Epidemiology is the study of the distribution and determinants of disease. It is based on the assumptions that disease does not occur randomly and that both causal and preventative factors can be identified. The evaluation of both human disease and pharmacological therapy goes through a specific sequence of events. Initially, there is suspicion of influence (eg, environmental, genetic, behavioral, or therapeutic) on an individual or disease. Next, a hypothesis is formed, followed by the systemic collection of data that includes an appropriate comparison group. Statistical analysis of the data will determine whether outcomes associated with the risk factor or intervention are different than outcomes in their absence. Finally, the validity of the statistical analysis is assessed by accounting for chance, biased data interpretation, and the presence of any confounding variables. Only then can judgment be made as to the importance of the risk factor or intervention under review.
MEASURES OF DISEASE FREQUENCY AND ASSOCIATION
Incidence and prevalence are the two basic measures of disease frequency used to qualify disease in a population. Incidence rates are designed to measure the rate at which previously healthy individuals develop the disease within a specific period of time, that is, the number of new cases in a population over a period of time. Prevalence rates measure the number of people in a population who have the disease at a given point in time.
CLASSIFICATION OF STUDY DESIGN
There are many different types of epidemiological study design ranging from observational to interventional. The two basic types of observational studies are the cohort study and the case–control study. The cohort, or prospective, study classifies patients based on the presence or absence of a risk factor and follows them through time to determine when and if they develop disease. Prospective studies have limitations; they take many years to complete and require many subjects leading to high cost and attrition. Since these studies are less susceptible to bias, they can obtain a true measure of incidence, leading to an accurate relative risk.
On the other hand, a case–control, or retrospective, study compares the proportions of patients with various exposures in a group of patients with a disease to a group without the disease. Such a study is most useful for diseases with a low incidence and for a group representative of the general population. These studies are usually performed quickly, easily, and inexpensively. The major limitation of a retrospective study is that the descriptive statistic, odds ratio, is only an ...