Although randomized trials remain the gold standard for determining the effects of different interventions on patient outcomes, they usually compare treatments that do not change over time. Further, in some cases randomized trials may not be feasible, ethical, or timely. The g-methods studied in this PCORI PILOT can help researchers use observational data to look at questions of effectiveness of dynamic treatment strategies, which are particularly important for many chronic conditions, when randomized trials are not a good option. The ‘product’ researchers propose to disseminate and implement is knowledge and application of g-formula methods. Using these methods and applying them to Medicare claims data, the research team obtained estimates of treatment effects that were similar to findings from a randomized clinical trial that addressed the same research question indicating the validity of g-methods.
This dissemination and implementation aims to increase the access, adoption, and uptake of g-methods among researchers and statisticians who are using observational data to answer questions regarding the comparative effectiveness of alternative treatment approaches. By promoting broader use of advanced analytical methods, the project will allow researchers to provide important scientifically valid information needed to help patients and providers make informed decisions. The team is partnering with 4 sites—Nationwide Children’s Hospital/The Ohio State University, University of Utah School of Medicine, UConn Health, and University of Florida—that have a total of 11 funded CER projects on the topics of pediatric triage of major trauma, pain relief medication use, spinal cord and traumatic brain injury, community-onset pneumonia, rheumatoid arthritis progression, and health disparities in maternal and child health. Researchers at these sites will demonstrate the application of g-methods in these studies.