Measures of Frequency and Occurrence
Measures of frequency and occurrence are the quantities epidemiology uses to count how much disease, death, or other health events exist in a population. They translate raw case counts into interpretable figures by relating the number of events to the size of the population and the time over which it was observed, providing the numerical foundation on which comparisons of risk and rate are built.
Definition
Measures of frequency and occurrence are epidemiologic quantities that express the amount of a health-related state or event in a population, by relating a count of cases to the population at risk and, for incidence measures, to the time at risk.
Scope
This area orients the reader to the family of basic occurrence measures: prevalence (how much disease is present), incidence (how fast new disease arises), and the two principal ways of expressing incidence — cumulative incidence (risk) and incidence density (rate). It also covers mortality measures, which apply the same logic to death as the event of interest. The treatment is conceptual and methodological, not a guide to clinical decisions.
Sub-topics
Core questions
- How much of a disease or condition is present in a population at a given time?
- How quickly do new cases arise in a population over time?
- When should occurrence be expressed as a proportion (risk) versus a rate?
- How is the population denominator, including person-time, defined and counted?
Key concepts
- Numerator (cases) and denominator (population at risk)
- Prevalence versus incidence
- Risk (proportion) versus rate
- Person-time at risk
- Point versus period measures
- Mortality as a frequency measure
Mechanisms
Every occurrence measure is a count of events placed over a denominator that captures the population from which those events arose. Prevalence uses a static population denominator and counts existing cases at a point or over a period; incidence uses a denominator that reflects who is at risk and for how long, counting only new cases. When the denominator is people followed over a fixed interval, incidence is a proportion (cumulative incidence, or risk); when the denominator is accumulated person-time, incidence is a rate (incidence density). Mortality measures apply the same numerator-over-denominator structure with death as the event. Choosing the correct denominator — and, for rates, correctly accruing person-time — is what makes these measures comparable across populations of different sizes and observation periods.
Clinical relevance
These measures underlie nearly all quantitative statements in the health sciences about how common or how rapidly occurring a condition is, and they are prerequisites for understanding measures of association and effect. Reading them correctly is part of evidence appraisal; they describe population-level occurrence and are not instructions for individual diagnosis or treatment.
Epidemiology
Frequency measures are reported throughout descriptive epidemiology, surveillance, and the introductory sections of analytic studies, where they characterise the burden of a condition before associations are examined. Prevalence is typical of cross-sectional surveys; incidence requires follow-up of an at-risk population, as in cohort studies; mortality measures are staples of vital statistics and public-health monitoring.
History
Counting deaths and cases relative to population has roots in the seventeenth-century bills of mortality and the nineteenth-century vital-statistics work that accompanied the rise of public health. The modern, explicitly defined distinctions between prevalence and incidence, and between risk and rate, were consolidated in twentieth-century epidemiologic textbooks and dictionaries, which standardised the denominators and terminology now in routine use.
Related topics
Seminal works
- grimes-descriptive-2002
- rothman-2008
- porta-2014
Frequently asked questions
- What is the difference between a measure of frequency and a measure of association?
- A measure of frequency counts how much disease or death occurs in a single population (for example prevalence or incidence), whereas a measure of association compares frequencies between groups (for example a risk ratio or rate ratio). Frequency measures are the building blocks from which association measures are computed.
- Why does the denominator matter so much in these measures?
- The denominator defines the population the count is relative to. Using the wrong denominator — counting people who are not at risk, or ignoring the time they were observed — distorts the measure and makes comparisons between populations invalid.