VAPIR contributes to large VUMC study

On Feb. 27, 2018, the New England Journal of Medicine published a research study conducted by a team at Vanderbilt University Medical Center which found that patients who receive a balanced fluid that closely resembles the liquid part of the blood have better outcomes than patients who receive a saline solution containing only sodium chloride. The team examined more than 15,000 intensive care patients and more than 13,000 emergency department patients who were assigned to receive saline or balanced fluids if they required intravenous fluids.

If that sounds like it would be a lot of data to analyze, you would be correct. Thankfully, the Vanderbilt Anesthesiology and Perioperative Informatics Research (VAPIR) Division were a part of this study. Led by Jonathan Wanderer, MD, MPhil, the team took on the task of writing the code that extracted the data needed for the study.

According to Wanderer, the code generated a massive data set on a weekly basis and it required a lot of work to make sure the output was correct, comprehensive and formatted in a way that would allow for statistical analysis in an easy way.

Senior Database Administrator Karen McCarthy explained that the unique challenge of this project was the amount of time it spanned. The project started in 2015 and grew, requiring multiple code revisions. “Even though we keep all our code and documentation with it, when somebody leaves, some knowledge leaves as well. When somebody new picks it up, it takes some time for them to get into the project,” she said. Other challenges included pulling information from multiple sources and reaching out to different areas for information.

Business Intelligence Analyst Frank Aline revealed that the team collaborated to check the integrity of the data upon export. His role specifically was to review the revision needs, make changes to the code accordingly and ensure the data was exported correctly. 
The team had to perform a lot of data analysis and understand where data is missing or potentially erroneous and, according to Wanderer, the team needed to have confidence in the data that was collected.

As for why VAPIR was picked for this project, Wanderer explained that while there are many teams across the organization that work with large data sets, VAPIR has a particular expertise in acute care as well as perioperative areas and the team was familiar with these types of data. “I think we have a good reputation for doing this sort of detailed, high quality work that you need to do to get a data set of this size together,” he said.

The project has received a lot of attention from being published in the New England Journal of Medicine to being featured on National Public Radio. As a result, VUMC encouraged its medical providers to stop using saline as intravenous fluid therapy for most patients. Aline explained that the work done on a daily basis by the VAPIR team, whether big or small, is pleasing and satisfying when they can see improvements to patient care.

“It’s really cool to see a project that has taken the number of years it has taken to have that outcome at the end where you are seeing the benefits of it or reading about the benefits of it,” McCarthy said.

Wanderer further expressed that it’s great to see what VUMC can do as a community. This project required different skill sets from a lot of different people across the whole organization. “We can really accomplish a lot working together and it’s exciting that the trajectory in the future is really embracing this approach to pragmatic clinical trials and helping to fulfill the vision ins becoming a learning healthcare system,” he elaborates.

As for the future of VAPIR and how it relates to this project, Wanderer said the team is in the process of trying to take some of the new things developed for this group of patients and generalize it to ICU patients at large for future retrospective work.

Wanderer revealed that VAPIR has already started on another project and is in the grant proposal period. It will also use large data sets. “We are already plugged into what might be the next big thing,” he explains.