Future Thinking in Diabetes Care: The Impact of New Practice Models

On the second day of the second annual meeting on diabetes management held by The American Journal of Managed Care in Princeton, NJ, the keynote address was delivered by Robert Gabbay, MD, PhD, chief medical officer and senior vice president at the Joslin Diabetes Center.

Gabbay focused on discussing quality issues associated with patient care, using steps successfully implemented at Joslin as examples. He discussed the influence of payment reforms by the Affordable Care Act on improved quality in patient care. He also addressed patient engagement, particularly the influence of digital solutions.

Diabetes and its associated secondary health issues are a huge drain on the healthcare system. Over 8% of the population in the United States has diabetes, and the number keeps increasing with an average of 2 million new cases annually. More than 40% of end-stage renal disease and kidney failure is diabetes-associated, and diabetes results in about 65,000 amputations per year. “If you think our health system is broken now, imagine one-third of all patients having diabetes,” said Gabbay.

Almost a quarter of a trillion dollars are spent in direct healthcare costs for diabetes, and a typical person with diabetes costs twice as much in healthcare as one without diabetes. The bulk of the costs are hospital-related, with provider and pharmaceutical costs comprising a relatively small percentage. An evidence base developed with clinical trials has linked most diabetes-related costs to preventable complications, complications that can be controlled by adhering to standards such as getting glycated

hemoglobin (A1C) to <7 for most patients, keeping blood pressure at <140/90 mmHg, and using statins, aspirin, and ACE inhibitors as preventive therapy.

Gabbay believes that early screening for complications can lower long-term costs. “Only 14% of people with diabetes are at goal for A1C, blood pressure, and cholesterol,” he said. “That is an improvement from a few years ago when it was only 8%, but there are still 86% who are not where they need to be. So what is the problem? Why is this happening?”

He said a major change in physician outlook is necessary—shifting from thinking of a single patient at a time to thinking about the entire population of patients associated with a practice. The electronic health record (EHR) system can prove to be a tremendous resource in such a population analysis. Essentially, the capability exists to build a diabetes registry through EHRs that can offer valuable feedback on disease-related quality metrics. Gabbay acknowledged the incredible vision of the founder of the Joslin Diabetes Center, Elliott Joslin, who maintained meticulous records of patients with chronic diseases, essentially fostering the epidemiology of patient disease.

Gabbay believes that measuring quality is the first step to improving it. “Most providers overestimate the effectiveness of their care, and that is why we go through a little process when we share those data. So the idea is to measure quality looking at a population and then measure that by provider, by practice, and then ultimately by region and making those data ultimately transparent to drive improvement,” he explained. Providers, Gabbay continued, need to realize and accept the fact that quality measures, though not perfect, serve the ultimate goal of recognizing gaps and improving treatment. The competitive nature of physicians can be used to drive improvement by comparing patient data within or between practices.

This was the model adopted at Joslin to improve quality of care. Patients enrolled at Joslin for treatment, usually those patients who are most unwell, were followed for 1 year in the system. At the end of the year, a 230% improvement in the attainment of A1C, blood pressure, and cholesterol goals was documented. The drop in A1C was from 8.5 to 7.6—since a majority of the patients suffer from significant kidney disease, cardiovascular disease, and hypoglycemia unawareness, it is not safe to drive their A1C to <7. Additionally, the percentages of patients with A1C >9, probably the biggest group to target in any quality improvement initiative, as well as those with very high low-density lipoproteins, were both reduced by about 50%.

In treating chronic diseases like diabetes, more frequent patient monitoring is essential. The easiest steps to improve treatment quality would involve reaching patients identified as having or at risk of having high A1C values, who have not been seen in the clinic for 6 months. Existing treatment strategies are not working well. Ideally, each patient would be connected with a diabetes educator or care manager via an outreach program. The benefits of such a program would be trackable at predetermined intervals, providing, for instance, a monthly report for A1C or percentage of patients at the target A1C. Subsequently, comparing the data among patient populations—those under the care of various physicians, for example&mdash;could provide data to fuel further improvements in care.

Gabbay went on to emphasize the importance of microchanges in improving care, based on a model developed at the Institute for Healthcare Improvement. The model emphasizes implementing small changes in care for fewer patients, at first, to examine the effect of the change before implementing it in all patients, a method called the rapid-cycle test. Based on the results of the change, the procedure can then be tweaked as needed and retested in another small patient population. “It is a series of these small changes that builds, ultimately, toward improvement,” said Gabbay.

According to Gabbay, diabetes professionals have always been at the forefront in healthcare innovation. Out of diabetes management, he noted, came the idea of team-based care (physician, dietitian, diabetes educator); this approach is now widely practiced in treating numerous other diseases. “A plethora of patient-centered medical home initiatives have really focused around diabetes as an initial target disease. It really is the way to reorganize healthcare at the primary care level right now,

and it is our best hope for improving primary care delivery,” he said. Gabbay posited that although the National Center for Quality Assurance (NCQA) recognizes medical homes that meet certain standards, that designation alone may not be sufficient to determine if a practice is truly a patient-centered home.

For the most effective cost control, Gabbay said, a new role should be defined for hospital-based specialists, those based at long-term care facilities within the medical neighborhood. Patients would be stratified: at the bottom would be those with pre-diabetes, for whom lifestyle interventions could suffice. For type 2 diabetes mellitus (T2DM) patients, the specialists and dietitians could help provide primary care. Additionally, a short burst of engagement with a specialist diabetes educator could prove very effective, and pharmacists could also help. For T2DM patients with multiple complications and comorbidities or patients with type 1 diabetes mellitus, the specialists would be the most useful.

The Diabetes Practice Liaison Program, developed at Joslin, provides diabetes around the country, to educate practice staff at various levels on diabetes basics. The more that all staff members know and the closer to the top of their license they can practice, the more effective they can prove to be in helping patients.

There may or may not be value in traditional professional education, typically not perceived as a top priority by many healthcare providers. Developing the most effective method of professional education is the biggest challenge. Academic detailing is one potential solution: short snippets of information can be delivered to providers in a way that can really impact practice. Gabbay discussed an example of an approach recently implemented in Pennsylvania that included academic detailing. This statewide and state-led multi-payer patient-centered medical home initiative brought together 17 payers and over 150 practices that included about 1000 providers. The data from the first rollout in southeastern Pennsylvania were recently published in JAMA; lessons learned from that data were then implemented in other regions, where significantly different results are expected within about 6 months.

“The practices were guided to change through the learning collaborative model using a rapid-cycle test, plan-do-study act (PDSA) cycles, monthly registry quality reporting, practice coaches, and then reimbursement changes,” said Gabbay. These reimbursement changes were tied to NCQA recognition, specifically the 2008 recognition. Gabbay said that an important lesson learned from the initiative was that receiving NCQA recognition doesn’t necessarily equate to improved care. Additionally, improvements in foot exams, eye exams, and screening for nephropathy were observed, as was identifying self-management goals in a negotiated manner with patients, using some simple motivational interviewing techniques.

Gabbay then shifted gears to specifically discuss reimbursement changes that are focused on quality improvement. The goal: for diabetes care centers to be cost-saving institutions, rather than cost centers that need to be subsidized. The bundled payment program, developed by Joslin in collaboration with Michael Porta at the Harvard Business School, allows for outreach activities and efficient and valuable delivery of care without worrying about fee-for-service.

However, substantial challenges exist, such as defining the risk groups, data sharing between payers and providers, inclusions in the bundled payment program, and turnaround time for the investment. The primary concerns for the provider are developing different care models around bundles, and Joslin is actively developing care pathways for patients in different risk groups.

Gabbay ended his keynote address by highlighting the importance of patient engagement in improving the quality of treatment. Joslin’s approach to the process includes motivational interviews, monitoring medication adherence using virtual methods like Skype, and using digital technology such as smartphones to control such patient devices as insulin pumps and glucose monitors. A patient’s smartphone can share collected data with a cloud-based server; following data analysis, the patient receives targeted messages based on his or her status.

EBDM

“At Joslin, we have an affiliate network across the county and in some international locations where we’re using this same kind of approach as a learning network to be able to push out innovation and share those innovative approaches,” concluded Gabbay.

Download PDF: Future Thinking in Diabetes Care: The Impact of New Practice Models