Types of Mortality Measures: Crude and Specific Mortality Rates
Mortality measures quantify deaths in populations and help describe health status, compare groups, identify priorities and evaluate services. Every measure requires a clearly defined numerator, denominator, place and time period.
Learning objectives
- Calculate crude, specific and cause-specific mortality rates.
- Calculate infant, neonatal, under-five and maternal mortality measures.
- Differentiate mortality rate, case-fatality proportion and proportionate mortality.
- Interpret crude versus standardised comparisons.
- Identify limitations of mortality data.
Basic structure
Mortality rate = Deaths in a defined population during a period ÷ Population exposed to the risk of death during that period × multiplier
The mid-year population is commonly used as an approximation of the average population for annual rates. The multiplier depends on frequency and convention, often 1,000 or 100,000.
1. Crude mortality rate
The crude mortality rate includes all deaths from all causes in the total population.
Crude mortality rate = All deaths during the year ÷ Mid-year population × 1,000
Worked example
A district records 1,200 deaths in a mid-year population of 200,000.
1,200 ÷ 200,000 × 1,000 = 6 deaths per 1,000 population per year.
Crude rates summarise overall mortality but are strongly affected by population age structure. An older population can have a higher crude rate even when age-specific survival is better.
2. Specific mortality rates
Age-specific mortality rate
Deaths in an age group ÷ Population in the same age group × multiplier
Sex-specific mortality rate
Deaths among a sex group ÷ Population of the same sex × multiplier
Cause-specific mortality rate
Deaths from a specified cause ÷ Population at risk during the period × 100,000
Cause-age-sex-specific rate
Restricts both numerator and denominator to the same cause and demographic subgroup, enabling more precise comparisons.
| Measure | Numerator | Denominator |
|---|---|---|
| Crude mortality | All deaths | Total population |
| Age-specific | Deaths in age group | Population in age group |
| Sex-specific | Deaths in specified sex | Population of that sex |
| Cause-specific | Deaths from specified cause | Total population at risk |
3. Proportionate mortality
Proportionate mortality describes the percentage of all deaths caused by a specified condition.
Proportionate mortality = Deaths from a cause ÷ Deaths from all causes × 100
It is a proportion, not a population mortality rate, because the denominator contains deaths rather than the population at risk.
Example
If 240 of 1,200 district deaths were due to cardiovascular disease, proportionate mortality is 240 ÷ 1,200 × 100 = 20%. This does not mean 20% of the population died from cardiovascular disease.
4. Case-fatality proportion
Case fatality measures severity among diagnosed cases.
Case-fatality proportion = Deaths from a disease among cases ÷ Total diagnosed cases of that disease × 100
If 8 of 200 confirmed cases die, case fatality is 4%. It is not the same as cause-specific mortality, whose denominator is the population.
5. Infant and child mortality measures
| Measure | Numerator | Denominator | Multiplier |
|---|---|---|---|
| Neonatal mortality rate | Deaths before 28 completed days | Live births in same period | 1,000 |
| Post-neonatal mortality rate | Deaths from 28 days to under 1 year | Live births | 1,000 |
| Infant mortality rate | Deaths under 1 year | Live births | 1,000 |
| Under-five mortality rate | Deaths before age 5 | Live births | 1,000 |
| Stillbirth rate | Stillbirths meeting the adopted definition | Total births (live births + stillbirths) | 1,000 |
Although traditionally called rates, several of these are probability-like measures. Use the current national definitions for gestational age, birthweight and reporting period.
6. Maternal mortality measures
Maternal mortality ratio
Maternal deaths during a period ÷ Live births during the same period × 100,000
The denominator is live births, so this is termed a ratio. It approximates obstetric risk per live birth and reflects both the risk of pregnancy and quality/access of maternal care.
Maternal mortality rate
Maternal deaths ÷ Women of reproductive age × multiplier
This incorporates both the chance of becoming pregnant and the risk of dying once pregnant.
Worked maternal example
A region records 24 maternal deaths and 40,000 live births.
Maternal mortality ratio = 24 ÷ 40,000 × 100,000 = 60 maternal deaths per 100,000 live births.
7. Perinatal mortality
Perinatal mortality combines late fetal deaths and early neonatal deaths according to the adopted definition.
Perinatal mortality rate = Stillbirths meeting definition + early neonatal deaths ÷ Total births × 1,000
It reflects maternal health, antenatal care, intrapartum care and immediate newborn care.
8. Standardised mortality
Standardisation improves comparison when populations differ in age or another major characteristic.
- Direct standardisation: applies each population's specific rates to the same standard population.
- Indirect standardisation: applies standard rates to the study population to calculate expected deaths.
- Standardised mortality ratio (SMR): observed deaths ÷ expected deaths × 100. An SMR above 100 indicates more deaths than expected.
Years of potential life lost
Years of potential life lost gives greater weight to deaths occurring at younger ages. For each death, subtract age at death from a selected reference age, then sum across deaths. It highlights causes responsible for premature mortality.
Sources of mortality data
Routine sources
- Civil registration and vital statistics
- Medical certification of cause of death
- Hospital and mortuary records
- Maternal and perinatal death surveillance
- Community reporting
Additional sources
- Census and household surveys
- Demographic surveillance sites
- Verbal autopsy
- Disease registries
- Special epidemiological studies
Data-quality problems
| Problem | Effect | Improvement |
|---|---|---|
| Incomplete death registration | Mortality is underestimated | Strengthen civil registration and community notification |
| Ill-defined cause of death | Cause-specific patterns become unreliable | Improve certification and coding |
| Wrong denominator | Rate is biased | Use matching population and period |
| Small numbers | Rates fluctuate widely | Pool periods or present confidence intervals |
| Different definitions | Comparisons are invalid | Apply standard definitions consistently |
Interpretation cautions
- Compare crude rates only when population structures are sufficiently similar.
- A high case-fatality proportion may reflect virulence, delayed care, poor access or case ascertainment limited to severe cases.
- Proportionate mortality can rise because other causes decline, even when the cause-specific rate is unchanged.
- Facility deaths do not represent all community deaths.
- Always state units, multiplier, geography and time.
Exam-focused summary
- Crude mortality uses all deaths and the total population.
- Specific rates restrict numerator and denominator to matching groups or causes.
- Case fatality measures severity among cases; cause-specific mortality measures population impact.
- Infant mortality uses live births; maternal mortality ratio uses live births and a multiplier of 100,000.
- Standardisation is needed for fair comparison of populations with different age structures.
Review questions
- Calculate a crude mortality rate from a district dataset.
- Differentiate cause-specific mortality, proportionate mortality and case fatality.
- Write formulas for neonatal, infant and under-five mortality.
- Calculate and interpret a maternal mortality ratio.
- Explain direct and indirect standardisation.
- Describe five limitations of mortality data.
References and further reading
- WHO Global Health Observatory indicator definitions.
- CDC Field Epidemiology Manual.
- Clinical Medicine Year 2 curriculum.
Educational use only. Use current Uganda civil-registration, surveillance and maternal/perinatal death-review definitions.
