GALVESTON, Texas – Knowing which patients are more likely to be readmitted to the hospital after a procedure would be helpful to both patients and their health care providers. Doctors would be able to customize care for high-risk patients, which could result in reducing the personal and financial burdens associated with hospital readmission.
However, current models designed to predict hospital readmission need to be more accurate before they can be useful in hospital settings, said Suresh K. Bhavnani, associate professor from the Institute for Translational Sciences at The University of Texas Medical Branch at Galveston.
“We believe current models can be improved because they assume that all patients admitted for a specific condition like a hip fracture can be described by a single model,” Bhavnani said. “However, numerous studies have shown that most humans tend to share key characteristics such as gene profiles or pre-existing conditions, forming distinct patient subgroups. Predictive models that take such patient subgroups into consideration should be more accurate compared to the current models that do not.”
To achieve that goal, the Patient-Centered Outcomes Research Institute awarded Bhavnani and his multidisciplinary team of data scientists and clinicians $525,000 to develop a new visual analytical method which will automatically identify patient subgroups in large datasets such as Medicare data, and test whether those subgroups can help to improve the prediction of hospital readmission in the elderly. This is the first PCORI award that UTMB has received, and is in the funding category of improving methods for PCORI.
“We are honored to be recommended for funding by PCORI and are excited to develop big data visual analytical methods focused towards improving patient-centered outcomes such as hospital readmission,” Bhavnani said.
Bhavnani knows the importance of reducing unplanned hospital readmissions. His mother recently suffered a hip fracture, but despite the surgery being successful, she was back in the hospital a few weeks after discharge. This was because the surgery, bed-rest and modified nutrition interacted poorly with her pre-existing conditions, resulting in a costly hospital readmission.
His mother’s experience demonstrates how seemingly unrelated factors can result in a patient having to be readmitted. Bhavnani, who heads the Discovery and Innovation through Visual Analytics lab at UTMB, will use an advanced form of visual analytics to automatically identify patient subgroups that share similar characteristics – those with diabetes and renal failure, for example – and build prediction models that are targeted to those subgroups.
“A key goal of precision medicine is to identify patient subgroups, comprehend the reasons for their negative outcomes, and design interventions targeted toward those patient subgroups. Our visual analytical method for automatically identifying patient subgroups and their characteristics in large datasets therefore has direct relevance to precision medicine,” said Bhavnani.
The study will focus on three conditions common in elderly patients: chronic obstructive pulmonary disease, congestive heart failure and hip/knee surgeries. His team will include clinicians and researchers to develop and test the predictive models which will use information about patient subgroups.
“This is UTMB’s first funding award from PCORI and we share Dr. Bhavnani’s enthusiasm for using innovative visual analytical methods to improve patient care,” said Dr. Danny O. Jacobs, executive vice president, provost and dean of the School of Medicine at UTMB.
PCORI is an independent, nonprofit organization authorized by Congress in 2010. Its mission is to fund research that will provide patients, their caregivers and clinicians with the evidence-based information needed to make better-informed healthcare decisions. For more information about PCORI’s funding, visit www.pcori.org.
“This project was selected for PCORI funding not only for its scientific merit and commitment to engaging patients and other stakeholders, but also for its potential to fill an important gap in our health knowledge and give people information to help them weigh the effectiveness of their care options,” said PCORI Executive Director Joe Selby. “We look forward to following the study’s progress and working with UTMB to share the results.”