Health Care Transition to Adult Care in Type 1 Diabetes: Associations With Student and Employment Status—The SEARCH for Diabetes in Youth Study Free

The SEARCH for Diabetes in Youth (SEARCH) study is a multicenter study aimed at understanding the burden and clinical course of diabetes among youth in the United States. SEARCH recruited youth based on geographical sites in the United States. Sites were located in California, Washington, Colorado, Ohio, and South Carolina (18). SEARCH began conducting population-based study in those diagnosed with diabetes who were <20 years of age in 2000. As part of the SEARCH protocol, a brief survey was first conducted, after which participants with type 1 or type 2 diabetes who were diagnosed with diabetes in 2002–2006, 2008, or 2012 were invited to participate in follow-up research study visits. Participants completed informed consent and assent (when applicable), had blood samples drawn, completed a physician exam, and completed questionnaires. Questionnaires included demographic and clinical questions pertinent to their age-group and length of time in the study. Individuals in selected incident years were invited for follow-up visits. These follow-up visits included additional questionnaires, blood samples, and a physical exam. Institutional review boards for each of the study sites approved this study protocol.

Inclusion criteria for this study was SEARCH participants who were diagnosed with type 1 diabetes between 2002 and 2005 and completed surveys at a follow-up study visit in 2016–2019, when they were between 18 and 26 years of age, during which they provided information about their current diabetes provider, student status, and employment status (Figure 1). We excluded 63 participants with incomplete information. We also excluded 206 participants who stated that they were seeing a provider who treats all ages or answered “don’t know” to whether they changed from a pediatric to an adult diabetes provider. Overall, data for 1,043 participants with type 1 diabetes were included in this study. Eighty participants were excluded from further analysis because they had an unknown transition date or did not report a transition date (either stated that they had always seen their adult provider, did not answer any questions about transition, or could not describe how old they were when they transitioned to an adult provider). Therefore, analysis of diabetes care and complications between transition groups (those who transitioned >2 years ago, those who transitioned within 2 years, and those who remained with a pediatric provider) was performed on a total of 963 participants for whom completed transition status was known.

Demographic characteristics include age at diagnosis, age at study visit, age at transition to adult care (if transitioned), and sex. Age at transition in the survey was asked as a multiple-choice question (option broken down by year for those aged 15–21 years, or “before age 15,” “22 and older,” or “don’t know”). Age of transition was coded as 14 years for anyone ≤14 years of age when transitioned and as 22 years for anyone who transitioned at the age of ≥22 years. Self-reported race and ethnicity, highest education level, combined household income, current health insurance, and current living situation were also obtained.

Participants were divided into four groups depending on their student and employment status: 1) employed student, 2) unemployed student, 3) employed nonstudent, and 4) unemployed nonstudent. Participants were considered students if they identified as “student not seeking employment” or answered positively to the question “Are you currently attending a college, university, or vocational/technical school where you take courses for academic credit?” Participants were considered employed based on their answer to “What is your current employment status?” We included those who indicated that they were full-time, part-time, or selected “other” or “don’t know” and worked ≥10 hours/week. Participants who reported only having seasonal employment were not categorized as employed. Time since transition was calculated by subtracting the age at transition from the age at the study visit when applicable. Transition status was divided into three groups: 1) those who transitioned within the past 2 years, 2) those who transitioned >2 years ago, and 3) those who were still with their pediatric provider. Two years was chosen as the cutoff to compare groups of recently transitioned versus established post-transition status.

Insulin regimen, continuous glucose monitoring (CGM) use, and diabetes provider type were self-reported. A1C level was obtained by blood sample at the study visit.

Complications included DKA events and SH events in the past 12 months, as well as emergency department (ED), urgent care, or hospitalization events for any reason in the past 12 months. These were self-reported. The degree of diabetes care independence was obtained by asking the question “How much of your diabetes care do you do for yourself?” Responses were multiple choice and included “all” or percentage ranges from 0 to 99% self-care, but were dichotomized for purposes of analysis as either fully independent (100%) or partially independent (<100%). Participants who stated that they had changed from a pediatric diabetes provider to an adult diabetes provider were also asked if they have gone ≥3 months without obtaining diabetes care at any time since transitioning to an adult diabetes provider.

Descriptive statistics for baseline demographic and clinical characteristics are presented for the total sample and stratified by student and employment status. Bivariate analyses examined the relationships between student and employment status and baseline characteristics using χ² tests for categorical variables and one-way ANOVA for continuous variables. We then examined the relationship between student and employment status and the odds of transitioning to adult care with and without adjustment for age using logistic regression models. Differences in the prevalence of diabetes complications and independence in performing all diabetes care were compared across adult care transition status groups using χ² tests. Mean A1C levels were compared using one-way ANOVA. Finally, we fit a series of logistic regression models to examine the relationship between student and employment status and likelihood of experiencing diabetes complications and independently performing all diabetes care among the subsample of 258 participants who had transitioned to an adult care provider within the past 2 years. First, we examined the unadjusted main effects of student and employment status in two separate models (Model 1). Next, we fit a model with both student and employment status (Model 2). Finally, we adjusted Model 2 for sex, race, ethnicity, age, education level, insurance type, diabetes duration, and SEARCH study site (Model 3). Interactions between student and employment status were then tested in the fully adjusted model. Of note, this may have limited the ability to detect significant differences between groups given the multiple variables adjusted for.

We included 1,043 participants (56.6% female; mean age 22.2 ± 2.5 years at survey visit [range 18–26 years]; 74.6% non-Hispanic White; mean diabetes duration 11.7 years) with type 1 diabetes in the initial analysis (Table 1). Among them, 313 (30.0%) were employed students, 181 (17.4%) were unemployed students, 424 (40.6%) were employed nonstudents, and 125 (12.0%) were unemployed nonstudents. There were differences between these groups in age, sex, race/ethnicity, diabetes duration, education level, insurance status, marriage status, and whether they were living with parents or relatives at the time of data collection.

Table 2, which includes data for the 963 participants with known transition status, shows diabetes care and complications differences between health care transition status groups. There was no significant difference in A1C between those who transitioned within the past 2 years versus ≥2 years ago or those who had not transitioned care (P = 0.73). Among the subsample who transitioned to adult care (n = 672), those who transitioned ≥2 years ago were more likely to go without diabetes care for ≥3 months (P = 0.005) compared with those who had transitioned within the past 2 years. There was a difference between the three health care transition status groups in terms of being fully independent with diabetes care (P <0.001). Pairwise tests revealed that those who transitioned ≥2 years ago were more likely to perform all diabetes care compared with those who transitioned within the past 2 years (P = 0.012) and compared with those who were still receiving care from a pediatric provider (P <0.001). Rates of DKA, SH, ED and urgent care visits, or hospitalizations in the past 12 months were high in those who transitioned within the past 2 years (22% DKA; 9.8% SH; 45.7% ED, urgent care, or hospitalizations), but did not differ significantly between groups.

In unadjusted logistic regression models testing for main effects of student and employment status, those who were nonstudents were significantly more likely (odds ratio [OR] 2.4, 95% CI 1.8–3.2) to have transitioned to adult care relative to participants who were students. Similarly, participants who were employed were significantly more likely (OR 1.7, 95% CI 1.3–2.3) to have transitioned to seeing an adult diabetes care provider compared with participants who were unemployed. These findings, however, did not retain significance after adjusting for age (P = 0.196 and P = 0.965, respectively). Test of the interaction between student and employment status was not significant (P = 0.708).

When looking at diabetes outcomes based on student and employment status in those who transitioned care within the past 2 years, nonstudents were significantly less likely to be performing all of their diabetes care compared with students (OR 0.5, 95% CI 0.3–0.9, P = 0.023) (Table 3). Student and employment status did not predict the likelihood of going for ≥3 months between diabetes care visits. There was no significant main effect of student or employment status on DKA or SH in any of the models for those who transitioned to an adult diabetes provider within the past 2 years. However, there was a significant interaction in this group between student and employment status when looking at the association with DKA (P = 0.04). Among nonstudents, those who were employed were significantly less likely to experience DKA than those who were unemployed (OR 0.3, 95% CI 0.1–0.8, P = 0.02; data not shown).

This study adds to the literature investigating the association of student and employment status on type 1 diabetes care during the transition from pediatric to adult care and in the post-transition period. Employment and education status were not associated with health care transition status at the study visit. Within the group who transitioned to adult care, those who transitioned from pediatrics recently (within the past 2 years) did not have higher rates of DKA or SH compared with other studies looking at post-transition care (5,9,10). However, rates of DKA and SH were still quite high in these groups. Among nonstudents who transitioned within the past 2 years, those who were unemployed were more likely to have had DKA compared with those who were students and employed. When looking at the whole study population, nonstudents who were unemployed were more often uninsured and had lower rates of CGM and insulin pump use compared with other groups, which may have contributed to increased rates of acute complications in those who had recently transitioned.

Young adults report time constraints, transportation, and cost as barriers to obtaining care (19). Students who are not currently employed may have a combination of these factors as barriers, which may result in decreased care and increased risk of acute complications. Specific focus on addressing barriers in this group of patients may help to decrease rates of acute complications. One study investigated the use of home telemedicine visits in young adults and found an increased number of diabetes visits and decreased hospitalizations in those who participated in the program (20,21). Telemedicine is becoming increasingly common compared with the era before the coronavirus disease 2019 pandemic. Programs involving telemedicine and group visits such as these could help address the barriers of time, transportation, and cost in young adults.

Young adults who transitioned ≥2 years ago were more likely to independently perform all diabetes care. They were also more likely to go ≥3 months without seeing a diabetes provider compared with those who had transitioned within the past 2 years. Other studies have shown a decrease in diabetes care and visits in the early period after transition (6–8). Studies have also shown an association between clinic attendance and glycemic control and DKA (22,23). Our study shows that a decrease in accessing diabetes care persists beyond the immediate post-transition period. This is the first study to identify continued decreased visit attendance even after the immediate post-transition period, suggesting that continued monitoring and assistance may be needed even years after the initial post-transition period to assess the necessary frequency of visits on an individual basis.

Among those who transitioned within the past 2 years, those who were nonstudents were less likely to be independently performing all of their diabetes care compared with students. This study found that those who were unemployed nonstudents were more likely to be living at home with parents, but those who were employed nonstudents were least likely to be living at home with parents, which makes it difficult to discern whether parental help explains the lower rates of autonomous care in nonstudents.

To our knowledge, this study is the first in the literature to examine the transition from pediatric to adult diabetes care in the context of student and employment status in individuals with type 1 diabetes. Limitations of this study include that, given the nature of the survey process, participants at the time of interview were not at a uniform number of years since transition. We also did not have student or employment status at the time of transition to adult care in participants who had already transitioned; we only had their current status, which may have changed over time. We could not identify A1C values at the time of transition to adult care. Additionally, there is no uniform transition process or program among all SEARCH sites, so different sites may have varying degrees of transition support for their adolescent and young adult patients. However, differences between groups analyzed in this study remained even after controlling for the SEARCH site.

This study found that nonstudents with type 1 diabetes who were unemployed had the highest risk of DKA after transitioning to adult care, specifically in the first 2 years after transitioning. It is unclear what additional factors (e.g., behavioral or mental health issues or additional chronic conditions) may have prevented them from participating in work and school activities, which may be related to the increased DKA risk around the transition of care. Additional strategies to ensure continued access to care may be needed in this group to improve diabetes management and decrease rates of DKA. Additionally, it is important that diabetes care teams inquire about student and employment status to help identify susceptible individuals. Further work is needed to evaluate student and employment status and improve processes for transitioning young adults with type 1 diabetes from pediatric to adult care to increase the frequency of diabetes visits and identify successful interventions to decrease rates of acute complications.

The SEARCH for Diabetes in Youth study is indebted to the many youth and their families and health care providers, whose participation made this study possible.

The SEARCH for Diabetes in Youth Cohort Study (1R01DK127208-01, 1UC4DK108173) is funded by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, and supported by the Centers for Disease Control and Prevention. The Population Based Registry of Diabetes in Youth Study (1U18DP006131, U18DP006133, U18DP006134, U18DP006136, U18DP006138, and U18DP006139) is funded by the Centers for Disease Control and Prevention (DP-15-002) and supported by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention and the National Institute of Diabetes and Digestive and Kidney Diseases.

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

S.M. wrote the first draft of the manuscript. S.M., A.J.R., R.P.W., and C.P. designed the present research question. C.K.S. and B.A.R. analyzed the data. A.J.R., C.K.S., B.A.R., F.S.M., S.M.M., S.C., K.R., G.I., R.P.W., and C.P. reviewed, critically revised, and approved the manuscript. S.M. 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.

Parts of this study were presented in poster form at the virtual 81st Scientific Sessions of the American Diabetes Association, 25–29 June 2021.