A Systematic Review of Factors Associated With Transition Readiness and the Transfer to Adult Health Care for Young People With Diabetes

Abstract

This systematic review describes factors associated with transition readiness and the transfer of care from pediatric to adult health care services among adolescents and young adults with diabetes. Overall, thirty-three international nonintervention, nonqualitative studies were identified between 2018 and 2023. Collectively, this review’s findings identify targets for intervention and may inform policies and care practices related to transition, such as transition readiness, family and peer support, mental health support, and engagement in pediatric care.

As the medical community aims to improve diabetes-related outcomes, an important strategy for accelerated change is to target efforts to vulnerable subgroups of people with diabetes. One group of people with diabetes who have historically experienced suboptimal glycemic outcomes and high rates of acute complications are those in late adolescence and early adulthood, specifically as they navigate the transition to adult care models and new adult care providers (1–3). Indeed, many people with diabetes in emerging adulthood experience a decline in clinic attendance as well as increased psychosocial challenges during this time (4,5). The need to understand and improve outcomes in this age-group of people with diabetes is increasingly recognized as a priority (6,7).

Within the literature on adolescents and young adults (AYAs) with diabetes, three important concepts are transition of care, transfer of care, and transition readiness. “Transition of care” is defined as “the process by which adolescents and young adults with chronic childhood illnesses are prepared to take charge of their lives and their health in adulthood” (8). Meanwhile, “transfer of care,” a part of the transition process, “defines an event or series of events through which adolescents and young adults with chronic physical and medical conditions move their care from a pediatric to an adult health care environment” (8). In other words, transfer of care includes the establishment in adult health care and typically the end of the transition process. Finally, “transition readiness” represents a multicomponent and multisystemic construct that includes diabetes-specific skills (e.g., adjusting insulin doses) and skills for navigating the health care system (e.g., making appointments) that are believed to prepare AYAs for successful transfer to adult-oriented care (9,10).

Despite the clear need for interventions aimed at AYAs with diabetes, a recent systematic review reveals a lack of consistent evidence for efficacious interventions to improve transition outcomes (11). The interventions included in the review varied in design, but often involved care coordinators, transition-specific clinics, or transition readiness educational programs. Most showed some positive impact on either care satisfaction, clinic visit attendance, or transition readiness, but few had an impact on glycemic outcomes (11). To design and implement effective interventions, it is important to understand factors associated with transition and transfer-related outcomes. To our knowledge, there has been no systematic review of factors associated with transition and transfer-related outcomes among AYAs with diabetes within the past 10 years.

In this review, we aim to describe factors associated with transition readiness and the transfer of care from pediatric to adult health care services. In summarizing what is known about these factors, we aim to support the development of well-informed and more impactful transition-related interventions and policies. We also identify gaps in the current literature and offer suggestions for future research.

Research Design and Methods

A systematic review was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist to promote rigor in review methods and dissemination (12). Authors searched Ovid MEDLINE, PubMed, CINAHL (Cumulative Index to Nursing and Allied Health Literature), Embase, and PsycInfo databases for peer-reviewed articles published between 2018 and 2023 (inclusive) related to transition readiness or the transfer to adult diabetes health care. The time range was chosen to identify new studies published since a related review on diabetes transition was conducted in 2017 (13). Search terms included the following MeSH (medical subject headings) terms and keywords: (type 1 diabetes OR type 2 diabetes OR diabetes mellitus) AND (transition readiness OR transition to adult care OR transfer to adult care).

Inclusion Criteria

Studies were included in the review if they were peer-reviewed articles written in English, focused on type 1 or type 2 diabetes (studies that focused on general chronic illness were eligible for inclusion only if people with diabetes were explicitly included), and described intervention or nonintervention, nonqualitative studies related to either transition readiness or the transfer from pediatric to adult diabetes health care.

Study Design

This article on nonintervention, nonqualitative studies focused on factors associated with transition readiness or the transfer to adult diabetes care. A previous article (11) reported on the intervention studies identified through this review. The remaining publications (i.e., qualitative studies focused on factors associated with transition) were beyond the scope of the current article and will be reported elsewhere. Studies were categorized into those that examined factors associated with 1) quantifiable transition readiness or 2) other metrics of transition or transfer to adult diabetes care.

Study Selection

References from the initial literature search were downloaded to EndNote software and then uploaded to Covidence, a web-based collaboration software platform that streamlines the production of systematic reviews (14). Two of three authors (J.L.P., S.D., and J.W.-B.) independently screened all titles and abstracts. For articles potentially meeting inclusion criteria, full-text articles were retrieved, read, and checked for inclusion criteria. Once eligible studies were identified, they were reviewed by two authors from the larger author group to extract relevant data using a standardized extraction form that was developed by all authors for the purpose of this review. Any discrepancies in data extraction were discussed between the two reviewers, and, if consensus was not reached, a third reviewer was included to reach consensus.

Quality Assessment

To assess study quality, applicable Joanna Briggs Institute checklists were used, including the Cohort Studies (n = 19) and Analytical Cross-Sectional Studies (n = 14) checklists (15). Two authors independently assessed quality, discussed any discrepancies, and deferred to a third reviewer if needed to reach consensus. Supplementary Table S1 displays results of the quality assessment. For cohort studies, scores ranged from 3 to 10 out of 11 (median 8, mean 7.1). For analytical cross-sectional studies, scores ranged from 4 to 8 out of 8 (median 6, mean 5.9). No studies were excluded based on their quality metrics.

Results

Included Studies

As shown in Figure 1, after importing 1,423 articles into Covidence for screening, 661 were determined to be duplicates. The titles and abstracts for the remaining 762 articles were then screened, and 520 were found to not meet the selection criteria. The full texts of 242 articles were then assessed for eligibility, and 159 were excluded, resulting in 83 articles representing 83 separate studies. These were reviewed and categorized as intervention studies (n = 26), qualitative studies (n = 24), or nonintervention, nonqualitative studies (n = 33), with the latter being the focus of this article. Among the 33 nonintervention, nonqualitative studies, 11 were specifically focused on and directly measured transition readiness, and 22 focused on other metrics of transition or transfer.

Figure 1.

PRISMA search strategy.

Studies Focused on Transition Readiness

Table 1 summarizes study characteristics and main findings of the 11 studies that focused on transition readiness. Seven studies were conducted in the United States (16–22), two in Canada (23,24), and one each in Japan (25) and South Korea (26). Most studies were cross-sectional (16,17,19,21–26), whereas two collected longitudinal data (18,20). Sample sizes ranged from 44 to 1,826. Type 1 diabetes was the sole focus for eight studies (16,18–21,23,24,26), and three studies also included youth with other chronic illnesses (17,22,25). Eight studies only gathered data prior to the transfer to adult care (16,18,19,21,23–26). One study focused on caregiver transition readiness (25), whereas the others focused on adolescent and/or young adult transition readiness. When considering measures used to assess transition readiness, four studies (16,17,20,22) used the Transition Readiness Assessment Questionnaire (TRAQ) (27), two (21,26) used the Successful Transition to Adulthood with Therapeutics = Rx (STARx) questionnaire (28), one (19) used the Readiness for Emerging Adults with Diabetes Diagnosed in Youth (READDY) questionnaire (29), one (18) used both the READDY (29) and the Transition Readiness Measure (TRM) (30), one (24) used the Am I ON TRAC? For Adult Care (TRAC) measure (31), one used the On TRAck (23), and one (25) used an unnamed measure developed for the study.

Table 1.

Studies Focused on Transition Readiness (n = 11)

Article	Country, Sites	Study Type, Design	Participants, n	Participant Group	Timing in Relation to Transfer	Transition Readiness Measure	Key Transition Readiness Findings
Al Khalifah et al. (23)	Canada, single site	Observational, cross-sectional	118	Adolescents 13–18 years of age with type 1 diabetes, caregivers, providers	Prior to transfer	On TRAck	Adolescent-self-reported transition readiness and health care provider assessment of transition readiness were positively associated with age, negatively associated with A1C, and not associated with diabetes duration. Health care provider assessment of transition readiness was associated with less diabetic ketoacidosis.
Alwadiy et al. (24)	Canada, single site	Observational, cross-sectional	74	Adolescents 17 years of age with type 1 diabetes	Prior to transfer	TRAC	Lower overall transition readiness was associated with less optimal glycemic control (A1C <9%). Transition readiness knowledge was not associated with A1C.
Chan et al. (16)	United States, multisite	Chart review, cross-sectional	95	Adolescents 14–19 years of age with type 1 diabetes	Prior to transfer	TRAQ	For every 1-year increase in age, there was an increase in overall transition readiness score of 0.03 after adjusting for A1C and sex. With an increase in A1C of 1%, the overall transition readiness score was increased by 0.06 after adjusting for age and sex. Females had higher transition readiness than males after adjusting for age and A1C.
Dattilo et al. (17)	United States, single site	Other, cross-sectional	153 (7 with diabetes)	YAs 18–20 years of age with type 1 diabetes or other chronic illness*	Not specified	TRAQ	Transition readiness indirectly mediated the association between greater perceived maturity and better mental quality of life, but not physical quality of life.
Gutierrez-Colina et al. (18)	United States, single site	Observational, longitudinal	44	YAs 18–23 years of age with type 1 diabetes	Prior to transfer	READDY/TRM	Transition readiness was greater for YAs compared with adolescents and was associated with lower self-management. Transition readiness health behaviors and transition readiness insulin self-management domains were negatively associated with concurrent A1C.
Kamoun et al. (19)	United States, single site	Chart review, cross-sectional	805	AYAs 15–24 years of age with type 1 diabetes	Prior to transfer	READDY	Higher transition readiness confidence was associated with lower A1C. Documentation of provider counseling on certain topics was associated with higher scores for corresponding items on the survey.
Kelly et al. (20)	United States, multisite	Observational, longitudinal	217	Adolescents 17–18 years of age with type 1 diabetes	Prior to and after transfer	TRAQ	Transition readiness self-management was associated with higher friend knowledge/helpfulness. Parental knowledge, patient-centered communication, and friend knowledge/helpfulness were not associated with transition readiness self-advocacy. There was no association between transition readiness and A1C. There were no significant interactions between transition readiness and transfer status.
Kim et al. (26)	South Korea, multisite	Observational, cross-sectional	87	AYAs 16–24 years of age with type 1 diabetes	Prior to transfer	STARx	Family support, self-management competency, and patient age were significantly associated with health care transition readiness.
Ozawa et al. (25)	Japan, multisite	Observational, cross-sectional	179 (26 with diabetes)	Caregivers of AYAs with type 1 diabetes or other chronic illness	Prior to transfer	Unnamed; designed for study	Transition readiness was higher among parents of children with diabetes and cardiovascular disease than those of children with cancer and blood diseases. Transition readiness was not associated with parental health literacy or child’s age.
Richards et al. (21)	United States, multisite	Observational, cross-sectional	165	Children and adolescents 6–17 years of age with type 1 diabetes	Prior to transfer	STARx	Youth who completed chores had higher transition readiness than youth who completed no chores. Higher chores frequency was not more predictive of transition readiness above and beyond the effects of having chores at all.
Ruppe et al. (22)	U.S., single site	Observational, cross-sectional	1,826 (21 with diabetes)	YAs 18–20 years of age with type 1 diabetes, other chronic illness, or other nonchronic illness*	Not specified	TRAQ	College students with a chronic illness reported greater transition readiness, greater depressive symptoms, and a higher percentage of past-year cannabis use compared with peers without a chronic illness.

*Findings were not reported separately for participants with diabetes. YA, young adult.

Only one study examined transition readiness in relation to transfer status and found no association (20). Four studies found positive associations between transition readiness and age (16,18,23,26), and one found greater transition readiness among those identifying as female (16,22). Four studies found that greater transition readiness was associated with optimal glycemic outcomes (18,19,23,24), whereas one study found that it was associated with suboptimal glycemic outcomes (16), and one found no association (20). Greater transition readiness was associated with better self-management behavior in one study (26) and less optimal self- management in another (18). Other constructs that were found to be positively associated with greater transition readiness included perceived maturity (17), responsibility for household chores (21), family support (26), friend support (20), and quality of life (17). Overall, there were no remarkable patterns in findings based on whether the study used a diabetes-specific measure of transition readiness (n = 3 studies) (18,19,23) or a generic measure of transition readiness (n = 8 studies) (16,17,20–22,24–26). A minor pattern that emerged was that studies that used a diabetes-specific measure (18,19,23) all assessed A1C and found a negative association between A1C and transition readiness, whereas the other studies (16,17,20–22,24–26) did not consistently assess A1C, and, of those that did, the associations with transition readiness were mixed.

Studies Focused on Other Factors Related to Transfer of Care

Table 2 details study characteristics and main findings of the 22 studies that focused on factors related to other metrics of the transfer to adult diabetes care. Eight studies were conducted in the United States (32–39), four in Canada (40–43), three in the United Kingdom (44–46), and two in Germany (47,48), and one study each was conducted in Australia (49), Norway (50), Singapore (51), South Korea (52), and the United Arab Emirates (53). Eleven studies collected longitudinal data (32,37,40,42–46,48,50,51), four collected cross-sectional data (33,35,52,53), and seven were retrospective (34,36,38,39,41,47,49). Sample sizes ranged from 54 to 14,616. The majority of studies focused on AYAs with diabetes, although three studies included AYAs with other chronic illnesses (44,45,53), and two each included caregivers (44,45) and diabetes providers (35,52). One study collected data prior to transfer only (52), five studies collected data after transfer only (35,41,48,49,51), and the rest of the studies collected data both prior to and after transfer.

Table 2.

Studies Focused on Factors Associated With Transition or Transfer (n = 22)

Article	Country, Sites	Study Type, Design	Participants, n	Participant Group	Timing in Relation to Transfer	Study Aim; Main Outcomes Examined	Main Findings
Agarwal et al. (32)	U.S., multisite	Observational, longitudinal	182	AYAa 18–25 years of age with type 2 diabetes	Prior to and after transfer	Examine outcomes associated with transfer; age, duration of diabetes, and A1C	Of those between 18 and 25 years of age, 56% reported transfer to adult care, 15% reported no care, and 29% did not transfer. Duration of diabetes and age at diagnosis predicted leaving pediatric care. The status of transfer to adult provider or having no care was associated with a higher likelihood of having an A1C ≥9.0% at follow-up controlling for demographic factors.
Al-Yateem et al. (53)	United Arab Emirates, multisite	Observational, cross-sectional	576 (219 with diabetes)	AYAs 14–22 years of age with type 1 diabetes, type 2 diabetes, or other chronic illness*	Prior to and after transfer	Examine quality of care related to transition; whether provider discussed support, expectations, and options for transition	Participants described their health care providers as doing the following at least some of the time: 41.7% asked if patient had family support for the transition, 38.2% asked about expectations for adult care, and 38.9% asked if patient knew adult care options.
Baker et al. (33)	U.S., multisite	Observational, cross-sectional	247	AYAs 17–19 years of age with type 1 diabetes	Prior to and after transfer	Examine associations with maternal relationship quality; perceptions of maternal relationship quality, PCC, diabetes-related self-efficacy, adherence, A1C, and transfer status	Higher maternal relationship quality was associated with higher PCC, higher PCC was associated with higher diabetes-related self-efficacy, higher diabetes-related self-efficacy was associated with higher adherence, and higher adherence was associated with lower glycemic control. Higher maternal relationship quality was indirectly associated with better adherence through higher PCC. Higher PCC was indirectly associated with better A1C through better adherence. Transfer status moderated the association of PCC with diabetes-related self-efficacy, indicating stronger associations among those who had transferred to adult care. Transfer status did not moderate the associations of PCC with adherence or A1C.
Colver et al. (44)	United   Kingdom,  multisite	Observational, longitudinal	374 (128 with type 1 diabetes)	AYAs 14–18 years of age with type 1 diabetes or other chronic illness and caregivers	Prior to and after transfer	Examine exposure to key transition elements and associations with outcomes; YA clinic attendance, meeting adult provider before transfer, promotion of health self-efficacy, written transition plan, parent involvement, having a coordinator, having a coordinated team, receiving life-skills training, and having a clinical team transition manager	Findings specific to participants with type 1 diabetes: Exposure to recommended features (at the optimal level) was 97% for AYA clinics, 65% for met adult team before transfer, 68% for promotion of health self-efficacy, 29% for written transition plan, 32% for both YA and parent happiness with parent involvement, 50% for care coordinator, 93% for coordinated team, 66% for holistic life-skills training, and 24% for transition manager for clinical team. Appropriate parent involvement, promotion of health self-efficacy, and meeting the adult team before transfer were all strongly associated with improved outcomes.
Colver et al. (45)	United   Kingdom,  multisite	Observational, longitudinal	304 (150 with type 1 diabetes)	AYAs 14–18 years of age with type 1 diabetes or other chronic illness and caregivers	Prior to and after transfer	Examine provider and patient reports of exposure to key transition elements; YA clinic attendance, meeting adult provider before transfer, promotion of health self-efficacy, written transition plan, parent involvement, having a coordinator, having a coordinated team, receiving life-skills training, having a clinical team transition manager, satisfaction with services, mental well-being, and autonomy	Findings specific to participants with type 1 diabetes: 83% experienced an AYA clinic, 30% met adult team before transfer, 95% experienced promotion of health self-efficacy, 26% had a written transition plan, 76% of YAs were happy with parent involvement, 76% of parents were happy with parent involvement, 65% had a care coordinator, 89% had a coordinated team, 56% had holistic life-skills training, and 17% had a transition manager for the clinical team. Over time, satisfaction with services decreased overall but remained stable. Mental well-being was steady overall. Participation and autonomy in appointments increased. Young people with type 1 diabetes reported better quality transitional health care compared with those with cerebral palsy or autism.
Garvey et al. (34)	U.S., multisite	Other (claims data), retrospective	14,616	AYAs 16–24 years of age with type 1 diabetes	Prior to and after transfer	Examine outcomes during adolescence and adulthood; age, gender, race, income, outpatient endocrinology care utilization, ED utilization, and hospitalizations	Endocrinologist visits declined from age 16 to age 22. ED rates increased from age 16 to age 20, then decreased by age 24. Hospitalizations per year increased from age 16 to age 19, then decreased by age 24. YAs experienced higher rates of ED visits and hospitalizations than adolescents. Predictors of ED visits and hospitalizations included female sex and Black race. Individuals with two or more endocrinologist visits per year and those with higher income were less likely to have ED visits and hospitalizations.
Goethals et al. (35)	U.S., multisite	Observational, cross-sectional	531	Providers of YAs with type 1 diabetes	After transfer	Describe provider practices and perceptions; medical records review and access to mental health resources	Although 88% of providers reported reviewing pediatric records as important or very important, only 22% often or always reviewed them. 58% reported easy access to mental health care for YAs, and 50% stated a need for additional resources. Diabetes care and education specialists without easy access to mental health professionals were significantly more likely to report barriers to diabetes management for YAs with depression, substance abuse, eating disorders, and developmental disabilities.
Gray et al. (46)	United   Kingdom,  multisite	Observational, longitudinal	108	Adolescents 14–18 years of age with type 1 diabetes	Prior to and after transfer	Examine associations with clinical course during transition; transfer status, A1C, clinical course, mental well-being, and satisfaction with health care services	Of adolescents who had not transitioned (mean age 16 years), by 4 years later, 81.5% had left pediatric health care services. Median A1C increased significantly from 8.5% at baseline to 9.0% at year 4. Mental well-being was comparable with those in the general population at baseline and was stable over the study period. Satisfaction with health care services during transition was stable. By year 4, the majority had a suboptimal clinical course. There were no significant differences in average well-being, satisfaction, or transfer status between the optimal and suboptimal clinical course groups.
Hodnekvam et al. (50)	Norway, multisite	Observational, longitudinal	321	AYAs 20–26 years of age with type 1 diabetes	Prior to and after transfer	Examine experiences during transition period; age at transfer, contact with health care personnel, continuity of care, satisfaction, and perceived preparedness	The mean age at transfer was 18.0 years (range 15.0–23.5 years). Significant differences in patient experiences were found between pediatric and adult diabetes care in several areas, including contact with health care personnel, continuity of care, interval between consultations, and overall satisfaction. Registry and medical records data confirmed the patient-reported experiences. The longitudinal analyses identified two groups with distinctly different trajectories of glycemic outcomes over time. Patient-provider continuity and perceived preparedness for transfer were the most influential predictors.
Hu et al. (36)	U.S., single site	Chart review, retrospective	237	YAs of unspecified age with type 1 diabetes	Prior to and after transfer	Examine associations between mental health and diabetes care; mental health diagnosis, A1C, time in pediatric care, diabetes-related hospitalizations, days hospitalized, successful transfer, and post-transfer follow-up	Of YAs who transferred care, 70% had ongoing engagement with an adult provider. The median age of transition was 20.9 years, and the median interval between pediatric and first adult visit was 0.64 years. A1C prior to transition was not associated with successful transfer to adult care, but longer time in pediatric care was. 42% were diagnosed with a mental health disorder during pediatric care. Presence of a mental health disorder was associated with higher A1C prior to transition, increased rates of diabetes-related hospitalizations during the transition period, and less consistent follow-up after transition.
Kapellen et al. (47)	Germany, multisite	Chart review, retrospective	1,283	AYAs of unspecified age with type 1 diabetes	Prior to and after transfer	Compare outcomes before and after transfer; A1C, DKA, severe hypoglycemia, microvascular complications, and hospitalizations	A1C increased significantly from 8.95% before transition to 9.20% in the first year after transfer. Frequency of DKA with hospitalization and severe hypoglycemia doubled during transition. Microvascular complications increased based on time between last pediatric and first adult visits.
Kim and Yoo (52)	S. Korea, multisite	Observational, cross-sectional	72	Type 1 diabetes care providers	Prior to transfer	Describe provider perspectives on transitioning youth to adult care; reasons for transfer, impediments to transfer, facilitators of transfer, and communication with adult providers	The main reasons for transfer to adult care were request from a patient or family (69.6%) and age ≥18 years (42.0%). Impediments to transition were long-lasting therapeutic relationship with patient/family (72.9%), lack of adult specialists in type 1 diabetes care (62.9%), no established protocol for transition care (55.7%), no information on adult providers (30.0%), and not having enough time because of low priority (24.3%). Identified facilitators of care included development of transition protocols (79.2%), a multidisciplinary team approach for transition care (52.8%), education of adult and pediatric providers (52.8 and 45.8%, respectively), development of tools for self-care evaluation (40.3%), and provision of experts for adult type 1 diabetes care (29.2%). Communication between pediatric and adult endocrinologists was via nonstructured patient summary (68.6%) and telephone or e-mail (27.1%), overlap of pediatric and adult practice (12.9%), a structured document (7.1%), or without any document or contact (12.9% at the time of transition).
Perry et al. (49)	Australia, multisite	Chart review, retrospective	172	YAs of unspecified age with type 1 diabetes	After transfer	Examine outcomes the first 2 years after the last planned pediatric appointment; number of planned and unplanned health care contacts, specialist care, acute services care, insulin delivery method, patient residential area, and A1C	21% had no planned specialist care, and 49% accessed acute services for diabetes. Those in metropolitan areas and users of continuous subcutaneous insulin infusion therapy were more likely to access specialist care and were less likely to use acute services for unplanned care. Those achieving at least nine planned care contacts in 2 years had a shorter duration between the last pediatric and first adult health care contact, but there were no significant differences in A1C. A1C was associated with number of unplanned health care contacts, but not number of planned health care contacts.
Pundyk et al. (40)	Canada, multisite	Observational, longitudinal	652	Adolescents 16–18 years of age with type 1 or type 2 diabetes	Prior to and after transfer	Compare outcomes between youth with type 1 diabetes and those with type 2 diabetes; medical visits and hospitalizations	76% of youth with type 2 diabetes attended a follow-up visit within 2 years of transition compared with 97% of youth with type 1 diabetes. Youth with type 2 diabetes had higher rates of hospitalization pre- and post-transition and fewer medical visits compared with youth with type 1 diabetes. Achieving 4 visits in 2 years post-transition was predicted by the number of visits pre-transition and diabetes type.
Robinson et al. (41)	Canada, multisite	Chart review, retrospective	1,772	AYAs of unspecified age with type 1 or type 2 diabetes	After transfer	Examine associations between transfer and psychiatric disorders; gap in care, age at transfer, all acute emergency visits, and psychiatric diagnoses	42% had a gap (>180 days) in establishing adult diabetes care; there were no significant associations between gap in care and psychiatric disorders diagnosed in the ED or hospital. Older age at transfer and higher number of all-cause ED visits in the year before transfer increased the risks of psychiatric disorders.
Schweizer et al. (48)	Germany, single site	Observational, longitudinal	190	YAs 17–42 years of age with type 1 diabetes	After transfer	Examine outcomes up to 20 years post-transfer; adult care type, A1C, diabetes-associated complications, family status, parenthood, and living situation	Most patients (92.6%) were in diabetes-specific care during the first year after transfer. About 40% of patients noted a change in the medical institution, sometimes up to three times, in the first 4–5 years after the transfer. 6.1% reported diabetes-related complications. Patients in diabetes-specific care had lower A1C levels. A1C during the year before transfer predicted A1C after transfer. A1C did not differ between those living alone and those living with parents. Married patients and patients with children had lower A1C levels than patients who were single and did not have children. There was an improvement in A1C between transfer and when patients got married or became parents.
Shulman et al. (42)	Canada, multisite	Observational, longitudinal	2,525	AYAs 15–26 years of age with type 1 diabetes	Prior to and after transfer	Examine outcomes during transition age (17–18 years of age) and early adulthood (19–26 years of age); diabetes care visits, primary care visits, DKA, and death	During transition age, 47.0% had a >12-month gap in diabetes care, and 9.5% had no primary care visits. A gap in diabetes care of >12 months and no primary care visits during transition age were associated with an increased risk of DKA or death. 17.7% had at least one admission for DKA during early adulthood (19–26 years of age).
Shulman et al. (43)	Canada, multisite	Observational, longitudinal	2,618	AYAs 15–20 years of age with type 1 diabetes	Prior to and after transfer	Compare outcomes between two provinces that deliver care differently; diabetes-related hospital admissions	Diabetes-related hospital admissions were stable post-transition in both provinces: 13.5% at 15–17 years of age and 14.4% at 18–20 years of age in Ontario (a population with lower socioeconomic status), and 11.4% at 15–17 years of age and 10.5% at 18–20 years of age in Newfoundland and Labrador.
Soliman et al. (37)	U.S., single site	Observational, longitudinal	214	AYAs ≥12 years of age with type 1 diabetes	Prior to and after transfer	Examine associations with gaps in care and transfer; time to transfer, age at transfer, gap in care, mental health provider visits, diabetes educator visits, ED visits, and hospitalization rates	The median time to transition and age at transition were 8.0 months and 21.5 years, respectively. The duration of the gap in care more than tripled for those who did not see a mental health provider pre-transfer. Those who did not see a diabetes educator in pediatrics had an increase in mean age at transition of 2.62 years. Increased gap in care was associated with post-transfer ED visit rate. Increased age at transition was associated with decreases in post-transfer hospitalization and ED visit rates.
Tilden et al. (38)	U.S., single site	Chart review, retrospective	449	AYAs of unspecified age with type 1 diabetes	Prior to and after transfer	Examine outcomes associated with transfer; A1C, days hospitalized, and latency between pediatric and adult visits	Transfer required a median of 5 months and was strongly associated with increased A1C within 1 year of transfer for each 6 months of latency. Transfer latency exhibited a significant association with days spent hospitalized.
Walch et al. (39)	U.S., single site	Chart review, retrospective	54	AYAs 16–21 years of age with type 1 diabetes	Prior to and after transfer	Compare outcomes before and after transfer; age at transfer, length of transition process, A1C, hypertension, and adherence to screening	Mean age at transfer was 17 years. Mean length of transition process was 7.8 months. When comparing before and after transfer, there were no significant changes in A1C, hypertension diagnosis, and lipid or retinopathy screening; adherence to neuropathy and microalbuminuria screening increased.
Zhu et al. (51)	Singapore, single site	Observational, longitudinal	131	AYAs 17–25 years of age with type 1 or type 2 diabetes	After transfer	Examine associations with mental health; A1C, diabetes type, anxiety, and depression	Mean A1C at time of transfer was 8.6%. A1C was higher for those with type 2 diabetes than for those with type 1 diabetes. Persistent anxiety decreased over the first year post-transfer. Those who were persistently anxious had considerably higher A1C over the first year post-transfer than those who were persistently nonanxious or had fluctuating anxiety. Those with depression had lower A1C across the first year post-transfer than those without depression.

*Findings were not reported separately for participants with diabetes. DKA, diabetic ketoacidosis; ED, emergency department; PCC, patient-centered communication; YA, young adult.

The aims of the studies varied. Although many described the characteristics of those who had transferred (e.g., age and type of diabetes) or the transition process (e.g., time between last pediatric visit and first adult visit), others examined predictors of transfer status (32,36,37,40,41,46,49) or compared outcomes prior to and after transfer (39,47). Six studies examined, at least in part, patient, caregiver, or provider perceptions of transition or transfer (e.g., services received and satisfaction) (35,44,45,50,52,53). The main outcomes examined across studies also varied. Transition- and transfer-related variables included transfer status, age at transfer, time between pediatric and adult care, transition-related care and services received, and barriers to and facilitators of transfer. Many studies included diabetes-related medical outcomes (e.g., diabetes type, diabetes duration, A1C, diabetic ketoacidosis, and severe hypoglycemia), other medical outcomes (e.g., hypertension), and health care utilization outcomes (e.g., hospitalizations, emergency department visits, primary care visits, pediatric provider visits, adult provider visits, diabetes educator visits, and adherence to screening). Six studies examined associations between mental health variables (e.g., psychiatric diagnoses, mental well-being, and access to mental health resources) and transfer (35–37,41,45,46,51). Finally, relatively few studies (34,48) examined demographic factors (e.g., age, gender, race, and income) in their primary analyses (34,48).

Given the variability in their aims, outcomes, and participant samples, there were no clear and consistent findings across studies. For example, whereas one study found that A1C increased after transfer (47), another found that it did not increase (39). However, several studies identified factors that were associated with more favorable outcomes during or after transfer (e.g., a shorter gap in care and less acute care utilization), such as older age at transfer (37), more pediatric visits prior to transfer (49), meeting the adult team prior to transfer (44), more frequent follow-up post-transfer (40), parent involvement (44), and promotion of self-efficacy (44). Factors that were associated with less favorable outcomes during or after transfer (e.g., less consistent follow-up) included the presence of a mental health disorder while receiving pediatric care (36), lack of a transition protocol (52), and a longer therapeutic relationship between the pediatric provider and patient (52).

Discussion

This systematic review identified 33 nonintervention, nonqualitative studies published between 2018 and 2023 (inclusive) on factors associated with transition readiness (n = 11) and the transfer to adult health care (n = 22) for AYAs with type 1 diabetes or type 2 diabetes globally. Collectively, this review’s findings identify targets for intervention and may inform policies and care practices related to transition. Numerous gaps in the literature were also identified and highlight critical areas for future research.

Despite the recognized importance of transition readiness (11,54,55), only one-third (n = 11; Table 1) of the 33 included studies directly measured transition readiness. A decade ago, there were at least seven validated generic measures of transition readiness for youth with chronic health conditions (10). Since then, several diabetes-specific measures have been developed, including the READDY (29) and On TRAck (23) instruments, both of which were used in studies included in this review. Future research on diabetes transition should incorporate psychometrically validated measures of transition readiness. Doing so will aid our understanding of how transition readiness is related to transition-related outcomes and can inform how interventions target transition readiness. Research and intervention efforts would also benefit from greater consistency in measure selection, given the challenges of synthesizing results across studies that use different measures. Interestingly, only one (20) of the 11 studies that measured transition readiness examined its association with transfer status. Given that the inherent interest in measuring and improving transition readiness is to facilitate the transfer to adult health care and improve post-transfer outcomes during young adulthood, it is imperative that we build a greater knowledge base of longitudinal data that can inform theoretical models for how transition readiness may lead to successful outcomes. Indeed, transition readiness proves to be a somewhat complex construct, as evidenced by this review’s finding that greater transition readiness was usually, but not always, associated with lower A1C.

Other notable findings include that none of the studies that measured transition readiness included youth with type 2 diabetes. Youth-onset type 2 diabetes is increasing at alarming rates, with rising incidence and worsening severity of disease presentation (56). There is a concerning dearth of research on transition readiness and transition-related experiences among youth with type 2 diabetes that demands greater attention (57). Indeed, there is no validated tool to measure transition readiness specifically among youth with type 2 diabetes. The tools that are diabetes specific—the READDY (29) and On TRAck (23) instruments—contain items that may not be relevant to everyone with type 2 diabetes (e.g., items about insulin management). In addition, although two studies from this review found that family and peer support were associated with greater transition readiness, only one study considered parent readiness. Given the importance of family and parent factors for transition readiness and outcomes (58), better understanding of parents’ readiness for their child’s transition is worthy of additional research.

A range of research has been conducted on factors related to the transfer to adult care for those with diabetes (n = 22; Table 2). Half of all identified studies (11 out of 22) were observational, longitudinal studies and several more examined longitudinal data in the context of a retrospective chart review. This finding is encouraging because the transition to adult health care is, by definition, a process that unfolds over time, and singular snapshots either before or after the transfer to adult health may not expose the nuances of this complex experience for young people with diabetes.

The 22 identified studies focused on the transfer to adult care were conducted across nine countries. There is strength in the international organizations and initiatives dedicated to improving outcomes for people with diabetes around the globe (59). Still, the variability in health care systems within and between countries, along with many other factors (e.g., countries’ standards of living and access to diabetes supplies), creates challenges for synthesizing research across countries. What may help in this regard is more transparent reporting on the health care systems from which participant samples are derived, to provide context for interpreting results (60). Attempts to measure, control for, and compare health care system factors may also be fruitful (61). Several studies in this review attempted to do just that, by asking adult providers about their care practices (53) and assessing AYAs’ perspectives on the specific care elements they receive (44,45) or whether they have access to specialists and resources (37). What will also aid in understanding results across countries is greater attention to and reporting of demographic factors and social determinants of health. Although there have been clear calls for addressing health inequities in diabetes (62), few studies in this review considered sociodemographic factors in their analyses, and none were conducted in low-income countries.

Perhaps some of the most compelling findings from this review are those that identify factors that may contribute to successful post-transfer outcomes because these may be factors targeted in intervention efforts or considered in transition policies. Such factors include those that are specific to individuals and families (e.g., age, mental health, health self-efficacy, and parental involvement) (36,37,44) and those that are specific to the health care system (e.g., number of pediatric visits, relationship between the pediatric provider and the patient, presence of a transition protocol, meeting the adult team prior to transfer, and follow-up post-transfer) (40,44,49,52). Notably, many of the aforementioned factors can be considered to be modifiable and thus promising targets for interventions. A recent systematic review on transition-related interventions (11) found that the majority of existing interventions target health care utilization and metabolic outcomes, with fewer focused on psychosocial outcomes and transition-related knowledge or skills. Findings from the current review suggest that a variety of factors may be related to post-transfer outcomes and that interventions that target both individual-level factors (e.g., health self-efficacy) and system-level factors (e.g., transition protocols) may be the most effective. Future work can help validate these findings, given that there was a lack of consistent outcomes and measurement approaches used across studies in this review.

Limitations

A limitation of this systematic review includes that the classification system used to describe and summarize results was novel to this work and may not have reported features or findings of the included studies. There are likely many nuances to the reviewed studies that may be meaningful but were not reported on given the scope of this review. Another limitation is that articles not published in English were excluded, potentially missing important contributions to the literature.

Conclusion

There is broad recognition that AYAs with diabetes are vulnerable as they navigate the transition to adult health care. Although considerable clinical and research efforts have been made to improve outcomes, this review demonstrates that an ongoing commitment to improving care and outcomes is warranted. The review suggests that, clinically, focusing on transition readiness, family and peer support, mental health support, and engagement in pediatric care would benefit AYAs’ transition to adult health care. Specific recommendations for future research include more longitudinal studies to capture the complexity of how the transition process unfolds over time and how exactly it contributes to the transfer to adult care. More consistency in measurement is also needed, including which outcomes are assessed and which measures are used; more consistent measurement of transition readiness of youth and their caregivers is especially needed. In addition, more attention is needed to transition for AYAs with type 2 diabetes. Finally, more attention to health care system factors can improve integration of research findings across studies and more directly lead to practice guidelines and policies that can improve health care delivery. All individuals with diabetes will benefit if local and global initiatives prioritize efforts to improve the transition to adult health care for AYAs with diabetes.

This article contains supplementary material online at https://doi.org/10.2337/figshare.29803586.