News & Events

Le Bonheur researchers using sociomarkers to predict hospital revists due to pediatric asthma

Published On 10/17/2018

Environmental and socioeconomic sociomarkers can predict the likelihood of a pediatric asthma patient’s return visit to the hospital, according to new research from the Center for Biomedical Informatics at the University of Tennessee Health Science Center.

“Health is a social as much as a biological matter,” said researcher Arash Shaban-Nejad, PhD, MPH. “Our research directly shows the importance of living environment and social and economic conditions for their contributions to individual health outcomes.” He and fellow researchers  Oguz Akbilgic, PhD, Ruhi Mahajan, PhD and Eun Kyong Shin, PhD, published research in Nature Digital Medicine

Their research introduced the idea of sociomarkers – a measurable indicator of social conditions in which a patient is embedded. The sociomarkers used in this study included the proportion of individuals living below the federal poverty level, blight prevalence, neighborhood quality, trash dumping presence and broken window pervasiveness.  The research focused on pediatric asthma, one of the major chronic conditions in the U.S. and worldwide with high sensitivity to the environment and sociomarkers.

The study compiled information from three data sources – pediatric asthma encounter records from 2016 at Le Bonheur Children’s Hospital in Memphis, Tenn., the 2010 U.S. Census and the neighborhood and quality survey data by Memphis Property Hub. By integrating these datasets and performing analytics, researchers were able to examine the effect of social features in identifying the patients who visited the hospital more than once during the observation period.

The results of the study revealed the importance of social conditions on determining a pediatric asthma patient’s health. By using demographic information and sociomarkers alone, researchers could predict which patients would return to the hospital with 61% accuracy. “Without using any symptom related predictors, information gathered from a ZIP code level and demographic characteristics still allow us to predict two out of three patients who will revisit the hospital due to asthma,” said Eun Kyong Shin. In comparison, using demographics and biomarkers alone predicted patient return visits with 65 percent accuracy and using all three data sets – demographics, biomarkers and sociomarkers – could predict with 66 percent accuracy.

Sociomarkers’ ability to predict health outcomes upholds the theory of non-genetic components of health outcomes. This theory says that health outcomes are related to a wide variety of factors – 15 percent social circumstance, 5 percent environmental exposure, 10 percent health care, 40 percent behavioral patterns and 30 percent genetic predisposition. “The bottom line is – environment matters,” said Shaban-Nejad. “Chronic disease explanations need to be widened to include not only genetics but also lifestyle and environment.”

Because of the importance of sociomarkers in health, the study proposes further exploration into the effect of environment in other areas of medicine so that healthcare providers can predict which patients may be prone to certain health outcomes. “The ultimate goal is to provide care and narrow health disparity,” said Shin. “Multidimensional interventions are needed to care for the health of patients, and detailed neighborhood-level information can help us unpack the pathways of social inequalities to health disparities.”