Premature mortality: how education, occupation and deprivation shape inequalities

The premature mortality rate – the number of people dying before age 75 each year – is an important public health outcome measure that highlights disparities in preventable ill health. Our 2025 analysis on geographic inequalities in premature mortality, in partnership with the Office for National Statistics (ONS), found premature mortality rates varied substantially across England and Wales (after accounting for differences in age and sex). People were almost twice as likely to die before age 75 in areas with the highest rates compared with the lowest.

We found socioeconomic factors (level of education, type of occupation and area-level deprivation), ethnicity and being born outside the UK accounted for almost two-thirds (61%) of the variation in premature mortality rates across England and Wales. 

We partnered with ONS again to learn more about how each of the socioeconomic factors contributes to inequalities in premature mortality. We explored how premature mortality rates vary across demographic and socioeconomic groups, and how much of this variation persists after accounting for other contributing factors. To do so, we used a unique linkage of Census data – containing personal information such as ethnicity, occupation and education level – and death records. 

Our previous analysis found that ethnicity and being born outside the UK accounted for 7% of the variation in premature mortality rates. We have partnered with the Race Equality Foundation to publish an accompanying report, which looks in more detail at how these factors relate to premature mortality across ethnic groups.

There are large inequalities in premature mortality rates between different socioeconomic groups (Figure 1). There is a clear trend towards higher premature mortality rates as levels of education decrease, occupation type shifts and deprivation increases, with substantially higher rates among people with no qualifications, those who have never worked or are long-term unemployed, and those who live in the most deprived areas.

• Females with no qualifications have a premature mortality rate 151% higher than females with a degree or higher (564 premature deaths per 100,000 people, compared with 225 per 100,000). A similar extent of inequality is seen in males with no qualifications, at 137% higher rate than those with a degree or higher (818 vs 345). 
• Among those who are long-term unemployed or who have never worked, rates are 250% higher for males (1,069 vs 306) and 231% higher for females (677 vs 205), compared with those who are in higher managerial, administrative and professional occupations. 
• For those living in the most deprived areas, males have a rate 131% higher (815 vs 353) and females 128% higher (556 vs 243) than those living in the least deprived areas.

While achieving a higher level of education or working in managerial and professional occupations is associated with a lower risk of dying prematurely, this protective effect is substantially reduced in more deprived areas. As shown in Figure 2, rates increase as the extent of deprivation increases, with a larger difference between the most deprived 20% of areas across England (quintile 1) and the second most deprived 20% of areas (quintile 2). 

Males with a degree or higher who live in the most deprived areas have a premature mortality rate 106% higher than males with a degree or higher who live in the least deprived areas (556 premature deaths per 100,000 compared with 269). Similarly, males working in higher managerial, administrative and professional occupations in the most deprived areas have a rate 93% higher (492 vs 255) than males in similar roles in the least deprived areas. 

Inequalities in premature mortality rates across education levels, occupation types and extent of area-level deprivation partly reflect other demographic and socioeconomic factors associated with differences in health outcomes. 

To better understand the role that each factor plays, we compared the inequalities before and after accounting for other factors. For example, when looking at inequalities across education levels, we have accounted for differences in occupation type, area-level deprivation, age, ethnicity and being born outside the UK. We also accounted for the upper tier local authority in which people lived to capture differences in local-level factors. 

Education

Even after accounting for differences in these other factors, we found that more than two-fifths (females 46%, males 44%) of the variation in premature mortality rates remains between those with no qualifications and those with a degree or higher (Figure 3). After accounting for these factors, females with no qualifications continue to see a premature mortality rate 68% higher than females with degree or higher. This is down from 151% higher before accounting for these other factors. For males, the rate is 59% higher, down from 137% higher.

These persistent inequalities suggest that education level plays an important part in shaping health. Higher qualifications often lead to higher-quality jobs and larger incomes and lifetime earnings – these enable a better standard of living. A good education can also have social and psychological benefits – building confidence, developing problem-solving skills and growing social networks that support mental and emotional wellbeing. Finally, education can improve people’s understanding and use of health-related information. Being able to access and use information and services to maintain good health and wellbeing can help people to choose healthier options. 

Inequalities across education levels might also reflect broader disadvantage experienced throughout a person’s life. Childhood hardship, such as experiencing child poverty, impacts both educational attainment and health. For example, children who were eligible for free school meals (often used to identify lower-income households) are 64% less likely to meet the expected educational attainment by their GCSE exams (aged 15–16 years) than those who didn’t have free school meals. 

Poorer health or disability may explain some of the differences in premature mortality rates between those with no qualifications and those with a degree or higher. Disabled people have a mortality rate more than twice as high as non-disabled people; they also are twice as likely to have no qualifications and less likely to have a degree.

Occupation type

We found that over half (females 57%, males 55%) of the difference in premature mortality rates between those who have never worked or are long-term unemployed and those employed in higher managerial, administrative and professional occupations remains after accounting for differences in demographic and socioeconomic factors (Figure 4). After accounting for these factors, the premature mortality rate for males who have never worked or are long-term unemployed is 137% higher than males employed in higher managerial, administrative and professional occupations. This is down from 250% higher before accounting for demographic and socioeconomic differences. For females, the premature mortality rate is 131% higher, down from 231% higher.

Persistent inequality in premature mortality rates between occupation groups points to the important part that work plays as a building block of health. Being long-term unemployed or economically inactive can negatively affect physical and mental health; employment can provide financial security and a sense of purpose and social connections, which all support good health. While being employed is good for health, job quality – which includes things such as adequate pay, secure work and reliable working patterns – is also important. Lower quality jobs are more common in some types of occupations, and poorer working conditions can be detrimental to health.

Some of the inequality in premature mortality rates that remains after accounting for other factors might also reflect the interplay between work and worse underlying health or disability. Poor health can limit the type or amount of work people can do. People with long-term health conditions are more likely to experience other barriers to employment, in addition to health-related barriers. Alongside their higher mortality rate, disabled people are also more likely to work in routine occupations or have never worked or are long-term unemployed than non-disabled people.

Deprivation

Three-fifths (61%) of the inequality in premature mortality rates between people who live in the most and least deprived areas of England remains even after accounting for differences in demographic and socioeconomic factors. Females who live in the most deprived areas have a premature mortality rate 77% higher than their peers in the least deprived areas, down from 128% higher before accounting for other factors. This is similar for males, at 79% higher, down from 131% higher. This tells us that the socioeconomic characteristics of the areas in which people live in play a major part in shaping their chances of dying prematurely.

Area-level deprivation is a composite measure which covers several factors across the building blocks of health. People living in areas of higher deprivation often have poorer employment prospects and lower earning potential. They also often experience environmental disadvantages, characterised by higher levels of air pollution or lower-quality housing. 

Deprivation is a major contributor to premature mortality, and can reduce the protective effect of some individual factors.

There are differences between regions in premature mortality rates for people who otherwise have the same education level, occupation type or live in areas with the same level of deprivation. For example, the rate for females who have never worked or are long-term unemployed is 796 deaths per 100,000 people in the North West – 72% higher than London (462 deaths per 100,000).

There are also differences in the extent of within-region inequalities. Inequalities are highest in the North East, North West, and Yorkshire and The Humber, and lowest in London. For example, females who live in the North West and have never worked or are long-term unemployed have a premature mortality rate 247% higher than females who work in higher managerial, administrative or professional occupations. In contrast to most other regions, in London this inequality divide is much lower, at 141% higher. Factors that may contribute to lower inequality in London include population composition, selective migration and the city’s social and economic context, but we need to learn more about how much these factors contribute.

As seen for England as a whole, inequalities by education level, occupation type and area-level deprivation within each region persist after accounting for other factors. Differences in within-region inequalities also persist after accounting for other factors. This suggests that there are other, region-specific factors – not accounted in this analysis – that reduce or increase within-region inequalities across England.

There are large inequalities in premature mortality rates across education levels, occupation types and extent of area-level deprivation in England. Rates are higher for people with fewer qualifications, who are either not in employment or work in routine occupations, and who live in more deprived areas. 

Although achieving a higher level of education or working in certain occupations is associated with a lower risk of dying prematurely, living in a more deprived area can limit the protective health benefits of education and good work. Females with a degree or higher who live in the most deprived areas have a premature mortality rate 90% higher than females with a degree living in the least deprived areas.

This analysis shows that these large inequalities in premature mortality rates across education levels, occupation types and extent of area-level deprivation persist even when accounting for differences in other demographic and socioeconomic factors. 

Previous analysis by the Health Foundation has shown that a range of socioeconomic factors are major contributors to geographic inequalities in premature mortality across the country. This new analysis confirms that a range of socioeconomic factors are strongly linked with the risk of dying early, but also that each factor interacts with others. The significant part that area-level deprivation plays in shaping health – even after accounting for differences in characteristics between people – underlines the need for urgent action across the building blocks of health to reduce inequalities between places. Action is also needed on the individual factors that shape health, including removing barriers to achieving a good education and to finding and maintaining good quality and stable work. 

As this research only looks at a specific point in time, some of the remaining differences in premature mortality rates between socioeconomic groups are likely due to differences in exposure to factors detrimental to health across the life course that are not captured here. These include relative opportunities in childhood that affect people’s chances to achieve higher education and a high-quality job, and which impact their health. 

Some factors not considered in this analysis, such as risk factors for ill health (smoking, obesity, alcohol and drug use) and strength of social connections, play a part in the remaining differences in premature mortality rates. There is more to learn about the extent to which these factors further account for differences in premature mortality rates between socioeconomic groups.