Epidemiology is the study of the occurrence, frequency, and distribution of diseases in a given population. As part of this study, epidemiologists—scientists who investigate epidemics (widespread occurrence of a disease that occurs during a certain time)—attempt to determine how the disease is transmitted, and what are the host(s) and environmental factor(s) that start, maintain, and/or spread the epidemic.
The primary focus of epidemiology are groups of persons, rather than individuals. The primary effort of epidemiologists is in determining the etiology (cause) of the disease and identifying measures to stop or slow its spread. This information, in turn, can be used to create strategies by which the efforts of health care workers and facilities in communities can be most efficiently allocated for this purpose.
In tracking a disease outbreak, epidemiologists may use any or all of three types of investigation: descriptive epidemiology, analytical epidemiology, and experimental epidemiology.
Descriptive epidemiology is the collection of all data describing the occurrence of the disease, and usually includes information about individuals infected, and the place and period during which it occurred. Such a study is usually retrospective, i.e., it is a study of an outbreak after it has occurred.
Analytical epidemiology attempts to determine the cause of an outbreak. Using the case control method, the epidemiologist can look for factors that might have preceded the disease. Often, this entails comparing a group of people who have the disease with a group that is similar in age, sex, socioeconomic status, and other variables, but does not have the disease. In this way, other possible factors, e.g., genetic or environmental, might be identified as factors related to the outbreak.
Using the cohort method of analytical epidemiology, the investigator studies two populations, one who has had contact with the disease-causing agent and another that has not. For example, the comparison of a group that received blood transfusions with a group that has not might disclose an association between blood transfusions and the incidence of a bloodborne disease, such as hepatitis B.
Experimental epidemiology tests a hypothesis about a disease or disease treatment in a group of people. This strategy might be used to test whether or not a particular antibiotic is effective against a particular disease-causing organism. One group of infected individuals is divided randomly so that some receive the antibiotic and others receive a placebo—a "false" drug that is not known to have any medical effect. In this case, the antibiotic is the variable, i.e., the experimental factor being tested to see if it makes a difference between the two otherwise similar groups. If people in the group receiving the antibiotic recover more rapidly than those in the other group, it may logically be concluded that the variable—antibiotic treatment—made the difference. Thus, the antibiotic is effective.
Although the sudden appearance of dreaded diseases has plagued humanity for millennia, it was not until the nineteenth century that the field of epidemiology can be said to have been born. In 1854, the British physician John Snow (1813-1858) demonstrated the power of epidemiologic principles during an outbreak of cholera in London. Snow discovered that most of the victims of cholera he questioned obtained their drinking water from a well on London's Broad Street. Moreover, most of the people afflicted with the disease drank from the polluted section of the Thames River, which ran through London. Snow arranged to have the Broad Street pump closed, preventing people from drinking water from that well. Subsequently, the cholera epidemic subsided.
Since the days of Snow, epidemiology has grown into a very sophisticated science, which relies on statistics as well as interviews with disease victims. Today, epidemiologists study not only infectious diseases, such as cholera and malaria, but also noninfectious diseases, such as lung cancer and certain heart disorders.
In the process of studying the cause of an infectious disease, epidemiologists often view it in terms of the agent of infection (e.g., particular bacterium or virus), the environment in which the disease occurs (e.g., crowded slums), and the host (e.g., hospital patient). Thus, beta-hemolytic streptococci bacteria are the agent for acute rheumatic fever; but because not all persons infected with the organism develop the disease, the health of the host helps to determine how serious the disease will be for a particular person, or even, whether it will occur.
Among the important environmental factors that affect an epidemic of infectious diseases are poverty, over-crowding, lack of sanitation, and such uncontrollable factors as the season and climate.
Another way epidemiologists may view etiology of disease is as a "web of causation." This web represents all known predisposing factors and their relations with each other and with the disease. For example, a web of causation for myocardial infarction (heart attack) can include diet, hereditary factors, cigarette smoking, lack of exercise, susceptibility to myocardial infarction, and hypertension. Each factor influences and is influenced by a variety of other factors.
By identifying specific factors and how they are ultimately related to the disease, it is sometimes possible to determine which preventive actions can be taken to reduce the occurrence of the disease. In the case of myocardial infarction, for example, these preventive actions might include a change in diet, treatment for hypertension, eliminating smoking, and beginning a regular schedule of exercise.
Epidemiologic investigations are largely mathematical descriptions of persons in groups, rather than individuals. The basic quantitative measurement in epidemiology is a count of the number of persons in the group being studied who have a particular disease; for example, epidemiologists may find 10 members of a village in the African village of Zaire suffer from infection with Ebola virus infection; or that 80 unrelated people living in an inner city area have tuberculosis.
Any description of a group suffering from a particular disease must be put into the context of the larger population. This shows what proportion of the population has the disease. The significance of 10 people out of a population of 1,000 suffering tuberculosis is vastly different, for example, than if those 10 people were part of a population of one million.
Thus one of the most important tasks of the epidemiologist is to determine the prevalence rate—the number of persons out of a particular population who have the disease:
Prevalence rate = number of persons with a disease / total number in group. Prevalence rate is like a snapshot of a population at a certain point in time, showing how many people in that population suffer from a particular disease. For example, the number of people on March 15 suffering infection from the parasite cryptosporidium in a town with a polluted water supply might be 37 out of a population of 80,000. Therefore, the prevalence rate on March 15 is 37/80,000.
A prevalence rate can represent any time period, e.g., day or hour; and it can refer to an event that happens to different persons at different times, such as complications that occur after drug treatment (on day five for some people or on day two for others).
The incidence rate is the rate at which a disease develops in a group over a period of time. Rather than being a snapshot, the incidence rate describes a continuing process that occurs over a particular period of time.
Incidence rate = total number per unit developing a disease over time / total number of persons. For example, the incidence rate of prostate cancer among men in a particular country might be 2% per year; or the number of children getting measles in a town might be 3% per day. Once a person has developed a lifelong disease, such as AIDS, he or she cannot be counted in the denominator of the incidence rate, since these people cannot get the disease again. The denominator refers only to those in the population who have not yet developed the disease.
Period prevalence measures the extent to which one or all diseases affects a group during the course of time, such as a year.
Period prevalence = number of persons with a disease during a period of time / total number in group. In the case of a year, such as 1995, the period prevalence equals the prevalence at the beginning of 1995 plus the annual incidence during 1995.
Epidemiologists also measure attributable risk, which is the difference between two incidence rates of groups being compared, when those groups differ in some attribute that appears to cause that difference. For example, the lung cancer mortality rate among a particular population of non-smoking women 50 to 70 years old might be 20/100,000, while the mortality rate among woman in that age range who smoke might be 150/100,000. The difference between the two rates (150-20 = 130) is the risk that is attributable to smoking, if smoking is the only important difference between the groups regarding the development of lung cancer.
Epidemiologists arrange their data in various ways, depending on what aspect of the information they want to emphasize. For example, a simple graph of the annual occurrence of viral meningitis might show by the "hills" and "valleys" of the line in which years the number of cases increased or decreased. This might provide evidence of the cause and offer ways to predict when the incidence might rise again.
Bar graphs showing differences in rates among months of the year for viral meningitis might pinpoint a specific time of the year when the rate goes up, for example, in summertime. That, in turn, might suggest that specific summertime activities, such as swimming, might be involved in the spread of the disease.
One of the most powerful tools an epidemiologist can use is case reporting: reporting specific diseases to local, state and national health authorities, who accumulate the data. Such information can provide valuable leads as to where, when, and how a disease outbreak is spread, and help health authorities to determine how to halt the progression of an epidemic-one of the most important goals of epidemiology.
Cohn, Victor. News and Numbers. A Guide to Reporting Statistical Claims and Controversies in Health and Related Fields. Ames: Iowa State University Press, 1989.
Nelson, K.E., C.M. Williams, and N.M.H. Graham. Infectious Disease Epidemiology: Theory and Practice Gaithersburg: Aspen Publishers, 2001.
Perera, F.P., and I.B. Weinstein. "Molecular Epidemiology: Recent Advances and Future Directions." Carcinogenesis 21 (2000): 517-524.
"Pidemics Of Meningococcal Disease. African Meningitis Belt, 2001." Weekly Epidemiological Record / World Health Organization 76, no. 37 (2001): 282-288.