Basic Principles of Epidemiology and Biostatistics

Abstract

Epidemiology is a discipline that studies the distribution and determinants of disease. The applications of epidemiology are diverse and include identifying causal mechanisms of disease, diagnostic testing, determining prognosis, and testing new treatments. The goal of an epidemiological study is valid measurement. To understand this goal, it is necessary to understand what it is that is being measured and what is meant by the term validity. The epidemiologist is usually interested in measuring the relationship or association between some exposure and some outcome. The first step is to measure the frequency of the exposure and/or the frequency of the outcome (e.g., disease) of interest. It is then possible to compare these measures of frequency between two populations. The demonstration that disease occurs more frequently within a population exposed to some risk factor than within a population without such exposure provides insight into the association between the exposure and the disease. “Validity” describes the extent to which a measurement is correct (i.e., reflects the truth). A major difficulty is that we typically do not know the truth. Ensuring validity, therefore, requires that we take precautions to minimize error. Broadly speaking, epidemiological studies are susceptible to two types or error—systematic error (also known as bias) and random error. Error is reduced by an awareness of its origins and by the use of appropriate study design.