Glossary

Key Points

Basic concepts	Epidemiology is one of several sources of knowledge for Public Health. Electronic health records and genome wide association studies provide sources of Big Data for today’s epidemiologists. A population is a group of people that share a common characteristic. A population can either be fixed, or dynamic. Useful examples include catchment populations and candidate populations Members of a sample are chosen at random from a population, so that the statistics of the sample are suitably close to the statistics of the population. frequency measures estimate how many members of a population have a certain health-related state. The geographical or temporal variation of frequency is called distribution. Disease control can be achieved through surveillance programs. Newborn screening programs and national disease registries are important epidemiological contributions to sickle cell disease management.
Study designs	In experimental studies, an *investigator actively manipulates a therapeutic agent. Informed consent* is critical for experimental studies. In cohort studies, the starting point is *exposure* and the endpoint is *outcome. In case-control studies, the starting point is outcome* and endpoint is *exposure. Case definition* is an important aspect of case-control studies. *Cross-sectional studies* take a snapshot of disease prevalence in relation to exposure prevalence at a particular time. In *ecological studies*, the units of analysis are populations rather than individuals.
Data collection and Management	The scope and objectives of the study are essential in determining the data to be collected. There are two main categories of epidemiological data sources: primary and secondary data. Primary data is collected by an investigator for their specific study. Secondary data is that which has already been collected by others for different purposes. Data management involves all activities pertaining to the transfer and maintenance of data in a central database. The dimensions of data quality are Accuracy, Completeness, Reliability, Validity, and Timeliness. A study should have in place guidelines that ensure adherence to the general governing principles for data protection. It is very important to ensure that re-identification of study subjects is prevented.
Measuring and Comparing Disease Frequencies	To estimate the frequency of a disease in a population, epidemiologists must: (1) develop a concrete definition of the disease, (2) count the number of people affected by the disease, (3) determine the size of the population from which the disease cases arose, (4) and account for the passage of time. Location of residence, such as a country, state, city, or neighbourhood, is one of the most common ways to define a population. Incidence is defined as the occurrence of new cases of disease that develop in a candidate population over a specified time period. Cumulative incidence is defined as the proportion of a candidate population that becomes diseased over a specified period of time. Incidence rate is defined as the occurrence of new cases of disease that arise during person-time of observation. Prevalence measures the frequency of existing disease. Prevalence is defined either at a single point in time (point prevalence) or over a period of time (period prevalence). Examples of common disease frequency measures are crude mortality (or death) rate, cause-specific mortality rate, morbidity rate, and survival rate. If there is no relationship between the exposure and disease, the numeric value for the relative measure is 1.0 (\(RR = 1.0\)). If there is a positive relationship between the exposure and disease (i.e., the exposure increases the rate of disease), the numeric value is greater than 1.0 (\(RR > 1.0\)). If there is a negative relationship between the exposure and disease (i.e., the exposure decreases the rate of disease), the numeric value is less than 1.0 (\(RR < 1.0\)). A relative risk of 0.5 means that the exposed group has one-half the risk of the unexposed group, or a 50% decreased risk of disease. A relative risk of 0.33 means that the exposed group has one-third the risk, or a 67% reduction in risk, and so on. The relative measure of comparison can range from zero to infinity.
Critical appraisals of Journal articles	Critical appraisal is an important element of epidemiology and evidence based medicine.

public health: a multidisciplinary field whose goal is to promote the health of the population through organised community efforts.
epidemiology: the study of the distribution and determinants of disease frequency in human populations and the application of this study to control health problems.
population: a set of people that share one or more common characteristics, such as gender, place of residence, age, or use of certain medical services.
fixed population: A population whose membership is permanent. See also: dynamic population.
dynamic population: a population whose membership can change in time as members enter or leave. See also: fixed population.
catchment population: of a health facility is a population that consists of the people who use the facility’s services.
candidate population: a population whose members are at risk of an adverse change in health-related state, such as contracting a certain infectious disease.
sample: (of a population) a subset of the population where each member in the sample is chosen at random from the population, so that the statistics of the sample are suitably close to the statistics of the population.
health-related state: (of a member of a population) a state of a person that is health-related. Sometimes referred to by epidemiologists as a “disease”.
frequency: (of a health-related state) a measure of how often a health-related state arises in a population. Also called a frequency measure.
frequency measure: see frequency.
distribution: (of a health-related state) the change in frequency of a health-related state in a population with respect to the characteristics of person, the place, or time.
determinant: (of a health-related state) a factor that brings about a change in a person’s health-related state.
incidence: (of a health-related state) the proportion of a candidate population that changes their given health-related state over a specified period of time (or simply the occurrence of new cases).
prevalence: (of a health-related state) the proportion of the total population that has a given health-related state). See also: incidence.
point prevalence: (of a health-related state) the proportion of the population that has a given health-related state) at a single point in time. See also: prevalence, period prevalence.
period prevalence: (of a health-related state) the proportion of the population that has a given health-related state) during a specified duration of time. See also: prevalence, point prevalence.
exposure: a characteristic for which people are compared in an epidemiological study.
absolute comparisons: the subtracted measures of disease comparisons. Also called risk difference rate difference, incidence rate difference, cumulative incidence difference, and prevalence difference.
relative comparisons: the divided measures of disease comparisons. Also called risk ratio or relative risk.
randomization: an act of of assigning or ordering that is the result of a random process.
placebo: an inactive drug that is given to the group of participants in a trial that are compared with the group that receives the active drug.
noncompliance: the failure to observe the requirements of an experimental protocol.
cohort: a group of people with a common characteristic or experience.
fixed cohort: a cohort formed on the basis of an irrevocable event such as undergoing a medical procedure.
closed cohort: a fixed cohort in which there is no loss to follow-up.

Common measures of disease frequency

There are many measures of disease frequency that are commonly used in the public health disciplines. Some are incidence measures, some are prevalence measures, some are ratios. Descriptions and examples of the major measures follow. Note that the word rate is often used incorrectly to describe a proportion or ratio.

Crude mortality (or death) rate: Total number of deaths from all causes per 100,000 population per year. The term crude means that the rate is based on raw data. In 2015 the crude mortality rate in the United States was 844.0/100,000 population/year.
Cause-specific mortality rate: Number of deaths from a specific cause per 100,000 population per year. In 2015, the cause-specific mortality rate from heart disease in the United States was 197.2/100,000/year.
Age-specific mortality rate: Total number of deaths from all causes among individuals in a specific age category per 100,000 population per year in the age category. In 2015, the age-specific death rate was 589.6/100,000/year among U.S. children under the age of 1 year.
Years of potential life lost: The number of years that an individual was expected to live beyond his or her death. In 2015, a total of 957 years were lost from heart disease, 1,283 years were lost from cancer, and 1,172 were lost from unintentional injuries before age 75 per 100,000 population younger than 75 years of age in the United States. 18 The number of years of potential life lost reflects both the number of individuals who died of a particular cause and the age at which the death occurred. For example, a cause of death that is more common among children and young adults (such as unintentional injuries) will result in more years of life lost per individual than a cause of death that is common among the elderly (such as heart disease).
Livebirth rate: Total number of livebirths per 1,000 population per year. A livebirth is a pregnancy that results in a child who, after separation, breathes or shows any other evidence of life. Sometimes, the denominator includes only women of childbearing age. In 2015, the crude livebirth rate among women who were residents of the United States was 12.4/1,000/year.
Infant mortality rate: Number of deaths of infants less than 1 year of age per 1,000 livebirths per year. This statistic is often divided into neonatal deaths (those occurring during the first 27 days following birth) and post neonatal deaths (those occurring from 28 days through 12 months). In 2014, the infant mortality rate in the United States was 5.8/1,000 livebirths/year, the neonatal mortality rate was 3.9/1,000 livebirths/year, and the post neonatal death rate was 1.9/1,000 livebirths/year.
Birth defect rate (also called congenital anomaly or malformation rate): Number of children born with defects, usually per 10,000 births. The numerator and denominator often include both livebirths and stillbirths. In 2016–2017, the prevalence of brain malformations, including microcephaly, was 5% among women with evidence of recent possible Zika virus infection.
Morbidity rate: Number of existing or new cases of a particular disease or condition per 100 population. The time period that is covered and the population size in the denominator vary. Morbidity is a general word that can apply to a disease, condition, or event. For example, from 2011 to 2014, the prevalence of physician-diagnosed diabetes among U.S. adults aged 65 years and over was 20.6%.
Attack rate: Number of new cases of disease that develop (usually during a defined and short time period) per the number in a healthy population at risk at the start of the period. This cumulative incidence measure is usually reserved for infectious disease outbreaks. For example, the 24-hour attack rate for food poisoning was 50% among people who ate chicken salad at the banquet.
Case fatality rate: Number of deaths per number of cases of disease. Note that this measure is a type of cumulative incidence and therefore it is necessary to specify the length of time to which it applies. For example, in 2014 in the Democratic Republic of Congo, the 5-month case fatality rate among individuals with Ebola virus disease was 74.2%.
Survival rate: Number of living cases per number of cases of disease. This rate is the complement of the case fatality rate and is also a cumulative incidence measure. Five-year relative survival rates for cancer compare people with a particular cancer to similar people in the general population. For example, from 2007 to 2013, 5-year relative survival rates for prostate cancer were 100% among men diagnosed while the tumour was still confined to the prostate or had spread only to the regional lymph nodes and 29.8% among men whose tumour had metastasized to distant sites.

Epidemiology: Glossary

Key Points

Glossary

Common measures of disease frequency

Further reading

Epidemiology and Big Data

Precision public health