Many patients remain at high risk for diabetes complications because of poor glycemic control.1–4 Case management, defined as “the assignment of authority to a professional (the case manager) who is not the provider of direct health care, but who oversees and is responsible for coordinating and implementing care,”5 is an effective intervention to improve glycemic control.6–8
The use of nurses as case managers (NCMs) for patients with poor glycemic control follows the Chronic Care Model (CCM) of collaborative care in that a proactive approach is undertaken by the health care team to improve outcomes.9 Similarly, the use of NCMs is aligned with the core principles of the Patient-Centered Medical Home (PCMH) model (e.g., enhanced access and coordinated and comprehensive care).10,11 However, research findings have not always shown that NCMs improve clinical outcomes.12
A recent evaluation of the Kaiser Permanente Northern California's care management program suggests that an important consideration for achieving success in clinical outcomes is ensuring that the NCM program encourages needed intensification of medication regimens for patients.13 However, finding and hiring nurses previously trained in glucose pattern management, including having the knowledge to make specific recommendations about adjustment of hypoglycemic medications, may present a barrier to health care organizations seeking to implement an effective NCM program.
This article describes an internal training program for NCMs to improve glycemic control for patients in the Cleveland Veterans Administration (VA) health care system. This quality improvement (QI) training project allowed existing nursing staff members to become diabetes NCMs by providing them with the necessary skills to help address the growing gap between the care needs of patients with diabetes and the level of diabetes expertise available.
The Cleveland VA operates 12 community-based outpatient clinics in northeast Ohio, serving more than 20,000 veterans with diabetes. We recruited community-based outpatient clinic nurses who expressed a special interest in diabetes education and case management and tracked their utilization and patients' outcomes via an electronic diabetes registry.14
Study Focus
We hypothesized that NCM care would achieve greater success than usual outpatient care in improving A1C levels among patients with persistent poor glycemic control.
QI Intervention: NCM Program
Table 1 summarizes the rationale, role definition, and patient-centered goals that were the foundation for obtaining buy-in from key stakeholders. Table 2 outlines elements of the program and highlights components of the CCM and PCMH model addressed.
An experienced certified diabetes educator (CDE) who is a nurse practitioner in the endocrinology section of the Cleveland VA trained each NCM in weekly sessions that included review and discussion of actual patient charts and glucose data. The CDE mentored each NCM in the core principles of diabetes self-management education (DSME) and glucose pattern management, including medication adjustment.15 NCMs were encouraged to study for the CDE exam16 and were given time to attend an annual day-long Cleveland VA retreat on DSME.
Integral to all NCM-patient encounters was DSME and discovery of barriers to self-care. NCMs were trained in clinical protocols developed by the endocrinology section in concert with the Cleveland VA Skeggs Diabetes Center (Tables 2, 3 and 4). Although the targeted population was patients whose A1C levels were ≥ 9.0%, the NCMs also worked with any patients as requested by primary care providers.
Research Design and Methods
This was a retrospective, observational study of a QI intervention. The data source was the VA Integrated Service Network 10 (VISN 10) electronic diabetes registry. The study was approved by the Louis Stokes Cleveland Department of Veterans Affairs Medical Center Institutional Review Board.
Patient groups
Patients with diabetes enrolled in the Cleveland VA Medical Center were defined by at least one diabetes-specific ICD-9-CM diagnosis code (250.xx, 357.2,362.0, 366.41) listed as an active problem in the electronic chart or a diabetes-specific medication dispensed from a VA pharmacy on at least two separate dates between 1 October 2001 and 31 December 2006.14 Insulin-taking patients were defined by having insulin dispensed from a VISN 10 pharmacy in either the baseline or follow-up periods.
The NCM patient group included all diabetic patients with at least one visit to an NCM between April 2003 and March 2005. The concurrent usual care (UC) patient group included all diabetic patients with no NCM visits through October 2005. Patients whose baseline mean A1C was < 7% or who lacked A1C test results at baseline or follow-up were excluded from the comparison analysis but were counted in the general descriptive data.
Time-dependent variables
For NCM patients, the index date was the date of each patient's first visit to the NCM. For UC patients, the index date was defined as 1 October 2004 —the median of all index dates for the NCM group. The mean baseline A1C value was the average of all A1C values for each patient for 12 months up to and including the index date. The mean follow-up A1C value was the mean of all A1C values within the 12 months following the index date.
Statistical methods
Each patient's A1C values were averaged across the 12-month baseline and follow-up periods to minimize bias induced by seasonal shifts in glycemic control17 and regression to the mean. Statistical analysis used Student's t test and χ2 analysis. Statistical software was SAS (Cary, NC), V 9.1.
Study Results
Forty-one percent (647/1,567) of NCM patients were taking insulin at baseline compared to 25% (2,080/8,290) of UC patients (P < 0.01). During follow-up, an additional 14% of the NCM patients started insulin compared to 6% of UC patients (P < 0.01). A greater percentage of NCM than UC patients had mean baseline A1C levels ≥ 9% (39 vs. 25%, respectively). Among patients whose mean baseline A1C was ≥ 9%, NCM patients did not differ from UC patients in mean baseline A1C value (10.4 ± 1.4 vs. 10.3 ± 1.2%, P = 0.14) (Figure 1). However, during follow-up, NCM patients improved their A1C values to a slightly greater extent (9.1 ± 1.6 vs. 9.4 ± 1.6%, P < 0.01). By October 2006, the NCMs had seen a cumulative total of 3,842 unique patients at least once, representing 22% of patients with diabetes in the Cleveland VA system.
Interviews with the NCMs indicated improved job satisfaction; they appreciated the new NCM role as providing a meaningful intervention for their patients. Most part-time NCMs have now been converted to full-time NCMs. From 2003 to 2007, the program grew from 1 to 10 NCMs, all of whom have achieved CDE status. (Only two were CDEs before the NCM program.)
In addition, registered dietitians at the clinic sites who also team-taught the diabetes classes and saw patients for medical nutrition therapy were motivated by the success of the program. An additional six dietitians have become CDEs as well. Together with the primary care providers, they developed a comprehensive team approach to the delivery of diabetes care.
Conclusions
To serve the increasing number of patients with high-risk diabetes within the Cleveland VA, a system-level change was needed consistent with CCM and PCHM models. We demonstrated that NCMs who were recruited and trained from within our own pool of clinic nurses achieved meaningful improvement in glycemic control despite caring for patients with greater severity of diabetes. The NCM program is well accepted, as demonstrated by the growth in numbers of NCMs and patients served. As reported by others,18,19 we found NCMs to be enthusiastic about their new role and CDE status. NCMs have now been trained to perform diabetic foot screenings and to monitor and refer patients for annual eye exams, enabling our system to implement new aspects of comprehensive and coordinated care.
Although the success of our NCM program depended on the talent and leadership of the nurse practitioner CDE who led the program, the administrative support for necessary training and mentoring time for each NCM was essential. One of the benefits of an internal training program is that the NCMs not only form natural teams with the primary care providers in their clinics, but also function as a highly motivated team of experts within the Cleveland VA. We conclude that the diabetes NCM program has enhanced access to intensified care for diabetic patients with poor glycemic control and has improved glycemic outcomes in the Cleveland VA system.
ACKNOWLEDGMENTS
The authors wish to acknowledge the support of the administration of the Cleveland VA Medical Center (William Montague, former chief operating officer; Susan Fuerher, chief operating officer; Murray Altose, chief of medical staff; Scott Ober, medical director of community-based outpatient clinics; Jan Solomon, nursing director of community-based outpatient clinics; Janet Anselmo, RD, CDE, director of the diabetes self-management education committee, and the 10 dedicated NCMs for diabetes.
Dr. Kern was supported by the National VA Quality Scholar Fellowship Program during part of this study. Support for Dr. Lawrence was provided by a System Redesign Improvement Capability Grant from the U.S. Department of Veterans Affairs, VHA Systems Redesign (awarded to Brook Watts, MD, MS). The views expressed in this article are those of the authors and do not necessarily represent the views of the U.S. Department of Veterans Affairs.
