PCORI implementation projects promote the use of findings from PCORI-funded studies. This project focuses on implementing findings from the completed PCORI-funded research project: Comparing Two Research Methods to Measure How Well Dynamic Treatment Works.
This project is in progress.
What were the results from the original PCORI-funded research study?
People who have the same long-term health problems often receive different treatments. Doctors pick treatments based on things like the patient’s age, sex, and treatment history. Doctors often change treatment over time based on how the patient responds. When treatment changes like this, it’s called dynamic treatment. Researchers find it difficult to study the effectiveness of dynamic treatments because of the way the treatments change over time.
The research team used two types of data analysis methods to address the difficulty of studying dynamic treatments. One was the parametric g formula; the other was inverse probability weighting. The team tested these “g-methods” and found that both methods work equally well.
Why is this research finding important?
Randomized controlled trials are the gold standard for learning how different treatments affect patients. But these trials don’t usually compare treatments that change over time. The analysis methods developed in this PCORI pilot study can help researchers study dynamic treatments using patients’ medical records.
What is the goal of this project?
This project aims to increase use of g-methods among researchers and statisticians who use medical records data to compare how well treatments work. These methods can help researchers provide useful evidence to help patients and clinicians choose treatments based on what is most important to them.
What is the project team doing?
The team is working with researchers in four hospital systems who are comparing the effectiveness of different treatments in 11 studies. The studies involve issues such as pain medicine, brain injury, and pneumonia. The team is helping researchers at these hospitals design and improve their analyses and training them to use g-methods. In addition, the team is making software programs that use the g-methods available for other researchers to use in their own projects.
How is the team evaluating this project?
The 11 study teams are giving feedback about how well g-methods work for their studies and whether they are easy to use. The project team is following up to see if hospital systems that try out the g-methods use them correctly and continue to use them after the project is complete.
How is the project team involving patients and others in making sure the findings reach people who can use them?
Training for the 11 study teams includes ways to work closely with patients, patient advocates, and clinicians in designing studies and selecting study outcomes that matter to patients.