Between 250,000 and 500,000 people worldwide suffer a spinal cord injury (SCI) each year. These are life-changing injuries. Only about one third of patients recover enough to return to work. Research has suggested links between the type of injury, methods of acute care, and rehabilitation outcomes, but there remain gaps in connecting specific early treatment events to long-term patient outcomes.
A study led by Shivayogi Hiremath, assistant professor in the Department of Health and Rehabilitation Sciences, will apply new methodologies to examine patient data, seeking links between treatment and recovery that could inform acute care approaches to SCI and improve long-term outcomes.
“If we can show relationships, we can go to trauma centers and highlight variables that might help them focus on certain types of care,” Hiremath says. Temple is partnering with Thomas Jefferson University and Virginia Commonwealth University in the two-year study, funded by a $331,040 grant from the Eunice Shriver National Institute of Child Health & Human Development.
Hiremath and the research team will link data from two sources: the National Spinal Cord Injury Model Systems (SCIMS) database and the Pennsylvania Trauma Systems Outcomes Study (PTOS) database. These databases, each containing thousands of records, provide complementary information, allowing researchers to connect SCI cases from early “acute-phase variables” through inpatient rehabilitation facilities and reintegration into the community. Acute-phase variables, Hiremath explains, can include the cause of an injury, the type of trauma, and the process of care.
“How were they treated in the hospital? What procedures were performed? What was the outcome during acute care?,” he says.
The research team is adapting machine learning algorithms to analyze and compare the datasets in search of useful and previously undiscovered connections. Patient records will be linked "probabilistically" between the databases, matching characteristics such as age, sex, year of injury, and length of hospital stay, rather than using identifiable information such as names or social security numbers. This allows patients to be traced from acute hospital stays to one-year post-discharge from inpatient rehabilitation facilities without using any personal information.
“We hope results of this research can inform clinical decision-making at the system level for trauma care and inpatient rehabilitation,” Hiremath says. “The long-term goal is to improve the overall quality of life of individuals with spinal cord injury.”