Prospective Research Designs
Analyze cohorts as they move forward in time and collect longitudinal outcome data
Prospective research designs are the most powerful types of observational designs. When using prospective research designs, researchers start in the present, and build a cohort with participants that are exposed or non-exposed to a given phenomenon or treatment. The study participants are then followed forward in time (into the future) to assess how exposure causes outcomes of interest.
What differentiates prospective studies from retrospective studies is when the data is collected. In most retrospective studies, the data already exists and needs to be mined in order to test associations. Researchers "go back" in time and make statistical inferences based on existing data.
Prospective research designs have many advantages in relation to retrospective research designs. Prospective research designs yield measures of incidence, or the number of new cases that come into being within a population over a specified time frame. Prospective research designs allow for researchers to calculate relative risk rather than traditional odds ratios. Relative risk is much stronger mathematical calculation in that the risk ratios for exposure and non-exposure groups are directly compared. Relative risk tends to yield a much more conservative and applicable measure of association, with increased credibility and interpretability. Lastly, prospective research designs give researchers the opportunity to collect longitudinal data related to disease states, allowing for more understanding of etiological and temporal aspects of disease.
There are several caveats associated with using prospective research designs. They can be very time-intensive studies and they can cost much more money that retrospective research designs. This is especially true as it relates to following participants forward into the future. Follow-up with study participants is absolutely necessary to generate valid prospective findings, but it is often the hardest part of conducting these types of designs. Participants will fall out of the study due to any number of reasons and attrition can have a negative impact on the generalization of statistical findings. These observations can be censored in survival analyses so that participants with different levels of follow-up.
What differentiates prospective studies from retrospective studies is when the data is collected. In most retrospective studies, the data already exists and needs to be mined in order to test associations. Researchers "go back" in time and make statistical inferences based on existing data.
Prospective research designs have many advantages in relation to retrospective research designs. Prospective research designs yield measures of incidence, or the number of new cases that come into being within a population over a specified time frame. Prospective research designs allow for researchers to calculate relative risk rather than traditional odds ratios. Relative risk is much stronger mathematical calculation in that the risk ratios for exposure and non-exposure groups are directly compared. Relative risk tends to yield a much more conservative and applicable measure of association, with increased credibility and interpretability. Lastly, prospective research designs give researchers the opportunity to collect longitudinal data related to disease states, allowing for more understanding of etiological and temporal aspects of disease.
There are several caveats associated with using prospective research designs. They can be very time-intensive studies and they can cost much more money that retrospective research designs. This is especially true as it relates to following participants forward into the future. Follow-up with study participants is absolutely necessary to generate valid prospective findings, but it is often the hardest part of conducting these types of designs. Participants will fall out of the study due to any number of reasons and attrition can have a negative impact on the generalization of statistical findings. These observations can be censored in survival analyses so that participants with different levels of follow-up.
Click on a button below to continue.
Hire A Statistician
DO YOU NEED TO HIRE A STATISTICIAN?
Eric Heidel, Ph.D., PStat will provide you with statistical consultation services for your research project at $100/hour. Secure checkout is available with Stripe, Venmo, Zelle, or PayPal.
- Statistical Analysis on any kind of project
- Dissertation and Thesis Projects
- DNP Capstone Projects
- Clinical Trials
- Analysis of Survey Data