What Is Mortality?

Mortality, expressed statistically as a death rate, is the frequency of death in a defined population over a defined period. As a statistical quantity it is a ratio: deaths in the numerator, person-time or midyear population in the denominator, conventionally scaled to deaths per 1,000 or per 100,000. The apparent simplicity is deceptive, because both terms require careful definition. Which deaths count depends on residence rules, on how the underlying cause is assigned, and on the completeness of registration; the denominator depends on population estimates that are themselves modeled between censuses.

Mortality measurement belongs to demography and biostatistics and underpins actuarial science, epidemiology, and public health surveillance. It is one of the three components of population change, alongside fertility and migration, and the input from which life expectancy and survival curves are computed.

Crude, Specific, and Standardized Rates

The crude death rate divides all deaths by the total population. It is easy to compute and almost useless for comparison, because a population with more old people will show a higher crude rate even if health at every age is identical. Specific rates address this by restricting both numerator and denominator to a stratum: an age-specific rate, a sex-specific rate, or a cause-specific rate such as deaths from ischemic heart disease.

Standardization then recombines specific rates against a common reference population so that two groups can be compared without age structure confounding the result. In direct standardization, each age-specific rate is weighted by the proportion of that age group in a standard population and the weighted rates are summed. The age adjustment methods used by the National Center for Health Statistics use the projected year 2000 US standard population, which means an age-adjusted rate is a comparison index rather than a count of anything real. Indirect standardization inverts the procedure and produces a standardized mortality ratio, preferred when the study population is small and its age-specific rates are unstable.

Life Tables and Derived Measures

A life table converts a schedule of age-specific mortality rates into a synthetic cohort, tracking how many of an initial 100,000 births would survive to each exact age if current rates persisted. From it come the probability of dying between ages, the number of person-years lived, and life expectancy at any age. Related measures extend the same machinery: infant and maternal mortality rates, case fatality ratios, years of potential life lost, which weights deaths by how early they occur, and excess mortality, which compares observed deaths against a counterfactual baseline fitted to prior years. Excess mortality became the preferred summary during the COVID-19 pandemic precisely because it sidesteps disagreement over cause-of-death attribution.

Data Sources and Cause Coding

In most high-income countries the primary source is civil registration. The US National Vital Statistics System compiles death certificates from every state and territory, roughly 3 million records a year, and codes the underlying cause using the International Classification of Diseases. Annual summaries such as the report on mortality in the United States for 2023 publish age-adjusted rates, leading causes, and life expectancy from these files. Where registration is incomplete, demographers fall back on sample registration systems, sibling survival histories from household surveys, and verbal autopsy. Statistical modeling fills remaining gaps, and methods for modeling age-adjusted mortality address the smoothing and forecasting problems that arise when rates are sparse at older ages.

Applications

Mortality statistics have applications across several fields, including:

• Public health surveillance and outbreak detection
• Actuarial pricing and reserving for life insurance and annuities
• Clinical trial endpoints and hospital outcome benchmarking
• Population projection and pension system planning
• Environmental and occupational epidemiology, including exposure risk estimates
• Health system performance comparison across countries and regions