Health Care Transition in Young Adults With Type 1 Diabetes: Perspectives of Adult Endocrinologists in the U.S.

The young adult period presents special challenges for patients with type 1 diabetes, a chronic illness that requires intensive daily self-management and close medical follow-up (1,2). As young adults with type 1 diabetes experience competing life priorities and decreasing parental involvement in diabetes care, treatment adherence and glycemic control may decline. Young adults with type 1 diabetes, like adolescents (3), have been shown to be at risk for poor glycemic control and adverse health outcomes, including acute diabetes complications such as ketoacidosis and severe hypoglycemia, chronic microvascular complications, and early mortality (4–10).

Health care transition has been defined as “the planned, purposeful movement of young adults from child-centered to adult-oriented health-care systems” (11). There is broad consensus that a lack of effective transition from pediatric to adult diabetes care may contribute to fragmentation of health care, decreased frequency of clinical follow-up, and increased risk for adverse outcomes in young adults with type 1 diabetes (2,12).

Empiric data are limited on best practices in transition care, especially in the U.S. (10,13–16). Prior research, largely from the patient perspective, has highlighted challenges in the transition process, including gaps in care (13,17–19); suboptimal pediatric transition preparation (13,20); increased post-transition hospitalizations (21); and patient dissatisfaction with the transition experience (13,17–19). Previous studies (22–24) have also identified differences between pediatric and adult diabetes care environments as perceived by patients, including, for example, an increased focus on the family in the pediatric setting and an increased focus on diabetes complications in the adult setting.

To advance improvements in care, the national landscape of health care transition and post-transition care for young adults with type 1 diabetes requires greater understanding of the barriers, especially with respect to receiving physician perspectives. Available data suggest that adult endocrinologists care for many young adults with type 1 diabetes following their transfer from pediatric endocrinologists. Among a cohort of participants ≥18 years old in the SEARCH for Diabetes in Youth Quality-of-Care Survey sample, 45% received care from an adult endocrinology care provider, compared with 17% from a generalist (25). However, in order to identify the best practices to enhance the transition process, it is critically important to evaluate the experiences of the adult endocrinologists accepting these transitioning young adult patients with type 1 diabetes.

The objectives of this study were to describe experiences reported by a national sample of adult endocrinologists caring for young adults with type 1 diabetes transferring from pediatric to adult diabetes care and to assess the clinical resources and barriers to care for young adults with type 1 diabetes reported by adult endocrinologists.

We developed a structured survey to characterize the clinical experiences, resources, and barriers reported by adult endocrinologists caring for young adults with type 1 diabetes. Academic literature review, patient survey results, and pediatric and adult provider interviews guided survey development. Initial revisions were made following individual and group feedback from pediatric and adult diabetes providers. The survey was then administered to a small convenience sample of eight endocrinologists for cognitive testing prior to final revisions.

The final survey included 60 items and required ∼10 min for completion. The survey was divided into the following domains: 1) practice characteristics, 2) physician demographics, 3) health care transition components and their perceived importance, and 4) description of diabetes care provided for young adults, including resources, barriers, and recommendations for improvement. The survey ended with an optional open-ended question asking whether respondents wanted to share anything else regarding experiences caring for young adults with diabetes. A secure electronic version of the survey was created using Research Data Electronic Capture, or REDCap (26), hosted by the Joslin Diabetes Center in conjunction with user support from Harvard Catalyst. A copy of the survey is available upon reader request to the corresponding author. This study was approved by the Committee on Human Studies at the Joslin Diabetes Center, Boston, MA.

For assessment of components of transition care, the survey asked: “In your experience, how often do the following occur when patients with type 1 diabetes are transitioning to you from pediatric care?” Components included receipt of patient summary, review of pediatric records, communication with pediatric providers, parent involvement in first adult visit, combined pediatric/adult diabetes visit, and participation in a transition program. Response options included the following: never, rarely, sometimes, often, or always. Respondents then indicated the perceived importance of each component, with the following response options: not at all important, somewhat important, important, or very important.

To evaluate clinical resources, the survey asked respondents to first indicate resources to which they had easy access (diabetes educator, dietitian, mental health provider, exercise physiologist, care coordinator, and interpreter) and then to specify resources to which they still needed additional access to care effectively for young adults with diabetes.

For the assessment of barriers, the survey presented a series of clinical scenarios involving a 22-year-old patient with type 1 diabetes. Several scenarios focused on mental health topics (e.g., clinical depression, eating disorder, alcohol/drug abuse, severe fear of hypoglycemia, and developmental disability), while others focused on medical management (continuous glucose monitoring, elite athlete, obesity, and poor glycemic control with recurrent ketoacidosis). For each scenario, we asked respondents to describe the barriers to clinical diabetes management; response options included the following: 1) requires too much additional time, 2) inadequate clinical resources, 3) lack clinical expertise in this, 4) other, 5) none.

Eligible subjects included physicians with a valid e-mail address with specialty listed as “Endocrinology” or “Diabetes” in the American Medical Association (AMA) Physician Masterfile, excluding trainees or physicians with pediatrics or pediatric endocrinology listed as the primary specialty. The AMA Physician Masterfile is a record of current and past data from physicians in the U.S. A physician’s profile is created upon entrance to medical school and is updated with information collected from board certification and state licensure programs, annual AMA census surveys, and an annual online profile update.

We sent an electronic survey to eligible physicians in four waves between July and September 2012. Responses were anonymous. For respondents to proceed with the survey, they had to report caring for five or more young adults with type 1 diabetes, and for inclusion in analyses, respondents had to complete all items related to health care transition. Subjects were offered a choice of a $10 donation to either the American Diabetes Association or JDRF upon survey completion.

All statistical analyses were conducted using SAS version 9.2 (SAS Institute Inc., Cary, NC). Descriptive statistics were presented as the mean and SD or proportions. χ² tests evaluated associations of practice (academic setting, yes/no) and physician (>10 years of experience, yes/no) characteristics with reported components of transition and clinical resources. Because of multiple comparisons across survey questions, a P value of <0.01 was used as the threshold for statistical significance. Following the recognition that mental health services were the only desired clinical resource that was unavailable to >50% of respondents, additional analyses explored factors associated with the absence of mental health services. For the clinical scenarios created to assess barriers, multivariable logistic regression (controlling for years of physician experience and practice setting) provided the odds (with 95% CI) of endorsing barriers in each specific scenario (e.g., depression) according to reported lack of mental health resources.

Thematic analysis was undertaken of the responses to the open-ended question inviting further input from respondents about their experiences caring for young adults with type 1 diabetes. This analysis was iteratively conducted by three members of the research team (K.C.G., G.H.T., and L.M.L.). The team members independently read all free-text comments and marked and categorized key words and phrases to generate initial codes. Codes were organized using Microsoft Excel. Initial codes were discussed by the group, and discrepancies were resolved through consensus on each comment, ultimately generating a list of second-cycle codes. Each team member then applied the revised coding framework to all free-text comments, and the team met to reach consensus on six final themes.

A total of 6,398 physicians in the nonterritorial U.S. were listed in the AMA Physician Masterfile in July 2012 with Endocrinology or Diabetes as the primary specialty (excluding trainees or pediatric physicians). For comparison, the American Board of Internal Medicine recorded 6,384 valid nonterritorial certifications in Endocrinology, Diabetes, and Metabolism as of February 2013 (www.abim.org). We sent the electronic survey to the 4,275 eligible physicians with an available e-mail address. Of these, 4,215 surveys were successfully delivered.

We received 536 responses (13% response rate). Of these, 29 surveys were minimally complete (according to predetermined criteria), 64 physicians sent messages to report their ineligibility (e.g., geriatric providers or working in industry), and 25 physicians cared for fewer than five young adults with diabetes. In sum, 418 surveys were analyzed.

Table 1 displays respondent and practice characteristics. The majority of endocrinologists responding to the survey were male (57%), Caucasian (79%), >45 years old (59%) and had been practicing adult endocrinology/diabetes for >10 years (64%). Sex and age comparison data are available from a recent Endocrine Society analysis (27) of data from the 2011 AMA Physician Masterfile, in which 62% of endocrinologists were male and the mean age was 51 years. Half of the respondents worked in an urban setting, and 42% worked at an academic medical center. A total of 46 states plus the District of Columbia were represented in the sample; only Idaho, Nebraska, South Dakota, and Wyoming were not represented. Examining responses by U.S. Census region, 30% of respondents practiced in the Northeast, 30% in the South, 23% in the Midwest, and 17% in the West. On the basis of the regional breakdown of American Board of Internal Medicine endocrinology certifications as of February 2013, the sample represented between 5.5% and 7.5% of endocrinologists for each of the four census regions (www.abim.org).

Endocrinologists reported multiple referral sources for young adults with type 1 diabetes entering their practice. Referral sources often/always generating new patients in this population included referral by pediatric providers (43%), referral by family or friend (40%), self-referred (28%), referred by health insurance (11%) or student health (9%), or other (10%).

Figure 1 shows the proportion of respondents reporting an occurrence of health care transition components (often/always) for young adults with type 1 diabetes entering their practices. None of these components was endorsed as often/always occurring by more than half of respondents. Having the patient’s parent at the first adult visit was most commonly endorsed (47%), and approximately one-third of respondents endorsed having the opportunity to review pediatric records (36%). Very few endocrinologists reported receiving patient summaries, direct communication with pediatric providers, joint pediatric-adult provider visits, or patient participation in a transition program (all ≤12%). While occurring infrequently, half of the transition components were nonetheless rated as important/very important by the majority of respondents, including patient summary (73%), pediatric record review (72%), and parent presence at the first visit (56%). One-third of respondents rated direct communication with pediatric providers and transition programs as important/very important, while only 10% endorsed the importance of a joint pediatric-adult provider visit.

χ² comparisons showed no significant differences in occurrence of transition care components for endocrinologists with >10 years of experience or those practicing in academic settings.

Figure 2 depicts the availability of clinical resources for young adult diabetes care and reports the need for additional access; availability of and need for additional access were not mutually exclusive. Most endocrinologists reported easy access to diabetes educators (94%), dietitians (95%), and interpreter services (61%). Fewer (42%) reported access to mental health providers (e.g., social worker, psychologist, or psychiatrist). Endocrinologists who reported easy access to mental health referrals for young adults with type 1 diabetes were more likely to practice at academic medical centers (52% vs. 35%, P = 0.0006). Very few endocrinologists reported easy access to care coordinators (15%) or exercise physiologists (16%). Regarding the need for additional resources, mental health (54%) was the only need endorsed by the majority of respondents.

In nine patient scenarios focusing on barriers to clinical diabetes management, specific barrier options included inadequate clinical resources, lack of clinical expertise, and too much additional time required. Of these, the resource barriers were the most highly endorsed, especially for the scenarios involving mental health issues, including substance abuse (47% endorsed the lack of resources barrier), eating disorder (39%), depression (38%), and developmental disability (31%). Resource barriers were less frequently endorsed (10–16%) for scenarios focused on medical management or technology. Similarly, for the mental health scenarios, 30% of endocrinologists endorsed lack of expertise as a barrier for substance and eating disorder cases (both 30%), followed by developmental disability (24%) and depression (15%), in contrast to minimal expertise barriers (1–3%) for recurrent ketoacidosis, obesity, and continuous glucose monitoring. Figure 3 summarizes the report of resource and expertise barriers.

In general, time barriers were less frequently selected. Time barriers were endorsed by 24% of respondents for developmental disability; by 11–13% of respondents for substance abuse, eating disorders, depression, continuous glucose monitoring, recurrent ketoacidosis, and elite athlete; by 9% for fear of hypoglycemia; and by 7% for obesity.

Logistic regression models calculated the odds of endorsing barriers for each specific scenario according to reported lack of mental health resources, controlling for physician experience and practice setting. All model P values (likelihood ratio test) were <0.0001. Endocrinologists without easy access to mental health referrals were most likely to report barriers to diabetes management for young adults with depression (odds ratio [OR] 5.3; 95% CI 3.4, 8.2), substance abuse (OR 3.5; 95% CI 2.2, 5.6), and eating disorders (OR 2.5; 95% CI 1.6, 3.8). In addition, endocrinologists without easy access to mental health referrals were also more likely to report barriers to management for fear of hypoglycemia (OR 2.5; 95% CI 1.4, 4.4) and developmental disability (OR 2.3; 95% CI 1.5, 3.6). Barriers to management for all other scenarios were not significantly increased in endocrinologists without easy access to mental health referrals.

Overall, 75% of endocrinologists reported that they enjoyed seeing young adults with type 1 diabetes (agree/strongly agree), and 56% endorsed wanting to see more young adults with type 1 diabetes in their practice. About half felt that treating young adults with type 1 diabetes required more time (45%) and more resources (45%) compared with older adults with type 1 diabetes. To foster improvements in young adult diabetes care, respondents recommended patient support groups (82% helpful/very helpful), improved reimbursement rates (76%), online provider resources about young adult diabetes management (60%), and continuing medical education about young adult behavioral issues (57%).

A total of 153 of 418 of the respondents (37%) provided comments to the open-ended question inviting further input regarding their experiences caring for young adults with type 1 diabetes. During the coding process, the team reached consensus on six final themes. More than 80% of narrative comment codes were encompassed by the following three themes: 1) intensive time and care coordination requirements of young adults with type 1 diabetes (36%): adult endocrinologists endorsed a need for increased time to spend with young adults as well as a need for increased resources to address young adult social, emotional, and family issues; 2) challenges with nonadherence among young adults with type 1 diabetes (30%): adult endocrinologists frequently cited competing priorities of young adults and their lack of acceptance of the potential severity of type 1 diabetes as major barriers to adherence); and 3) divergent approaches to care by pediatric and adult diabetes providers (18%): adult endocrinologists perceived “coddling” and “excessive” focus by pediatric endocrinologists on avoidance of hypoglycemia at the expense of glycemic control). Table 2 includes representative quotations from each of these three central themes, encompassing 84% of the comments.

The remaining narrative comment thematic categories included financial aspects of young adult diabetes care (7%), the role of family and social support in young adult diabetes care (6%), and (6) other/miscellaneous comments (3%).

To our knowledge, this is the largest study to date examining the experiences of adult endocrinologists caring for transitioning young adults with type 1 diabetes in the U.S.

Results from our sample, representing 418 adult endocrinologists practicing in 47 states, underscore a number of major challenges in the health care transition process.

Only one-third of adult endocrinologists reported the opportunity to review pediatric records of young adults entering their practice, although three-quarters felt that it was important to do so. A minority (<15%) described direct communication (e.g., e-mail or phone) with pediatric diabetes providers or the receipt of a formal transition medical summary. Further, adult endocrinologists in our sample noted that a majority of young adult patients with type 1 diabetes appeared without any physician referral.

These findings of deficiencies in transition care coordination reinforce results from young adult patient (13,20) and pediatric provider (28) surveys showing suboptimal transition planning. Joint expert consensus guidelines from the American Academy of Pediatrics, American Academy of Family Physicians, and the American College of Physicians state the importance of medical record review by the receiving adult provider as well as direct communication between pediatric and adult providers (29). A position statement of the American Diabetes Association, in collaboration with many professional societies, emphasizes the value of a pediatric care summary document, with recommended components including assessment of diabetes self-care skills, summary of past glycemic control and diabetes-related comorbidities, and summary of mental health issues (2). The Endocrine Society has created materials to help ease this transition process, including a comprehensive care summary template (30). Nonetheless, the gaps between national consensus recommendations and current physician practices, as demonstrated in our study, support the need for additional collaborative efforts at individual, practice, and systems levels to enhance provider hand-offs for transitioning young adults with type 1 diabetes.

A number of studies document deficiencies in provider hand-offs across other chronic conditions and point to the broader relevance of our findings. For example, in two studies of inflammatory bowel disease, adult gastroenterologists reported inadequacies in young adult transition preparation (31) and infrequent receipt of medical histories from pediatric providers (32). In a study of adult specialists caring for young adults with a variety of chronic diseases (33), more than half reported that they had no contact with the pediatric specialists.

Importantly, more than half of the endocrinologists in our study reported a need for increased access to mental health referrals for young adult patients with type 1 diabetes, particularly in nonacademic settings. Report of barriers to care was highest for patient scenarios involving mental health issues, and endocrinologists without easy access to mental health referrals were significantly more likely to report barriers to diabetes management for young adults with psychiatric comorbidities such as depression, substance abuse, and eating disorders.

Prior research (34,35) has uncovered the lack of mental health resources in diabetes care. In the large cross-national Diabetes Attitudes, Wishes and Needs (DAWN) study (36), which was not specific to type 1 diabetes or young adults, diabetes providers often reported not having the resources to manage mental health problems; half of specialist diabetes physicians felt unable to provide psychiatric support for patients and one-third did not have ready access to outside expertise in emotional or psychiatric matters. Our results, which resonate with the DAWN findings, are particularly concerning in light of the vulnerability of young adults with type 1 diabetes for adverse medical and mental health outcomes (4,34,37,38).

In a recent report from the Mental Health Issues of Diabetes conference (35), which focused on type 1 diabetes, a major observation included the lack of trained mental health professionals, both in academic centers and the community, who are knowledgeable about the mental health issues germane to diabetes. Our results support a need for increased clinical training programs for mental health providers focusing on the mental health needs of young adults with diabetes, including depression, substance abuse, and eating disorders. In addition, continuing medical education programs for adult endocrinologists on mental health topics may foster an integration of mental and physical health care for young adults with type 1 diabetes.

Our thematic analysis also highlighted provider perceptions of the intensive psychosocial needs of young adults with type 1 diabetes, along with concerns about time constraints in clinical care. Challenges regarding the psychosocial needs of young adult patients have similarly been reported in qualitative work with general internists (39). The third most common theme in our study of adult endocrinologists related to perceived divergent approaches to care by pediatric and adult diabetes providers. This observation calls for future study and an open dialogue between pediatric and adult providers. Qualitative research (24,40) has suggested that pediatric and adult care systems represent two different medical “subcultures” and that the young adult’s lack of preparation for successful independent participation in the adult health care culture may contribute to transition challenges. Pediatric diabetes providers may consider discussing with transitioning patients that adult providers may address glycemic control and diabetes complications in a different manner than experienced in pediatrics.

There are several limitations to this study. The study was limited to physicians with valid e-mail addresses listed in the AMA Physician Masterfile. Nonresponse bias is a concern, though the response rate is comparable with other studies using electronic physician surveys (28,41). Moreover, data were not available regarding demographic characteristics of the nonrespondents. Given the low proportion of respondents (9%) practicing in rural areas, future study is needed to capture the unique needs of transitioning young adults in rural areas. In addition, the results may be biased toward physicians who are interested in thinking about young adults with type 1 diabetes and thus may represent a “best case” scenario of transition care practices. Self-reported survey data are subject to recall bias and, perhaps, to a desire to provide socially desirable answers (though the anonymity of the survey should mitigate this). The survey was conducted in 2012, and the results may not reflect the latest advances in care following publication of expert consensus guidelines in recent years. Finally, this study focused solely on endocrinologists and did not evaluate the experiences of primary care physicians or nurse professionals, who may provide diabetes care for transitioning young adults with type 1 diabetes. Perceptions of young adult diabetes care from different types of adult diabetes care providers, including primary care physicians as well as certified diabetes educators and advanced practice registered nurses, require further study.

In conclusion, our results provide important information about the experiences of adult endocrinologists caring for young adults with type 1 diabetes in the U.S. Our findings support the high importance of enhanced information transfer and direct communication between pediatric and adult diabetes providers, along with efforts to increase mental health provider training and access and to implement educational opportunities for adult endocrinologists on behavioral health topics specific to young adults with type 1 diabetes.

Funding. This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases (K12DK094721, K23DK102655, and P30DK036836), the National Institute of Child Health and Human Development (K24HD060786), and the William Randolph Hearst Foundation. This work was conducted with support from Harvard Catalyst | The Harvard Clinical and Translational Science Center (National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health Award UL1 TR001102) and financial contributions from Harvard University and its affiliated academic health care centers.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. K.C.G. researched the data and wrote the article. G.H.T. and P.F. analyzed the data and reviewed and edited the article. J.S.N. and L.M.L. researched the data and reviewed and edited the article. J.A.F. contributed to the discussion and reviewed and edited the article. K.C.G. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. Parts of this study were presented in abstract form at the 73rd Scientific Sessions of the American Diabetes Association, Chicago, IL, 21–25 June 2013.