attack rate, in epidemiology, the proportion of people who become ill with (or who die from) a disease in a population initially free of the disease. The term attack rate is sometimes used interchangeably with the term incidence proportion. Attack rates typically are used in the investigation of acute outbreaks of disease, where they can help identify exposures that contributed to the illness (e.g., consumption of a specific food). The attack rate is calculated as the number of people who became ill divided by the number of people at risk for the illness.
In order to calculate an attack rate, a case definition, or set of criteria to define the disease of interest, must first be developed. Case definitions may be based on a constellation of clinical signs (e.g., fever with vomiting or diarrhea) or on serology (e.g., the presence of antibodies to the etiologic agent). The number of people who meet the case definition is represented in the numerator of the attack rate. The denominator of the attack rate is the number of people who are at risk of becoming ill. At-risk individuals are those persons who had the opportunity to be exposed to the disease—for example, all individuals who ate a certain food item.
The time over which case numbers are collected is defined by the specifics of the outbreak (e.g., those persons who develop the defined symptoms over a set period after an event). The period over which new cases occur can be a clue to the identification of the causative agent. In addition, individuals who did or did not become sick are interviewed to determine their exposures, and attack rates for those who were or were not exposed are calculated. It may be impossible to determine the causative exposure from exposure-specific attack rates alone, since many of the subjects will have been exposed to multiple potential agents. A relative attack rate must be calculated for each exposure.