in mental health.

People on an average income who had a physically demanding job before they retired and those on a high-income who retire at a later age are at risk of feeling low during the retirement process, experts say.

Retirees who are female, unmarried and on a low-income at the time of leaving work also reported significantly lower levels of mental health, the study found.

Researchers say the report could help guide targeted policy interventions and shows that people with a low income are a group at risk and require greater policy attention.

The study is among the first to assess the three phases of retirement – before, after and during leaving work – and to measure the mental health outcomes associated with factors including pre-retirement job demands and age

A University of Edinburgh team assessed a dataset of 1583 people from the Netherlands, where participants completed surveys on health, work, education, family and personality from 2007 to 2023.

Data was included from those who reported being retired and not undertaking any paid work. The average retirement age in the sample was 66 to 67 years.

Statistical models were used to assess people’s mental health five and a half years before and after retirement, and upon retirement.

Mental health was measured by the Mental Health Inventory, which records psychological well-being, depressive symptoms and anxiety symptoms.

The researchers found participants with a low income – earning below the minimum wage – reported the lowest levels of mental health during retirement. Although their mental health generally improved, there was a decline after an initial post-work ‘honeymoon’ phase, with dips emerging around two and a half years after retiring.

For retirees with an average income there was a substantial improvement in mental health before retirement, followed by a relatively modest increase.

Within the average income group, those who held more physically demanding jobs tended to report lower overall mental health throughout the retirement process.

Retirees with high-income showed no change in mental health before and after leaving work, but reported a significant sudden increase upon retirement. Those who retired later showed a slower improvement in mental health during the retirement year.

The study took into account other factors such as education, marital status, and retirement age.

Xuefei Li, of the University of Edinburgh’s School of Philosophy, Psychology and Language Sciences, who led the study, said: “This study identified distinct mental health trajectories across different income groups. We can assume that people's immediate health changes at the year they retire and longer-term health changes are different. Understanding these phases sheds light on the adjustment processes that people undergo when faced with the internal and external challenges.”

Professor Aja Murray, of the University of Edinburgh’s School of Philosophy, Psychology and Language Sciences, said: “All income groups showed a general improvement in mental health during the transition to retirement, yet there were multiple phases of development where mental health can take a dip. During these vulnerable times people may benefit from targeted support.”

Future studies could assess the impact of voluntary retirement to provide insights on the effect of personal choice on leaving work, researchers say.

The study is published in SSM - Mental Health. Link to study: Mental health trajectories surrounding retirement: A longitudinal perspective - ScienceDirect

The work was funded by the Wellcome Trust.  

The study used data from 17 waves of Longitudinal Internet studies for the Social Sciences (LISS) panel administered by Centerdata at Tilburg University in the Netherlands.

For further information, please contact: Joanne Morrison, Press and PR Office, joanne.morrison@ed.ac.uk