What is a Retrospective Study and When is it Used in Clinical Trials?

Clinical trials can vary widely in their design and scope depending on their objectives, structure, and the research methods used. One such type of clinical trial is known as a retrospective study. But what exactly is a retrospective study, when is it used, and how does it work?

In this article, we’ll explore everything you need to know about retrospective trials, including the benefits of them and how they compare to other types of clinical studies.

What is a retrospective study?

A retrospective study can be defined as a type of clinical study that uses information on events that have taken place in the past. Retrospective studies typically look back and examine exposures to suspected risk or protection factors concerning an outcome that would have been established at the beginning of the study.

In most retrospective studies, much of the data has already been gathered and stored in the registry. In this case, researchers can use this data to analyze past events and circumstances. Sometimes, individuals may be sampled and have information about their past collection. They would then be interviewed and asked to recall important events, with the information then examined to establish whether there are relationships between exposure and outcome.

Retrospective study types

There are a few different retrospective study types available to researchers, with the main two types being cohort and case-control studies. While each study design differs in detail, they both share a similarity in that each compares subjects with and without a condition to establish how they are different.

Retrospective cohort study

A cohort study can be retrospective or prospective. This type of clinical study compares groups of subjects who are similar overall but have one different characteristic, which is usually their exposure to a risk factor. It aims to determine whether exposure to a certain risk or protective factor will affect an outcome.

In retrospective cohort studies, the exposure and outcomes happened in the past. Retrospective cohort studies are conducted on data and existing medical records (usually from prospective studies) and the exposures are defined before the study begins. This means researchers can establish whether exposure to a risk factor is associated with a statistically significant difference in the outcome development rate.

Case-control study

Case-control studies are generally retrospective studies. They are comprised of two groups which are defined at the beginning of the study. There is usually one group with the outcome/disease and one without.

In a retrospective case-control study, researchers define a risk factor between the groups and look at past data and information to assess whether there is a statistically significant difference in exposure rates. Many case-control studies evaluate a particular incident, using an exploratory design to identify potential risk factors.

The difference between retrospective case-control studies and cohort studies

While both types of studies use historical data and are conducted retrospectively, there are differences between the two. A case-control study uses information on subjects who either have a particular outcome or don’t. In retrospective case-control studies, researchers look back to establish what the exposure was. In retrospective cohort studies, historical data explains the exposure level at some baseline in the past, before explaining the subsequent outcome/disease in the present.

What is the main purpose of a retrospective study?

Retrospective studies might be used for a variety of reasons, but one of their key purposes is to help inform a prospective experimental design. Retrospective cohort studies in particular will often be used as an intermediate step between a weaker preliminary study and a prospective cohort study. This is because the results from a retrospective study can strengthen any assumptions behind prospective cohort studies of the future.

Advantages of retrospective studies

A retrospective clinical trial has many advantages, from greater cost efficiencies to quicker completions and the ability to address rare or complex diseases more easily.

Cost-effective

Retrospective studies are more cost-effective than traditional studies as there is no requirement for a lab or equipment. In addition, the data has already been collected and is readily available for analysis. This means a smaller research team is required and fewer resources are needed.

More efficient

In line with lower costs, retrospective studies are also efficient in terms of time. Retrospective cohort studies, in particular, require fewer subjects than other methods and use preexisting secondary research data. Because the data has already been gathered, studies can be completed much more quickly. While a study period might span several years, completing a retrospective study is only as long as it takes to collate and analyze the existing data.

Greater outcome analysis

Retrospective studies, especially cohorts, can analyze multiple outcomes simultaneously because a single exposure was identified in the beginning. This can lead to greater outcome analysis and the ability to identify the natural history of a disease.

Able to address rare diseases

When it comes to rare disease research, retrospective studies are particularly beneficial because affected people are already identified. When studying rare or unusual exposures, retrospective studies can be more successful than prospective studies which would need a longer study period to be able to form conclusions. Prospective cohort studies would also need a much larger group of participants, which for rare diseases, is difficult to do.

Helpful for ethical considerations

If a study has ethical considerations or concerns about participants that could prevent the study from following a traditional design, retrospective studies can be used. Researchers can hand-pick the data they need for the study, based on what is already available.

Disadvantages of retrospective studies

Just as there are many advantages to retrospective studies, they can also have challenges such as inconsistencies, bias, and credibility.

Possibility of inconsistencies

As retrospective studies use data that was measured for other purposes, it’s possible that inconsistencies could occur. When the data was originally gathered, it may have been recorded by different people, who could have used different procedures or equipment. Likewise, measurements of data could have occurred under different conditions, making the data inconsistent.

Not all key statistics can be measured

Looking back at historical data could mean that certain key statistics were not measured at the time, which means some elements of the study may not be completed accurately. This lack of data can also lead to significant biases that could affect the selection of controls (the comparison group that is unaffected or unchanged by the other variables).

Exposure or outcome assessment cannot be controlled

Because researchers in retrospective studies rely on others for accurate record-keeping, they cannot control exposure or outcome assessment. When relying on individual recall in particular, recounts may be inaccurate or subject to biases. It can be very difficult to make accurate comparisons between the exposed and non-exposed when asking people to recall their former exposure to risk variables.

Potential bias

We’ve already touched on bias, and while biases are common in several observational studies, retrospective studies are at a particularly high risk for many research biases. Recall bias and observer bias are prominent because researchers rely on a person’s memory and self-reported data. There could also be selection bias and mis-classification or information bias due to the retrospective element.

Low internal and external validity

Validity can be a problem with retrospective studies as they can never fully establish casualty. When attempting to evaluate the efficacy of an intervention in a retrospective study, several threats could limit the interpretation of the results. These include a lack of comparison or control group, high rates of attrition, statistical regression, or social interactions.

Issues with reliability and credibility

With historical data, there is always a risk of reliability and credibility. Furthermore, as many patients will have been studied by different healthcare professionals throughout a study or illness, there could be variability in the measurement of risk factors and outcomes.

When to use a retrospective clinical study

Because retrospective clinical studies use secondary data, they are usually quantitative. This is different from many other observational studies which tend to be qualitative. A retrospective study can be used to conduct both exploratory research and explanatory research. This type of study works well for research if the following are applicable:

• A prospective study is not feasible for the variables being investigated
• The examination of the effect of an exposure, outbreak, or treatment on an outcome is needed quickly
• Research is needed to investigate an early-stage hypothesis or potential association between the agreed variables of interest

Prospective study vs retrospective study

One of the key comparisons made in clinical research is between a retrospective and prospective study. But before we can get into the battle of retrospective study vs prospective study, we need to establish a definition for the latter.

What is a prospective study?

A prospective study is carried out to learn more about a particular medical topic. It will watch for outcomes throughout the study duration and relate this to factors such as suspected risk or protection factors. Prospective studies typically last over a long period. Sometimes, researchers may design studies, recruit participants, and collect baseline exposure data before a medical condition has developed.

How do retrospective and prospective studies differ?

There are several differences between retrospective and prospective studies in clinical research, from the way data is collected to the costs involved in running the study.

Data collection

This is one of the key differences between retrospective and prospective studies. While retrospective studies involve the investigation of existing data around exposure factors and their impact, prospective studies do not have a record of the incident or interest because it is concerned with events that have not yet happened.

Data analysis

As the data collection methods are different in prospective and retrospective studies, data analysis is often different too. Researchers may use a variety of methods to analyze data in both types of study, including descriptive, diagnostic or predictive analysis.

Use and purpose

Both types of study are used for different purposes too. A retrospective study is typically used to add to existing research or provide further context on a recent discovery. Whereas prospective studies are concerned with the future and will be used to discover the likelihood of an event happening. If a study is designed to find the cause of a disease outbreak, a retrospective study is more likely to be used.

Time and cost

We’ve already mentioned that lower costs are one of the benefits of retrospective studies, but they also typically take less time than a prospective study too. As retrospective studies don’t involve observation of participants, there will be less money and time spent on data collection. In addition, prospective studies are often carried out over a long period of time, so it will take much longer to measure and analyze the effect of exposure on participants.

Examples of retrospective studies in statistics

Retrospective studies are common in fields like medicine, epidemiology, and healthcare. Below are some instances when a retrospective study might be used.

Tanning beds and skin cancer

Researchers might decide to use a retrospective study to look at the relationship between the incidence of skin cancer diagnosis and the use of tanning beds. In this case, they would collect data from people with skin cancer who used tanning beds.

They might look at how many times participants visited tanning beds, how long they were there, and the type of bed used. This information would then be analyzed to establish the potential risk factors of tanning beds for skin cancer.

Smoking and lung diseases

Researchers may hypothesize that smoking increases a person’s risk of lung diseases or cancer and use a retrospective study to confirm this. They would potentially establish two groups of people, one with a lung disease or cancer and one without. The prevalence of smoking could then be analyzed in each group to establish whether there are risk factors for smoking on lung diseases.

Real-life example: COVID-19 in children

While there have been many retrospective studies to note, COVID-19 has been a particularly interesting time for retrospective studies. Compared to adults, there have been very few studies on COVID-19 in children. There have been even fewer studies focusing on the unique features of COVID-19 on children concerning lab findings, locations of computerized tomography (CT) lesions, and the role of CT in clinical recovery.

A retrospective study in China collected the clinical, imaging, and laboratory data of 76 children and found that there were few differences between COVID-19 children and COVID-19 adults in terms of lab findings and CT characteristics. Without the ability to retrospectively study the data collected, this study would not have found the differences and similarities highlighted in the results.

Conclusion

Retrospective studies can be extremely beneficial in clinical research, especially when prospective studies are not possible, or information needs to be gleaned from historical information. However, it’s still important to weigh up the pros and cons before embarking on a retrospective study as some studies are better suited to other types of study.