How Is Life Expectancy Calculated? Life Tables and Mortality Rates Explained

Life expectancy is a statistical estimate derived from patterns of death at different ages. It does not predict the exact age at which any particular person will die. National statistical offices use population and mortality data to build life tables, which show how a hypothetical group would survive under specified conditions. Understanding the calculation helps students interpret health statistics, compare countries and avoid misleading claims about average lifespan.

Key takeaways

Life expectancy is calculated from age-specific mortality rates, not by averaging only the ages of people who died.
A period life table applies mortality rates from a particular period across a hypothetical lifetime.
Cohort life expectancy can incorporate projected changes in mortality over time.
Life expectancy at birth differs from remaining life expectancy at older ages.
Data quality, population size and uncertainty matter when comparing places.

What does life expectancy mean?

Life expectancy is the average number of years a person would be expected to live under a defined set of mortality assumptions. It is a population statistic, not a promise or prediction about an individual. When a report says life expectancy at birth is 80 years, it does not mean that most people die on their eightieth birthday. People have widely different lifespans, and the statistic summarises a distribution of possible outcomes.

Life expectancy is closely connected to the study of global population change over time, because births, deaths and migration shape population size and age structure. But it measures something different from population growth. A country can have improving life expectancy while its population falls if births remain low or emigration is high.

Why simply averaging ages at death is misleading

Suppose a town records deaths during one year and calculates the average age of those who died. That average depends heavily on the town’s age structure. A young community with few elderly residents may record a low average age at death even if people of each age face relatively low risks. An older community may show a high average simply because many residents are already elderly.

Life tables avoid much of this problem by examining mortality at each age. They ask what would happen to a hypothetical group if it experienced the measured pattern of death risks. This approach allows more meaningful comparisons than a raw average of death certificates.

What is a life table?

A life table is a statistical model that follows a hypothetical group, often starting with 100,000 births. At each age it applies a probability of death and calculates how many people would survive to the next age. It also estimates the total years lived by the group. From those values, statisticians calculate the expected remaining years of life at different ages.

The UK Office for National Statistics publishes national life tables based on deaths and population estimates. The tables are not records of 100,000 actual people followed from birth to death. They are a mathematical way to summarise observed or projected mortality conditions.

How age-specific mortality rates are estimated

Statisticians need the number of deaths at each age and an estimate of the population exposed to the risk of death at that age. A central mortality rate is broadly the number of deaths divided by the relevant population exposure. More detailed calculations account for the timing of deaths and special issues at infancy and the oldest ages.

The underlying data come from systems such as death registration and population estimates. Census results help establish population counts and characteristics. Earth Site’s article on how a population census is conducted explains how a census is organised and why the quality of the count matters. If populations are undercounted or ages are recorded inaccurately, calculated rates can be distorted.

A simple hypothetical example

Imagine a hypothetical group of 100,000 people all starting at birth. Suppose 500 die before reaching age one. Then 99,500 survive to their first birthday. At each subsequent age, the life table applies the corresponding probability of death to the number still alive. The calculation continues through older ages until the hypothetical group has no survivors remaining.

The total number of person-years lived is then divided by the initial number of people to estimate life expectancy at birth. A person-year means one person living for one year. This method incorporates the timing of deaths, not just the final number who died. Real life tables use far more detailed rates than this simplified illustration.

Period life expectancy explained

Period life expectancy uses the age-specific mortality rates observed during a particular period, such as a recent three-year interval. It applies those rates to the hypothetical group throughout its entire lifetime. This produces a useful snapshot of current mortality conditions, but it does not assume that the real-world future will remain unchanged.

For example, a period life expectancy calculated from 2022–2024 mortality rates describes what would happen if those rates continued at every age. It does not literally forecast how long a child born in 2024 will live. Medical progress, epidemics, social conditions and other factors may change mortality over the following decades.

Cohort life expectancy explained

Cohort life expectancy follows a group born in a particular year and can use different mortality rates as that group ages. For future ages, statisticians often need projections about how mortality might change. A cohort estimate can therefore differ from a period estimate, sometimes substantially, if future improvements in survival are expected.

The ONS explains this distinction in its guidance on past and projected life tables. Projections depend on assumptions, so they become more uncertain further into the future. When reading a headline, students should always check whether the number is period or cohort life expectancy. Comparing one type with the other without explanation can produce misleading conclusions.

Life expectancy at birth and at age 65

Life expectancy can be calculated at any age. Life expectancy at age 65 means the average additional years expected for someone who has already reached 65 under the stated mortality assumptions. It is not calculated by subtracting 65 from life expectancy at birth. People who survive to older ages have already passed through earlier risks, so their remaining expectation must be calculated separately.

This is particularly important when comparing historical populations. High infant and childhood mortality could reduce life expectancy at birth sharply, even when some adults lived to advanced ages. The claim that people in the past generally died at 30 because life expectancy at birth was around 30 confuses a population average with an individual’s likely adult lifespan.

Why countries and regions differ

Life expectancy varies with many factors, including health services, sanitation, nutrition, income, education, working conditions, violence and environmental exposures. Differences can also reflect past events and population characteristics. A national average may hide substantial variation between neighbourhoods or social groups. It should therefore be interpreted alongside measures of inequality and health.

Earth Site’s guide to how population pyramids show age structure explains how population age structures differ. Age structure matters when interpreting crude death rates, but life tables use age-specific risks to improve comparability. Even so, comparisons require consistent data definitions and careful consideration of uncertainty.

How shocks such as pandemics affect the figures

A major epidemic, war or disaster can increase deaths at particular ages and reduce period life expectancy for the affected years. This does not mean that every child born during the event will necessarily have a shorter eventual lifespan. If mortality conditions improve afterwards, later period estimates may recover. Cohort outcomes depend on the rates experienced over many years.

The ONS uses three-year periods for national life tables partly to reduce random year-to-year fluctuations. However, a sustained or very large shock can still change the result significantly. When comparing trends, students should check the dates, the length of the period and whether methods have changed.

Why small populations create uncertainty

In a small area, one unusual cluster of deaths can cause a noticeable change in measured rates. Estimates may therefore fluctuate more than national figures. Statistical offices sometimes combine several years or use specialised methods to produce more stable local estimates. Confidence intervals help show how precise the reported value is likely to be.

A ranking of small districts based on tiny differences can be misleading if the uncertainty ranges overlap. A responsible comparison asks whether differences are large enough to be meaningful and whether population estimates are reliable. Numbers with many decimal places are not automatically more accurate.

What life expectancy cannot tell us

Life expectancy does not directly measure quality of life, disability, mental wellbeing or access to care. Two populations can have similar average lifespans but very different experiences of illness. Measures such as healthy life expectancy attempt to incorporate years lived in good health, although these require additional data and assumptions.

It also cannot explain causation on its own. A difference between regions may be associated with deprivation, healthcare access or environmental conditions, but identifying the causes requires further research. Students should distinguish a descriptive statistic from an explanation of why the pattern exists.

A checklist for interpreting life expectancy headlines

Start by identifying the country or region, the years covered and whether the estimate is at birth or another age. Next check whether it is a period or cohort measure. Ask which population groups are included and whether the data are comparable with earlier releases. Finally examine uncertainty, especially for small areas, and consider the social context behind the numbers.

A headline such as “life expectancy has fallen” may describe a real change in period mortality, but it does not automatically predict the future of every person born that year. Careful interpretation makes the statistic more useful for public debate, education and health planning.

A worked example using survival probabilities

Suppose a hypothetical group begins with 1,000 people. If the probability of dying before age one is 0.005, the expected number of deaths in that interval is five, leaving 995 survivors. If the next age interval has a death probability of 0.001, roughly one more person is expected to die. The life table repeats this calculation across all ages using appropriate rates.

The final life expectancy depends on the total person-years lived by the group, not merely the number surviving at one age. This example uses simplified probabilities and a small starting group to make the arithmetic visible. Official calculations use more detailed methods and much larger statistical datasets.

Why period and cohort results can differ

A period table applies mortality rates from a particular time to every age in a hypothetical life. A cohort table instead follows a birth group through changing conditions, using observed rates for past years and projections where future years are unknown. If mortality is expected to improve, cohort life expectancy may exceed the period estimate.

Neither figure is inherently the only correct answer. They address different questions. Period estimates describe current mortality conditions, while cohort estimates are more directly connected with a generation’s potential future experience but rely on assumptions. Clear reporting should name the measure and explain its limitations.

Life expectancy and inequality

A national average can conceal large differences in health outcomes between communities. Income, occupation, housing, education and environmental conditions may all be associated with mortality risks. Comparing life expectancy across neighbourhoods can reveal important inequalities, but it does not by itself establish which factors caused them.

Students should also distinguish life expectancy from healthy life expectancy, which estimates years lived in good health using additional survey or health information. A population can live longer while still experiencing substantial illness or disability. Looking at several measures together produces a fuller picture of wellbeing than relying on a single headline number.

Conclusion

Life expectancy is a carefully constructed statistical measure, not a simple average or personal forecast. Life tables combine mortality rates across ages to reveal patterns that raw death counts can obscure. By distinguishing period and cohort estimates, checking data quality and considering uncertainty, readers can use life expectancy figures more accurately when comparing health and development across places and time.

Frequently asked questions