Imagine this scenario: A person comes to a physician, complaining of severe stomach pain. The doctor enters the symptoms into the electronic health record (EHR) system. As the doctor tries to determine what's causing the pain, an alert from the health department pops back up on the screen -- public health surveillance data advises that there's been a salmonella outbreak in the region. The patient receives the proper medication and avoids an overnight stay because of the prompt action.
This is the type of situation providers in Chicago are hoping to create as they test whether data standards and clinical decision support tools can be transformed into real-time, actionable alerts that affect patient care and improve the population's health.
The Alliance of Chicago, GE Healthcare, and the Centers for Disease Control and Prevention (CDC) have begun a joint six-month prospective study to determine whether they can generate alerts that are useful to doctors, and whether physicians will act on the information in the alerts. The project will follow 10 to 14 providers who will serve as principal investigators. The organizations hope to have preliminary results ready to share by the fall.
Dr. Fred RachmanCEO, Alliance of Chicago
The alliance, a network of Chicagoland community health centers that share an EHR system and other technology resources, sees the project as the first step toward broad public health surveillance. "This is the prototype for things that are changing in real time," said Dr. Fred Rachman, CEO of the alliance, which also is a founding member of the Chicago Health Information Technology Regional Extension Center.
It's a lofty goal, but one with implications for meeting federal meaningful use requirements, which include public health reporting and public health surveillance criteria.
Tapping the CDC's vast public health surveillance database
Currently, it might take several days for the CDC to collect enough information from providers to determine that there's a food-borne illness problem and alert caregivers. Achieving the scenario described above -- in which a physician can be informed almost immediately that one sick patient might be part of a bigger trend -- requires a significant operation behind the scenes. That's where the EHR system, a retrospective data set of 18 million de-identified patient records, the local health department and the CDC's expansive public health surveillance guidelines all meet.
The key is that the EHR system provides the information to the CDC, which can filter data based on its guidelines, according to Mark Dente, chief medical informatics officer for GE Healthcare. GE's Centricity, the EHR system employed for the study, is used by the Alliance of Chicago's clinicians. GE Healthcare's Medical Quality Improvement Consortium database will be used to spot trends.
Patient symptoms and findings are collected at the point of care then sent in real time through GE Healthcare’s database to the CDC -- both anonymized and compliant with federal privacy mandates -- where it is measured against a disease profile. The CDC matches those profiles to other patient information with similar symptoms in a concentrated area and issues an alert. The alert is sent back through Centricity to the physician, who will have the opportunity of clicking a button to open the alert, or ignoring it. The study will analyze how often the alert is opened and whether the information is utilized in care.
To work, Centricity must collect the correct data elements and provide standardized language so the CDC's rules engine can identify common terms, according to Dente. Information entered by the physician is mapped automatically to a vocabulary set so that the CDC system knows what to look for among its food-borne illness guidelines. The program will use Continuity of Care Document and Health Level 7 International standards.
Interop, clinical decision support push CDC guidelines to providers
Receiving alerts via an EHR system is not new. Doctors routinely see pop-up screens for a variety of reasons. For example, the system might send a medication alert if the doctor tries to prescribe a drug to which a patient is allergic.
Those types of guidelines, however, are embedded right in the EHR system, within the hospital's walls. The project with the Alliance of Chicago and GE seeks to demonstrate the data exchange and interoperability capabilities of clinical decision support, the alliance's Rachman said.
Hospitals lack the manpower and bandwidth to monitor conditions that change constantly in real time, such as food-borne illnesses and outbreaks of infection. "This is not something that's very manageable at the EHR level," Rachman said. By reaching into the CDC's guidelines, clinicians have access to wider public health surveillance information to help them make diagnoses.
And getting the right feedback to the right population quickly improves overall patient care, according to GE Healthcare's Dente. "We're very interested in the whole concept of population health."
The partners believe this type of real-time analysis is scalable as well. GE Healthcare is talking to the CDC about using its technical infrastructure for other areas of public health surveillance. "We need to make sure our assumptions are correct," Dente said. "But this could easily capture a whole slew of events."
Edited April 8, 2011 to reflect the focus of the pilot program on primary care settings, not hospitals.
Let us know what you think about the story; email Jean DerGurahian, Executive Editor.
How clinical decision support can improve medication management
How meaningful use could drive clinical data analytics adoption