The Maestro Project: A Patient Navigator for the Transition of Care for Youth With Type 1 Diabetes

The Diabetes Education Resource for Children and Adolescents (DER-CA), in Winnipeg, Manitoba, Canada, was started in 1985. It was the first centralized, regional pediatric program in Canada that was funded outside of a global hospital budget, yet physically located in the Children's Hospital. The program has a case management system and single reporting structure to a program director. The DER-CA team currently includes two diabetes nurse educators, two dietitians, a social worker, three pediatric endocrinologists, and five support staff members. The current caseload includes 550 youth with type 1 diabetes and 175 youth with type 2 diabetes.

The concept of a patient navigator program arose from the cancer care literature. The first patient navigation program was created in 1990 in Harlem, N.Y., by Harold P. Freeman, MD. This was an innovative program that assisted low-income women in overcoming barriers to breast cancer screening and follow-up care.¹¹  Since this pioneering work, there has been an expansion in programs, especially in the United States. In 2007, the American Cancer Society funded more than 60 patient navigation services across the country. Although several definitions of patient navigation have been published, common functions include assisting patients with accessing and coordinating services, identifying individual barriers to care, and providing emotional support and advocacy.¹² 

The Maestro Project evolved from this literature as a unique program designed to help young adults with type 1 diabetes navigate through the complex adult health care system. Before 2002, there were no reports in the literature that described a comparable navigation service during transition from pediatric to adult diabetes care. The Maestro Project started as a 2-year pilot project. It was initially funded by the Lawson Foundation, a family foundation that provides grants for sustainable community-based programs that improve quality of life for Canadians. In 2001–2002, the foundation introduced a nationwide diabetes funding strategy that included a grant to the Maestro Project.

The patient navigator is not a health professional and provides no medical advice or diabetes education. She works closely with community-based health centers in all Manitoba Regional Health Authorities involved in diabetes education to promote community linkages and improved communication.

The navigator has initiated several alternate methods of service delivery for young adults, including a comprehensive Web site (www.maestroproject.com) and a resource book that is given to all youth at their last visit to the DER-CA. There is also a newsletter that is e-mailed to participants, a casual evening drop-in coffee group, and evening group educational events. These events are designed to encourage socialization with peers and to introduce and facilitate relationships with community services and care providers. Events are open to all participants and a guest of their choosing, which helps to increase participants' comfort level and facilitate ongoing education about and awareness of key diabetes issues for the support people in their lives.

Approximately 60–70 youth are transferred annually at 18 years of age from the DER-CA to various private practice adult medical clinics in Winnipeg and rural Manitoba. There are many options for diabetes education for adults with diabetes in Manitoba, including two tertiary care hospitals and various community-based hospitals and health centers throughout the province. The level of experience and expertise with type 1 diabetes varies among these resources.

Since 1995, the DER-CA has had a formal, content-based transition program for adolescents aged 13–18 years in the pediatric clinic to prepare them for transfer to adult care. This includes individual education sessions, written information, and checklists to ensure that all content is taught and reviewed before formal transfer. Expectations for self-care are reinforced, and competency skills are assessed.

However, there was increasing evidence that this was not enough to ensure a successful transfer to adult care. Previous research in Manitoba and other centers in Canada confirmed that young adults were dropping out of care at an alarming rate soon after transfer from pediatric to adult programs.^13–16  Preventable deaths of young adults with type 1 diabetes became the catalyst for change.

We hypothesized that a project designed to enhance transition would improve medical follow-up and reduce adverse outcomes for young adults with type 1 diabetes. The primary aim was to test the feasibility and acceptability of the patient navigator model by this population. The secondary aims were to 1) determine the clinical effectiveness of the Maestro Project on medical and educational follow-up, 2) track the effectiveness of the Maestro Project on medical outcomes, and 3) identify barriers to care for young adults in the adult diabetes care system.

The evaluation study design was a pre- and post-test design. A central database was created in August 2002 for two cohorts of participants in the Maestro Project. An older group and a younger group of participants were identified. The older group (19–25 years of age; n = 164) were graduates of the DER-CA between January 1995 to July 2002. They were transferred to adult care before the inception of the Maestro Project but were later enrolled in the program after August 2002, up to 7 years after graduation from pediatric care. The younger group (18 years of age; n = 84) were graduates of the DER-CA between August 2002 and August 2004. They had the assistance of the patient navigator as they transferred from pediatric to adult care.

Different methods of data collection were used to maximize validity and ascertainment. These included extraction from the Maestro Project database, an audit of hospital and ambulatory clinic records, and participant self-report by survey. This study was approved by the University of Manitoba Health Research Ethics Board.

Evaluation was completed after December 2005, comparing the year before the Maestro Project was created to the year after the Maestro Project was in place. The patient navigator collected information from the participants by telephone every 6 months regarding medical or diabetes education visits, diabetes-related hospitalizations for diabetic ketoacidosis (DKA) or severe hypoglycemia, the presence of chronic complications, and barriers to accessing care in the adult diabetes system. Pre- and post-test results and between-groups comparisons were analyzed by χ² tests of association.

The Maestro Project database provided complete follow-up information for 101 participants in the older group and 64 participants in the younger group, which was used for analysis. The percentages of young adults receiving follow-up in the two groups before and after the patient navigator intervention are shown in Table 1.

The patient navigator helped the older group reconnect with adult medical services; in the year before intervention, only 60% had seen a doctor at least once, and this increased significantly to almost 75% after 1 year (P < 0.001). For education, only 25% had seen a diabetes educator before the intervention; this increased nonsignificantly to 40% after the intervention.

The patient navigator model also helped the younger group reduce their first year dropout rate after transfer from pediatric care. They maintained a similar percentage of medical visits but significantly fewer educator visits (P < 0.001). Before intervention from the navigator, > 40% of the older group had dropped out of adult medical care completely. This dropout rate was significantly lower among members of the younger group, who had access to the navigator at the time of transfer from pediatric care (P < 0.001). Less than 11% of the younger group members dropped out of adult medical care completely after transfer.

Both groups reported acute complications requiring visits to the Emergency Department or admission to a hospital, although no significant differences were found between the two groups. The younger group had 7.9 cases/100 patient-years of DKA and 4.7 cases/100 patient-years of severe hypoglycemia. The older group had 3.0 cases/100 patient-years of DKA and 2.4 cases/100 patient-years of severe hypoglycemia. The younger group reported no long-term complications. The older group reported pregnancy loss of 38%; one case each of heart failure, legal blindness, and an amputation; four cases of proliferative diabetic retinopathy; and four deaths (two from DKA and two unrelated to diabetes). No child under the age of 18 years has died from DKA in the past 25 years in Manitoba.

The record audit was planned not only to verify self-reported information and to determine the extent of recall bias, but also to obtain information about participants' glycemic control. However, we received very few participant consents to access medical records.

Of the audited charts, there were nine females and eight males, with an average age of 25.7 years (range 22–29 years). The average age at diagnosis of type 1 diabetes was 9.5 years (range 5–15 years). Only 5.9% of participants achieved an average A1C of ≤ 7%. The average A1C overall was 8.8% (range 7.0–11.4%). Only 11.8% of participants had medical follow-up three or more times per year after transfer from pediatric care, and only 17.6% had diabetes education follow-up at least once yearly. There were four cases of DKA and two cases of severe hypoglycemia in four individual participants. None of these events were reported to the Maestro Project for inclusion in the database, reflecting recall bias on the part of participants. The majority of participants audited (76%) had large gaps of care in their records, with an average gap of 3.8 years between visits. For half of these participants, it was impossible to tell where or if they had received care during these times.

A survey was mailed to all participants. This survey examined participants' experiences with the health care system after they were transferred from pediatric care and explored whether or not participants felt that there was a need for the services offered by the Maestro Project. Direct quotes from anonymous participant statements written in the surveys are presented in italics below and are used to highlight key themes throughout this section. Only 14% (35/248) of participants returned the surveys: 9 from the younger group and 26 from the older group. This was a very poor response despite much effort by the navigator to try to increase the return rate. All nonresponders received follow-up e-mails and phone calls, and advertisements were placed on the Maestro Project Web site and in the e-mailed newsletters.

Thirty-three percent of the younger group and 42% of the older group experienced difficulties or frustrations with establishing regular follow-up with their new health care team after transfer from pediatric care. Almost half of those surveyed from both groups had a period of time without medical care for > 1 year, and 40% of the older group had gone for ≥ 3 years without medical care.

Reasons for dropping out of care included lack of time, difficulty scheduling appointments and knowing whom to see, difficulty establishing relationships with the new health care team, feeling overwhelmed and lost in the system, and a lack of perceived value. “For a person without diabetes, trying to become independent, move out on your own, go to university, etc., can be difficult and stressful. Unfortunately, with diabetes, I tend to put my health and blood sugars second and focus on all my new challenges, which then makes my diabetes worse and adds to the stress.”

When asked if there was a single factor that motivated those who reconnected with medical care, most reported a fear of chronic complications. Some participants also showed insight regarding the need for self-care and discipline. “I grew up. I finally realized that if I just ignored it, no one would take care of it for me, so I realized that it was my responsibility.”

One hundred percent of participants in the older group and 78% of the younger group felt that there was a need for a transition service. Most participants felt the Maestro Project was valuable because it answered their questions and helped them access services. Many also liked the opportunities to meet other young adults with type 1 diabetes for peer support in a casual, nonjudgmental, open, and respectful environment. “It helps young adults not only make connections with health care teams but also with other young adults.”

Seventy-one percent of the older group and 44% of the younger group felt that the Maestro Project was helpful in motivating or encouraging them to do better with their diabetes management. “I found it easy to put my diabetes on the back burner—just put those appointments off until a better time. Just receiving that newsletter reminds me that it is my responsibility now. Also just knowing that The Maestro is an e-mail away if I have questions is a real relief.”

We acknowledge a number of limitations related to bias, generalizability, and the unique features of our setting. First, the patient navigator, who has a nonprofessional advocacy role with all of the participants, collected data for the database. There may have been recall bias in the information provided by the young adults. Second, participation in the Maestro Project is voluntary, leading to potential selection bias. Third, there was a low response rate to the survey. The results of the study may not be generalizable to all young adults with type 1 diabetes. The young adult population is transient and difficult to contact even with current modes of communication (e-mail and cell phone calls), which resulted in high rates of attrition throughout the study period. Given the quality improvement nature of the project and short duration of the study period, it was not possible to find a cause-and-effect relationship between participation and rates of chronic complications. This study was not designed as a randomized, controlled trial. Therefore, there may be important confounders or effect-modifiers that are not controlled in the comparison of the two groups of participants.

As of June 2010, there were 655 active young adult participants. A total of 877 young adults have participated in the Maestro Project since its inception; some have “aged out” at 25 years of age, and some are no longer participating for various reasons (moved, deceased, or nonresponsive). Since the beginning of the program, 423 referrals have been made for 191 participants to endocrinologists, diabetes educators, mental health professionals, and eye specialists. A total of 226 individual participants have contacted the Maestro Project 595 times for information.

The Maestro Project was recognized as a Good and Leading Practice by the Canadian Council of Health Services Accreditation Team in 2006 and attained program status within the Winnipeg Children's Hospital in 2009.

To be effective, a transition support program needs to have overlap between pediatric and adult services; each system must be able to recognize the differences in the other's health care delivery model, while still working collaboratively.¹⁷  The patient navigator now attends weekly pediatric diabetes clinics at the DER-CA to meet with 16- to 17-year-olds and their families. The navigator introduces the program, works on building a relationship, and obtains their consent for participation. Pre-transitioned youth are now enrolled earlier in the Maestro Project and are invited to use the service and participate in program events before formal referral at 18 years of age.

The DER-CA team, along with the patient navigator, now meets every 6 months with a local adult diabetes education team from a community-based diabetes center in Winnipeg to discuss pending young adult transfer referrals and case conference. In January 2006, a pilot Young Adult Diabetes Clinic was initiated at this adult center as a joint partnership between pediatric and adult endocrinology. The Youville Diabetes Centre is a health resource that offers a coordinated, interdisciplinary approach to diabetes self-management education with nurses and dietitians and includes onsite access to a mental health professional and point-of-care laboratory testing. It is one of several advanced practice adult education centers in the region that accepts referral of youth transitioning from pediatric diabetes services at the DER-CA.

The Young Adult Diabetes Clinic was established to serve 16- to 25-year-olds with type 1 diabetes who are at high risk for defaulting from care, have a history of poor metabolic control, or are disenfranchised and without access to specialty services. It is offered every Thursday evening with a dedicated team of diabetes educators and an adult endocrinologist who attends the clinic one evening per month to provide medical follow-up. The clinic is centrally located and on major bus routes and offers free parking and evening appointments. Walk-ins are permitted, and the clinic provides reminder phone calls to all clients. In the past 4.5 years, there have been almost 3,000 visits, with 200 distinct clients, although this program has yet to be formally evaluated.

Some of the challenges have included a no-show rate of 20–25% and many last-minute cancellations leading to crisis-driven contact, such as for DKA or an unplanned pregnancy. Wait times are increasing as the program enrollment accumulates. The wait time to see the education team is ∼ 3 weeks; for the endocrinologist, wait times can be up to 7 months pending the cancellation list.

It is unknown at this time whether this project will be sustainable. It is also unknown whether it will be possible to broaden the program's availability to include all interested young adults with type 1 diabetes in Manitoba if it is proven to be successful in its current form.

The Maestro Project helped to reconnect the older group with medical care, and it reduced the younger group's first-year dropout rate. However, there is no evidence to suggest that this will translate into improved medical outcomes for participants. Participants of the Maestro Project are willing to take part in this initiative and appreciate the access to support, information, and services that the project provides, including the opportunity for social contact with other young adults with type 1 diabetes.

The Maestro Project recognizes the need to evolve to continue to best meet the needs of young people with diabetes at this stage of their life. This is an ongoing discussion among all of the stakeholders involved. One challenge that has been identified is the need to explore the changing needs and preferred communication styles of the upcoming younger cohort of participants. In an attempt to reach this group, the Maestro Project is now on Facebook. The program is also working to provide a similar navigation service for youth with type 2 diabetes in Manitoba.

The health care system also needs to evolve. There is growing recognition of the vulnerability of young people with diabetes. Instead of requiring young adults to conform to the requirements of the adult system, there is a need for creative ideas for making innovations in service delivery that meet the needs and experiences of these young adults. The young adult clinic in Winnipeg that was described above is one such example of this, despite its challenges.

In conclusion, the patient navigator model has been a successful stimulus for systems change in Manitoba; it has improved services for young adults with type 1 diabetes and reduced rates of those lost to medical follow-up within the first few years after transfer from pediatric to adult diabetes care.

The authors thank Colleen Rand, a chronic disease specialist for the Winnipeg Regional Health Authority, and the staff members at the Diabetes Education Resource for Children and Adolescents and the Youville Diabetes Centre. This project received financial support from the Children's Hospital Foundation of Manitoba, the Lawson Foundation, the Canadian Diabetes Association, and the Manitoba Institute for Child Health.