Caring for the Whole Patient With Diabetes Drives ChristianaCare Pilot Success

Published on: 

Collaboration enabled ChristianaCare to significantly grow its population of patients with well-controlled disease, thanks to a diabetes control roadmap that ensured both physical and emotional health needs were met through interprofessional partnerships.

An innovative pilot program instituted by ChristianaCare in 2019 at 4 of its primary care sites targeted patients with poorly controlled diabetes, as gauged by uncontrolled glycated hemoglobin (A1C). Combining primary and supportive specialty care from an endocrinologist and behavioral health counselor, along with deploying a registry of patients with diabetes, allowed ChristianaCare to grow its population of patients with well-controlled diabetes by 16%.

The original goal had been to increase this population of patients by 10%, which means the midsize health care system surpassed its original hoped-for outcome by 60%. There were also reductions in patients with A1Cs above 9% (from 13% to 12%), those who were overdue for A1C testing (from 20% to 9%), and those missing data (from 5% to 1%). (internal data)

Collaboration and cooperation, from clinicians and patients alike, enabled ChristianaCare to see such results by following a diabetes control roadmap that ensured patients’ physical and emotional health needs were met. That guidance consisted of 5 major steps:

  1. Knowing who were their diabetic patients
  2. Monitoring patients’ A1C
  3. Titrating medications as needed
  4. Assessing barriers to diabetes control
  5. Addressing those barriers

Headquartered in Wilmington, Delaware, ChristianaCare counts close to 25 primary care practices, 3 hospitals, a Level 1 trauma center, and a comprehensive stroke center among its numerous facilities.1 Services offered by the health care provider include behavioral care, neurological care, wellness, and more, meeting the needs of patients in Delaware and parts of Maryland, New Jersey, and Pennsylvania.2

Diabetes care, as one of the specialty services provided by ChristianaCare, was chosen as the focus of the pilot program due to the sizeable portion of the patients seen each year who have diabetes. Greater accountability for both quality of care and the costs related to managing the disease were also focal points, as part of ChristianaCare’s move toward a value-based model of care.

“We found at ChristianaCare that 1 out of every 6 patients that we see in our ambulatory settings is diabetic,” said Marina V. Zeltser, MD, MBA, assistant chief medical information officer for population health, director of population health analytics, and a primary care physician at ChristianaCare, in an interview with The American Journal of Managed Care® (AJMC®). “And ChristianaCare is moving very aggressively toward a value-based population health model of care. We’re involved in a lot of arrangements with health plans to take on financial risk for our populations, so much of quality and cost is impacted by how well our diabetics can manage their blood sugar over time.”

The ChristianaCare program’s success builds on an idea originally pioneered by Wayne Katon, MD, in 1995, then of the Department of Psychiatry and Behavioral Sciences at the University of Washington Medical School, who many considered the “father of collaborative care.”3 In study findings published in JAMA that year from a randomized controlled trial, he introduced the idea of integrating behavioral health services into the primary care setting, “to compare the effectiveness of a multifaceted intervention in patients with depression in primary care with the effectiveness of ‘usual care’ by the primary care physician.”4

Following a year-long study of 217 patients classified as having major depression (n = 91) or minor depression (n = 126), Katon and his team saw that a greater proportion of the intervention group (primary care plus psychiatric care) compared with the control group (primary care only) had a higher degree of antidepressant medication adherence and rated that medication as helping somewhat or a great deal.4


Fifteen years after that initial study, The New England Journal of Medicine published the results of a single-blind randomized controlled trial, led by Katon, of 214 patients with a poorly controlled comorbid condition (diabetes, coronary heart disease, or both) and depression.5

Over the 12-month follow-up at the 14 participating primary care clinics, A1C, low-density lipoprotein cholesterol, systolic blood pressure, and Symptom Checklist-20 scores improved more overall in the intervention group (guideline-based care from a medically supervised nurse plus the primary care physician) vs the usual-care group. Better quality of life and a higher degree of satisfaction with their diabetes, coronary heart disease, and depression care were also reported in the intervention cohort participants.5

For over 30 years before his death in 2015, Katon continued to advocate for a collaborative approach to care. “I think we have to move away from the idea that we’re going to train enough psychiatrists or child psychiatrists to treat all people with mental illnesses,” he said during a December 2014 AJMC® panel discussion on mental health. “We do need team-based approaches.”6

The tripling of diabetes cases7 over the last 20 years and the significant increase in pharmacologic treatment options have necessitated multifaceted partnerships “to establish an approach that provides timely access and respects specialty practice capacity,” ChristianaCare reports (internal data).

Having employed the Cerner HealtheRegisteries tool since 2017 to gauge provider and practice performance on key metrics, ChristianaCare used it in 2019 to identify the 4 primary care practices that would deploy a diabetes registry as part of the pilot program:

  • Primary Care at Concord (Chadds Ford, Pennsylvania)
  • Primary Care at Limestone (Wilmington, Delaware)
  • Primary Care at Smyrna (Smyrna, Delaware)
  • Primary Care at Wilson (Wilmington, Delaware)

This registry identifies patients with diabetes by examining data on claims, problems, medications, and lab values, defining those with poorly controlled disease as having a hemoglobin A1C above 9% (internal data).

“We know that if their hemoglobin A1C is above 9%, then we can say unequivocally that their diabetes is out of control and they’re at much higher risk of poor outcomes, organ damage, infections, different reasons for hospitalization,” Zeltser stated.

A root cause analysis soon followed at the 4 practices, incorporating feedback from physicians, medical and clerical assistants, social workers, and psychologists to decipher the top barriers and issues to address for these patients, “really figuring out how do we organize ourselves to meet our patients where they’re at to match their particular needs,” she added.

These data were synthesized into 3 focused areas of improvement:

  1. Optimizing the diabetes registry, by increasing the accuracy of identifying patients with diabetes
  2. Patient outreach, by routinely listing all patients with poorly controlled diabetes and contacting those overdue for routine visits
  3. Transforming care delivery, by integrating with specialty care (endocrinology and behavioral health)

Each of the practices then went one step further by implementing and testing additional measures. Concord utilized the Problem Areas in Diabetes (PAID) questionnaire8 to make referrals to its embedded behavioral health counselor; Limestone placed resource folders in each exam room; Smyrna tested group visits with its endocrinologist; and Wilson prepped cases, via a rounding tool, for discussions with its endocrinologist (internal data).

“We did our best to try and follow the lead of [the] organization and execute the plan of refining our approach to diabetes, some of our goals, our treatment aspects, to better improve this population so that less of our patients are uncontrolled [and more] have better care, better outcomes, and overall feel better,” Raymond E. Carter, MD, clinical leader of primary care at Primary Care at Concord, told AJMC®. “This is an opportunity to really change the way we think about health care and the way we’re approaching it.”

He reiterated the importance of A1C that Zeltser stressed previously, noting that it can act as both a short- and long-term indicator of a patient’s diabetes care continuum, “which can be checked and should be checked with regularity for our patients who have diabetes.”

Something his practice did with the diabetes registry data from ChristianaCare was to involve its first-line workers—their medical assistants, nurses, front-desk staff who greet the patients—in a group exercise to “tease down” what was preventing their patients from receiving optimal care. “We tried to think outside the box,” Carter stressed, “to bring in all aspects of patients’ lives and their care and their situation into how they could or could not [achieve] the desired goal that we wanted for them.”

This involved not only making the registry information useful to the practice’s care providers, but also launching an extensive outreach effort to bring back into the practice fold the patients who hadn’t been seen in a while, to uncover the reasons why, and to personalize the response. Were the patients missing routine lab work? Were recent results not as good as they should be? Were there unknown barriers the patients did not know how to address with their primary care physicians? These were just some of the questions Carter’s practice attempted to address, and they did so with the help of their behavioral health counselor, Edward Feuer, MA.

“A lot of the patients knew a lot about their medical problem, but there were these intangible things that were preventing them from getting some of the care that they needed,” Carter told AJMC®. “Home life, transportation, other psychosocial things… Edward helped guide us in some of these intangibles about home life. It just really taught us that we have to meet the patients on their level.”

Decreasing the stigma still sometimes associated with being a mental health worker drives Feuer’s work at the practice. He helps the providers help their patients manage their conditions better, “which is why I think ‘behavioral health’ sounds better,” he emphasizes.

Often acting as an intermediary, Feuer noted that although he may only see a patient once before referring them out—or when he does see them again, it’s during a visit with their primary care provider—he was reaching out to patients as part of the pilot program, too. The theme of a multifaceted, collaborative approach to diabetes care can be seen in his many calls to his patients with diabetes in between their visits, when he would ask if they are okay or if anything needs to be tweaked.

“They wouldn’t have to wait until the next visit, which I think closed the gap a little bit,” he told AJMC®.

When prompted to further explain how a behavioral health specialist works with patients with uncontrolled diabetes, Feuer told of how he uses education to try and dispel the myths and stigma that continue to surround the chronic condition. He also highlighted the practice’s use of the PAID form, underscoring its importance in assessing diabetes-related stress and measuring progress on the individual patient level, emphasizing that every patient deals with the disease in their own way.

“I think part of what I try to do is to educate people that there’s a lot of new information, a lot of medical care that can be provided now that may not have been provided in one’s parents’ generation,” he said. “Diabetes requires an awful lot of attention from the individual that has it, and being able to have somebody else participate in that process is really important.”

Collaboration continues to be of benefit here, too, because time is such a valuable commodity. “Our emotional health and our physical health are one and the same,” he concluded. “Having a behavioral health consultant in the practice gives patients an opportunity to talk about things that they might not have talked about, or that there might not have been enough time for.”

The ChristianaCare pilot program focusing on diabetes is an ongoing effort. Continuously updated data are now being analyzed according to race, ethnicity, emergency department utilization, and medication optimization for comorbid chronic conditions that may necessitate switching diabetes prescriptions. The new target is to have 80% of patients with well-controlled diabetes, and the program leaders hopes to expand to more primary care sites, to double to 8 the practices that have an endocrinology partner, and to deploy wellness and chronic condition registries to all primary care practices (internal data).

“I think our journey with diabetes and optimizing diabetes care continues indefinitely,” Zeltser concluded. “We’re curious and continue to look for ways to innovate. We need to innovate.”


1. About ChristianaCare. ChristianaCare. Accessed February 10, 2021.

2. Service categories. ChristianaCare. Accessed February 10, 2021.

3. Wayne Katon, MD. AIMS Center/University of Wisconsin, Psychiatry & Behavioral Sciences, Division of Population Health. Accessed February 12, 2021.

4. Katon WJ, Von Korff M, Lin EHB, et al. Collaborative management to achieve treatment guidelines. Impact on depression in primary care. JAMA. 1995;273(13):1026-1031.

5. Katon WJ, Lin EHB, Von Korff M, et al. Collaborative care for patients with depression and chronic illnesses. N Engl J Med. 2010;363(27):2611-2620. doi:10.1056/NEJMoa1003955

6. Dr Wayne Katon, pioneer for collaborative care in mental health, dies at 64. The American Journal of Managed Care®. March 4, 2015. Accessed February 12, 2021.

7. Tran CS, Liddy CE, Liu DM, Afkham A, Keely EJ. eConsults to endocrinologists improve access and change primary care provider behavior. Endocr Pract. 2016;22(10):1145-1150. doi:10.4158.E161321.OR

8. Appendix C. Problem areas in diabetes questionnaire. The Royal Australian College of General Practitioners. Accessed February 17, 2021.