Outbreak epidemiology is the investigation of a disease cluster or epidemic with the goal of controlling or preventing further disease in a population. The word epidemic—defined as an increase in the number of cases of a disease above what is expected—is derived from the Greek epi and demos, meaning “that which is upon the people.” This meaning enriches our understanding of epidemics by emphasizing the burdensome toll they have on a population or “a people.” This definition also allows us to realize that epidemics are not always caused by infectious agents. Many other hazards, such as chemicals or physical conditions, can cause unusually high numbers of cases of disease in a given population. Regardless of the etiology of the disease, the techniques used to investigate and control outbreaks are generally similar.

Accounts of outbreaks have been recorded throughout the centuries. Cholera, influenza, malaria, smallpox, and the plague are extensively documented as causes of epidemics and pandemics that have altered the outcome of wars, disrupted political structures, killed millions of people, and affected the lives of countless others.¹ The prevention of disease through measures such as improved sanitation, as well as isolation and quarantine, was recognized even before the causes of these diseases had been identified.

New challenges continue to arise in the control of infectious diseases. Public health officials and healthcare providers face the emergence of new diseases and the reemergence of diseases that were no longer thought to be a threat to the public’s health.² The threat of the use of biologic agents or their toxins (e.g., anthrax, plague, botulism, and smallpox) as weapons of bioterrorism has resulted in increased attention being paid to the possibility of outbreaks caused by intentionally released agents.³ Changes in the environment; industrial practices; agriculture and food processing; international transportation of people, foods, and goods; and changes in human behaviors have increased the risk of disease and the speed at which communicable disease can spread.

Concurrently, the number of people at increased risk of infection and the density of human populations have increased. People with immune dysfunction (e.g., because of human immunodeficiency virus infection, rheumatologic conditions, hematologic conditions, organ transplantation, cancer chemotherapy, or chronic use of corticosteroids), infants, and the elderly are more susceptible to infectious diseases, including those caused by infectious agents that may not have been medical concerns in the past. These compromised individuals may present with lower temperature or with unusual symptoms that complicate the diagnosis, and their infections may be more difficult to treat. Immunosuppression may also increase the infectiousness of the individual, either by increasing the number of infectious organisms that the individual sheds or by increasing the length of time during which the individual is infectious.⁴ Meanwhile, the increasing density of the human population, especially in some developing nations, creates situations that foster the spread of new and old infectious diseases.

Nevertheless, the advantage is not all to the microbes. Advances in laboratory techniques, medical interventions (such as antibiotics), sanitation, technology, and epidemiology have increased our ability to identify and control infectious diseases.⁵

The news media and other media such as the Internet and social media have enhanced outbreak awareness. Names of organisms and events such as Escherichia coli O157:H7, Ebola virus, Cryptosporidium, group A Streptococcus (“the flesh-eating bacteria”), antibiotic-resistant organisms (methicillin-resistant Staphylococcus aureus [MRSA]), severe acute respiratory syndrome, and
pandemic influenza conjure up reports of outbreaks that have been brought to the attention of the public through the media. That E. coli O157:H7 can cause fatal hemolytic uremic syndrome, especially in children, was demonstrated in U.S. outbreaks caused by contaminated ground beef and fresh spinach and in the Japanese outbreak caused by contaminated radish sprouts.⁶, ⁷ and ⁸ The speed at which international diseases could become threats to U.S. citizens was underscored by the reports of Ebola virus isolated in monkeys in Reston, Virginia.⁹ Reports of outbreaks involving strains of Mycobacterium tuberculosis, MRSA, and Streptococcus pneumoniae that are resistant to antibiotics have raised public awareness of the danger of inappropriate or excessive antibiotic use. Investigation of such events allows epidemiologists to identify risk factors and to determine preventive measures that will limit and control the spread of disease.

An outbreak may occur for a variety of reasons, including poor food handling practices and personal behaviors, environmental contaminants, and intentional acts of terrorism. An epidemiologist must investigate cases of disease with an awareness of all these possibilities. However, the basic techniques in investigation remain the same. This chapter reviews the methods used to investigate and control outbreaks of infectious disease.

Outbreaks may come to the attention of health professionals through a report from a doctor’s office, a hospital, a nursing home, a laboratory, or even a patient’s call to the health department. Alternatively, outbreaks may be recognized through analysis of routine public health surveillance reports of individual cases of disease or through active surveillance for specific syndromes. Refer to the chapter on surveillance for additional information and methods relating to surveillance of diseases.¹⁰

Each state has its own individual laws, regulations, or both that govern which diseases and conditions are reportable and specify the method and timing of reporting; thus there are some variations across states. Nevertheless, reporting of outbreaks is generally included in most states’ disease reporting systems. The list of diseases for inclusion in national surveillance is established by the Council of State and Territorial Epidemiologists (CSTE) and the federal Centers for Disease Control and Prevention (CDC) and is updated yearly.¹¹ States then voluntarily report to the CDC cases of diseases that are on the “nationally notifiable” list.

In traditional passive public health surveillance, reports can originate from a variety of sources, including physicians, laboratories, hospitals, schools, childcare centers, vital records departments, and other facilities (Figure 5-1).¹² Cases are generally reported to state or local health departments with information such as diagnosis, name, age, gender, address, and date of onset. Reports may also include other information, such as laboratory results, treatment, occupation, setting of occurrence, and risk factors (Figure 5-2 and Figure 5-3).¹³, ¹⁴ Case reports, without personal identifiers, are then transmitted weekly from states to the CDC for inclusion in national summary data published in Morbidity and Mortality Weekly Report.¹⁵

Information collected as part of the disease reporting system is compiled and evaluated at local, state, and federal levels. Outbreaks are often first identified at the local or state health department level. Multistate outbreaks can sometimes be detected at the federal level by identification of increases in routine surveillance reports to CDC, the Foodborne Disease Active Surveillance Network (FoodNet), or early detection systems such as PulseNet. FoodNet, a collaborative project among CDC, 10 state health departments, the U.S. Department of Agriculture (USDA), and the U.S. Food and Drug Administration (FDA), consists of active surveillance for foodborne diseases; surveys of laboratories, physicians, and the general population; and population-based epidemiologic studies. Personnel in state health departments contact local laboratories to obtain reports of selected foodborne infections diagnosed in residents of these areas.¹⁶ PulseNet, whose activities are also coordinated by CDC, is a network of public health and food regulatory agency laboratories that perform molecular subtyping of foodborne bacteria; this information is submitted electronically to CDC and maintained in a standardized database. This national database allows for rapid comparison of molecular patterns.¹⁷ Detection of outbreaks can also occur through specialized analytic routines for aberration detection or other systems that use early indicators such as “syndromic surveillance”¹⁰, ¹⁸ or surveillance for syndromes of diarrhea, rash, or respiratory symptoms.

The information collected as part of the surveillance network is made available for additional epidemiologic evaluation. These data represent years of continuous data collection and, therefore, can be used to examine disease trends in a community.
However, these data are not collected primarily for the conduct of research or study. Rather, surveillance data are collected to detect disease, to describe cases found in a specific community, and to guide public health actions. Ideally, epidemiologic research data would be collected to address specific hypotheses and to produce generalizable knowledge. Surveillance data are collected according to those procedures that maximize consistency and minimize the barriers to reporting. For example, surveillance databases often include data collected by passive reporting and have only minimal information on cases. In contrast, more detailed data collection may be required for research purposes to examine specific hypotheses, such that employees may be dedicated to managing complex data collection systems. Because of these differences, surveillance data may be inadequate to answer some epidemiologic questions. Even so, surveillance data are an excellent source of information to establish baseline rates and detect outbreaks, to identify new problems or trends in a community, to guide public health actions, to evaluate programs, to assist health professionals in estimating the magnitude of a health problem, and to identify possible hypotheses that can be explored by enhancing surveillance or by a research study.

Because outbreak epidemiology is concerned primarily with the control of disease, the sensitivity of the surveillance or reporting network is of paramount importance. If cases of disease are not reported, an outbreak may not be detected or may continue unabated.

Based on experience with outbreak investigations, a series of steps has been identified that can be used to guide any epidemiologic field investigation.¹⁹, ²⁰, ²¹, ²², ²³ and ²⁴ The outbreak epidemiologist is the “disease detective” of public health. Outbreak investigation is a systematic process of evaluating data to form hypotheses, and then collecting additional data to test the hypotheses. An understanding of the basic steps of outbreak epidemiology can guide the investigator in determining which types of data to collect and how to collect them; each outbreak is unique, however, so it is also important to be aware of how an outbreak
under current investigation differs from previous outbreaks.

Most commonly, outbreak investigations, including laboratory support, occur at the local or state public health level. Generally, the CDC is consulted in multistate outbreaks or in outbreaks of diseases that require resources or skills that state and local agencies are unable to provide. The laboratory and epidemiologic capabilities of the CDC are enlisted to assist with domestic and international disease outbreaks of unusual etiology or of major public health significance.

Local and state health departments decide to conduct an outbreak investigation based on health regulations and/or policies and on the professional judgment of staff regarding the outbreak and its
public health impact or implications. Regardless of the etiologic agent, the setting in which the disease may have been transmitted, or the population at risk, it is possible to summarize the basic steps and goals of the investigation. The primary motivation of any outbreak investigation is to control the spread of disease within the initial population at risk or to prevent the spread to additional populations. Many outbreaks occur in defined populations, groups, or settings, such as a foodborne outbreak at a wedding banquet or a potluck dinner. Fortunately, these outbreaks generally have a single exposure, where the contaminated vehicle may have been consumed or discarded before the first cases are apparent and no more people are at risk. In other outbreaks, where transmission may be ongoing, such as legionellosis among hospitalized patients or disease caused by commercial products or fruits and vegetables, the epidemiologic investigation must be initiated quickly and control measures implemented to halt further transmission. Prevention of disease requires that the investigation identify the etiologic agent, its source,
the mode of transmission, and the vehicle. The information learned during the course of investigation is important in preventing and controlling outbreaks of the same disease in the future and may be useful in linking sporadic cases to the same source.

The steps for conducting an outbreak investigation are outlined in Figure 5-4.²⁴ A hallmark of outbreak epidemiology is that these steps do not necessarily proceed in a specified sequence. In actuality, several steps in the investigation usually occur simultaneously. These steps are tailored to the situation and depend on factors such as the urgency of implementing control measures; the availability of staff, resources, and time; and the difficulty of obtaining the data. Multiple persons may perform activities concurrently. Action and reaction proceed based on new and cumulative information. Because implementation of control measures is central to the goal of any outbreak investigation, measures to control the spread of disease must be implemented early in the investigation and may be altered as data are collected and analyzed.