- EHR data analytics, real-time clinical dashboards, and a comprehensive population health management program are helping Arrowhead Regional Medical Center (ARMC), a safety net medical system in San Bernardino County, California, improve patient outcomes and raise the level of quality care.
In a case study on population health management presented at HIMSS15, Chief Medical Information Officer Vikram Kumar, MD, MBA, FAAP, described how participation in a patient-centered medical home model that leverages dedicated care teams and a robust health IT infrastructure has helped better serve the patients in one of California’s largest counties that nonetheless has some of the poorest outcomes.
ARMC is a 456 bed, county-owned safety net medical center with more than 250,000 outpatient visits and 130,000 emergency department visits per year, Kumar said. With two-thirds of its patients identifying as ethnic minorities and 80 percent of its patient population uninsured or insured by Medicare or Medicaid, its population health management challenges, are significant and varied.
ARMC’s dedication to ensuring access to primary care services, preventative care, and chronic disease management led the organization to turn to data analytics and an innovative patient registry that supplements its patient-centered care teams. Within the framework of the PCMH, care teams are organized into “pods” and “podlets” that are each assigned a specific patient population. These teams include nurses, physicians, clinical assistants, and residents who work together to provide services to patients they learn to know well.
Each team uses ARMC’s patient registry to perform risk stratification, conduct care coordination, reach out to patients who are overdue for services or have abnormal test results, and measure their own effectiveness through reporting on outcomes and quality measures. “Our nurses serve as primary contact for the patient,” Kumar explained. “They assist with care coordination and patient education.”
“They also help with self-management goals,” he added. “These are subjective; they are goals the patient picks in coordination with their primary care provider, and it may be something specific such as quitting smoking, or something like committing to walking three times a week or losing five pounds within some timeframe. All these things are small steps towards achieving better population health.”
Education, goal-setting, and patient reminders are driven by the data stored in and reported from the organization’s SQL-based patient registry, which is connected to its enterprise EHR. The registry collects clinical and demographic information, including lab results, vital signs, imaging studies, age, condition, medications, and even addresses and phone numbers to help care coordinators get in touch with those patients who are due for an appointment.
Providers can also run reports using multiple data elements to identify those patients at risk of heart failure who also have diabetes, for example, or those with abnormal weight issues who are also taking a thyroid medication that may need adjustment. Services can be targeted more appropriately to patients with complex chronic care needs when the data is easily available.
“We believe that access to care needs better data analytics,” Kumar said. “This includes sharing of health information both within and without the county, such as clinics, hospitals, behavioral health, and so on. The patient is the center of the interaction, but in the background is the EHR and patient registry which collect data and report it to the stakeholders, so the stakeholders can react appropriately to that.”
Each “pod” is also given a report card derived from patient outcomes and clinical quality measures included in the health system’s data analytics dashboards and registries. Teams may be driven towards improvement through comparisons with other care teams, bringing a healthy element of competition to the population health management process. “It gets them to own the data and react to it accordingly,” Kumar says.
This type of motivation towards continuous improvement isn’t just important for maintaining recognition as a patient-centered medical home, Kumar says. While monthly PCMH workshops and registry training meetings can help fine-tune workflow processes and highlight needs for improvement, reimbursement is also on the line. Between five and ten percent of ARMC’s reimbursements are currently tied to value-based quality and outcomes measures, and that number is set to grow in the near future.
ARMC has already made significant progress towards meeting clinical quality measures and outcome goals that will position it well for value-based reimbursement, Kumar said.
Preliminary data shows that the organization has increased mammogram rates by 20 percent in a short time frame, now reaching around 80 percent of eligible patients. Thanks in part to a mobile pediatric asthma care center, 50 percent of children with the respiratory disorder are now receiving recommended care, up for 20 percent at baseline. “We consider this a big success for us,” Kumar said.
On the inpatient side, ARMC has nearly eliminated the occurrence of pressure ulcers since 2011, when close to 5 percent of patients experienced the painful condition thanks to a stringent focus on quality management. Central line infections have been cut from 4.8 percent in 2011 to just 0.2 percent in 2014. “We’ve seen these infection rates plummet with better data and better analytics,” said Kumar, adding that quality management staff walking the floors has also helped to increase communication, ensure staff compliance, and keep clinical leaders aware of opportunities for continued improvement.
The transition has not been without its challenges, however. The process of pulling unstructured data from process notes and other documents for inclusion in the patient registry has presented its difficulties, while agreeing on which clinical quality measures and technical capabilities to implement required some time and compromises.
Time was also a factor in convincing clinical staff to take on population health management tasks in addition to typical patient care activities, especially when wait times within the safety net system were already long. “It’s very hard to do two jobs at the same time, which essentially population health management becomes,” Kumar acknowledges.
Other staff resistance came from physician partners who are still primarily operating in a fee-for-service atmosphere. Without a financial incentive to go above and beyond the already difficult processes of ensuring patients receive high quality care, asking swamped providers to participate in expanded population health management is a difficult proposition. But ARMC is committed to using its health IT infrastructure and data analytics capabilities to navigate the inevitable transition to value-based reimbursement.
In the future, the organization hopes to expand its clinical systems to other care sites in an effort to manage additional chronic conditions and continue to help the county meet its population health management goals. With a sophisticated analytics background and a firm grounding in patient-centered care, ARMC is well prepared for shouldering additional financial risk through accountable care arrangements.
“We’re going to be going on this journey whether we want to or not, because medicine is moving towards value-based purchasing,” Kumar says. “The Affordable Care Act has placed the onus on providers to do a better job of improving quality and showing that through better outcomes. We need to do a better job of managing the patient’s health and proving to our payers and patients that we can produce better outcomes before we get reimbursed. I think data analytics will be a huge tool for making that happen.”