Measures of Disease Frequency in Epidemiology
The core objective of descriptive epidemiology is to quantify the occurrence of disease or other health-related events within a defined population. This quantification allows public health professionals and researchers to understand the burden of disease, allocate resources, and, most critically, investigate potential causes (etiology). The measures used to accomplish this are collectively termed measures of disease frequency, and they transform simple observations (counts) into meaningful, comparable statistics (proportions and rates) by incorporating the size of the population at risk and the element of time.
Before diving into the two principal measures, prevalence and incidence, it is necessary to establish the foundation of all frequency measures: counts, ratios, and proportions.
Fundamental Components: Counts, Ratios, and Proportions
A simple count—the absolute number of cases of a disease—is the most basic measure. While useful for local health officials to determine resource needs (e.g., the number of hospital beds or vaccine doses), a count is limited for comparison because it does not account for the size of the population from which the cases arose. One thousand cases of flu in a small town is a much greater public health concern than 1,000 cases in a major metropolitan area.
A ratio is a fraction in which the numerator is not necessarily included in the denominator. For example, the ratio of males to females with a specific illness is calculated by dividing the number of male cases by the number of female cases. It simply compares the magnitude of two separate quantities.
A proportion, on the other hand, is a type of ratio where the numerator *must* be included in the denominator. Proportions are dimensionless, ranging from 0 to 1, and are often expressed as a percentage. The proportion of a population who are vaccinated is a simple example. Both prevalence and cumulative incidence (incidence risk) are essentially proportions.
A rate is a measure of frequency that explicitly includes a measure of time in the denominator. True rates measure how quickly something is happening. In epidemiology, incidence rate is the primary measure that utilizes the concept of “person-time,” reflecting the dynamic nature of a population where individuals are at risk for different lengths of time.
Prevalence: The Measure of Existing Disease Burden
Prevalence is the proportion of a population that has a disease or other health condition at a specified point in time or during a specified period. It provides a snapshot of the disease burden and is crucial for planning health services and resource allocation.
There are two main types of prevalence. Point prevalence refers to the proportion of people with the disease at a single, defined point in time (e.g., on January 1, 2025). The formula is the number of current cases divided by the total population at that point in time. Period prevalence, a broader measure, refers to the proportion of people who had the disease at any time during a specified time interval (e.g., over the course of a calendar year). This includes cases that existed at the start of the period and any new cases that developed during the period.
A critical characteristic of prevalence is that its numerator includes both recently diagnosed (new) cases and individuals who have lived with the condition for a long time (old cases). Consequently, prevalence is influenced by both the rate at which new cases occur (incidence) and the duration of the disease. Diseases of long duration, such as diabetes or chronic arthritis, will have a higher prevalence than short-term illnesses, such as the common cold, even if their incidence is similar. For diseases that are incurable but not immediately fatal, improved treatments that extend survival will *increase* prevalence, demonstrating its complexity as a summary measure.
Incidence: The Measure of New Disease Occurrence
In contrast to prevalence, incidence is a measure of the number of *new* cases of a disease that develop in a population at risk during a specified time period. Because incidence measures the transition from a disease-free state to a diseased state, it is the fundamental measure for studies of disease causation (etiology), allowing researchers to link exposure to a risk factor with the subsequent development of the disease.
To measure incidence, the population must be “at risk,” meaning all individuals in the denominator must be capable of developing the disease. For instance, in a study of prostate cancer incidence, women and men who have already had a prostatectomy would be excluded from the population at risk.
Incidence Proportion (Cumulative Incidence or Risk)
Incidence proportion, also known as cumulative incidence or risk, is the most straightforward measure of incidence. It is calculated as the number of new cases during a specified period divided by the size of the population at risk at the start of that period. It is a proportion, meaning it ranges from 0 to 1 and is often expressed as a percentage, such as “a 5% risk of developing the flu during the winter season.”
The incidence proportion is interpreted as the average risk of developing the disease over a fixed time interval. It makes the assumption that the entire population at risk at the beginning of the study has been followed for the entire duration. This measure is highly valuable in fixed cohorts (groups of individuals followed over time), but it becomes less accurate in dynamic populations where individuals enter or leave the population during the study period through births, deaths, or migration, leading to unequal observation times.
Incidence Rate (Incidence Density)
The incidence rate, or incidence density, is a more precise and sophisticated measure of incidence that addresses the challenge of varying follow-up times. Instead of simply using the number of people at risk as the denominator, the incidence rate uses the total amount of observation time contributed by all individuals who remained at risk, measured in “person-time” units (e.g., person-years, person-months). The formula is the number of new cases divided by the total person-time at risk.
This approach accurately accounts for the fact that participants may be lost to follow-up, refuse to continue participating, die from another cause, or enter the study at different times. Each person contributes only the time they were truly observed and at risk. The resulting incidence rate is expressed as an event per unit of person-time (e.g., 15 cases per 1,000 person-years). This measure reflects the speed at which new cases are occurring in the population and is mathematically a true rate.
The Relationship Between Prevalence and Incidence
Prevalence and incidence are closely linked, and for rare, stable diseases, their relationship can be approximated by a simple equation: Prevalence ≈ Incidence × Duration. This formula highlights how a disease’s occurrence in a population is a function of how often new cases arise (incidence) and how long people live with the condition (duration). Consequently, a change in prevalence can be due to a change in incidence (e.g., a new epidemic leads to higher incidence) or a change in duration (e.g., a new drug extends the life of patients without curing them, increasing duration and therefore prevalence).
Understanding this relationship is crucial for interpreting public health data. If prevalence is rising, an epidemiologist must determine if it is due to more people getting sick (higher incidence, indicating a need for prevention) or people living longer with the disease (higher duration, indicating a need for chronic care and support services). The choice between measuring incidence and prevalence depends entirely on the research question: incidence for etiological studies and prevalence for burden assessment and resource planning.
In summary, the measures of disease frequency—counts, ratios, proportions, prevalence, and incidence—are the essential tools of epidemiology. They provide the necessary context to move beyond anecdotal observation to systematic, comparable data, forming the foundation of all efforts to monitor, understand, and control disease within populations.