Incidence

Incidence is an epidemiological measure that quantifies the frequency of new cases of a disease, injury, or other health event arising within a defined population during a specified period. Unlike prevalence, which counts all existing cases at a given point or over a period, incidence focuses exclusively on the transition from a nondiseased to a diseased state.

Incidence is one of the most fundamental measures in epidemiology and public health. It reveals the rate at which new disease occurs, providing direct evidence about risk and about the forces driving disease in a population. Public health officials use incidence data to detect outbreaks, evaluate the effectiveness of prevention programs, and allocate resources to areas with the highest disease burden. Researchers rely on incidence estimates when calculating sample sizes for clinical trials, when comparing disease risk across populations, and when modeling the future burden of disease. Without accurate incidence data, it is hard to distinguish whether a high disease burden reflects a large number of new cases or a long duration of existing cases.

Incidence can be expressed in two main forms. Cumulative incidence (also called incidence proportion or risk) is the proportion of an initially disease free population that develops the condition over a defined time period. It is calculated as the number of new cases divided by the number of individuals at risk at the start of the period. Cumulative incidence assumes that the entire population is followed for the full period, making it most appropriate for closed cohorts with complete follow up.

Incidence rate (also called incidence density or person time rate) accounts for the fact that individuals contribute different amounts of follow up time due to late entry, loss to follow up, or competing events. It is calculated as the number of new cases divided by the total person time at risk. Person time is the sum of the time each individual in the study population is observed and remains at risk. The units are typically expressed as cases per 1,000 person years or per 100,000 person years.

Both measures require a clear and consistent case definition, a method for ascertaining new cases (such as surveillance systems, disease registries, or prospective cohort follow up), and a well defined population denominator. Incidence can be further stratified by age, sex, geography, or other characteristics to identify high risk subgroups and to calculate age standardized rates that allow valid comparisons across populations with different age structures.

Use cumulative incidence when the population is fixed and all individuals are followed for the same duration, such as in a cohort study with a defined enrollment period and complete follow up. Use incidence rate when follow up times vary across individuals, as is common in large cohort studies, registry analyses, and surveillance systems. Incidence measures are appropriate whenever the research question concerns risk, etiology, or the dynamics of disease occurrence. They are less informative when the goal is to understand the total burden of disease at a given moment, for which prevalence is more appropriate.

A frequent error is confusing incidence with prevalence. Prevalence reflects the stock of disease (how many people have it at a given time), while incidence reflects the flow (how many people newly develop it). Another mistake is calculating cumulative incidence in an open cohort without accounting for varying follow up times, which underestimates the true risk. Researchers also sometimes fail to restrict the denominator to individuals at risk, including people who already have the disease or who are not susceptible to it. Finally, comparing crude incidence rates across populations without adjusting for differences in age structure can produce misleading conclusions.

A national surveillance program tracks new diagnoses of type 2 diabetes among adults aged 30 to 65 who had no diabetes diagnosis at the start of 2024: 12 million people at risk, 8 million aged 30 to 44 and 4 million aged 45 to 65. During 2024, 180,000 new cases are identified. Treating the group as followed for the whole year, the one year cumulative incidence is 180,000 divided by 12,000,000, or 1.5%. Because people stop contributing time at risk once they are diagnosed, die, or emigrate, the group contributed about 11.9 million person years at risk, so the incidence rate is about 15.1 new cases per 1,000 person years. When the data are stratified by age, the cumulative incidence is 0.8% (64,000 cases) among those aged 30 to 44 and 2.9% (116,000 cases) among those aged 45 to 65, revealing that older adults face more than three times the risk of new onset diabetes.