Outbreak Investigation

Outbreak investigations are a subgroup of epidemiologic studies called ‘field investigations.’ When the numbers of persons affected by a particular disease, usually infectious, exceeds the number of cases expected in a given place during a given time period, it may be said that there is an outbreak of that disease. Epidemiologists may then conduct targeted investigations to (a) determine the cause and etiology of the disease, (b) to limit the spread and severity of illness of the disease, and (c) to prevent future outbreaks. In addition, investigations of this sort can serve to identify new modes of transmission of illnesses, identify new pathogens, and monitor the effectiveness of prevention activities. Collectively, these activities make up an outbreak investigation. Investigations of this type require epidemiologists to seek out and collect information (via interviews, lab studies, etc.) from persons affected by the disease.

Though the precise order may vary, most outbreak investigations share many steps and tasks in common. First, investigators must determine that an outbreak actually exists. The presence of a potential outbreak can be detected by any of several sources, including health care workers, laboratory workers, the general public, formal disease surveillance systems, or other health data. Investigators then compare the numbers of currently observed cases with historical data for the similar time period in previous [Page 770]years to determine that the observed cases represent an actual outbreak. Several sources of data adequate for this sort of comparison exist and may include disease surveillance records, birth certificates, death certificates, hospital discharge information, and so on. Changes in observed numbers of cases may be due to reasons other than the presence of an outbreak, for instance if there was an underlying change in case ascertainment or a change in the population at risk for the disease.

In addition to confirming the presence of an outbreak, investigators must also confirm the diagnosis and generate a case definition. The case definition, ideally consisting of the simplest, most concrete criteria possible, will help investigators and health care workers identify persons to be included in the investigation, and may be refined as new information about the illness and at-risk populations come to hand. Interviews with or surveys of cases will help place each case within the epidemiologic triad of person, place, and time, and can help refine ideas and initial hypotheses regarding who is at risk as well as beginning to address the issues of how and why the outbreak began. In an iterative process, investigators will continue to refine their case definition and explanatory hypotheses as new information comes to hand via interviews and surveys.

Once some of this preliminary descriptive work has been completed, investigators will plan and conduct an analytic study to further identify the source of the infection. Most commonly, case-control study designs are used, though cohort studies can also be used. As in more controlled study settings, both study designs have their advantages and disadvantages. Case-control studies are often used in the context of a large outbreak, where relative efficiency in both time and cost is important. In addition, in many outbreaks, the entire cohort is often not clearly defined, making a case-control study approach more appropriate. Case-control studies may also be nested within larger cohort studies, where testing a specific hypothesis on the entire larger cohort is not feasible. Cohort study designs have the advantages that investigators can evaluate multiple disease outcomes and can directly measure attack rates.

...