Mental health trends among working-age people
About 15 mins
Key points
- The mental health of the working-age population appears to be getting worse. Over 10% of working-age people report signs of poor mental health across a range of data sources, including self-reported survey measures, screening tools and clinical diagnoses.
- While this rise is seen across people of all ages, the greatest increase has been for people of younger ages (16–34 years). Multiple data sources indicate that rates have at least doubled over the past decade.
- People with lower-level qualifications, those outside the labour market and people living in Scotland and northern England are more likely to report mental health conditions. Regions in the north of England also tend to have the greatest proportion of the population receiving benefit payments due to mental health.
- Behind these trends, the data also show the varying nature and severity of mental health conditions. While 60% of mental health conditions are work-limiting, the greatest rise over the past two decades has been in non-work-limiting mental health conditions, which have risen 12-fold.
- People with work-limiting mental health conditions are half as likely to be in work compared with people with no health conditions, although higher qualifications can reduce this gap. While people with non-work-limiting mental health conditions have higher employment rates that match those of people with no health conditions, they are more likely to have lower pay once in work.
- These trends likely reflect a combination of rising prevalence, higher reporting and evolving definitions. Our blogs accompanying this analysis suggest the rise in mental health challenges among young people reflect long-term changes in young people’s development and call for a broader understanding of people's capacity for work.
- Better understanding the drivers of poor mental health is key so that government and, where relevant, employers, can:
- design better practices to support the significant number of people in work reporting poor mental health
- act early to help ensure people with mental health challenges remain in work
- redesign health, work and welfare systems to embed mental health best practice as a default, not as an exception, to reflect that a higher number of working-age people have poor mental health.
- These issues, and potential solutions, are being considered by the Commission for Healthier Working Lives, supported by the Health Foundation. The Commission will publish recommendations for long-term action on workforce health in its final report in spring 2025.
‘Poor mental health’ includes a wide range of conditions with different levels of severity. Different types of data can be used to understand these trends, including:
- Self-reported conditions: surveys such as the Labour Force Survey and Health Survey for England rely on individuals to identify certain long-term health conditions they may be experiencing. The Labour Force Survey has two condition categories that are used to capture poor mental health – ‘depression, bad nerves and anxiety’ and ‘mental illness, phobias and panics’. These conditions can be further classified as ‘work-limiting’ or ‘non-work-limiting’, where a condition is considered work-limiting if it limits either the type or amount of work an individual can do. In the Health Survey for England, individuals are able to select ‘mental disorders’.
- Psychometric tools: there are a number of screening tools that can be used to identify common psychiatric conditions. The Understanding Society survey uses the General Health Questionnaire (GHQ-12) with 12 questions such as, ‘Have you recently been able to concentrate on whatever you are doing?’ and ‘Have you recently been able to enjoy your normal day-to-day activities?’ Individuals are required to answer on a four-point scale ranging from ‘Not at all’ (equal to a score of 0) to ‘Much more than usual’ (equal to a score of 3). The scores to the 12 questions can be combined to give an overall scale ranging from 0 to 36, where a higher score indicates worse mental health. In our analysis, we use a score of 15 or more to indicate ‘some psychological distress’ and a score of 20 or more to indicate ‘severe psychological distress’.
- Health records: clinically diagnosed mental health conditions can be captured using data derived from the Clinical Practice Research Datalink and Hospital Episode Statistics. These are based on attendances at GP and hospital settings and in this analysis we use a category that captures people who are diagnosed with anxiety or depression.
- Benefits claims: people with mental health conditions may be entitled to a number of welfare benefits such as Universal Credit, Employment and Support Allowance (ESA) and Personal Independence Payment (PIP). PIP onflow data are the only benefits data that can be broken down by an individual’s primary health condition, including a range of psychiatric disorders such as anxiety disorders, mood disorders and eating disorders. In our analysis, we focus on all psychiatric disorders.
There are concerns about the quality of recent Labour Force Survey data due to issues with sampling and response rates. However, it remains the most comprehensive source linking health and labour market outcomes. In this analysis, we focus on changes over the past decade or more and the composition of the population when using this data. Notably, the worsening trends observed in the Labour Force Survey, as well as the broader impacts and characteristics of the population with mental health conditions, are reflected in other data sources.
Figure 1 shows that since 2010 the proportion of the working-age population indicating mental health conditions has risen to at least 10% for all age groups across four key data sources, apart from for people aged 45–64 years in the Health Survey for England. This rise has been particularly stark for younger ages, with all data sources indicating at least a doubling in proportion of those aged 16–34 years with poor mental health, reflecting 12% to 14% of the age group, depending on the data source.
Figure 1
There is mixed evidence on the severity and types of conditions
Although data consistently show an overall rise in mental health conditions, evidence on trends in the types and severity of conditions is mixed but overall indicates growth in different levels of severity.
Labour Force Survey data shows that around 60% of primary mental health conditions (those that people report as their main condition) were work-limiting by 2019. This has remained broadly similar since. However, Figure 2 shows that the largest proportional growth has been in non-work-limiting mental health conditions. Since 2003, this group has grown 12-fold to include over 1.2 million people, while the number of people with work-limiting mental health conditions has tripled. The difference in magnitude of these trends could point towards a broader range of conditions being captured in the data that might not have been captured previously.
Figure 2
Currently, 64% of those aged 16–64 years with mental health conditions report experiencing depression, bad nerves or anxiety; 12% experience mental illness, phobias or panics; and 24% experience a combination of both. This distribution has not changed significantly over the past decade.
More than 70% of working-age people with a mental health condition also report at least one other health condition. However, this proportion has fallen slightly over the past decade, from 78% in 2013. At the same time, the proportion of individuals reporting a mental health condition as their primary health problem has risen from 51% in 2012 to 57% in 2023.
Psychometric scores, such as GHQ-12, provide further insight by assessing poor mental health on a continuous scale, with higher scores indicating greater psychological distress. Figure 3 shows the distribution of GHQ-12 scores in 2012 and 2022, with a noticeable shift towards higher scores over this period, reflecting the rising trend in mental health conditions. Overall, fewer people are now reporting scores in the 0–9 range, while more are reporting scores between 10–36. Notably, the proportion of people reporting ‘severe psychological distress’ at 13% (score higher than 20) is now slightly higher than those with ‘some psychological distress’ at 12% (score between 15 and 20) – the opposite was true a decade ago. The share of people with scores in the 10–14 range, just below these categories, has also increased over time.
Taken together, these data points show a clear rise in prevalence of mental health conditions. Severity appears to be increasing for some groups, as indicated by higher psychological distress scores. However, the growth in non-work-limiting conditions and a slight decline in co-occurring conditions suggest a more complex picture. The rest of this analysis aims to explore what these trends mean for people’s day-to-day lives, including how these trends vary for different groups and the implications for labour market outcomes.
Figure 3
Figure 4
There are also marked geographic disparities. Figure 5 shows that working-age people in Scotland and northern England are more likely to report mental health conditions. In the North East, for example, 19% report having mental health conditions, compared with 10% of people in London. The same pattern was true in 2013, although London and the West Midlands have seen the greatest rise in mental health conditions over the past decade. These patterns are also reflected in the proportion of people reporting work-limiting mental health conditions. When looking at non-work-limiting conditions there is a little more variation, however, Scotland does have the highest proportion of people reporting non-work-limiting mental health conditions.
Figure 5
Regions in the north of England also tend to have the greatest proportion of the population receiving PIP benefit payments due to mental health (captured under all psychiatric disorders) – in the North East, the proportion of the working-age population receiving PIP benefits for mental health reasons is double that in London. Meanwhile, NHS data indicate that people living in the most deprived areas of England are more than twice as likely to be in contact with mental health services as those in the least deprived areas.
People outside the labour market (economically inactive) are the most likely to report any mental health condition – three times as likely as those in work (28% compared with 9%). While the same is true for work-limiting mental health conditions, people who are unemployed are the most likely to report non-work-limiting mental health conditions. Recent research has further highlighted the links between employment and mental health, with job losses or gains being found to impact mental health beyond any direct impacts due to changes in income.
Among employed people, there is also considerable variation across sectors: over 10% of workers in sectors like public administration, education and health, and distribution, hotels and restaurants, report mental health conditions, compared with just 5% in the construction sector. The same broad patterns are seen when looking at work-limiting and non-work-limiting primary mental health conditions separately. These patterns have remained consistent over the past decade and likely reflect a combination of workforce composition, workplace pressures and industry culture. Recent research by the Institute for Employment Studies highlights how job quality impacts health with, for example, burnout being a particular challenge in health and education.
Figure 6
Qualification levels not only influence the likelihood of experiencing mental health conditions but also have a strong impact on employment outcomes. The trend in employment rates for people with non-working limiting mental health conditions is similar across all qualification types, reflecting the overall pattern shown in Figure 6. Therefore, we focus on the difference in employment outcomes by qualification for people with a work-limiting mental health condition. Figure 7 shows that higher qualification levels appear to mitigate the negative employment effects of mental health conditions, while those with lower qualifications face significantly worse employment outcomes.
Over the past two decades, improvements in employment outcomes for people with work-limiting mental health conditions have been most significant among those with higher qualifications, such as degrees or A-levels. Among people with degree-level qualifications, the employment rate gap between those with work-limiting mental health conditions and no health conditions has narrowed by two-thirds, from 53 percentage points in 2003 to 18 percentage points in 2023. In contrast, progress has been far more limited for people with no qualifications, where the equivalent gap has reduced by less than one-fifth.
Of people out of work (unemployed or economically inactive) with a primary mental health condition, 22% have no formal qualifications (compared with 11% of people with no health conditon), while 67% hold qualifications below degree level (64% with no health condition).
Figure 7
Earnings
Mental health conditions influence not only whether people participate in the workforce but their experiences when in employment. People with mental health conditions tend to earn less, with average hourly pay at just 79% that of workers with no health conditions. The pay gap is widest for those with work-limiting mental health conditions, but even individuals with non-work-limiting mental health conditions earn only 82% of the hourly pay of their peers without health conditions. These disparities have been consistent over the past decade and Figure 8 shows that this gap exists across all qualification levels. The pay gap is largest for people with GCSE qualifications or equivalent, with people with work-limiting mental health conditions earning just 73% that of those with no mental health conditions.
Figure 8
Wellbeing at work
Data from Understanding Society also show that people with mental health conditions report consistently worse work outcomes across various other measures, such as job wellbeing, satisfaction, autonomy and security. For example, just 60% of people with severe psychological distress report being satisfied in their job, compared with almost 90% of people with no psychological distress.