Nonexperimental studies in which cases of a particular disease are selected and the patient’s exposure in the past is compared with that of people without the disease. This type of study is suitable for studying rare diseases. The numbers of subjects needed are small compared to those needed in cohort studies. Since the cases are selected without knowing the size of the source population at risk from which they arose, no information on the incidence rate of the disease in the population is obtained in these studies. Consequently, the relative risk cannot be calculated but is approximated by calculating an odds ratio. The advantage of this study design is that exposure and disease are both measured at the same time, and therefore one does not have to wait as long as in the cohort design. In this type of study, however, valid assessment of exposure may be a problem, since exposure in the past is measured after the disease has occurred. The disease may have affected recollection of the exposure by the subject. For diseases with a long latency period information on exposure in the distant past is needed. This may be impossible.
A design for human research that involves matching a group of people (known as the case patients) who have a particular disease, chronic condition, or injury with another group of people (known as the controls) why lack the disease, condition, or injury in order to detect the differences between them. Although case control studies are considered statistically crude, they are often the most practical method for studying harmful outcomes, such as cancer caused by exposure to an environmental pollutant or industrial chemical.
Research in which ‘cases’ — that is, persons with a particular condition or receiving a particular medication — are compared with a group similar in age, sex distribution and social class, etc. who did not have the condition or who are not receiving the drug.