Epidemiology: define the population, time and comparison
Epidemiology studies patterns of health and disease in populations and investigates factors associated with them. A count becomes more informative when its population, time period and definition are clear. Start with those details before interpreting a headline about increasing risk.
A case count needs a definition and a denominator
A case definition defines the people included in an investigation for a certain disease. The population refers to the group of individuals where the cases arose. If you do not have a denominator (population size), a larger number of cases could indicate a larger population size rather than an increased risk of developing the disease among individuals. The method of defining cases and gathering cases will determine comparisons between areas or time frames.
Assume two cities have the same total number of cases but totally different populations. In this instance, although both numbers are equivalent, the percentages are likely to vary. Prior to making a comparison, make sure that the case definitions and time frames utilized for each location match. Comparing numerically-based results when utilizing different types of cases or allowing one city to develop more cases over a longer time frame makes the comparison less relevant.
New cases and existing cases answer different questions
Incidence pertains to new occurrences emerging within a population at risk of developing the condition over a given time frame. Prevalence pertains to the cases of a condition present in a population at a particular point in time or over a designated period. Although many diseases can lead to a high prevalence due to the fact that individuals can live with the disease for extended periods, prevalence is not merely another way of saying incidence rates of diseases.
To assist in revising, create a timeline representing multiple individuals falling ill at various times. Calculate new cases occurring during an interval. Identify all individuals impacted by the condition on a single chosen date. Define the reason why the two responses differ. Leave the time reference in the last line; deleting it may convert a valid calculation into an ambiguous statement concerning the population.
An association needs interpretation beyond the calculation
There may exist an association between two variables based upon comparing values across groups; however, the mere existence of such an association does not imply causality. Other potential explanations for associations include random chance, differences in the manner in which participants were selected for inclusion in the sample, errors in measurement, and other confounding factors. A confounder exists when another variable affects the association being examined by introducing bias into the comparison being made between the groups under examination.
Consider an example of comparing fictitious groups. One group is older on average than the other. If age is correlated with the outcome measured, the crude comparison may not adequately address the question regarding an exposure. Specify what further information or additional analyses would provide insight into addressing this concern. The above illustrates why even a precisely calculated estimate can be misleading when the comparison is poorly designed.
Surveillance and analytic studies serve related but distinct purposes
Health surveillance is concerned with collecting and reviewing health information over time to facilitate public health actions. Analytic epidemiologic investigations examine specific research questions through an appropriate comparative strategy. Both require reliable definitions and high-quality data. Changes in reported cases may reflect changes in either disease occurrence, reporting behavior, testing practices, etc.; thus, interpretation requires knowledge of how the data were collected.
A useful study summary names the population, exposure or question, outcome, time period and main limitation. End with a conclusion no stronger than the evidence supports. Population findings can guide services and prevention, but they do not diagnose an individual or replace clinical care. Ask what was compared before interpreting the size of the difference.
Check your understanding
A short quiz, with an explanation after each answer.
What is needed to interpret a case count as a proportion?
Return to the section you found hardest and explain its example without looking at the page. Check the explanation, then repeat the quiz later. Being able to describe the reasoning is more useful than remembering the position of a correct answer.