There is a type of research design that makes it possible to formulate hypotheses about possible associations between an outcome and an exposure and to investigate further the possible relationships that exist, it is the so-called retrospective study.
This type of observational study is used above all in the health sector, for example to obtain information from participants who have a disease or condition. However, in a retrospective study, a causal statement about this association is not usually made.
Let’s learn more about the characteristics of this type of study.
What is a retrospective study?
A retrospective study is one that aims to find out what potential risk factors or other associations and relationships a group has in common. The opposite of a retrospective study is a prospective study in which participants are enrolled before any of them have the disease or outcome being investigated.
When conducting a retrospective study, an investigator often uses administrative databases, medical records, surveys, or interviews with patients who are already known to have a disease or condition.
There are those who also call it a control study, this is mainly due to the fact that, when it comes to diseases and conditions, you always want to carry out a control and follow-up. A historical epidemiological study without a control would be unthinkable, and perhaps even useless.
Steps to do a retrospective study
These are the steps you have to follow to carry out your retrospective study:
- Define the population that will be the object of your study
- Select the time period back from where you will get the data
- Limit the results you want to obtain, that is, if you will be studying the result of a particular disease or the occurrence of certain events.
- Perform data collection
- Analyze the results carefully
Advantages and disadvantages of a retrospective study
These are some of the benefits and drawbacks that you may have when conducting a retrospective study:
- Useful for studies on rare diseases or unusual exposures.
- Smaller sample sizes .
- Studies take less time because data is readily available (you just have to collect and analyze it).
- The costs are usually lower.
- Lack of data: For example, it may happen that information on the status of a person’s exposure to a disease is not available and data that may be useful has not been collected.
- Forgetfulness bias: Participants may not be able to remember whether or not they were exposed to certain factors.
- Confounding variables are difficult or impossible to measure.
As this is a relatively weak type of study, causal claims cannot be made, although correlations can be made.
Retrospective cohort study
In a retrospective cohort study, researchers focus on a certain period of time back to find the exposure of a group to the same risk factor.
A cohort is a defined group, such as “nurses,” “people 30-39 years old,” or “high school students.” Participants are chosen for a reason, and not at random.
For example, researchers may want to investigate whether exposure to commonly used carpentry glues increases the risk of developing COPD. A cohort consisting of retired carpenters could be selected. A control group is also chosen. This could be made up of delivery drivers and administrative workers who have not been exposed to such material.
Health records and employment records are used as data sources. Because data is collected retrospectively in an uncontrolled environment, it is not possible to make claims about causation.
Although claims about causality cannot be made, associations and possible relationships can be found,