Healthy Life Expectancy

Healthy Life Expectancy in Population Health and Policy

Healthy life expectancy splits life expectancy into the years a population can expect to spend in good health and the years it can expect to spend in poorer health. The figure comes from joining a period life table with data on how healthy people are at each age, so it changes when either death rates or ill health change. The measure matters to health policy because longer lives are a mixed gain if the extra years are spent unwell, and gaps between areas and social groups are far wider in healthy years than in total years. Two different measures share the name, the self-rated measure published by the Office for National Statistics (ONS) and the severity-weighted HALE published by the World Health Organization, and their numbers should not be compared with each other.

Two measures that share one name

Health expectancies divide expected lifetime by health state, and members of the family differ in how the healthy state is defined. Murray, Salomon and Mathers divide summary measures of population health into two broad families, health expectancies and health gaps. Healthy life expectancy belongs to the first family. The DALY works the other way round, adding years of life lost to years lived with disability so that it counts health lost rather than healthy years achieved.

The ONS asks survey respondents to rate their general health on a five-point scale running between "very good" and "very bad", and combines the top two answers into "good" health. The WHO definition counts the average number of years a person can expect to live in "full health" if exposed to the age-specific mortality and morbidity prevailing at the time. WHO first published such estimates for 191 countries for 1999, labelled DALE (disability-adjusted life expectancy), with Japan highest at 74.5 years at birth. Other health expectancies, such as disability-free and active life expectancy, define the healthy state by limitation or dependency instead.

The Sullivan calculation

Both the ONS and the WHO figures are produced by the Sullivan method. Jagger, Van Oyen and Robine trace the first example to a 1971 US Department of Health, Education and Welfare report by Sullivan. The method needs only age-specific prevalence of the healthy and unhealthy states, often from cross-sectional surveys, and age-specific mortality from a period life table.

$$\text{HLE}_x = \frac{1}{l_x} \sum_{i \ge x} L_i \,(1 - \pi_i)$$

where $\text{HLE}_x$ is healthy life expectancy at age $x$, $l_x$ is the number of life-table survivors at exact age $x$, $L_i$ is the person-years lived in age band $i$, $\pi_i$ is the proportion of people in band $i$ who are not in good health, and the sum covers every band at or above age $x$, the oldest band being open-ended. Dividing the same sum without the health adjustment by $l_x$ gives ordinary life expectancy at age $x$.

The WHO version replaces the yes or no health state with a fraction. Its indicator metadata describe an "equivalent lost healthy year fraction" for each age, estimated as the all-cause rate of years lived with disability per capita, adjusted for independent comorbidity.

$$\text{HALE}_x = \frac{1}{l_x} \sum_{i \ge x} L_i \,(1 - w_i)$$

where $\text{HALE}_x$ is the WHO healthy life expectancy at age $x$, $w_i$ is the equivalent lost healthy year fraction in age band $i$, and $l_x$ and $L_i$ are as above. The fractions rest on years lived with disability, which in the Global Burden of Disease study are prevalence multiplied by disability weights. Step-by-step mechanics are on the HALE calculation page.

How the ONS produces its figures

The ONS uses five-year age bands and abridged life tables, because subnational samples are too small for single years of age, and pools three years of data. Prevalence of "good" health comes from the Annual Population Survey. The survey does not cover people under 16 and has sparse data above 84, so census-based adjustment factors impute prevalence at those ages. Since December 2024 a logistic regression model has predicted prevalence to offset falling survey sample sizes. The February 2026 bulletin carries the label "official statistics in development" for that reason.

Worked example with an illustrative life table at age 65

All numbers in this example are illustrative. A period life table starts with 100,000 people alive at exact age 65 and groups later life into four bands, the last of them open-ended. The survey gives the share of each band not in good health, and a burden study gives the equivalent lost healthy year fraction for the same bands.

1. Life expectancy. Total person-years divided by survivors at 65 gives remaining life expectancy.

$$e_{65} = \frac{1{,}680{,}000}{100{,}000} = 16.8$$

where $e_{65}$ is remaining life expectancy at 65 in years.

2. Healthy life expectancy. Each band's person-years are multiplied by the share in good health, for example 480,000 × 0.70 = 336,000. The four products sum to 976,000.

$$\text{HLE}_{65} = \frac{976{,}000}{100{,}000} = 9.76$$

where $\text{HLE}_{65}$ is healthy life expectancy at 65 in years. Of 16.8 remaining years, about 9.8 are expected in good health, 58.1% of the total (9.76 / 16.8 = 0.581), and 16.8 minus 9.76, or 7.04 years, in poorer health.

3. Severity-weighted version. Weighting by the lost fractions instead, for example 480,000 × 0.88 = 422,400, gives 1,370,300 person-years and $\text{HALE}_{65}$ = 1,370,300 / 100,000 = 13.703, about 13.7 years. The weighted figure is higher because a year in mild ill health counts as most of a healthy year, while the self-rated measure counts it as nothing.

4. A change in health with no change in mortality. Suppose prevention cuts the share not in good health at 80 and over to 0.50. Healthy person-years in that band rise to 420,000 × 0.50 = 210,000, the total rises to 1,018,000, and $\text{HLE}_{65}$ becomes 10.18 years, a gain of 0.42 years. Life expectancy stays at 16.8, so the whole gain is a shift of time between health states.

5. Interpretation. The same population can report 9.8 or 13.7 healthy years depending on the measure, so a figure means little without its definition of health. The 0.42-year gain is a population average spread over the whole cohort. The result does not say that each person gains five months of good health.

Inequality and policy targets in the UK

The ONS publishes healthy life expectancy for the UK, its four countries, English regions and local areas. In 2022 to 2024 males in the UK could expect 60.7 years in "good" general health, 77% of life, and females 60.9 years, 73%. The ONS reports that both fell to their lowest level since the series began in 2011 to 2013, although life expectancy had risen modestly since 2019 to 2021.

Gaps by deprivation are where the measure adds most to life expectancy alone. The April 2026 ONS bulletin puts healthy life expectancy at birth in the most deprived tenth of areas in England at 49.8 years for males and 48.2 for females, against 69.2 and 68.5 in the least deprived. Measured by the slope index of inequality, the gap is 19.3 years for males and 20.1 for females in healthy life expectancy, compared with 10.4 and 8.0 years in life expectancy. Poorer areas therefore have both shorter lives and a smaller share of life in good health, a pattern central to work on health inequality.

Governments have written targets in these terms. The Levelling Up white paper of February 2022 set a mission that by 2030 the gap in healthy life expectancy between local areas where it is highest and lowest will have narrowed, and that by 2035 healthy life expectancy will rise by 5 years. The ONS quality and methods guide reports that the Department of Health and Social Care uses healthy life expectancy as a high-level indicator in the Public Health Outcomes Framework and the 10 Year Health Plan for England. The Department for Work and Pensions uses it in State Pension age reviews, and both the statutory and independent reviews in 2022 drew on it.

Healthy life expectancy, QALYs and economic evaluation

Healthy life expectancy and the QALY both combine length and quality of life, but they answer different questions. A QALY in cost-utility analysis weights time by a preference-based health state utility value and is estimated for the patients affected by a decision. Healthy life expectancy describes a whole population under current period rates, and the ONS version uses a two-state self-rating with no valuation of how bad poorer health is. The wider family of time-weighted outcome units, the QALY and the DALY among them, is covered under the health-adjusted life year.

Quality-adjusted life expectancy bridges the two. Love-Koh, Asaria, Cookson and Griffin combined Health Survey for England quality-of-life data with ONS mortality to model quality-adjusted life expectancy by socioeconomic group, finding a socioeconomic inequality at birth of 11.87 QALYs. They present it as a refinement of binary measures such as disability-free life expectancy, suited to evaluating the health inequality impact of interventions within cost-effectiveness analysis, the use developed in distributional cost-effectiveness analysis. National healthy life expectancy is a poor outcome for a model of a single programme, because the population figure moves with every change in mortality and health across the whole population.

Assumptions and comparability

The Sullivan method gives a snapshot under current conditions, and several assumptions sit behind each figure. Each misreading below treats the number as more than that.

1. Treating the figure as a forecast. ONS estimates are period-based, the years a person would live if current age-specific rates applied throughout life.
2. Reading the years as a sequence. The ONS notes that the estimates do not imply that the first period of life is in "good" health followed by a period in "poorer" health.
3. Ignoring how prevalence lags incidence. Jagger, Van Oyen and Robine explain that prevalence at a given age reflects past onset, so the Sullivan method may misstate health expectancy when incidence changes quickly. A multi-state life table built on transition rates avoids this but needs longitudinal data.
4. Comparing across surveys. The ONS lists sensitivity to the measurement instrument, survey mode and whether people in institutions are included. Its Northern Ireland figures are not comparable with those of the Northern Ireland Department of Health, and WHO states that its HALE estimates for 2000 to 2021 are not directly comparable with earlier WHO estimates.
5. Applying the average to a person. The ONS describes the measure as reflecting the average experience of a population, not the health trajectory of any individual.

Sources

• Department for Levelling Up, Housing and Communities. Levelling Up the United Kingdom: executive summary. Published 2 February 2022. https://www.gov.uk/government/publications/levelling-up-the-united-kingdom/levelling-up-the-united-kingdom-executive-summary
• GBD 2021 Diseases and Injuries Collaborators. Global incidence, prevalence, years lived with disability (YLDs), disability-adjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990-2021. Lancet. 2024;403(10440):2133-2161. https://doi.org/10.1016/S0140-6736(24)00757-8
• Jagger C, Van Oyen H, Robine JM. Health Expectancy Calculation by the Sullivan Method: A Practical Guide. 4th ed. October 2014. https://ined-legacy.reves-network.org/fichier/s_rubrique/20182/sullivan.guide.pre.final.oct2014.en.pdf
• Love-Koh J, Asaria M, Cookson R, Griffin S. The social distribution of health: estimating quality-adjusted life expectancy in England. Value in Health. 2015;18(5):655-662. https://doi.org/10.1016/j.jval.2015.03.1784
• Mathers CD, Sadana R, Salomon JA, Murray CJL, Lopez AD. Healthy life expectancy in 191 countries, 1999. Lancet. 2001;357(9269):1685-1691. https://doi.org/10.1016/S0140-6736(00)04824-8
• Murray CJL, Salomon JA, Mathers C. A critical examination of summary measures of population health. Bulletin of the World Health Organization. 2000;78(8):981-994. https://pubmed.ncbi.nlm.nih.gov/10994282/
• Office for National Statistics. Healthy life expectancy by national area deprivation, England and Wales: between 2013 to 2015 and 2022 to 2024. Statistical bulletin, 15 April 2026. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthinequalities/bulletins/healthylifeexpectancybynationalareadeprivationenglandandwales/between2013to2015and2022to2024
• Office for National Statistics. Healthy life expectancy quality and methods guide. Last revised 15 April 2026. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthandlifeexpectancies/methodologies/healthylifeexpectancyqualityandmethodsguide
• Office for National Statistics. Healthy life expectancy, UK: between 2011 to 2013 and 2022 to 2024. Statistical bulletin, 19 February 2026. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthandlifeexpectancies/bulletins/healthstatelifeexpectanciesuk/between2011to2013and2022to2024
• World Health Organization. Healthy life expectancy (HALE). Global Health Observatory indicator metadata registry, accessed 25 October 2026. https://www.who.int/data/gho/indicator-metadata-registry/imr-details/66