Attack Rate

The attack rate is the proportion of persons exposed to a specific pathogen, toxin, or hazard who develop the disease or condition of interest within a specified time period.^[2][3] Despite its name, it is technically a proportion (risk) rather than a true rate, and it is most commonly used during outbreak investigations to quantify the risk of illness among exposed individuals.^[2][3]

The attack rate is a fundamental measure in outbreak epidemiology because it directly quantifies the risk of disease among people who were exposed to a particular source. Public health investigators use it to identify the source of an outbreak, evaluate the effectiveness of containment measures, and communicate risk to the public and to clinicians. Comparing attack rates across different exposure groups (for example, people who ate a specific food item versus those who did not) is a primary method for identifying the vehicle of transmission in foodborne or waterborne outbreaks.^[2] Attack rates also inform planning for medical countermeasures, such as determining how many doses of vaccine or antibiotic prophylaxis are needed.

The attack rate is calculated by dividing the number of new cases of disease among exposed individuals by the total number of exposed individuals, then multiplying by 100 to express the result as a percentage.^[2] For example, if 30 out of 200 people who attended a catered event develop gastroenteritis, the overall attack rate is 15%.

In outbreak investigations, food specific or source specific attack rates are calculated to compare the risk of illness among people exposed to different potential sources. A two by two table is constructed for each suspected exposure, with cells representing the number of ill and well individuals among the exposed and unexposed groups. The attack rate is calculated separately for each group, and the relative risk (the ratio of attack rates) points to exposures associated with elevated risk.^[2]

The secondary attack rate is a related measure that quantifies person to person spread. It is calculated as the number of new cases among contacts of primary cases divided by the total number of susceptible contacts, providing a measure of the infectiousness of a disease within households or other close contact settings.^[2][3]

Attack rates can be calculated for specific subgroups defined by age, sex, geographic location, or other characteristics to identify populations at particular risk. Time specific attack rates (the proportion becoming ill within defined time intervals) show how an outbreak unfolds, and the epidemic curve of cases by onset time helps distinguish point source outbreaks from propagated outbreaks.^[2]

Use the attack rate during outbreak investigations to quantify risk and identify the source of transmission. It is appropriate for closed or well defined populations (such as attendees of an event, passengers on a cruise ship, or residents of a facility) where both the number of exposed individuals and the number of cases can be ascertained. It is less appropriate for open populations where the denominator is difficult to define or when the exposure period is prolonged and variable across individuals.

A county health department investigates a norovirus outbreak at a wedding reception attended by 180 guests. Through telephone interviews, they identify 54 guests who developed vomiting or diarrhea in the 48 hours after the reception (overall attack rate: 30%). Food specific analysis reveals that 48 of the 54 ill guests ate the raw oyster appetizer, compared with 44 of the 126 well guests. The attack rate among oyster eaters is 48/92 (52%), compared to 6/88 (7%) among non eaters, yielding a relative risk of 7.7 (95% CI: 3.4 to 17.0). This strongly implicates the oysters as the outbreak vehicle, and testing confirms norovirus contamination.