A novel computer application developed by Northwestern Medicine scientists could change the way cities track disease and epidemiologists approach research in the future.
Led by Abel Kho, MD, MS assistant professor of Medicine-General Internal Medicine and Geriatrics and Preventive Medicine-Health and Biomedical Informatics, the collaborative effort was awarded a $7 million Patient-Centered Outcomes Research Institute (PCORI) grant at the start of the year.
“A great strength of the application was our ability to get both academic and nonacademic institutions onboard,” said Dr. Kho, a member of the Institute for Public Health and Medicine and Northwestern University Clinical and Translational Sciences Institute. “We proved with our eMERGE project that it was possible to cull information from electronic health records (EHR) to identify cohorts for genetic studies. This project relies on related methods to expand that effort, broadening the scope and scale across an entire city.”
The current dataset has linked more than 5 million de-identified records, combining individuals’ clinical data (diagnoses, medications, laboratory tests and vital signs) regardless of which participating provider they visit. Ten Chicago-based institutions, including Northwestern Medicine, Rush University Medical Center, Cook County Health and Hospital Systems and the Hines and Jesse Brown Veterans Administration Hospitals, are involved in the project.
In order to link EHR data from multiple sites, scientists and software engineers had to find an appropriate balance between data availability and risks to privacy. To maintain the anonymity of both patients and institutions, individuals’ common identifiers are encrypted. That tagging information is then replaced with a unique study ID.
The resulting database links EHR data by unique study ID across multiple healthcare providers. Because the effort was funded as a research project, these combined EHRs are not currently accessible by clinicians.
“The goal is to build a citywide system of electronic health record sharing,” Kho said. “Similar projects are underway in other cities, such as New York, but we’re not aware of anybody else who has gotten as much data aggregation across different care settings in as large of a city as Chicago.”
Dr. Kho launched a similar project while he was a National Library of Medicine Biomedical Informatics Fellow at the Regenstrief Institute in Indianapolis. There, he developed an electronic network that enabled infection data to be shared across the city’s major healthcare providers. The system currently tracks thousands of patients with drug-resistant infections and provides secure, real-time admission alerts. A study examining the system’s usefulness was recently published in Clinical Infectious Diseases.
After arriving at Northwestern University Feinberg School of Medicine in 2006, Dr. Kho’s first foray into EHRs in Chicago involved a collaboration with William Galanter, MD, at the University of Illinois College of Medicine at Chicago, and hospitalization data from five institutions. They then expanded the project to include clinical data before growing the effort to seven institutions and more than 5 million patient records in the Chicagoland area.
“Our success in building these relationships with multiple institutions over time and our ability to handle the growing data component led to this support from PCORI,” said Dr. Kho, who is also principle investigator and co-executive director of the Chicago Health IT Regional Extension Center, which assist more than 1,500 doctors in Chicago to achieve meaningful use of EHRs. “This pool of data is a unique resource for both scientists and public health officials.”
A third-generation software program is currently installed at seven participating institutions and allows for de-identified records to be searched by IRB-approved investigators. As part of the PCORI funded project, contact information from individuals who have given prior consent to be included in trials may be accessible to the scientists. Once completed, Dr. Kho envisions scientists throughout the city will be able to better identify and recruit potential study subjects.
“We would know how many people at each of our sites may be of interest to a scientist,” Dr. Kho said. “This bi-directional flow of information would mean that if a person fitting a specific request shows up at any of the healthcare providers, the system sends an alert, and if that person has already given appropriate consent, they can be contacted.”
The aggregation of so many EHRs also provides a database for mapping of disease throughout the city.
“When you pool all of these records you create a giant picture of what is going on in Chicago,” Dr. Kho said. “Overall figures of relatively common diseases are being released on the Chicago Health Atlas website through the Smart Chicago project, but I think tremendous additional value in this information comes with scientific interpretation rather than raw data.”
Over the next year, bioinformatics specialists will be working with the software to expand coverage to include patient records from all participating institutions. They’ll also look to streamline the task of inputting patient reported outcomes and merging these with EHR-derived data.
“If we can reduce the lag in uploading these records, soon after a person who may be eligible for a study arrives for care, they could potentially be recruited,” Dr. Kho said.
Initial efforts leading to the PCORI grant were supported by the National Institutes of Health and the Otho Sprague Institute.