Health Expectancies (HEs) divide predicted lifespan into time spent in given states of health. This adds a quality of life dimension to estimates of life expectancy (LE). The Office for National Statistics (ONS) routinely publishes two types of health expectancies. The first is Healthy Life Expectancy (HLE), which estimates lifetime spent in ‘Very good’ or ‘Good’ health based on how individuals perceive their health. The second is Disability-Free Life Expectancy (DFLE), which estimates lifetime free from a limiting persistent illness or disability. This is based upon a self-rated functional assessment of health.
HEs are used as high-level outcomes to contrast the health status of different populations at specific points in time. Changes in population health can be monitored over time, giving context to the impacts of policy changes and interventions, both nationally and locally. HEs are used increasingly in government and the private sector for developing policy and business planning to assess health and social care need, and gauge population level health improvement. They are also used for pension provision planning and state pension age reviews.
This bulletin is the second in the series of estimates of HLE at birth, across English regions and Upper Tier Local Authorities (UTLAs), for males and females. LE figures are presented alongside the HLE figures; these were previously published in the ONS sub-national LE release (ONS, 2013c). The next update of this time series will provide figures for the period 2011-13.
As life expectancy continues to increase in the UK, it is important to measure what proportion of these additional years of life are being spent in favourable states of health or in poor health and dependency. HEs help us to address this question by adding a dimension of quality of life to estimates of LE. They are estimates of the average number of years a person would live in a given health state if he/she experienced the specified population’s particular age-specific mortality and health status for that time period throughout the rest of his/her life.
The figures represent a snapshot of the mortality and health status of the entire population of a specified area in each time period. They are not, therefore, the number of years that a person will actually expect to live in the area in a given health state. This is because both mortality and health rates, and the exposure risks and treatment effects affecting them, are open to change in the future, and because of population movement into and out of the area.
Health expectancies are robust indicators of health related well-being and functional status. They are in part estimated through subjective self reports of general health, which can be influenced by an individual’s expectations. These are known to vary across socio-demographic factors such as age, sex, socio-economic position (749 Kb Pdf) (ONS, 2010) and area deprivation (ONS, 2013b).
Self-reported general health and limiting persistent illness are linked (ONS, 2012a, Manor et al., 2001), having some predictive value in subsequent health care need and usage, and risk of death. Research evidence has shown people with poor self-rated health (both general health and limiting persistent illness) die sooner than those who report their health more positively (Mossey and Shapiro, 1982; Idler and Benyamini, 1997; Miilunpalo et al, 1997; DeSalvo et al, 2006; Bopp et al, 2012; Ng et al, 2012).
In terms of morbidity the evidence is more limited. However, studies have shown that self-rated health, measured in terms of general health and limiting illness, has some predictive value in the subsequent use of health and social care services. This is shown in increased physician visits (Miilunpalo et al., 1997), hospital admission and nursing home placement (Weinberger et al., 1986). Studies have also shown that self-rated health correlates well with retirement due to disability or poor health (Pietilainen et al., 2011; Dwyer and Mitchell, 1999) and poor health outcomes (Lee, 2000).
Survey measurements of general health and limiting persistent illness are used globally to identify health inequality between administrative areas, inform unmet care and health service needs and to target and monitor the allocation of health care resources amongst population groups (Marmot, 2010). International organisations and networks such as the World Health Organisation (WHO, 2011), Eurostat (Eurostat, 2013) and the Reves Network on Health Expectancy (Reves) use this information to compare morbidity across countries and to monitor trends over time.
This bulletin focuses solely on HLE. These estimates are, in part, subjective and based upon the following survey question.
‘How is your health in general; would you say it was…’ – Very good, Good, Fair, Bad or Very bad?
The responses to this question are dichotomised to define ‘Good’ general health and ‘Not Good’ general health: ‘Good’ general health is the combination of the very good and good responses, and ‘Not Good’ general health is the combination of fair, bad and very bad.
HLE is included in both of the two overarching indicators for the Public Health Outcomes Framework (PHOF). The first indicator is increased HLE, taking account of the quality as well as the length of life. The second is to reduce differences in LE and HLE between communities, through greater improvements in more disadvantaged communities. The vision for the indicators is “to improve and protect the nation’s health and wellbeing, and improve the health of the poorest fastest” (page 9, DH, 2012).
The estimates of HLE at birth for England and for UTLAs are calculated using self-reports of general health status collected in the Annual Population Survey (APS). They rely on a method developed by ONS to derive general health state prevalence for persons under 16 years of age (ONS, 2013a) (see methods section). The England estimate in this bulletin is not comparable with the national health expectancy for the UK and Constituent Countries time series (ONS, 2012a). This is because the latter output uses a different data source to calculate general health prevalence, namely the General Household Survey (GHS), and the former imputes proxy responses of general health for those aged under 16. For UTLAs wanting to benchmark themselves against a national estimate of HLE, they should use the England estimate in this bulletin and future updates of this series. ONS will cross-validate these sub-national estimates using 2011 Census data, with findings due to be published in late 2014.
Changes in health expectancies over time are assessed by comparing non-overlapping time periods. Therefore estimates for 2010-12 should not be compared with estimates for 2009-11, as they will contain some of the same survey respondents.
Results are presented with 95% confidence intervals, to aid interpretation. Confidence intervals in this bulletin indicate the uncertainty surrounding LE and HLE estimates and enable more meaningful comparisons between authorities. When comparing the estimates of two areas, non-overlapping confidence intervals are indicative of statistical significance but to confirm this, a test of significance should be carried out. When the statistical significance is noted in the text, it is based on a statistical test of the difference (Jagger et al, 2007). All differences noted in the text have been calculated to more than one decimal place.
Quality information about ONS health expectancies (185.7 Kb Pdf) is available on the ONS website.
Healthy life expectancy (HLE) at birth in England was 63.4 years for males and 64.1 years for females. Across regions, a clear North-South divide was observed with regions in the South East, South West and East of England all having a significantly higher HLE than the England average. The West Midlands, North West, North East, and Yorkshire and The Humber all had significantly lower HLE than the England estimate (Table 1).
For both males and females, HLE was highest in the South East region (65.8 and 67.1 years respectively) and lowest in the North East region (59.5 and 60.1 years respectively). Therefore the inequality in HLE between regions was 6.4 years for males and 7.0 years for females. The inequality in LE between the North East and South East was much smaller than the inequality in HLE, at 2.5 years for males and 2.2 years for females.
HLE for males in the East Midlands, West Midlands, London, Yorkshire and the Humber, North East and North West was significantly below the male state pension age of 65. If we assess females against the same state pension age of 65, where it will be by 2018, the same is true.
The South East, South West and the East of England had a significantly higher HLE, compared with all other English regions for both males and females (Table 1). The North East was unique in that it had significantly lower HLE than all other regions for both males and females, including the North West and Yorkshire and The Humber. This suggests the health profile of the population of the North East remains distinct from other parts of England.
The inequality between the North and South can also be seen when looking at the proportion of LE spent in a favourable health state. Males in the South East are expected to live 82.0% of their life in ‘Good’ general health compared with 76.4% for men in the North East region. The same inequality was present among females, where the difference between the two regions was 6.4 percentage points: 80.0% of life expectancy spent in ‘Good’ general health in the South East, compared with 73.6% in the North East. For each gender, those living in the southern regions not only enjoy longer life expectancies but also greater proportions of their longer lives in a favourable health state compared with their counterparts living in the North.
|LE||HLE||Lower 95% confidence interval||Upper 95% confidence interval||Proportion of life spent in 'Good' health(%)|
|Yorkshire and The Humber||78.3||61.2||60.7||61.7||78.1|
|Yorkshire and The Humber||82.2||62.0||61.5||62.6||75.5|
Regions are presented by gender sorted by HLE.
For males, the UTLA with the highest HLE was Richmond upon Thames in the London region, at 70.0 years. For females, Wokingham in the South East had the highest HLE, at 71.0 years. The lowest HLE for males was in Tower Hamlets at 52.5 years, and for females in Manchester at 55.5 years. The absolute inequality in HLE between local authorities using the range is notably larger than it is for LE. It is also larger for males (17.5 years) than for females (15.4 years).
For female HLE at birth, 9 of the top 10 UTLAs were located in the South of England and London, and one in the East Midlands. When looking at the bottom 10, there was a greater geographical spread than with the top 10; 6 were located in the North , 1 in the Midlands and 3 in London. For male HLE at birth, all UTLAs in the top 10 were located in the South of England and in London, while 7 in the bottom 10 were located in the North.
For both males and females 41 UTLAs had significantly lower HLE estimates than the England estimate. 58 UTLAs for males and 60 UTLAs for females had a significantly higher HLE than the England average.
The within region UTLA inequality varies considerably across the regions. For both genders, London had the greatest within region inequality at 17.5 years for males and 14.2 years for females. The smallest within region inequality was in the North East for both males (7.0 years) and females (4.6 years). The greater population diversity present in London is likely to be a contributing factor to the wider inequality found in the capital.
|Local Authority||LE||HLE||Lower 95% confidence interval||Upper 95% confidence interval||Proportion of life spent in 'Good' health (%)||LE rank2||HLE rank2|
|Top 10||Richmond upon Thames||81.7||70.0*||67.4||72.7||85.8||3||1|
|Windsor and Maidenhead||81.1||68.4*||66.7||70.2||84.4||10||6|
|Blackburn with Darwen||76.5||56.3**||54.6||58.0||73.6||144||146|
**Significantly lower than the England male HLE at birth.
|Local Authority||LE||HLE||Lower 95% confidence interval||Upper 95% confidence interval||Proportion of life spent in 'Good' health (%)||LE rank2||HLE rank2|
|Richmond upon Thames||85.9||70.7*||67.8||73.5||82.2||1||2|
|Windsor and Maidenhead||84.6||69.9*||67.8||71.9||82.6||12||5|
|Bath and North East Somerset||84.4||69.0*||67.0||71.0||81.7||20||8|
|Barking and Dagenham||82.0||57.3**||54.5||60.1||69.9||114||142|
|Kingston upon Hull, City of||80.5||56.5**||54.5||58.4||70.2||145||148|
**Significantly lower than the England female HLE at birth.
The interactive maps published alongside this bulletin clearly show the North-South divide for both sexes. A darker shade on the maps represents lower HLE, and the lighter shades represent higher HLE.
Figures 1 and 2 show the relationship between LE and HLE. There is a linear trend between living longer (LE) and living longer in a ‘Good’ health state (HLE) for both males and females. If this pattern is put into context of the state pension age (using 65 for both males and females, where it will be by 2018), when using 95% confidence intervals, there were 77 UTLAs where male HLE was significantly lower than the current state pension age, and 68 where female HLE was significantly lower than their future state pension age. In fact only 17 UTLAs for males had significantly higher HLE than the state pension age, for females HLE was significantly higher in 27 UTLAs.
HLE in all UTLAs in the North East was significantly lower than the state pension age for both males and females, apart from Darlington for males. This emphasises the need to improve the health of those living in the North East.
In every UTLA, LE was longer for females than for males. The smallest gender inequality in LE was in Southend-on-Sea and Cheshire East, with females expected to live 2.9 years longer than males. The largest gender inequality in LE was in Blackpool, where females expect to live 6.0 years longer than males.
The gender differences in HLE were less regular in pattern than for LE: while in every UTLA LE for females exceeded that for males, HLE was higher for males in almost a quarter of authorities.
The largest difference in HLE between males and females was seen in Hammersmith and Fulham, where females can expect to live 5.2 years longer in a state of ‘Good’ general health than males. This was followed by Islington at 4.9 years. In Hammersmith and Fulham, it is the low HLE of males rather than the high HLE of females which drives this inequality.
The largest difference between males and females, where males have higher HLE, was in Waltham Forest where males can expect to live an additional 4.8 years in ‘Good’ health compared to females. Here, it is the lower HLE for females in Waltham Forest which drives this gender inequality.
Interestingly, although within gender inequality is greater for HLE (17.5 years for males and 15.4 years for females) than LE (8.1 years for males and 6.4 years for females); between gender inequality is greater for LE, at 6.0 years in Blackpool compared with 5.2 years for HLE in Hammersmith and Fulham.
The importance of HLE as a summary measure of population health is reflected in its inclusion in the two high-level outcomes in the Public Health Outcomes Framework.
The analysis provided in this bulletin adds a quality of life component to LE estimates. Results for UTLAs, regions and England are produced to allow comparisons across administrative areas at each time point. Producing the three levels of geography allows local areas to compare themselves to the England average, the regional average, with other local areas, and to assess their own progress over time.
This bulletin has shown a clear and persistent North-South divide in HLE, consistent with earlier findings in DFLE (ONS, 2014) at UTLA level. For both males and females, the longest durations of healthy life were found in prosperous parts of London and the South East region. The shortest durations remained in deprived parts of the North East, North West and London.
In addition, notable gender differences in LE and HLE were present although the pattern was less uniform in HLE. Female LE was higher than male LE across all the UTLAs analysed. However, gender differences in estimates of HLE were more varied, with males having a higher HLE than females in just less than a quarter of UTLAs.
Inequality in HLE across UTLAs was wider than it was for LE for both genders. For both males and females, the geographic inequality was more than twice as wide for HLE (17.5 years and 15.4 years respectively) than for LE (8.1 years and 6.4 years respectively).
It is important to track HLE as LE increases, to see whether these years of additional life are spent in states of good health or in poor health and dependency. Relevant to this are recent changes to state pension age in the UK, where people are expected to extend their working lives to take account of improvements in LE. The figures presented in this bulletin suggest the impact of increasing the state pension age differs greatly between populations in different areas of the country. Of the 150 UTLAs analysed, when assessed with 95% confidence intervals, males on average in 77 authorities had a significantly lower HLE than the state pension age (assessed at 65 for both genders, where it will be by 2018) while for females this was true in 68 authorities. In fact there were only 17 local authorities for men and 27 for females where HLE was significantly higher than the state pension age. Such information has relevance for healthy ageing and health care need, need for benefits during working age, and future pensions provision.
A consistent pattern of longer lives and smaller proportions of life spent in less favourable health states are associated with decreasing exposure to deprivation (ONS 2010, ONS 2013b). The English Index of Multiple Deprivation 2010 shows that the northern regions have higher concentrations of Lower Super Output Areas (LSOAs) in the most deprived decile (DCLG, 2011). This is consistent with the findings in this bulletin, which show those in the northern regions and northern authorities predominantly had lower LE and HLE than those in the South. The importance health agencies place on deprivation as a determinant of health is shown by its high prominence in health strategy documents, such as the Public Health Outcomes Framework. Tackling exposure to deprivation is therefore a key goal in reducing the health divide between the least and most advantaged areas, while continuing to bring about health improvement for all.
The data used in calculating the prevalence of good general health was obtained from the Annual Population Survey (APS). It was aggregated over a three-year period to achieve sufficiently large sample sizes to enable meaningful statistical comparison.
However, as the size of each UTLA varies, the number of respondents sampled in some UTLAs was a lot smaller than for others. For these authorities with relatively small sample sizes, such as the London borough of Tower Hamlets, the estimates are more susceptible to large fluctuations because of the impact of random variation.
The prevalence of ‘Good’ general health among males and females resident in private households in England was compared across regions and UTLAs, which include unitary authorities, London boroughs and metropolitan districts in England, but excludes the City of London and the Isles of Scilly. HLE was then calculated using the Sullivan method, which combines prevalence data with mortality and mid-year population estimates (MYPE) over the same period and geographical coverage to calculate estimates of LE and HLE at birth by sex ( ONS Life Table Template (192.5 Kb Excel sheet) , Jagger, 1999). The MYPEs used to estimate HLE for this bulletin are the revised backdated estimates based on the 2011 census (ONS 2013d).
The APS provides prevalence information for those aged 16 years and over. We are able to estimate HLE at birth by directly imputing health prevalence at age 16 -19 for those under 16 (ONS, 2013a).
The age band structure used for calculating HLE is not that outlined in the update to methodology to calculate health expectancies (ONS, 2013a) but the traditional age band structure of <1, 1-4, 5-9, 10-14, 15-19……85+.
Results are presented with 95% confidence intervals in reference tables to help interpretation. Confidence intervals in this bulletin indicate the uncertainty surrounding HLE estimates and allow more meaningful comparisons between areas. When comparing the estimates of two areas, non-overlapping confidence intervals are indicative of statistical significance but to confirm this, a test of significance should be carried out. When the statistical significance is noted in the text, this is based on a statistical test of the differences (Jagger et al., 2007). All differences noted in this text have been calculated to more than one decimal place.
HLE at a given age for a specific period and population, such as males and females at birth residing in private households in UTLAs in 2010-12, is an estimate of the average number of years a person would live in a state of ‘Good’ general health if he/she experienced the specified population’s age-specific mortality and health status rates for that time period throughout the rest of his/her life.
The figures reflect the mortality and health status of a population in a given time period residing in that area, rather than only those born in an area. It is not therefore the number of years that a person will actually expect to live in the various health states. This is because both the death rates and health status rates of the specified population are likely to change in the future, because of changes to health risk determinants such as smoking levels in the population, and some of those in the specified population may live elsewhere for part of their lives.
Health expectancies are indicators of health status that take into account the differences in the age structures of populations. Results are comparable by age, sex and between specified populations.
Bopp M, Braun J, Gutzwiller F, Faeh D and Swiss National Cohort Study Group (2012) ‘Health risk or resource? Gradual and independent association between self-rated health and mortality persists over 30 years’, PLoS ONE 7, available at: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0030795
DeSalvo KB, Bloser N, Reynolds K, He J, Muntner P (2006) ‘Mortality prediction with a single general self-rated health question. A meta-analysis’, Journal of General Internal Medicine 21, pp 267–275
Dwyer D S and Mitchell O S (1999) ‘Health problems as determinants of retirement: Are self-rated measures endogenous?’, Journal of Health Economics 18, pp 173-193
Eurostat (2013) Healthy Life Years and Life Expectancy at birth, by sex, available at: http://epp.eurostat.ec.europa.eu/portal/page/portal/product_details/dataset?p_product_code=TSDPH100
Idler EL and Benyamini Y (1997) ‘Self-rated health and mortality: a review of twenty-seven Community studies’, Journal of Health and Social Behaviour 38, pp 21–37
Jagger C (1999) ‘Health Expectancy Calculation by the Sullivan Method: A Practical Guide’, NUPRI Research Paper Series No 68, Tokyo
Jagger C, Cox B, Le Roy S and EHEMU (2007) Health Expectancy Calculation by the Sullivan Method: A Practical Guide Third Edition available at: http://www.eurohex.eu/pdf/Sullivan_guide_final_jun2007.pdf
Lee, Y (2000) ‘The predictive value of self assessed general, physical, and mental health on functional decline and mortality in older adults’ Journal of Epidemiology and Community Health, 54, pp 123-129
Manor O, Matthews S and Power C (2001) ‘Self-rated health and limiting longstanding illness: inter-relationships with morbidity in early adulthood’, International Journal of Epidemiology 30, pp 600-607
Marmot M (2010) ‘Fair Society, Healthy Lives, The Marmot Review, Strategic review of Health Inequalities in England post–2010’, The Marmot Review, available at: http://www.instituteofhealthequity.org/projects/fair-society-healthy-lives-the-marmot-review
Miilunpalo S, Vuori I, Oja P, Pasanen M and Urponen H (1997) ‘Self-Rated Health Status as a Health Measure: The Predictive Value of Self-Reported Health Status on the Use of Physician Services and on Mortality in the Working-Age Population’, Journal of Clinical Epidemiology 50, pp 517-528
Mossey J M and Shapiro E (1982) ‘Self-Rates Health: A Predictor of Mortality Among the Elderly’, American Journal of Public Health 72, pp 800-808
Ng N, Hakimi M, Santosa A, Byass P, Wilopo SA and Wall S (2012) ‘Is self-rated health an independent index for mortality among older people in Indonesia?’, PLoS ONE 7, available at: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0035308
ONS (2006) Health expectancies for local authorities in England and Wales, 2001. Health Statistics Quarterly 32. Pp.87-96
ONS (2010) Inequalities in healthy life expectancy by social class and area type: England 2001-03. (749 Kb Pdf) Office for National Statistics
ONS (2012a) Health Expectancies at Birth and at Age 65 in the United Kingdom, 2008–2010. Office for National Statistics
ONS (2013a) Update to the Methodology used to Calculate Health Expectancies. Health Statistics Quarterly Spring 56:1. Office for National Statistics
ONS (2013b) Inequality in Disability-Free Life Expectancy by Area Deprivation: England, 2003–06 and 2007–10. Office for National Statistics
ONS (2013c) Life Expectancy at birth and at age 65 by local areas in England and Wales, 2010-12. Office for National Statistics
ONS (2013d) Population Estimates for England and Wales, Mid-2002 to Mid-2012 Revised (Subnational). Office for National Statistics
ONS (2014) Sub-national health expectancies, Disability-Free Life Expectancy by Upper Tier Local Authority: England 2008-10. Office for National Statistics
Public Health Outcomes Framework (2013), available at: https://www.gov.uk/government/organisations/public-health-england/series/public-health-outcomes-framework
Pietilainen O, Laaksonen M, Rahkonen O and Lahelma E (2011) ‘Self-Rated health as a predictor of Disability Retirement – The contribution of ill-Health and Working Conditions’ PLoS ONE, 6 (9): e25004. doi: 10.1371/journal.pone.0025004
Reves (no date) http://reves.site.ined.fr/en/
Weinberger M, Darnell JC, Tierney WM, Martz BL, Hiner SL, Barker J and Neill PJ (1986) ‘Self-rated health as a predictor of hospital admissions and nursing home placement in elderly public housing tenants’ American Journal of Public Health, 76, pp 457–459
World Health Organisation (2011) World Report on Disability, available at: http://www.who.int/disabilities/world_report/2011/report/en/index.html
Figures in the text may not sum due to rounding.
The significance test refers to a one tailed Z- test of the difference of the estimates as detailed in (Jagger et al 2007).
Analysis has been carried out at the UTLA level, which includes counties, London boroughs, unitary authorities and metropolitan districts based on the 2009 reorganisation. Further information about the boundaries can be found on the ONS website. There are 152 UTLAs in England; ONS exclude City of London and the Isles of Scilly from the analysis due to small death and population counts. Therefore results are presented for the 150 remaining UTLAs.
This bulletin comments on a North-South divide. The North includes the North East, North West and Yorkshire and The Humber regions, and the South includes the South East, South West and East of England regions. London is not included in the South due to its differing characteristics which include its transient population, access to services and limited rural population.
For further information on the implications of differing survey sources, the continuity of the national health expectancies series and the child proxy imputation method, please see the update to methodology paper (ONS, 2013a).
Enquiries relating to these statistics should be made to:
Public Policy Analysis Division
Office for National Statistics
Tel: +44 (0) 1633 456396
Details of the policy governing the release of new data are available from the Media Relations Office.
Summary video podcasts explaining national and sub national health expectancies can be found on the ONS YouTube channel.
National Statistics are produced to high professional standards set out in the Code of Practice for Official Statistics. They undergo regular quality assurance reviews to ensure that they meet customer needs. They are produced free from any political interference.
The UK Statistics Authority has designated these statistics as National Statistics, in accordance with the Statistics and Registration Service Act 2007 and signifying compliance with the Code of Practice for Official Statistics.
Designation can be broadly interpreted to mean that the statistics:
meet identified user needs;
are well explained and readily accessible;
are produced according to sound methods; and
are managed impartially and objectively in the public interest.
Once statistics have been designated as National Statistics it is a statutory requirement that the Code of Practice shall continue to be observed.
A list of the names of those given pre–publication access to the statistics and written commentary is available in the pre–release access list. The rules and principles which govern pre–release access are featured within the Pre–release Access to Official Statistics Order 2008.
© Crown copyright 2014
You may re-use this information (not including logos) free of charge in any format or medium, under the terms of the Open Government Licence.
Or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU
Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: firstname.lastname@example.org
These National Statistics are produced to high professional standards and released according to the arrangements approved by the UK Statistics Authority.
|Kara Steel||+44 (0)1633 456396||Office for National Statistics||HLE@ons.gsi.gov.uk|