How does hospital length of stay affect a patient’s odds of surviving a hip fracture? Investigators from the University of Rochester Medical Center Department of Orthopaedics explore the topic in a new study published in the British Medical Journal. The paper is notable for the questions it asked and for the tool it used to find answers: in a first for URMC, the researchers built analytics software to mine New York’s Statewide Planning and Research Cooperative System (SPARCS) database. SPARCS contains all payers’ records of virtually every patient case in New York state, making it a rich source of information to study and compare patient treatments and outcomes.
Questions about fragility fractures and length of hospitalization are of particular interest at the University of Rochester Medical Center, home to an internationally renowned Geriatric Fracture Center that has improved patient outcomes while reducing patients’ time in the hospital. The center’s research over the past decade has suggested that patients do better with a shorter hospital length of stay. But a January 2015 study of Swedish hip fracture patients published in the British Medical Journal offered a seemingly opposite finding: After reviewing nearly 120,000 patients from 2006-2012, authors concluded that patients with shorter hospitalizations had an increased risk of death. For Swedish hip fracture patients hospitalized less than 10 days, each one-day reduction in length of stay increased their odds of death within 30 days of discharge by 8 percent in 2006, and the risk rose to 16 percent in 2012.
Hip fractures are among the most common and disastrous of orthopaedic injuries: more than 250,000 older adults suffer a hip fracture in the United States each year and the number is expected to rise as the population ages. It’s a given that the injury can compromise an older patient’s overall health and it greatly increases their risk of death within one year.
John C. Elfar, M.D., associate professor in the Department of Orthopaedics, was struck by the study findings. He had been leading Orthopaedics residents in a journal club dedicated to reviewing and critiquing emerging orthopaedics papers; when this one surfaced, the team launched a study of New York state patients to test its findings. Elfar is the paper’s senior author and the team included Lucas Nikkel, Stephen Kates, Michael Maceroli, Bilal Mahmood, and Michael Schreck.
Using the SPARCS database information, Elfar and his colleagues conducted a retrospective cohort study of 188,208 patients age 50 and older who were admitted to a hospital with a hip fracture in New York state from 2000-2011. They found that longer hospital stays spelled worse results for U.S. patients. Compared to patients with a hospital stay of 1-5 days, patients with a hospital length of stay of 11-14 days was associated with a 32 percent increase in odds of death within 30 days of discharge. The 30-day mortality risk rose to 103 percent for patients with length of stay of 14 days or more. Researchers concluded that decreased length of stay was associated with lower 30-day mortality rates.
“Patients in New York state spend far less time in the hospital than their counterparts in Sweden, but it does not mean that they are being released prematurely here,” Elfar said. “It also does not mean that being in the hospital for a long period of time in New York state is a cause of complications or a driver of poorer outcomes.”
“Our analysis shows that the difference in hospital stays and results between Sweden and the U.S. is related to a difference in health care systems,” Elfar said. “Patients do as well here with short hospital stays as they do with longer hospital stays in Sweden because U.S. hospitals focus on acute care and transfer patients to rehabilitation facilities as soon as possible. Such facilities are not available in Sweden, so patients rehab in the hospital setting and spend longer periods of time there.”
Finding the best approach to care is vital because hip fractures are so common and older patients are especially vulnerable, Elfar noted. “Women are more likely to die from a hip fracture than breast cancer. In older patients who suffer a hip fracture, perhaps one-third of them will die within a year’s time. In an institutional setting, they are 11 times more likely to die than a patient who has not broken a hip.”
Elfar said Peter Nordström, the corresponding author of the Swedish paper, reviewed URMC’s study and came to the same conclusion, “that the different systems and their outcomes are different for a good reason.” The new study findings validate the soundness of the Geriatric Fracture Center model, Elfar added: “Getting fracture patients out of the hospital quickly is the best approach provided you have a good place to send them for rehabilitative care, as we do here.”