What is epidemiology?
Although epidemiology is closely related to medicine, there are significant differences between the two fields. The main focus of medicine is to diagnose and treat diseases in individuals, while the core purpose of epidemiology is to identify factors that cause health problems and control diseases in populations. The health of a population is the responsibility of the field of public health, and epidemiology is a tool for public health. Epidemiology studies disease distribution in populations (for example, how often a disease occurs in different groups of people), examines determinants of diseases or risk factors that increase the risk for disease development, and evaluates strategies to prevent and control diseases in communities.
Diseases have certain patterns in populations. Some groups of people are at a higher risk for a particular disease. For example, smokers are at a higher risk for lung cancer. A key feature of epidemiology is the measurement of disease outcomes in relation to a population at risk. The concept of a population at risk can be explained by the traditional epidemiological triangle model. In this model, the three angles are agent, host, and environment. The interrelationship of these three factors is the basis of development of disease in the population.
In the triangle model, the agent is the cause of the disease and includes four main categories: biological, physical, chemical, and nutritive. Biological agents are often infectious. The common infectious agents that cause disease are bacteria, viruses, and parasites. Physical agents are related to mechanics, temperature, radiation, noise, and so forth. Chemical agents are often linked to poisons and air or water pollution. Nutritive agents are the macronutrients and micronutrients that the human body needs. Excess or deficiency in these nutrients can cause health problems.
The second aspect in the triangle model is the host—the intrinsic factors that influence exposure, susceptibility, or response of an individual to an agent. Such intrinsic factors include age, gender, ethnic group, immunity, heredity, and personal behavior. For example, older age increases the risk for many diseases, such as heart disease and stroke. Certain ethnic groups also have increased risks for certain diseases, such as a high incidence of breast cancer in Jewish women.
The third component of the triangle model is the environment, which consists of the surroundings and conditions external to the individual that allow disease transmission or occurrence. The environment consists of physical, biologic, and socioeconomic components. Geology and climate are some examples of physical environment. Biologic environment may include population density, age distribution, and food sources. Socioeconomic environment may include degrees of industrialization and urbanization, use of technology, job security, cultural practices, and the availability of health care.
The primary mission of epidemiology is to investigate the interrelationship among agent, host, and environment of a disease in a population and disrupt the connection at some point in the triangle, so that the disease can be prevented. Some typical epidemiological activities include identification and surveillance of individuals and populations at risk for diseases, monitoring of diseases over time, identification of risk factors associated with diseases, recognition of disease transmission mode, and evaluation of the effectiveness of public health programs.
A specialist of epidemiology is an epidemiologist, who usually possesses a graduate degree in epidemiology with additional training in disease, public health, and biostatistics. The main responsibility of an epidemiologist is to investigate all elements contributing to the occurrence or absence of a disease in populations. Epidemiologists may work at all levels of communities, including academic or research institutions, federal governmental agencies, state health departments, and local health organizations or medical centers.
The techniques or methods that epidemiologists use to investigate diseases in populations are epidemiological studies, which mainly consist of cross-sectional studies, case-control studies, and cohort studies.
Cross-sectional studies are also called descriptive epidemiology, because this method describes the distribution of diseases or health-related events and the exposure status of risk factors in terms of person, place, and time. Describing disease distribution by person allows epidemiologists to determine the disease frequency and which populations are at greatest risk. Populations at high risk for a disease can be identified by investigating such characteristics as age, gender, race, education, occupation, income, living arrangement, health status, smoking status, physical activity level, medication use, and access to health care. Disease frequencies can be observed specifically for any of these characteristics by different classifications. For example, hypertension occurrence can be observed by physical activity levels, such as low, medium, and high. Through comparisons of hypertension frequency among the three levels of physical activity, the group with the highest hypertension rate can be identified.
Describing disease distribution by place can provide information associated with the geographic extent of the disease. This information includes county, state, country, birthplace, and workplace. Identifying place allows epidemiologists to examine where the causal agent of disease resides and how the disease is transmitted and spread. Describing distribution by time can reveal any seasonality of the disease and trends over time. Some diseases may be more common during a certain season; for example, influenza is more likely to be seen in winter and early spring. By tracking disease trends over time, changes in disease distribution, either emerging or declining, can be documented, and corresponding measures can be taken in response to these changes.
From a cross-sectional study, a group of people with an increased risk for a disease may be identified. The next step is to ask why this group of people has a higher risk for the disease. To answer this question, epidemiologists use case-control studies and cohort studies. Both of these methods are considered analytical studies, as they examine the relationship between a disease and its possible risk factors.
A case-control study begins with the selection of a group of cases—the case group, individuals who have the disease or health-related outcome of interest. Then, through interviews or medical records, epidemiologists collect information about the previous exposure of case group members to possible risk factors. Because case-control studies obtain information about risk factors in the past, they are also called retrospective studies. Certain demographic variables, such as age, gender, race, occupation, education, and residence, are collected as well and are used as the criteria to select the control group by matching the control subjects to the cases as closely as possible with respect to the demographic variables. No individuals of the control group should exhibit the disease or health-related outcome under investigation. Information on previous exposure to risk factors is also collected from the controls.
Matching controls to cases allows the investigators to ignore the demographic variables and focus on risk factors in the analysis. Control subjects can match the cases individually or as a group. The ratio of cases to controls can be one to one, one to two, or more. Increasing the number of controls can increase the power of the study to detect the differences between cases and controls; however, a large number of controls can increase the cost of the study as well.
The case and control groups are then compared for previous exposure to the risk factors of the disease using statistical analyses. The association and the strength of association between risk factors and the disease under investigation are evaluated. The results of a case-control study may be a positive association, in which the risk factors increase the chance of seeing the disease; a negative association, in which the risk factors decrease the frequency of the disease; or no association, in which no relationship is found between the risk factors and the disease. For example, to study whether obesity is associated with type 2 diabetes, the researcher would select a group of diabetic cases and a group of controls who do not have diabetes but have similar demographic variables, such as age, gender, and occupation. Next, the history of weight would be assessed though interviews of both cases and controls. The weight history of the diabetic cases would then be compared to that of the nondiabetic controls. In this example, one would be likely to see a positive association between obesity and diabetes, which means that obesity is more frequently seen in the diabetic cases.
Cohort studies are used to examine the causal relationship between a disease or health-related outcome and its risk factors. The cohorts, or groups being studied, are identified by characteristics of risk factors exhibited by subjects prior to the appearance of the disease under investigation. Thus, one cohort may consist of subjects with risk factors for a disease, while another cohort may include subjects without such risk factors. In both cohorts, no subjects should have the disease under investigation at the beginning of the study. The research would then follow both cohorts for a set period of time; therefore, cohort studies are also called prospective studies or longitudinal studies.
During the follow-up period of a cohort study, the difference in the occurrence of the disease under investigation will be recorded and compared between the two cohorts. The results of a cohort study may also be positive, negative, or no association, as determined through statistical analyses. A positive association means the incidence of the disease is increased in the cohort with the risk factors. A negative association indicates that the incidence of the disease is decreased in the cohort with the risk factors, which can then be hypothesized to protect individuals from getting the disease—“good” risk factors. If no statistical differences are identified between the two cohorts, then the risk factors are not associated with the disease. A cohort study might study the relationship between cholesterol level and coronary artery disease. Individuals with a high cholesterol level would be included in one cohort, and individuals with a normal cholesterol level would be included in another cohort. Then, both cohorts would be followed up for a period of ten years. At the end of ten years, the incidence of coronary artery disease that has been diagnosed during that time would be evaluated and compared between the two cohorts. In this study, it is very likely that the cohort with a high cholesterol level would have a higher incidence of coronary artery disease during the ten years of follow-up. A cohort study with only two cohorts is the simplest design, but a study may use more than two cohorts, as long as each cohort has the unique risk factor characteristics.
There are advantages and disadvantages to both case-control studies and cohort studies. Cohort studies observe a disease from cause to effect and thus generate more accurate results; however, they are time-consuming and expensive. Case-control studies are quick and inexpensive, but their results are less accurate, since they are based on self-reported past experiences, which often encounter recall biases. In practice, epidemiologists often carry out a case-control study first. If the study shows a significant association, then a cohort study is used to confirm the association.
Literally translated from Greek, epidemiology means “the study of people”—the population-level study of disease. Epidemiology began with eighteenth-century London physician John Snow, who investigated an epidemic of cholera in the city. By observing and plotting the location of deaths related to the disease, Snow was able to demonstrate that cholera was spread through contaminated water and food.
In its early years, epidemiology was mainly used to study epidemics of infectious diseases, because infectious diseases were the major cause of death in populations at that time. Through improvements in nutrition, sanitation, and living standards, as well as advances in medicine, the major cause of death has shifted from infectious diseases to noninfectious or chronic diseases in developed countries. Epidemiology has now been applied to chronic diseases as well as conditions such as cancer, heart disease, diabetes, and injuries. The Framingham Heart Study is a famous epidemiological study of cardiovascular disease in residents of Framingham, Massachusetts. Epidemiological methods have been approved as a powerful tool to study diseases or other conditions in populations and have also been applied to other fields, such as sociology.
In the future, the use of epidemiological methods will continue to increase, allowing a better understanding of more human diseases and their causes. Because of improved medical technologies, epidemiology has been able to combine traditional observational methods with laboratory tests. New branches of epidemiology have been created, such as molecular epidemiology and genetic epidemiology. Research in these areas will yield knowledge about human diseases at a new level.
Centers for Disease Control and Prevention. Epidemiology and Prevention of Vaccine-Preventable Diseases. 12th ed. Atlanta: Centers for Disease Control and Prevention, 2012. Print.
Day, Ian N. M., ed. Molecular Genetic Epidemiology: A Laboratory Perspective. New York: Springer, 2002. Print.
Fletcher, Robert H., and Suzanne W. Fletcher. Clinical Epidemiology: The Essentials. 5th ed. Baltimore: Lippincott, 2014. Print.
Foxman, Betsy. Molecular Tools and Infectious Disease Epidemiology. Burlington: Academic, 2012. Print.
Friis, Robert H., and Thomas A. Sellers. Epidemiology for Public Health Practice. 5th ed. Burlington: Jones & Bartlett, 2013. Print.
Gordis, Leon. Epidemiology. 5th ed. Philadelphia: Saunders/Elsevier, 2013. Print.
Lilienfeld, David E., and Paul D. Stolley. Foundations of Epidemiology. 3rd ed. New York: Oxford UP, 1994. Print.
Newman, Stephen C. Biostatistical Methods in Epidemiology. New York: Wiley, 2001. Print.
Porta, Miquel, ed. A Dictionary of Epidemiology. 6th ed. New York: Oxford UP, 2014. Print.
Rothman, Kenneth J. Epidemiology: An Introduction. 2nd ed. Oxford: Oxford UP, 2012. Print.