Indiana University researchers to use social network analysis to study 'doctor shopping'
FOR IMMEDIATE RELEASE
BLOOMINGTON, Ind. -- Indiana University researchers have been awarded a $1.4 million grant from the National Institute on Drug Abuse for a study that will use social network analysis to examine "doctor shopping," in which patients visit multiple physicians to get prescriptions for controlled substances.
Brea Perry, associate professor of sociology in the College of Arts and Sciences, is the principal investigator for the four-year project. Also participating in the study are Yong-Yeol "Y.Y." Ahn, assistant professor in the IU School of Informatics and Computing, and researchers at the University of Kentucky. Perry and Ahn are affiliated faculty with the IU Network Science Institute.
The study will use advanced computational and statistical methods to develop better ways to identify patients who are doctor shopping and the physicians they are targeting. Doctor shopping is a key indicator of prescription drug abuse. Addiction to prescription painkillers, such as opioids and benzodiazepines, has grown rapidly in the United States, creating a public health crisis that includes increasingly widespread use of heroin.
"The goal is to identify doctor shoppers before they get to the point where they are engaging in behaviors that are likely to lead to overdose and death," Perry said. "We would like to be able to intervene early; and so far, the measures just don't exist that would allow us to do that."
The rate of drug overdose deaths in the U.S. more than tripled between 2000 and 2014, according to the Centers for Disease Control and Prevention. The center says 47,000 overdose deaths occurred in 2014, the highest number on record.
Health authorities have responded by creating prescription drug monitoring programs aimed at identifying doctor shoppers. But the programs typically rely on monitoring whether patients reach a certain threshold in the number of doctors they see or prescriptions they try to fill. That is not an accurate measure of whether someone is doctor shopping, Perry said.
For example, a cancer patient might legitimately see multiple physicians and have multiple prescriptions. Standard monitoring approaches would flag that person as a doctor shopper. Other patients may engage in moderate doctor shopping for a short period of time without triggering notice.
The study will use social network analysis to map connections between doctors and patients, with a link representing prescription of an opioid or other controlled substance. The researchers will apply the analysis to a large health records database containing information about physicians, patients and prescriptions, but from which personally identifying information has been removed.
They expect that patterns of connections between doctors and patients who are most active in the networks will paint a clearer picture of doctor shopping. In other words, a patient’s position in the prescription network should provide more precise information about who is actually engaging in this drug-seeking behavior than the number of prescriptions or number of prescribers alone.
"These measures, once they are developed and tested, could be applied to state prescription drug monitoring programs," Perry said. "They could be translated to effective interventions, both to help the patients who are doctor shopping and to provide education for doctors who are being targeted."
Perry's recent research has focused on the roles of social networks, peer and family relationships, and other factors in health, with a focus on mental illness and at-risk youth. Ahn is an expert in large-scale network analysis and machine learning who has developed mathematical models to map complex systems such as the brain, social networks and culture.
They joined forces through the IU Network Science Institute, which brings together over 100 researchers to work on scientific understanding of complex networked systems.
"The institute was really instrumental in bringing us together," Perry said. "This project really required a collaborator with Y.Y.’s expertise to manage and analyze data this large."
The National Institute on Drug Abuse is part of the National Institutes of Health in the U.S. Department of Health and Human Services. Its mission is to advance science on the causes and consequences of drug use and addiction and apply that knowledge to improve individual and public health.
