{
  "id": "cause-specific-mortality-rate",
  "title": "Cause-Specific Mortality Rate",
  "lang": "en",
  "source_lang": "en",
  "summary": "A cause-specific mortality rate measures the frequency of deaths attributed to a particular cause in a defined population over a specified period.",
  "category": "health",
  "subcategory": "public-health",
  "type": "concept",
  "aliases": [
    "Cause-specific death rate",
    "Death rate by cause"
  ],
  "generated_at": "2026-11-04",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/cause-specific-mortality-rate/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/cause-specific-mortality-rate/",
      "json": "https://aiwiki.page/zh/cause-specific-mortality-rate.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "A cause-specific mortality rate measures how frequently people in a defined population die from a particular disease, injury, or other specified cause during a stated period. It is a type of mortality rate used in epidemiology and public health to describe patterns of death, compare populations, and monitor changes over time. Its denominator is the population, or the time that population spends under observation—not merely the number of people diagnosed with the condition."
    },
    {
      "heading": "Definition and calculation",
      "text": "For routine annual population statistics, a cause-specific mortality rate is commonly calculated as:\n\n\\[M_c=\\frac{D_c}{P_{\\text{mid}}}\\times k\\]\n\nwhere:\n\n- \\(D_c\\) is the number of deaths attributed to cause \\(c\\) during the year;\n- \\(P_{\\text{mid}}\\) is the mid-year population of the same geographic area and population group;\n- \\(k\\) is a reporting multiplier, commonly \\(100{,}000\\).\n\nThe result is usually reported as deaths per 100,000 population per year. The mid-year population approximates the average population exposed to the possibility of death during the year.\n\nWhen individual follow-up time is available, the denominator can instead be expressed as person-time:\n\n\\[m_c=\\frac{D_c}{Y}\\times k\\]\n\nwhere \\(Y\\) is the total person-time observed while individuals remain alive and under observation. This formulation is especially useful in a cohort study, where participants may enter at different times or have different lengths of follow-up.\n\nThe numerator and denominator must refer to the same population and period. For example, deaths among residents should be paired with a resident population denominator, rather than with everyone physically present in the area.\n\nIllustrative example\n\nIf a population averaging 500,000 residents records 250 deaths attributed to stroke in one year, its annual stroke mortality rate is:\n\n\\[\\frac{250}{500{,}000}\\times100{,}000 =50\\]\n\nThis means 50 stroke deaths per 100,000 population during that year. It does not mean that 50 out of every 100,000 people with stroke died."
    },
    {
      "heading": "Assigning a cause of death",
      "text": "Routine mortality statistics usually classify each death by its underlying cause of death: the disease or injury that initiated the sequence leading to death, or the circumstances of the accident or violence that produced the fatal injury. This differs from an immediate cause, such as a terminal complication.\n\nInformation is generally recorded on a death certificate and coded using the International Classification of Diseases. A cause-specific rate may refer to a single condition, such as tuberculosis, or to a broad category, such as cancer or cardiovascular disease. Comparisons therefore require consistent cause definitions and code groupings.\n\nSome analyses count any mention of a condition on the death certificate rather than only the underlying cause. These multiple-cause mortality measures capture contributing conditions, but they answer a different question. Because one death may mention several conditions, counts across multiple-cause categories can overlap."
    },
    {
      "heading": "Crude, subgroup-specific, and standardized rates",
      "text": "A cause-specific rate can be crude, meaning that it covers the entire population without adjustment for its composition. “Cause-specific” does not imply “age-adjusted.”\n\nRates may also be calculated separately for age groups, sexes, geographic areas, or other population categories. For an age-specific rate, both the deaths and the population denominator are restricted to the same age group.\n\nAge is particularly important because mortality from many causes varies substantially across the life course. A population with more older people may have a higher crude cancer mortality rate even when its age-specific rates are lower.\n\nAge standardization reduces the influence of differences in age composition. With direct standardization, age-specific cause-of-death rates are weighted according to a common reference population:\n\n\\[M_c^{\\mathrm{std}}=\\sum_a w_a m_{c,a}, \\qquad \\sum_a w_a=1\\]\n\nHere, \\(m_{c,a}\\) is the mortality rate for cause \\(c\\) in age group \\(a\\), and \\(w_a\\) is that age group’s share of the standard population.\n\nA standardized rate is a comparative index, not the actual observed death rate. Comparisons require compatible age groups, methods, and standard populations. Crude rates and death counts remain important for describing the burden actually experienced by a population."
    },
    {
      "heading": "Distinction from related measures",
      "text": "Measure | Numerator | Denominator | Main interpretation\nCause-specific mortality rate | Deaths from a specified cause | Population or person-time | Frequency of death from that cause\n[[crude-death-rate | Crude death rate]] | Deaths from all causes | Total population\n[[case-fatality-ratio | Case-fatality ratio]] | Deaths attributable to a disease among its cases | Cases of that disease\nProportionate mortality | Deaths from a specified cause | Deaths from all causes | Share of deaths attributed to that cause\n[[incidence | Incidence rate]] | New cases of a condition | Person-time at risk\n\nThese measures can change independently. A cause may account for a rising proportion of all deaths while its mortality rate falls, if mortality from other causes falls faster. Similarly, high disease incidence does not necessarily imply high mortality when survival is good.\n\nA mortality rate is also not automatically a probability. The cumulative incidence of death from a particular cause measures the probability of dying from that cause within a specified interval and must account for deaths from other causes."
    },
    {
      "heading": "Competing risks",
      "text": "Death from one cause prevents a later death from another cause. This is the central issue in competing risks analysis.\n\nIn survival analysis, a cause-specific hazard describes the instantaneous rate of death from a particular cause among people who are still alive at a given time. It is related to, but not identical with, an annual population mortality rate.\n\nThe probability of eventually dying from a particular cause depends both on its cause-specific hazard and on mortality from competing causes. Consequently, a population rate or hazard cannot be interpreted directly as an individual’s lifetime probability of death from that cause."
    },
    {
      "heading": "Uses and limitations",
      "text": "Cause-specific mortality rates support public health surveillance, comparisons of disease burden, and evaluation of long-term changes in prevention and treatment. Death counts, crude rates, age-specific rates, and standardized rates provide complementary information.\n\nTheir interpretation depends on several data-quality and analytical issues:\n\n- Incomplete registration: Some deaths are never recorded, particularly where civil registration and vital statistics systems have limited coverage.\n- Cause misclassification: Incomplete medical information, diagnostic uncertainty, and certification practices can alter which cause is assigned.\n- Coding changes: Revisions to classification systems or coding procedures may create apparent trends unrelated to changes in mortality.\n- Population denominators: Migration, outdated population estimates, and inconsistent geographic boundaries can distort rates.\n- Small numbers: Rare causes or small populations can produce unstable rates. Confidence intervals, pooled years, or statistical models may be needed to characterize uncertainty.\n- Confounding: Differences in age, exposures, socioeconomic circumstances, and access to care can affect comparisons. Age standardization addresses age composition but not every source of difference.\n\nA change in a cause-specific mortality rate does not by itself establish why that change occurred. It may reflect changes in disease occurrence, survival after disease onset, competing causes of death, or the recording and classification of deaths. Attribution to a particular intervention requires additional evidence beyond the mortality trend alone."
    }
  ],
  "links": [
    "mortality-rate",
    "epidemiology",
    "public-health",
    "person-time",
    "cohort-study",
    "stroke",
    "underlying-cause-of-death",
    "death-certificate",
    "international-classification-of-diseases",
    "tuberculosis",
    "cancer",
    "cardiovascular-disease",
    "age-standardization",
    "crude-death-rate",
    "case-fatality-ratio",
    "incidence",
    "cumulative-incidence",
    "public-health-surveillance",
    "civil-registration-and-vital-statistics"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}