Neyman bias (also known as prevalence-incidence bias) is a type of bias that can occur in research studies in which extremely sick individuals or extremely healthy individuals are excluded from the final results of the study which may lead to biased results.
There are two ways in which this bias can affect the results of a study:
1. If extremely sick individuals are excluded from the study because they’ve died, then the disease will appear less severe.
2. If extremely healthy individuals are excluded from the study because they have recovered and been sent home, then the disease will appear more severe.
Examples of Neyman Bias
Here are two examples of Neyman Bias occurring in different scenarios:
Example 1: Sick individuals being excluded from a study.
Suppose a group of researchers at a hospital want to study the severity of a certain strain of flu. They randomly select a sample of 40 individuals in the area who contract that strain of flu and monitor their outcomes.
In this scenario, the individuals who contract a particularly severe case of the flu and happen to die from it will be excluded from the study. This means only individuals with mild cases will be included in the study, which will make the flu appear less severe.
Example 2: Healthy individuals being excluded from a study.
Suppose a group of researchers at a hospital want to study the severity of a certain seasonal cold. They randomly select a sample of 30 individuals in the area who contract the cold and monitor their outcomes.
In this scenario, the individuals who already contracted the cold and recovered will not be included in the study, which means only individuals with more severe cases who have not recovered will be included in the study. This could cause the cold to appear more severe.
In What Types of Studies Does Neyman Bias Occur?
Neyman bias occurs most often in studies in which there is a long time period between individuals contracting a certain disease and then being included in a study simply because this gives them more time to either (1) recover and not be included in the study or (2) die and not be included in the study.
Case-control studies are most susceptible to this type of bias, but it can also occur in cohort studies and cross-sectional studies.
How to Prevent Neyman Bias
There are two ways to avoid the pitfalls of Neyman bias:
1. Use incident cases rather than prevalent cases.
An incident case is a newly diagnosed case of a disease. A prevalent case is an existing case of a disease, in which an individual has typically had it for a longer period of time and thus have a more progressed and serious version of the disease.
By using incident cases, it’s less likely that individuals will be excluded from the study at some point since they’re a new case.
2. Use follow-up studies.
Another way to avoid Neyman bias is by using a follow-up study in which researchers follow up with individuals and examine their situation after the study is over.
This can be particularly useful for monitoring individuals who left a study because they recovered from the disease, which allows researchers to gain a better understanding of the long-term effects of a disease.
The following tutorials explain other common terms: