Tag Archives: Work

Are universal state pensions discriminating against those in lower-skilled jobs?

With the state pension age likely to rise further in coming years, are policymakers right to link pension eligibility to average life expectancy? In a one-size-fits-all system, which social groups will lose out? Dr Emily Murray and colleagues* used census data to look at who lives longest after leaving work.

In most industrialised countries, the eligibility age for state pensions is being increased. Between 2011 and 2018, the United Kingdom government raised the State Pension Age for women from age 60 to 65, to match that for men, and a further increase to age 67 for both genders is planned by 2028. A further increase to age 68 by 2039 has been mooted.

Yet our state pension system ignores some very basic facts – it doesn’t take into account the wide disparities in health and life expectancy between different social classes. Those in professional occupations can expect to live longer and to enjoy good health for longer than those in manual jobs. For example, the average 50 year-old man in a professional job can expect to enjoy a further 25 years of good health, while a man the same age in a manual occupation can only expect 18: a seven-year difference. That is why lower social class groups are more likely to find themselves on disability benefit.

We wanted to look more closely at these occupational social class differences in the amount of time older adults live after they stop work, and in particular at the extent to which these differences are due to health.

We used the Office for National Statistics Longitudinal Study, a one per cent representative sample of respondents to the English and Welsh censuses since 1971.  For our analysis, we included respondents who were aged 50-75 at the time of the 2001 census and who had stopped work by 2011 – the average age of stopping was 58 for women and 60.2 for men. These workers were born in 1951 or earlier, so men would have been eligible for state pension at 65 and women at 60.

That gave us a sample of 76,485 people, and over the next 10 years we were able to monitor deaths  – by 2011 14.6 per cent of the women and 25.1 per cent of the men had died.

We could see that for both genders, those in lower social classes tended to die younger – professional women lived two years longer than unskilled women, and professional men three years longer than unskilled men.

We estimated professional women in good health would live five years longer than unskilled women in poor health, while for men the gap would be five and a half years.

But despite these longevity gaps, those from lower social groups were facing more years between leaving work and being able to draw their state pensions – because they left work earlier.

We estimated that if two women were 65 in 2001, the woman who had worked in an unskilled occupation would live five years longer after leaving work than the professional woman with good health – because the unskilled woman would have left at a younger age. Two men in the same circumstances would live on average 25.0 and 19.5 years from stopping work to death.

The most likely explanation is that poor health has a greater impact on the ability of manual workers to continue working than it does on non-manual workers.  It is however important to note that associations between social class and post-work years were not entirely explained by health, and we feel more research is needed on this.

Poor health

But the conclusion is clear: our results show that a uniform state pension age disproportionately affects the poorest among us, because on average they must wait longer between stopping work and qualifying for their state pension, at a time when they are likely to be in poor health. This is despite the fact that they are likely to have started work younger and therefore to have worked and paid contributions for just as many years as their better-off peers.

The solution to this inequality is not straightforward. The preferred strategy for UK policymakers is to support individuals to stay in work for longer, and there is evidence that the average age of leaving work exit is increasing.  However, over half of women and two-fifths of men  still fall out of the labour market before state pension age.

Some researchers have suggested that pension ages should directly reflect life expectancy differences.  Alternatively the age requirement could be dropped and pension eligibility could be based solely on the number of years in work.

We believe a two-year earlier pension age may be more appropriate for individuals who work in manual occupations, given that they leave work earlier than professional workers not in good health.  With rises in pension age already in law, and evidence of stalling life expectancy, it is vital that researchers and policy-makers assess how these rises will influence financial security and health for the most vulnerable in society.

Inequalities in time from stopping paid work to death: findings from the ONS Longitudinal Study, 2001 to 2011 is by Emily T Murray,  Ewan Carr, Paola Zaninotto, Jenny Head, Baowen Xue, Stephen Stansfeld, Brian Beach and  Nicola Shelton.

*Emily T Murray, Ewan Carr, Paola Zaninotto, Jenny Head, Nicola Shelton and Baowen Xue are based at the Department of Epidemiology and Public Health, University College London.

Ewan Carr is also based at the department of Biostatistics and Health Informatics, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London.

Stephen Stansfeld is based at Queen Mary University of London, Wolfson Institute of Preventive Medicine, Centre for Psychiatry, London, EC1M 6BQ, UK

Brian Beach is based at the International Longevity Centre – UK, SW1P 3QB, London, UK.

Does caring for others damage your mental health?

This blog is dedicated to the memory of the paper’s lead author, the luminary Allison Milner, whose recent death has left the world a poorer place.

Health work and caring make up a significant proportion of jobs in high-income countries – but how does the mental health of people in those jobs compare with others? And how does the mental health of those in higher-status medical jobs compare with that of those in lower-skilled and lower-paid caring jobs? Tania King and colleagues* at the University of Melbourne turned to a major Australian longitudinal study for answers.

The health and social care workforce is very diverse, encompassing both those with the highest levels of skill and salary and those with some of the lowest. Yet all those workers have one thing in common: they work primarily in jobs which involve caring for others.

This means the emotional demands of the job are high: these workers often experience threats of or actual violence, for instance, and tend to have low levels of control over their work. This has been shown to lead to burnout, high levels of sickness absence and high job turnover.

We used the Household, Income and Labour Dynamics in Australia (HILDA) survey, a nationally representative study which has collected information from more than 13,000 people each year since 2001, to look at three questions:

  • Are job conditions in health and care more stressful than those in other sectors?
  • Do job stressors affect the health of health and care workers more than they affect others?
  • If health and care workers do suffer more than others, is that linked to the type of people who tend to go into that type of work and how they react to stress?

On the first question, we found that overall the working conditions of those in caring and support work jobs – the lower-paid, lower-skilled end of the sector – were worse than those in higher-status medical jobs, with care workers reporting a higher number of stress factors such as high demands, lack of control, job insecurity and perceived unfairness of pay.

Then we looked at workers’ mental health, using a widely recognised scale which assesses respondents’ symptoms of depression and anxiety in the past four weeks, along with more positive feelings such as calmness and happiness.

We found carers and support workers had poorer mental health scores than health workers. They also scored worse when compared to other workers who were not in health, care or other people-facing roles.

Other human service workers, such as hospitality workers, sales staff and personal assistants had worse mental health than those in roles which involved less personal contact. Of the four groups – carers, health workers, human service workers and others – carers had the worst mental health and health workers had the best.

Mental health effects

What were the biggest stress factors for the different groups? We found job insecurity had the greatest effect on all of them. But there were differences, too: In human service jobs, fairness of pay was the second-biggest factor, while in health care professions low control was associated with the second largest decline in mental health.

But we also found differences in the ways in which different groups reacted to stress: When carers and support workers reported three or more different stress factors, that resulted in a three-point decline on the mental health scale when compared to those reporting none. But when health care workers reported three or more stress factors that was associated with a lower, 2.65 point, decline in mental health.

Low-skilled is high-stress?

Our results suggest – perhaps surprisingly, given popular narratives about the stress suffered by senior executives – that lower skilled carers and support workers are more likely to be exposed to stress at work than higher skilled health workers such as doctors and nurses. They also have poorer mental health and a greater adverse reaction to stress.

There is other evidence to show low-income, less-educated workers suffer more from work stress. But our analysis adds another factor – the care workforce we looked at are mainly women with lower incomes and less education than either health workers or our reference group of other workers.

And even when this largely-female, low-status group of carers reports no stress factors at work, its mental health is worse than that of other workers – we found a three-point difference on the mental health scale between carers and either health staff or our reference category of non-health, non-people-facing staff.

Hence, people with lower levels of mental health face an accumulation of health hazards. They are likely to find themselves in stressful jobs which propel them into even worse mental health.

And so to our final question – is the poor mental health of care workers somehow intrinsic to them as people? Our methods enabled us to control for stable (time-invariant) characteristics such as gender, personality or country of birth (called within person effects), as well as things which might change over time, such as age, income and type of employment. The results showed that even after controlling for within person effects, as well as these other factors such as age and income, being exposed to these stress factors had a negative effect on the mental health of care workers.

We concluded that – while it’s important to note that health workers, too, experienced stress and poor mental health – low-quality working environments has a clear effect on the poorer outcomes of carers and support workers.

The impact of these environments does not stop with those who work in them:  this may also have knock-on effects on the quality of care provided. Earlier studieshave shown factors such as anxiety, stress and burnout  among healthcare workers are linked with poorer patient safety, too.

There is a dire need for more investigation into working conditions among health and human service workers. Such research could be critical in informing workplace design, particularly as the number of people employed as health and human service continues to grow. For that workforce to be sustainable, jobs need to be structured to promote health and productivity as well as a high quality of care.

The mental health impacts of health and human service work: Longitudinal evidence about differential exposure and susceptibility using 16 waves of cohort data, by Allison Milner, Tania L. King and Anne Kavanagh, was published in Preventive Medicine Reports, vol 14, June 2019.

Woman doing housework

Domestic work – why do women still do the lion’s share?

A recent international report suggests men need to increase their time spent doing unpaid care work by a minimum of 50 minutes per day in order to do 50 per cent of the work. The report calls for bold measures to help all men do their fair share of this work by 2030 and thus promote gender equality. So what do we know about how modern couples in the UK divide unpaid domestic work and the drivers behind that? A new study from Anne McMunn at the ESRC International Centre for Lifecourse Studies at University College London investigates why greater workplace equality has not yet been matched by a more equitable division of labour at home.

Some studies suggest education is the key to equality within the home – that women with better education, and therefore greater earning power, are in a stronger negotiating position when it comes to housework.

But that theory doesn’t entirely hold water – it’s clear that even when women are better educated than their partners, they’re still likely to bear the heaviest burden when it comes to domestic labour.

Similarly, it’s been suggested that domestic work is divided according to time available – so when a male partner works longer hours, the female does more housework. But again, the reverse doesn’t hold true and women who work more hours outside the home still do more housework too.

Longitudinal study

We used the UK Household Longitudinal Study, which has surveyed around 40,000 households since 2010, to examine a number of hypotheses about why greater workplace equality hasn’t been matched by a more equitable division of labour at home.

Does a shared belief in equality make a difference? As men are often asked, in the modern world, to do more domestic labour, are their beliefs a driving factor? What is the role of education, if any, in how couples divide up these tasks? If one partner is resistant to sharing the work, is the other empowered by having a higher level of education?

Using opposite-sex couples in the study who were aged 16-65 and had answered the relevant questions, gave us a sample of 8,513 couples. We looked at four types of work to give us a full picture of the labour those couples did: housework, paid employment, childcare and adult care – for instance, caring for an older relative.We used a technique that allowed us to see what groups emerged from the data to see how contemporary British couples share or divide these different types of work.

We also categorized our couples according to their answers to a series of ideological questions – were they both similarly traditional, egalitarian or middling, or was one partner more egalitarian while the other was more traditional?

Education levels

Then we looked at their levels of education – were both equally educated, was the woman more highly educated or was the man more highly educated?

Very few of the couples shared work equally. In just six per cent of couples, the woman was the main earner while domestic tasks were shared quite evenly. These women were likely to be more highly-educated than their partners. In a further one per cent of couples  the man remained at home and did more than 20 hours’ domestic labour. Even in those cases, around two thirds of the women also did some domestic work.

However, these stay-at-home men were quite likely to be caring for an adult – four out of ten of them provided more than 20 hours’ care per week. Only 30 per cent of female-earner couples had children under 16 living at home, and of those just a quarter of men had the main responsibility for childcare. Men in this group did more housework than in some other groups, but they still did less than their partners.

Almost half (49 per cent) of couples were dual-earner couples in which both members of the couple tended to be employed full-time but these couples were less likely than traditional couples to have children at home.

Traditional divisions of labour

About 30 per cent of couples were traditional in the division of work with men working full-time and women not employed or working short hours. These couples tended to have dependent children at home and women in this couples did large amounts of housework, and in a small minority of women in these couples doing over 20 hours of housework per week.

A small group of couples (two per cent) also emerged in which women worked part-time and men worked very long hours at 60 hours per week or more. These women had more responsibility for care, and more for housework.

The third most common group, at 13 per cent, was a slightly older group of couples in their fifties or early sixties, in which neither were working full-time and there was little or no care responsibility. Women in these couples did relatively high levels of housework, suggesting that these couples may have previously followed a traditional gender division of work.

Those couples who didn’t have shared egalitarian beliefs – either both had traditional views or one did but the other didn’t – were more likely to fall into a more traditional work pattern.

What about education?

Men who were better-educated than their partners were more likely to fall into the traditional working patterns, and men with lower educational qualifications than their partners were more likely to stay at home. And these traditional patterns were even more likely to pertain when the man was more highly-educated and both shared traditional views.

So, education makes a difference, and so does a shared ideology. But it is important to note that very few men, in any category, did longer hours of domestic work than their female partner. So when it comes to housework and caring, gender equality remains rare and gender norms remain strong. 

Our beliefs may form a starting point for shaping our behavior, but that’s only a starting point, and not a solution. The ‘bold’ 50 minutes for 50 percent suggestion in the State of the World’s Fathers report, certainly seems to be grabbing the bull by the horns and, like our research, points clearly to the need for urgent change in terms of who does the daily care work in our homes. That is if we genuinely want a significant shift in power relations between men and women and to bring about gender equality.

Gender divisions of paid and unpaid work in contemporary UK couples is research by Anne McMunn, Lauren Webb, Elizabeth Webb and Amanda Sacker is published in the journal, Work, Employment and Society.

 

Job prospects: does it matter where we live when we are young?

What are the influences on our employment prospects across our working lives? Could where we live when we’re young be important when we’re older – regardless of our social class or level of education? A new study by Emily Murray and colleagues from UCL, King’s College, London and Queen Mary, University of London, finds if we live in an area of high unemployment when we’re young, we’re more likely to leave the labour market at a younger age.

Which factors can help improve our prospects of employment – and of good health – in mid-life? One reason the question is important is that if we can stay healthy longer and work longer, we are less dependent on the state. And the cost of our ageing society is a major issue for governments and individuals across the developed world.

In the UK the state pension age will be raised to age 67 by 2028, but in fact most employees leave work well before they reach that stage. For some that’s a positive thing, but for others it’s forced upon them by unemployment or poor health – and that contributes to social inequality among older people.

Who is at risk? We know older workers in areas of high unemployment are more likely to be on disability benefits. And older people are more likely to leave work for non-health reasons, too – if they are made redundant, they find it harder than their younger colleagues to find another job.

But until now we didn’t know much about how unemployment and other factors experienced by the young might affect their prospects of being in work as they approach the state retirement age.

‘Scarring effect’

There are good reasons to suspect there might be an effect –research has shown that periods of unemployment in young adulthood can have a ‘scarring’ effect – so it stands to reason that early work experiences could set some people on good trajectories and others on less positive ones.

Higher-level job opportunities aren’t evenly distributed, and so we might speculate, for example, that workers in the north – where access to careers in finance, for instance, would be poorer than in the south – would be less likely to start out on one of those ‘good’ pathways.

And education might make a difference, too. We know those growing up in poorer areas are likely to end up with fewer qualifications, and therefore to be disadvantaged in the jobs market. That, too, could continue to affect them throughout their lives.

So health, employment status, occupational class and education might all influence the length of our working lives and lead to unequal retirement outcomes.

Survey of health and development

We used the Medical Research Council National Survey of Health and Development (NSHD), a sample of all births in one week in March 1946 across England, Scotland, and Wales, to test our theories.

This group of people have been questioned 24 times throughout their lives, most recently in 2014 when they were aged 68. We used data from when they were aged four, 26, 53, 60-64 and 68 – a total sample of 2526 people, all of whom had given information on their retirement age or were still in work at age 68 years.

We found there was a correlation between increased unemployment rates in the area a person lived in mid-life and the likelihood of an individual retiring earlier.  However, this relationship was explained by where people lived earlier in life.

For example, cohort members who lived in an area with higher unemployment when they were 26 were more likely to be outside the labour market at age 53. Compared to those who worked full-time, those who were unemployed at aged 53 retired on average 4.7 years earlier.

Similarly, mid-life health problems were more common among those who had lived in areas with high unemployment at age 26, even when taking account of age 26 health status.

We did not find a direct link between educational achievement by age 26 and retirement age. We did, however, find indirect links: for example, those who did not obtain any educational qualifications by age 26 were more likely to live in areas of high unemployment than those who gained degrees. There was no association between area unemployment at age 4 and educational achievement at age 26.

Adulthood is key

Our findings show for the first time that early adulthood is a key life stage at which local labour market conditions can affect our eventual retirement age. We found this happened through two interlinked factors – high area unemployment and worse health status at age 26.

So, there are clear messages for governments: strategies to extend the working lives of future generations will be most effective if they address youth unemployment rather than focusing on older workers in areas with high unemployment.

Policies to extend working life should focus not just on individuals but also on the wider labour market context in which those individuals reside. Maintaining employment and good health in mid-life are key to ensuring that individuals can work longer. And large-scale interventions that create new jobs in areas with high youth unemployment could bring long-term positive consequences for future generations’ extended working lives.

Linking local labour market conditions across the life course to retirement age: Pathways of health, employment status, occupational class and educational achievement, using 60 years of the 1946 British Birth Cohort, is published in Social Science & Medicine.

Emily T. Murray,  Paola Zaninotto, Maria Fleischmann, Nicola Shelton  and Jenny Head are based at the University College London Department of Epidemiology and Public Health.

Mai Stafford and Diana Kuh were based at the Medical Research Council Unit for Lifelong Health and Ageing at University College London when this paper was written. 

Ewan Carr is also based at the Department of Biostatistics and Health Informatics, Institute of Psychiatry, Psychology and Neuroscience, King’s College London

Stephen Stansfield is based at the Queen Mary University of London, Wolfson Institute of Preventive Medicine, Centre for Psychiatry.

Retirement: good or bad for your heart?

Across the globe, more people are spending more time in retirement than ever before. So staying healthy in later life is critical. Yet political debates on ageing tend to ignore a growing body of research on how retirement can affect our health. Baowen Xue and colleagues from the Department of Epidemiology and Public Health at UCL looked at links between retirement and cardio-vascular disease – and found unexplained differences between Europe and the US.

Is retirement good for your heart, or bad for it? The question is an important one because cardio-vascular disease (CVD) is the biggest cause of death globally and costs health services a huge amount of money.

Some studies have shown retired people have a higher risk of being diagnosed with CVD than those who are still working. But until now the evidence has been unclear.

We set out to review evidence from across the world, so that we could help to build a more accurate picture of whether, and how, retirement might affect our cardio-vascular health. As CVD is linked to our lifestyle, diet and other behaviour, there are lots of ways in which changes that take place in retirement might have an effect – both negative or positive.

Longitudinal studies

We looked for longitudinal studies that could help answer our questions, and found 82 which measured risk factors for CVD and 14 which looked at actual incidence of CVD. The second set of 14 papers provided the answer to our first question – does retirement affect our cardio-vascular health?

The answer revealed a major difference between the USA and Europe. Studies conducted in the US showed no significant effect, good or bad, on retirees’ cardio-vascular health. In Europe, meanwhile – with the exception of France – studies consistently showed a link between retirement and an increase in CVD.

Data from the British Regional Heart Study, for instance, showed that healthy men who retired before the age of 60 were more likely than others to die from circulatory disease within five and a half years. Fatal and non-fatal CVD was also more common among retirees in Denmark, Greece, Italy and the Netherlands.

Why might this be? Could there be cultural or lifestyle differences between Europe and the US which might cause this difference? We took a systematic look at the risk factors.

Weight gain

First, we looked at weight gain. If Americans were less likely to put on weight after retirement compared to Europeans, that might help to explain the difference. But when we looked at this, we found that body mass index (BMI) actually increased after retirement in the USA – and also Japan -but did not change in England, Denmark, France, Germany, Switzerland or Korea. While those who do physically demanding jobs are likely to put on weight after they retire, most people aren’t.

Could it be that retired people generally do less exercise – another risk factor – in Europe? The studies suggest that’s not the reason. While many retirees did more physical activities, they also spent more time sitting still – so the effect was a balanced one. For instance, a retiree might play more golf, but also watch more television.

Do retired people perhaps smoke more, we asked? Again, there were contradictory results but 12 out of 14 studies either showed no effect or showed retirement led to people smoking less.

Perhaps retired people in Europe drink more, then? Again, this couldn’t be identified as the reason. Studies in Australia, the UK, Japan and the USA suggested there was no association between retirement and alcohol consumption.

Diet is another possible cause of CVD, but again, there was no clear pattern of between retirement and diet emerged from reviewed studies.

No benefits

So the picture isn’t straightforward, and we don’t have answers as to why retirement might put Europeans at risk but not Americans. What we can say, though, is that none of the studies we looked at found any beneficial effects of retirement on CVD.

Apart from a decrease in smoking, there wasn’t evidence of any general ‘relief’ effect of retirement on people’s cardio-vascular health – so the supposition that working could be bad for our health and therefore retirement better for it doesn’t necessarily hold true.

However, studies that showed retirement brought negative health effects should be interpreted with caution. Many assessed the health effects of retirement by comparing retired people with employed people – and we know people who stay in the labour market are generally healthier than retirees. We do know people who have CVD, diabetes or hypertension are more likely to retire.

What our review has done is to reveal the complex nature of the underlying mechanism through which retirement might impact on the risk factors for CVD. Different people react differently to retirement, depending on their life experiences and the cultural and policy environments in which they live. So there isn’t one global solution to any of this – each country needs to plan its citizens’ retirement according to their individual needs.

The impact of retirement on cardiovascular disease and its risk factors: A systematic review of longitudinal studies, by Baowen Xue, Jenny Head and Anne McMunn, is published by The Gerontologist.

Is working flexibly good for your health?

Flexible working is considered good practice – and in England, most workers have the right to apply to work flexibly after they’ve been in their job six months. But what do we know about the benefits? A new study by Tarani Chandola and colleagues used biological measures to look at differences in stress markers among workers with reduced hours and those without.

In recent years many employees have been able to alter their work patterns to fit in with childcare and other responsibilities. Typically, this can mean working part-time, job-sharing, only working during school term-times or working from home some of the time.

It’s assumed this should help to relieve stress. But until now, we didn’t know whether this was necessarily the case. After all, there could be down-sides – for example working at home can mean a blurring of the boundaries between work and family time, part-time working can be a barrier to promotion and job-sharing can bring its own tensions.

Until recently we had to rely on workers’ own reports of how they felt in order to judge this interplay between work, family life and stress. But now a number of social surveys have begun collecting samples which allow us to measure biological changes which can indicate stress, too.

This is known as ‘allostatic load’ – when we’re repeatedly subjected to stress or trauma, this can lead to chronically heightened levels of stress hormones. And that is associated with all sorts of long-term health problems, such as heart disease, type-2 diabetes and depression.

We were able to use data from participants in the Understanding Society study, which began in 2009 and which follows more than 60,000 adults in 40,000 households. As well as responding to detailed questionnaires, many of them have been visited by nurses who have taken physical measurements and blood samples.

Blood-based markers

As well as blood-based markers such as insulin growth factor 1 and cholesterol, their pulse rate, blood pressure and waist-to height ratio were also measured.

After taking out those who weren’t employed, who didn’t have the nurse visits or for whom some measurements were missing, we had a sample of a little over 6,000 people.

All those people had been asked whether flexible working arrangements were available at their workplace, how many hours they worked and whether they were the primary carer for their children.

We categorised working hours into three groups, with different levels for men and women because they tend to have very different working patterns. So women were grouped into those working less than 24 hours per week, more than 25 hours and more than 37 hours; while men were grouped into those working less than 37 hours, 37-40 hours and more than 40 hours.

Unsurprisingly, we found more women than men had made use of flexible working  arrangements – almost no men in our sample were the main carers for two or more children.

Chronic stress

There were particularly high levels of biological chronic stress markers among women with childcare responsibilities who worked more than 37 hours per week. Those with similar childcare responsibilities but working fewer than 25 hours per week didn’t have any measurable effect on their stress levels.

Both men and women who had access to, and made use of, reduced-hours flexible working had lower levels of biological stress markers than those who didn’t have flexible working.

We found these types of reduced-hours arrangement were more common among those in lower-paid occupations, especially among men, and among older workers of both genders.

Other types of flexible working arrangements, such as working from home, were more common among those from more advantaged social groups. But we didn’t find any association between these types of working and lowered levels of stress.

So, what has our study told us? We’ve learned a good deal about the complex relationships between social and biological factors in our lives. And, crucially for policymakers, we can see that it’s particularly important for women with childcare responsibilities to be able to access shorter working hours when they need to. For employers, this isn’t just a matter of logistics and of ensuring a stable and happy workforce – it’s also a major factor in ensuring that workers live longer and healthier lives.

Are Flexible Work Arrangements Associated with Lower Levels of Chronic Stress-Related Biomarkers? A Study of 6025 Employees in the UK Household Longitudinal Study, is research by Tarani Chandola (University of Manchester and UCL), Cara Booker, Meena Kumari and Michaela Benzeval (University of Essex) and is published in Sociology.

Woman worker

Anti-social working hours: Are they making women depressed?

The rise of globalisation and the 24/7 economy are fuelling demands for people to work long hours and weekends.  But what’s the evidence about how these ways of working link with depression? Gill Weston and colleagues from the International Centre for Lifecourse Studies in Society and Health at UCL and Queen Mary University of London found such working conditions are linked to poorer mental health in women. 

Across the globe, the effects of overwork are becoming apparent.  In eastern Asian countries the risk of death due to overwork has increased.  In the UK, work-related stress accounts for millions of lost working days every year.  

Within the EU, a significant proportion of people have to work unsociable hours – with nearly a quarter working most Saturdays and a third working at least one Sunday a month.  But despite this, there isn’t much clear evidence about the links between work patterns and mental health. 

Some studies have found a connection between unsociable work patterns and depression.  But many of the studies only focused on men, some only looked at specific types of worker or workplaces and few took account of work conditions such as whether workers had any control over how fast they worked.  

To address these gaps, we set out to look for links between long or irregular hours and depression using a large nationally representative sample of working men and women in the UK.  We particularly wanted to look at whether there were differences between men and women because research has shown that work is organised, experienced and rewarded differently for men and women, and because men and women react differently to overwork and time pressure. 

We used data from Understanding Society, which surveys people living in 40,000 households across the UK.  In particular we focused on information about working hours, weekend working and working conditions collected from 11,215 working men and 12,188 working women between 2010 and 2012.  They had completed a questionnaire designed to study levels of psychological distress.

Who works the most? 

We found men tended to work longer hours in paid work than women, and having children affected men’s and women’s work patterns in different ways: while mothers tended to work fewer hours than women without children, fathers tended to work more hours than men without children.  

Two thirds of all men worked weekends, compared with half of all women.  Those who worked all or most weekends were more likely to be in low skilled work and to be less satisfied with their job and their earnings than those who only worked Monday to Friday or some weekends 

Which workers have the most depressive symptoms? 

Women in general are more likely to be depressed than men, and this was no different in our study. 

Independent of their working patterns, we also found that workers with the most depressive symptoms were older, smokers, on lower incomes, in physically demanding jobs, and who were dissatisfied at work. 

Are long and irregular hours linked to depression? 

Taking these findings and other factors into account, when we looked at the mental health effects of work patterns on men and on women, the results were striking: while there was little or no difference in depressive symptoms between men who worked long hours and those who did not, this was not the case for women.   

Those women who worked 55 hours or more per week had a higher risk of depression than women working a standard 35-40 hour week.   

Similarly weekend working showed differences for men and women.  Compared to workers who only worked on weekdays, men who worked weekends also had a greater number of depressive symptoms, but only if they had little control at work or were dissatisfied with work.  Whereas for women,  regardless of their control or satisfaction, working most or all weekends was linked to more depressive symptoms. 

Why might women suffer more than men while working these antisocial hours?  

There might be a number of reasons why women might be more affected than men: 

  • Women who work long hours are in a minority – just four per cent of them in our sample worked 55 hours or more per week. This may place them under additional pressure. 
  • Women working longer hours tend to be in male-dominated occupations, and this may also contribute to stress. 
  • Women working weekends tend to be concentrated in low-paid service sector jobs, which have been linked to higher levels of depression. 
  • Many women face the additional burden of doing a larger share of domestic labour than men, leading to extensive total work hours, added time pressures or overwhelming responsibilities.  

What should be done about these risks? 

Our findings should encourage employers and policy-makers to think about how to reduce the burdens and increase support for women who work long or irregular hours – without restricting their ability to work when they wish to.  More sympathetic working practices could bring benefits both for workers and for employers – of both sexes. 

Long work hours, weekend working and depressive symptoms in men and women: Findings from a UK population-based study by Gill Weston, Afshin Zilanawala, Elizabeth Webb, Livia Carvalho, and Anne McMunn is published in the  Journal of Epidemiology and Community Health, which is published by the BMJ. 

Can ‘nudge’ theory help extend working lives?

Government policy in the UK and other industrialised countries aims to increase the numbers of people staying on in work for longer – but there are significant differences between different groups. Can social and economic factors explain them? Or is there truth in the suggestion that some groups of workers are ‘resistant’ to staying in work, particularly in poorer areas? Nicola Shelton and colleagues from UCL looked at regional differencesin extending working lives and found policymakers may need to rethink their approach.

Despite the government’s stated desire for longer working lives, many workers still stop working before state pension age. The proportion of 60 year-olds in work in England and Wales is 20 per cent lower than the proportion of 50 year-olds, according to 2011 census data. 

And this drop in work participation rates isn’t uniform: Existing research tells us those with lower educational qualifications – particularly women –are more likely than others to leave work early.

So, why might that be?  Some official publicationshave suggested there may be resistance to continuing in work among some groups– perhaps in areas where there are fewer professional or skilled jobs, and where levels of deprivation and unemployment are high.

We wanted to find out more about this:  what regional differences are there in the age at which people leave work? Are there gender differences? Are there particular factors – working conditions, household or individual factors – which can promote extended working lives? And if there are, how do they affect any regional variations?  

There is some previous research on the subject. 

studyusing the ONS Longitudinal Study(ONS‐LS) and the English Longitudinal Study of Ageingfound those in lower-grade jobs, those previously unemployed, those with health problems and those with no dependent children tended to  stay longer in work, along with women from Pakistani or Bangladeshi backgrounds and single women. The study included just two areas, comparing the South, East and Midlands with the North of England and Wales.  

Similar evidence exists from other countries – In FranceNorwayand Great Britain, links have been found between unemployment and deprivation and retirement rates.

Census data

We used census data – a one per cent sample of the total population – to look at what happened to adults who were aged between 40 and 49 in 2001.  This ONS‐LS data covers more than 33,000 women and just under 32,000 men who were therefore aged 50-59 in 2011. 

We found men in the North East were significantly less likely to extend their working lives than others. Those in the South West and South East were 1.6 times more likely to stay on than those in the North East, in the East of England 1.5 times, and in the East Midlands 1.4 times. 

Women in all regions apart from London and Wales were significantly more likely to stay in work than those in the North East, with figures ranging from 1.15 times in the North West and West Midlands to 1.6 times in the South West.

But when we did further analysis, we found that for men at least, other social factors could explain these differences. Put bluntly, the reason men in the North East leave work earlier is because they tend to have fewer qualifications and less favourable employment status – both of which are associated with shortened working lives.

When we did the same analysis for women, we found some additional factors which affected their likelihood of staying on in work. Those in lower-skilled jobs were less likely still to be in work by 2011, along with those working for larger employers. Those who worked away from home were also more likely to have left, along with those who worked long hours. 

And again, – when we considered these factors along with prior employment, health, social status and caring responsibilities, and only those in the South West were significantly more likely to stay on than those in the North East. 

Working conditions

So, what can governments do? Given a good work environment, choosing to remain in work may have positive benefits such as maintaining good health and functioning and providing a sense of purpose- so working conditions are important.

The biggest single factor in determining whether workers stay on for longer is prior employment – and that is not likely to be changed by behavioural approaches such as the ‘nudge’ theory of behavioural economics which is popular with policy makers.

Policies that do not address issues such as low levels of education and high levels of unskilled employment can only be partially successful in enabling people to work for longer. Indeed, some groups who may have the most financial need to remain in work are most likely to leave earlier. This is particularly an issue for women.

Policies that increase skills and education in later life, rather than simply targeting those ‘receptive’ to extended working, will be more likely to make a difference.

Gender differences and individual, household, and workplace characteristics: Regional geographies of extended working lives, is research by Nicola Shelton, Jenny Head, Ewan Carr and Paola Zaninotto, and is published in Population Space and Place. 

Early retirement – can welfare systems help ease the transition?

The post-war baby boomgeneration in developed countries is reaching retirement age and this is placing strain on welfare systems. Sol Richardson and colleagues from the ESRC International Centre for Lifecourse Studies at UCL find the type of welfare system under which we live can affect our prospects of having a happy and fulfilled retirement.

We know stopping work can lead to changes in our sense of personal wellbeing both positive and negative. And we know this can be influenced by a range of factors, such as whether an individual has left work at the usual age or has stopped early.

There are other factors which can make a difference to how we fare after retirement, too: If we were dismissed, retired through illness or through unemployment, for instance, the change is more likely to hit us hard.  

But how much difference do the different types of welfare system which exist in different countries make to those who leave work early? Until now we havent had much clear evidence on this point.

Data

We looked at a sample of people from 16 countries, using data from the Study of Health, Ageing and Retirement in Europe (SHARE) between 2004 and 2013, and from the English Longitudinal Study of Ageing (ELSA) between 2002 and 2013 these are studies which revisit their participants over time.

Our subjects were aged 50 years and over and had been visited before and after they left work.

We looked at a total sample of 8037 respondents who had left work between 2002 and 2013 and for whom we had information not only on work history but also on personal wellbeing.

We categorised how people left work according to the type of benefit they received afterwards: Were they receiving disability benefit, unemployment benefit, sickness benefit, social assistance, early retirement pension, old age pension or none of those?

Retirement age is different in different countries it can depend on gender or on the number of years worked, and its been rising in many countries. So we defined retirement as the earliest age at which an individual can draw a full pension if he or she has been working since the age of 20.

And we looked at the wellbeing of our subjects, using a validated scale called CASP-12 (control, autonomy, self-realization and pleasure.)

And then we compared these findings according to the type of welfare regime the participants had in their home countries again using an internationally-recognised scheme that relates to how social benefits are granted and organized.

Negative effects

We found that those who left the labour market because of unemployment or disability and who left outside of the typical time-frame tended to experience negative effects on their personal wellbeing.

How much difference did country of residence make? We found this was significant, but that only a small proportion of the difference was down to the country itself. Almost two thirds of the wellbeing gaps we found between individuals in different countries could be accounted for, we found, by the type of welfare system they had.

Those living in Scandinavian social democratic welfare systems experienced the most positive transitions but this effect is unlikely to be down to expenditure alone. Other factors could be important for instance, employment rules guiding the ways people left work. Different finance mechanisms, the extent of benefit coverage and the eligibility regime for those benefits could also have an effect.  

When we looked at the different types of welfare system we found people in systems which could be described as Bismarckian,such as France or Germany, or  Scandinavian,such as Sweden or Denmark, did better than those in systems which could be categorised as Mediterranean,such as Italy or Greece.

As a generalisation, Scandinavian systems can be described as Social Democratic. They spend the most, they have high levels of cash benefits and a strong emphasis on services.

Bismarckian countries emphasise earnings-related cash benefits like pensions and they provide reasonable services, but not at the level of Social Democratic countries.

In Mediterranean countries, the pensions system is fragmented and services are rudimentary. People living in Mediterranean systems are more likely to rely on family and the voluntary sector for support.

Policy implications

What lessons should policy-makers draw from our study? We found that higher expenditure per head, particularly expenditure on non-healthcare services such as home help, did help our participants to feel better after they left paid work.

And our results have important implications for welfare policy: They underscore the importance of welfare services as greater numbers of workers approach retirement age and leave the labour market.

Country-level welfare-state measures and change in wellbeing following work exit in early old age: evidence from 16 European countries, by Sol Richardson, Ewan Carr, Gopalakrishnan Netuveli and Amanda Sacker, is published in the International Journal of Epidemiology, 2018, 113.

Working with a long-term illness – does employment status make a difference?

Across the developed world, a growing share of the population suffers from chronic disease such as diabetes, arthritis or heart problems – in the EU, around 45 per cent of those aged 55-64 had such a disease in 2015. And that affects their ability to work: just half of those with chronic illnesses are employed, compared to three quarters of those without. But how do the self-employed cope with such conditions, when compared with those in employment? Maria Fleischmann and colleagues from the RenEWL project find these differences in work status can make a major difference.

We know that when people become chronically ill, changes in their working conditions can help them to continue working. And we also know that good working conditions – being able to control how you use your time and how you do your job, whether you make the decisions and whether you feel valued, for instance – can help all of us to stay in paid work for longer.

And of course, if you become ill you’re more likely to feel the need to take time off or maybe to give up work altogether. We wanted to compare how the employed and the self-employed adjust their working conditions when facing a diagnosis of chronic disease.

Existing research tells us that many older people work for employers, and have to ask for their approval when it comes to making adjustments to their working conditions. The self-employed, meanwhile, are much more able to make their own decisions and tend to feel they have more control over their working lives.

We looked at people’s ability to control their work: physical demands, working hours, psychological demands such as how fast they had to work, and social aspects such as whether they felt valued.

And we had a great source of data for this – the English Longitudinal Study of Ageing (ELSA), which has followed a total of almost 9,000 over-50s since 2002.

From that group we were able to find and study 1389 participants who reported no chronic diseases when they were interviewed in 2004-5 – the second wave of interviews – and who were in work.  We were then able to look at what happened to them before the seventh wave of interviews, in 2013-14.

A little over 40 per cent of our sample were in managerial or professional occupations, a quarter in intermediate occupations, and almost 30 per cent in routine and manual occupations.

After diagnosis

At each interview, respondents were asked whether they had been diagnosed by a doctor with lung disease, asthma, arthritis, cancer, high blood pressure, diabetes or high blood sugar, stroke or heart problems.

During the study period 510 of the 1389 sample members were newly diagnosed with one of those conditions. We were able to look at how they fared at work for four years after that, and we found some striking differences between how the employed and the self-employed seemed to have been treated.

The physical demands of our participants’ jobs were pretty similar before their diagnosis, for instance. But afterwards significant numbers of those who were employed said that those demands had actually increased when they were diagnosed. The self-employed, meanwhile, told us the opposite had happened to them – they reported significantly lower physical demands at work immediately after diagnosis. This effect continued for some time, though it grew less pronounced.

How could that be? We think maybe the increase in physical demands among the employed could be due to perception – similar demands might be perceived as more strenuous by the chronically ill. The self-employed, meanwhile, have more freedom to adjust those demands when they feel they need to.

The self-employed reported that their working hours dropped by an average of 2.8 hours per week on diagnosis, while those who were employed did not see a change. This effect was not statistically significant, though.

Employees found that their level of autonomy at work also dropped marginally, while for the self-employed there was no significant change. We did not find any major changes in psychological or social conditions in either group.

So, what did we learn? Essentially, that improvements in working conditions after diagnosis of chronic illness were restricted to the self-employed. So employers may need to ask themselves some hard questions – do they want to hold on to workers who become unwell? If they do, then they should consider the levels of flexibility they offer, and they should think about making adjustments for those workers if they don’t want to lose them.

In an ageing society, older people are expected to work ever longer and therefore to remain at work even when they begin to suffer from health problems. Our findings should also encourage policy-makers and governments to think about how chronically ill older adults are treated at work.

Changes in autonomy, job demands and working hours after diagnosis of chronic disease: a comparison of employed and self-employed older persons using the English Longitudinal Study of Ageing (ELSA)by Maria Fleischmann, Ewan Carr, Baowen Xue, Paola Zaninotto, Stephen A Stansfeld, Mai Stafford and Jenny Head, is published in the Journal of Epidemiology and Community Health.