Population Based Estimates
Basic Tools of Epidemiologists: The development of a health information exchange network is well underway. Informaticians will be hard-pressed to meet all the needs of the various stake-holders. A significant stakeholder, the epidemiologists, will be anxious to start working with the new data that will soon be available to them.
As specifications are written for the various software parts, especially those that will service the public health epidemiologists, it will be advantageous and expected to implement functionality that either meets their needs or have interfaces with systems that do. These population-based mathematical tools are very familiar to the epidemiologist; indeed, without them, very little analytic data can be provided to improve population health. Knowledge of these basic data manipulations will help to prepare the informatics community for designing systems to meet the needs of the public health epidemiologist. 1
These basic statistics are used to summarize collected data so epidemiologist can learn about disease prevalence and rates. In the advent of full implementation of health records exchange, and even before that on a limited basis, these data will be used to analyze serious health conditions, and to evaluate and monitor outcomes. If causes for adverse events can be identified, then perhaps prevention and control measures can be identified also. It would serve the informaticist designing the new systems required by ARRA to be aware of the data needs of epidemiologists.2
Each population estimate description contains a purpose, definition, formula and an example.
Purpose: Collect information for estimation and hypothesis testing.
Definition: Prevalence of an outcome of interest is defined as the total number of cases in the population at a given time divided by the number of individuals in the population
Prevalence = All new and preexisting cases during a time period / population during the same time period
Example: The prevalence of high blood pressure in a population is the number of people with high blood pressure divided by the population who has high blood pressure added to those who don’t.
Purpose: Provide a measure of the number of new cases for the outcome of interest in a given time period for a population.
Definition: Incidence of an outcome of interest is defined as the measure of the risk of developing some new condition within a specified period of time.
Incidence = New cases occurring during a given time period / population at risk during the same time period
Example: The prevalence of high blood pressure in a population is the number of people who develop high blood pressure (a) divided by the population who has high blood pressure added to those who don’t (a + b) during 2010.
Purpose: Measures the degree to which the outcome of interest and the total population are related to one another.
Definition: Rate of an outcome of interest is defined as prevalence and incidence of the outcomes of interest during a specific time period
Rate = Number of cases occurring during a given time period / population at risk during the same time period
Example: The rate of high blood pressure in a population is the number of people with high blood pressure (a) divided by the sum of people with and without high blood pressure (a + b) during 2010.
Purpose: To measure the rate at which a subject at risk becomes a case during a brief period such as an outbreak.
Definition: Attack rate is a specific type of incidence rate. It is calculated for a narrowly defined population observed for a limited time, such as during an outbreak.
Attack rate = Number of new cases among the population during the period / population at risk during the period
Example: 50 people attend a picnic. 35 develop gastrointestinal distress. The attack rate is 35 (number of cases during the short duration period) divided by 50 (number in the population) times 100 or 70.0%.
Center for Disease Control
State of Missouri
State of Oregon
(1) Brian Edward Dixon. The Perceived and Real Value of Health Information Exchange in Public Health Surveillance. Indiana: Indiana University; 2011.
(2) Rabadan R, Calman MD N, Hripscsak G. Next Generation Syndromic Surveillance. Molecular Epidemiology, Electronic Health Records and the Pandemic Influenza A (H1N1) Virus. PubMed Central 2009 August 22.
Submitted by Isolde Knaap