StarTracker supports disease management and quality improvement
Up close, the health system can be quite impressive, but viewed from a distance it isn’t much of a system at all. When it comes to following widely recognized, evidence-based standards, health care is notoriously inconsistent from one doctor to the next, one hospital to the next, one region to the next.
High variability of practice has led to what experts of the National Academies Institute of Medicine have proclaimed to be a health system quality chasm.
For example, “Only about 30 percent of people with diabetes are receiving the agreed upon standard of care they need,” said Dr. Jim Jirjis, assistant professor of Medicine and director of the Adult Primary Care Center. “The big prize in medicine right now is to make the system work.”
In the future, computers will monitor disease across large patient groups, reminding doctors and their clinical teams when patients are due for testing, evaluation and treatment, and tracking the outcomes of care to support continuous improvement.
Through collaboration of the Adult Primary Care Center and the Informatics Center at Vanderbilt University Medical Center, this vision of the future is taking shape today.
StarPanel, the highly prized electronic medical records system developed at VUMC, is being adapted to identify and monitor patient groups. The system will report patient status to doctors and their teams, sending appropriate reminders when patients are due for further testing and evaluation. The name given this new tool by its Vanderbilt creators is StarTracker.
According to Jirjis, disease management programs can help close the quality chasm, and with new tools like StarTracker such programs will at last be achievable.
The project director for StarPanel and its profuse spin-offs is Dario Giuse, Ph.D., associate professor of Biomedical Informatics. Jirjis is working with Giuse and Dr. Bill Gregg, clinical fellow in the department of Biomedical Informatics, to develop and test StarTracker.
StarTracker is possible because, with the development of StarPanel, Vanderbilt Medical Group clinics recently removed paper-based processes in favor of a fully computerized clinical work environment. With this giant step accomplished, opening the door to disease management means adding special search and notification functions to identify and track high-risk groups.
The electronic record first let users gather up records with a keyword search, which was very handy, for example, in the event of a drug recall, but inadequate for definitive identification of disease groups. StarPanel allowed searches using keywords, demographic parameters and information from the clinic schedule; these search results create patient panels and subpanels that are a boon to clinic efficiency — for example, all my current patients, all my current patients with new lab results, all my current patients on today’s clinic schedule.
StarTracker adds the ability to create panels based on diagnosis codes and lab values, in combination with demographics and clinic scheduling information. This new search engine power lets doctors easily find chronic disease groups, and a StarTracker notification engine will let them monitor patient status.
Without any bother, StarTracker can tell you, for example, which of your coronary disease patients are due for cholesterol testing. Search terms may be strung together to answer very specific questions: Do I have any male patients from Kentucky between ages 60 and 65, with an HbA1c level greater than 10 percent and creatinine greater than 2 percent, who have been diagnosed with depression?
“StarTracker allows each physician to be an epidemiologist,” Gregg said. With StarTracker, the lines begin to blur between routine practice, clinical research and quality improvement.
If lab values alone aren’t adequate for tracking what’s happening with high-risk patients or for supplying data for a study, ad hoc fields can be added to the record to document findings from physical exams or other types of patient evaluation.
StarTracker search and notification currently includes sex, age, diagnosis codes (ICD-9 and CPT codes), and the most common labs for management of chronic disease. Gregg said the notification engine will eventually allow doctors and researchers to create automatic reports that focus on aspects of the patient record that are pertinent for the particular disease management program or clinical study. Notification can be limited to a single team or, as privacy considerations allow, it can be expanded to include selected teams throughout VMG.
Resources for disease
management
As a precursor to full-fledged disease management, Jirjis, Gregg, health systems researcher Dr. Russ Rothman and others are conducting a controlled clinical study of StarTracker’s impact on cholesterol management in high-risk patients.
The study tests what Jirjis refers to as just-in-time disease management. Patient calls to doctors’ offices far outnumber actual patient visits, and as calls come in that require clinical decisions, a messaging function within StarPanel lets staff and doctors efficiently share patient information on an as-needed basis in full compliance with patient privacy considerations.
StarTracker will maintain continuously self-updated panels of patients with diabetes or coronary disease who haven’t lately had their cholesterol checked. Anytime the record of one of these high-risk patients is opened, the system will automatically notify the user that the patient needs testing.
In the first few weeks of this test, six of Jirjis’s patients were brought in after StarTracker notified the team of the need for testing; two of these six were placed on cholesterol-controlling medication.
StarTracker will make it a snap to compare cholesterol management in the test group to that of other patients of the Adult Primary Care Center.
Just-in-time disease management is allowing the Adult Primary Care Center to build a disease management program at a manageable pace, until resources are available for more concerted efforts.
“StarTracker will allow us to apply evidence-based medicine disease management protocols that empower staff to help doctors manage chronic disease groups,” Jirjis said. “Doctors’ schedules will be less burdened with rote tracking functions that are central to disease management protocol, and they will be able to become more focused on acute and more serious complaints and illnesses.”
Most doctors are paid by the office visit or procedure. Money for additional staff follow-up of refractory cases has to come from somewhere, so Jirjis is seeking the funding from those who stand to benefit financially: the insurance companies and other payers.
Insurers already recognize disease management as an effective strategy for controlling morbidity and avoiding costly acute care episodes. Many insurers today use claims data to identify and monitor chronic disease groups among their enrollees; they’ve made it their business to call patients to recommend medical evaluation, tests, equipment and even treatment options.
Jirjis said calls from insurance companies to their ailing enrollees are inadvertently apt to undermine patients’ confidence in their doctors.
Talks have begun and Jirjis is confident that insurers will provide funding for physician-driven case management at Vanderbilt. “For chronic disease, the days of cowboy medicine are over,” he said. “The message to insurance companies is that we’re now ready to play ball.”
In addition to Jirjis, Giuse and Gregg, participants in the project include Dr. Russ Rothman, assistant professor, Health Services Research; Pat Covington, patient care manager, third floor round wing; Bob Carlson, administrator, medicine patient care center; and Dr. Jim Bracikowski, assistant professor of Medicine.