What Are Epidemiology Studies?
Epidemiology studies are conducted using human populations to evaluate whether there is a
correlation or causal relationship between exposure to a substance and adverse health effects.
These studies differ from clinical investigations in that individuals have already been administered the drug during medical treatment or have been exposed to it in the workplace or environment.
Epidemiological studies measure the risk of illness or death in an exposed population compared to that risk in an identical, unexposed population (for example, a population the same age, sex, race and social status as the exposed population).
Figure 1. Epidemiology studies tend to produce graphs and charts for data analysis and presentation
(Image Source: Adapted from iStock Photos, ©)
Types of Studies
There are four primary types of epidemiology studies. They are:
- Cohort studies — A cohort (group) of individuals with exposure to a chemical and a cohort without exposure are followed over time to compare disease occurrence.
- Case control studies — Individuals with a disease (such as cancer) are compared with similar individuals without the disease to determine if there is an association of the disease with prior exposure to an agent.
- Cross-sectional studies — The prevalence of a disease or clinical parameter among one or more exposed groups is studied, such as:
- The prevalence of respiratory conditions among furniture makers.
- Ecological studies – The incidence of a disease in one geographical area is compared to that of another area, such as:
- Cancer mortality in areas with hazardous waste sites as compared to similar areas without waste sites.
Cohort studies are the most commonly conducted epidemiology studies and they frequently involve occupational exposures. Exposed persons are easy to identify and their exposure levels are usually higher than in the general public. There are two types of cohort studies:
- Prospective, in which cohorts are identified based on current exposures and followed into the future.
- Retrospective, in which cohorts are identified based on past exposure conditions and study "follow-up" proceeds forward in time; data come from past records.
Common Statistical Measures
Standard, quantitative measures are used to determine if epidemiological data are meaningful. The most commonly used measures are:
- Odds Ratio (O/R) — The ratio of risk of disease in a case-control study for an exposed group to an unexposed group. An odds ratio equal to 2 (O/R = 2) means that the exposed group has twice the risk as the non-exposed group.
- Standard Mortality Ratio (SMR) — The relative risk of death based on a comparison of an exposed group to non-exposed group. A standard mortality ratio equal to 150 (SMR = 150) indicates that there is a 50% greater risk.
- Relative Risk (RR) — The ratio expressing the occurrence of disease in an exposed population to that of an unexposed population. A relative risk of 175 (RR = 175) indicates a 75% increase in risk.
When designing an epidemiology study, the most critical aspects include:
- An appropriate control group.
- An adequate time span.
- The statistical ability to detect an effect.
More specifically, the control population used as a comparison group must be as similar as possible to that of the test group, for example, same age, sex, race, social status, geographical area, and environmental and lifestyle influences.
Many epidemiology studies evaluate the potential for an agent to cause cancer. Because most cancers require long latency periods, the study must cover that period of time.
The statistical ability to detect an effect is referred to as the power of the study. To gain precision, the study and control populations should be as large as possible.
Epidemiologists attempt to control errors that can occur in the collection of data, which are known as bias errors. The three main types of bias errors are:
- Selection bias, which occurs when the study group is not representative of the population from which it came.
- Information bias, which occurs when study subjects are misclassified as to disease or exposure status. Recall bias occurs when individuals are asked to remember exposures or conditions that existed years before.
- Confounding factors, which occur when the study and control populations differ with respect to factors which might influence the occurrence of the disease. For example, smoking might be a confounding factor and should be considered when designing studies.
Finally, for consumer products, postmarketing epidemiological studies can be performed. Examples include studies developed by Poison Control Centers, companies, academia, and other sources to look at the "real world" health reports of effects associated with consumer use of a product or article under reasonably foreseeable conditions.