Observational Epidemiology

Epidemiological study designs also can be described according to a number of different aspects. One aspect is whether the study is observational or experimental. Investigators in observational studies may plan and identify variables to be measured, but human intervention is not a part of the process. Experimental studies, in contrast, involve intervention in ongoing processes to study any resulting change or difference. Epidemiological studies are also descriptive or analytical in nature. Descriptive studies attempt to uncover and portray the occurrence of the condition or problem, whereas analytical studies determine the causes of the condition or problem.

Observational epidemiology provides information about disease patterns or drug use problems by various characteristics of person, place, and time. This approach is used by public health professionals for efficient allocation of resources and to target populations for education, prevention, and treatment programs. It also is used by epidemiologists to generate hypotheses regarding the causes of disease or drug use problems. Some researchers do not consider experimental studies to be true epidemiological studies in the traditional sense because they follow clinical or planned research designs. Descriptive studies provide insight, data, and information about the course or patterns of disease or drug use problems in a population or group. Analytical studies are used to test cause–effect relationships, and they usually rely on the generation of new data.

There are five general types of epidemiological study design: (1) case reports and case series, (2) cross-sectional studies, (3) case-control studies, (4) cohort studies, and (5) experiments. A case report is a descriptive study of a single patient, and a case series is a collection of case reports. A cross-sectional study is a prevalence study—a basic descriptive study that examines relationships between a disease or drug use problem and other characteristics of people in a population at one point in time. A case-control study compares people who have the disease or problem (cases) to those who do not (controls) with respect to characteristics of interest (i.e., potential causes). A cohort study is an incidence study that measures characteristics or attributes in a population free of a disease or drug use problem and relates them to subsequent development of the disease in that population as it is followed over time. This type of study is also referred to as a longitudinal study. Experimental studies are clinical trials and intervention studies designed to compare outcomes between two or more treatment or intervention groups.

These different aspects and relationships of epidemiological study designs are summarized and portrayed in . At the basic level, investigators can exert control over the process by assigning subjects to study groups, such as in experiments, or they do not exert control and only observe what is occurring naturally. Experimental studies are further divided into studies that assess effects or outcomes in individual subjects or studies that assess effects in communities or large groups of people. Observational studies are divided into descriptive or analytical types, with the latter type being further defined by the focus of sampling for the exposure or disease.