“Veterans Administration Acute Care 30-Day Mortality Model: Development, Validation, and Performance Variation” poster wins honor.
by Richard Quinn
Eduard Vasilevskis, MD, is a member of SHM’s Research Committee and was a judge for the annual Research, Innovations, and Clinical Vignettes (RIV) poster competition at HM11 outside Dallas. But when the committee members discussed who would judge the HM12 contest, Dr. Vasilevskis passed. He had a poster he wanted to enter instead.
His poster took the top prize in the Research portion of the annual contest, as judged by a panel of hospitalist leaders and visiting professor Pamela Lipsett, MD, MHPE, FACS, FCCM, a professor in the department of surgery, anesthesiology, and critical-care medicine at Johns Hopkins University Schools of Medicine and Nursing in Baltimore.
“When you work in a specific research area and you’re really passionate about the subject you’re studying, you sometimes lose sight of how other people might care about it,” says Dr. Vasilevskis, a hospitalist and assistant professor of medicine in the section of hospital medicine at Vanderbilt University School of Medicine and Tennessee Valley-Nashville VA Hospital in Nashville. “It’s nice to have an audience to validate that.”
HM has “a really outstanding core of researchers,” he adds. “The people who are walking around in that group, I really look up to.”
The feeling was mutual, as a team of judges led by SHM Education Committee Chair Daniel Brotman, MD, FACP, FHM, set out to pick a winner in a contest that has grown so large, the Clinical Vignettes portion was broken into a separate session.
First, SHM selects which abstracts to accept. Then, 186 hospitalists or HM-led teams were invited to give poster presentations at the San Diego Convention Center. Then, 23 finalists were chosen for judging rounds. All of that happened before anyone got to the annual meeting. Once in the exhibit hall, two-clinician judging teams worked their way through two dozen research posters deemed the best of the best. Each team recommended a finalist except for one, which Dr. Brotman dubbed “the Russian judges.”
But from the get-go, it was clear that the work completed by Dr. Vasilevskis and his colleagues would be the winner. As the two-member teams went around naming the posters they wanted the full judging panel to visit, one of those who visited Dr. Vasilevskis was brief but enthusiastic in the description: “I really liked this one.”
So the judges went back to poster No. 63, “Veterans Administration Acute Care 30-Day Mortality Model: Development, Validation, and Performance Variation.” Dr. Brotman was impressed by the breadth of the study, which aimed to develop a hybrid administrative/clinical risk model to address the shortcomings of risk models that rely on only administrative data or clinical data.
The first number that caught the judges’ attention was 1,114,327. That’s how many patients were used in retrospective cohort study of 131 VA hospitals. Then it was the results, which were impressively accurate given the size of the sample.
“It was a strong survey with incredible methodological rigor, addressing an incredibly important issue,” Dr. Brotman says. Dr. Vasilevskis notes that few hospitals or health systems currently have access to the level of data available to his research team. But as electronic health records (EHR) capture more clinical and administrative data, the ability to build models like the one he presented will only increase. “The better your model, the more you can believe your results,” he adds. “And we want our clinicians and CEOs and hospital boards to believe those numbers.”
The Hospitalist newsmagazine reports on issues and trends in hospital medicine. The Hospitalist reaches more than 25,000 hospitalists, physician assistants, nurse practitioners, residents, and medical administrators interested in the practice and business of hospital medicine.