The Science of Diabetes Self-Management and Care 2025, Vol. 51(5) 497 –504 © The Author(s) 2025 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/26350106251361370 journals.sagepub.com/home/tde
Abstract
Purpose: The purpose of this study is to assess the effects of an intensive telehealth intervention on technology adoption and glycemic control in historically marginalized youth with diabetes mellitus on Medicaid.
Methods: This quality improvement project included youth (ages 12-18) with diabetes utilizing insulin therapy. Eligible participants were diagnosed at least 12 months prior to enrollment and had an A1C ≥9%, Medicaid insurance, and willingness to use a continuous glucose monitor (CGM) and an insulin pump. Participants received scheduled weekly contact (phone/video) with a certified diabetes care and education specialist and monthly video visits with a nurse practitioner.
Results: Youth (N = 18, 61% female, 78% non-White, diabetes duration 4.6±3 years) had baseline mean A1C of 11.4% ± 2.0%; 22% were on pump therapy. There was a sustained improvement between baseline A1C (mean 11.4% ± 2.0%) and 3 months (mean 10.5% ± 2.7%; P = .01) and 6 months (mean 9.8% ± 2.4%, 83.6 mmol/mol; P = .003). Significantly more participants used pump therapy by the end (n = 16) compared to baseline (n = 4; P < .0001). Participants wore CGMs more at 3 (P = .04) and 6 months (P = .0004) during the intervention compared to 3 and 6 months prior.
Conclusions: This telehealth intervention provided interim improvement in A1C and increased adoption of diabetes technology in a low socioeconomic status cohort from a historically marginalized population. Ongoing monitoring is needed to evaluate the durability of this intervention.
The prevalence of type 1 diabetes (T1D) is increasing nationally, with the greatest increase in non-Hispanic White (NHW) and non-Hispanic Black (NHB) youth.1 Managing diabetes requires that families quickly learn how to optimize all of the different factors that impact blood glucose levels, including food intake, physical activity, growth and development, illness, and school planning.
Advances in diabetes-related technologies hold promise to decrease the burden of disease and provide hope for optimizing glycemic management. Data from the Type 1 Diabetes Exchange study have shown that use of a continuous glucose monitor (CGM) in conjunction with multiple daily injections or insulin pump therapy can improve A1C.2 Despite these findings, youth with diabetes using CGMs and conventional insulin pumps meet glycemic targets less than 25% of the time.2 Novel automated insulin delivery (AID) systems are revolutionizing diabetes care with the ability to automatically adjust insulin dosing based on blood glucose sensor trends.3 AID technology also has the potential to improve diabetes quality of life related to sleep, anxiety, and diabetes-related disease burden.4
Disparities remaining within the area of diabetes technology utilization contribute to tech inequities.5,6 Systematic, institutional, operational, and individual factors impact the “digital divide” that leads to disparities within diabetes technology use.7 Socioeconomic status, race, ethnicity, and other social determinants of health may impact the uptake of technology and diabetes management.8-10 Research has shown that NHB youth are less likely to use an insulin pump or CGM compared to NHW.11 Data review from our clinic (urban academic center, Medicaid and private insurance) revealed similar results regarding differences in technology use in youth with insulin-dependent diabetes. However, opportunities exist to break down barriers to diabetes technology use. A recently published study investigating technology inequities in diabetes care found that assisting with insurance navigation and social needs were of paramount importance.12
Routine appointments with medical personnel are essential for achieving optimal glycemic management.13 In-person visits require transportation, school absence for the patient, and work absence for the parent. Making these arrangements can be time-consuming and burdensome. Telehealth visits provide the family with great potential for improved care.14 These visits allow the diabetes team to connect with patients to review glycemic control, provide educational support related to diabetes self-management tasks, and promote the adoption of diabetes technology, all at a fraction of the time required for an in-person visit.15 As a critical part of the diabetes team, diabetes care and education specialists have a key role in promoting and sustaining use of diabetes technologies, providing education and guidance, and advocating for digital health equity.16 The research team’s prior work demonstrated that frequent telehealth check-ins with a nurse and nurse practitioner (both diabetes care and education specialists) have the potential to improve glycemic management in a particularly vulnerable (low socioeconomic status, racial/ethnic minority, high A1C) cohort of youth with insulin-dependent diabetes.17
Given the potential benefit of telehealth for diabetes management, the research team developed a quality improvement telehealth intervention to increase uptake of diabetes-related technologies in a historically marginalized group of youth with presumed T1D under suboptimal control. The aim was to use telehealth to promote CGM and automated insulin pump adoption by providing support beyond routine quarterly clinic visits, with the hypothesis that increased technology use and access to the diabetes care team would help patients improve glycemic control (A1C) and reduce hospitalization/emergency department (ED) usage rates. Prior studies have shown that noncommercial insurance and elevated A1C values in youth with diabetes are risk factors for diabetes associated hospitalizations.18
This project, focused on improving patient care and technology adoption in a vulnerable group of youth, was designated as a quality improvement (QI) intervention.19 The research team sought to build on previous work that demonstrated frequent telehealth visits helped improve glycemic control for youth with insulin-dependent diabetes, on Medicaid insurance, and with a history of high A1C values.17 Due to the QI status, the Institutional Review Board determined the study exempt.
Eligible patients included youth, ages 8 to 18 years, with presumed type 1 insulin-dependent diabetes (as classified by the American Diabetes Association Standards of Care)20 of at least 12 months duration . All participants were established patients at a large urban academic medical center in the Southeast. Participants with negative pancreatic autoantibody titers were not excluded. Participants had Medicaid insurance and A1C ≥9% within the past year. Additionally, youth needed to be willing to wear a CGM and be open to considering the use of an insulin pump. The diabetes team presented information about available insulin pumps so that patients could make an informed, nonbiased decision when choosing a pump. The primary CGM used was the Dexcom G6 with the Dexcom Clarity cloud-based data portal.
The team recruited eligible participants by phone contact or in clinic. Patients received information about the project and then made the decision to participate. Because this was a QI intervention, a study size calculation was not used. Participants served as their own historical controls when evaluating prior A1C and hospitalization rates.
The 6-month intervention consisted of scheduled weekly telehealth (phone or video) check-ins with a certified diabetes care and education specialist (CDCES) nurse and monthly check-ins with a nurse practitioner. The weekly visits with the CDCES included blood glucose review and support of individual family goals focused on diabetes management. The CDCES also provided appointment reminders, care coordination with school nurses and other caregivers, and identification of needs related to social determinants of health. The monthly visits with the nurse practitioner included blood glucose and CGM review, navigation of prescription needs and insurance issues, and promotion of diabetes technologies. Participants were not removed for lack of contact or missed appointments, although staff followed up to reschedule.
Over the 6-month intervention, the study team used chart review to capture quarterly A1C data and diabetesrelated ED visits, hospitalizations, and visit adherence. Race and ethnicity data were also captured from chart review. The study team used the medical record to evaluate pump and CGM usage. Technology usage was considered consistent if a patient had been using the technology in the interval preceding the clinic appointment and at the time of the appointment. CGM and pump consistency were not otherwise tracked. Time in range (TIR) data were not separately tracked, with the definitive measure of glycemic control for this study set at visit-based A1C.
At baseline, families completed a needs assessment to help identify opportunities for clinical support and diabetes education, a 9-item “yes/no” response measure that assessed whether certain elements of diabetes care (eg, seeing social worker, dietitian, learning about pumps or exercise management) would be helpful. At baseline and 6 months, families completed a 10-item diabetes comprehension measure created for the study. This questionnaire asked families to evaluate their comfort with various components of diabetes management (eg, carb counting, sick day management) using a 5-point Likert scale ranging from 1 (not at all comfortable) to 5 (very comfortable). Higher scores indicated greater diabetes comprehension. These measures were used in previously published work by the study team.17 Youth and their caregivers completed the PedsQL Diabetes Module21 at baseline and 6 months to evaluate change in diabetes quality of life over the course of the intervention. The PedsQL Diabetes Module is a 33-item measure that is internationally validated to assess self-report diabetes-specific quality of life.21 Youth also completed the Diabetes Empowerment Scale (DES) at baseline and 6 months to measure change in diabetes selfefficacy over time.22 The DES is a short form and is a reliable and valid measure that includes 8 items using a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores reflecting greater self-efficacy related to diabetes management.22 Additionally, the Diabetes Distress Scale (DDS)23 was given to youth and their caregivers at baseline and 6 months. The youth version includes 17 items using a 6-point Likert scale ranging from 1 (not a problem) to 6 (a very serious problem), with higher scores indicating more diabetes distress. The parent version includes 20 items using a 5-point Likert scale ranging from 1 (not at all) to 5 (a great deal), with higher scores indicating more diabetes distress. The DDS has demonstrated internal reliability and validity and captures important elements of diabetes burden or distress that patients may experience.23 Finally, participants completed a satisfaction survey developed for the QI initiative and utilized in prior work at the same clinical site.17 This satisfaction survey evaluated various components of the project, including the intervention structure, technology, connection, assistance, and challenges, by asking families to indicate whether these elements were “helpful” or “not helpful.”
Statistical analyses were done using GraphPadPrism. Analyses included Fisher exact tests to identify associations between the categorical variable of pump transition and paired t tests to assess glycemic control, quality of life, diabetes distress, and diabetes empowerment changes over time. Descriptive results are reported as mean ± SD. A P value of <.05 determined statistical significance.
Eighteen participants completed baseline requirements, and 17 completed the study. One participant exited the study because the family felt they were doing well and did not need the ongoing intensive support. Participants were on average 14.3 ± 2.0 years (range 11-17) at baseline, with the majority (61%) of the cohort being female and of racial and ethnic minorities (78% self-identified with races other than White; 61.1% Black/African American, 16.7% White, 0.6% American Indian/Alaska Native, 16.7% other). Additionally, 11.1% identified as Hispanic and 88.9% as non-Hispanic. On average, they had a diabetes duration of 4.6 ± 3 (range 1-11) years. One patient did not have T1D antibody results, and 3 were antibody negative; all were insulin dependent. The majority (78%) were using a multiple daily injection (MDI) regimen. The 22% of participants who were using an insulin pump were using AID systems, including Insulet’s Omnipod 5 and Tandem’s t:slim X2 Control IQ. All participants were previously prescribed Dexcom prior to the study, but consistent use at the time of the baseline visit was variable. The mean baseline A1C was 11.4% ± 2% (8.3%-14.1%). See Table 1.
There was a significant cohort improvement in A1C over the course of the 6-month intervention. See Table 2. For patients who transitioned to an insulin pump during the intervention, the A1C significantly improved from baseline 12.18% ± 2.2% (range 7.5%-14%) to 11.0% ± 2.5% (range 7.6%-15.8%; P = .02) at 3 months and to 10.0% ± 2.6% (range 7.3%-14%; P = .004) at 6 months. There was no significant difference in A1C for participants who continued on MDI regimens or were previously on pump therapy.
There was no significant difference in outpatient clinic visit adherence at 3 and 6 months during the intervention compared to prior. There was a significant increase in technology adoption during the QI project. As noted previously, 22% (n = 4) of participants were utilizing AID technology at the beginning of the intervention. At the conclusion of the 6-month project, 89% (n = 16) of participants were utilizing or had transitioned to AID, (P < .0001). The majority transitioned to the Omnipod 5 system. See Figure 1. Of note, 3 participants who started AID systems during the intervention used the technology inconsistently, alternating between MDI regimen and AID. Additionally, participants wore a CGM more at 3 months (P = .04) and at 6 months (P = .0004) during the intervention compared to 3 and 6 months prior.
At the baseline needs assessment, the majority of participants reported that a visit with the diabetes educator (63%), communication with the school nurse (63%), child’s participation in the telehealth visits (69%), wearing a CGM (81%), using an insulin pump (69%), learning more about exercise (75%), and knowing how to contact the diabetes team (56%) would be helpful to their child’s diabetes management. A smaller percentage of participants endorsed a visit with the social worker (38%) or dietitian (50%) would be beneficial.
Families reported high diabetes comprehension at baseline, 3.7 ± 1.0 (range 1.8-5 on Likert scale 1-5), and increased at 6 months, 4.1 ± 0.8 (range 2.5-5), although the difference was not significant.
Quality of life improved for youth (74 vs 67, P < .01) and parents (76 vs 61, P < .0001) at intervention completion compared to baseline (Table 3). Similar trends were seen for diabetes distress in youth (1.5 vs 2, P < .05) and parents (1.8 vs 2.1, NS). There was no significant improvement in diabetes self-efficacy. Participants felt the intervention was helpful; over 85% of participants endorsed each element of the intervention as helpful (ie, appointment reminders, technology support, team support, assistance with prescriptions, school communication, and advocacy) on the patient satisfaction survey.
There was no change in ED visits or hospitalization rates during the intervention compared to 1 year prior. Although not statistically significant, there were more ED visits (a total of 6) in the year prior to the intervention and none during the intervention. In total, there were 8 diabetes-related hospitalizations 1 year prior to the intervention and 8 during the intervention. Two out of the 18 patients had 2 or more hospitalizations during the intervention and in the 1 year prior.
This 6-month intensive telehealth program demonstrated an improvement in glycemic control for a population of vulnerable youth with insulin-dependent diabetes. The results suggest that weekly virtual check-ins with a CDCES nurse specialist and monthly virtual check-ins with a diabetes nurse practitioner provided needed support to help families work toward their diabetes-related goals between quarterly clinic visits. One specific focus of this intervention was to promote diabetes-related technologies, including AID systems. It is well established that diabetes technology improves glycemic control. Historically marginalized individuals with socioeconomic challenges have lower uptake of insulin pumps. Through this intervention, we sought to encourage participating families to adhere to the use of CGMs and pursue insulin pump technology while supporting them closely in this process. The majority of families expressed interest in learning more about insulin pumps and CGMs. Participants wore a CGM more consistently at 3 and 6 months during the intervention than at those time points prior to the interventions. Subsequently, there was a significant increase in the number of youth who transitioned to an AID system, with A1C improvement.
Outcomes related to psychosocial aspects of diabetes also improved. Frequent diabetes team support helped to improve diabetes-related quality of life for families. Additionally, diabetes distress improved for youth over time, and although not significant, parental distress scores were lower at the end of the intervention. These improvements could alleviate some of the burden of living with this chronic condition and help efforts to optimize glycemic control. Interestingly, there was no significant difference in diabetes self-efficacy over time. This may be related to families reporting high levels of self-efficacy at the onset of the intervention. It may also be valuable to assess selfefficacy over a longer duration of time to see if use of AID technology impacts self-efficacy.
Despite these results, this project has limitations. The small sample size of this QI project and short duration of intervention may be limitations that impact the generalizability to the greater clinic population; however, there is potential to adapt and replicate this clinical approach and encourage the uptake of diabetes-related technology for historically minoritized populations of youth living with insulin-dependent diabetes. Variable participant use of CGM both with and without AID systems increased the complexity of attributing glycemic improvement to technology versus the intervention alone and is another limitation. When patients did not have access to a CGM, whether because of pharmacy shortage, early loss of sensor site, or choice not to wear, AID users frequently discontinued pump use as well. During these times, data from fingerstick blood glucose monitoring was not always available via patient report and was sometimes absent entirely. Because CGM use was a requirement and expectation of participation, no other method of blood glucose monitoring was remotely uploaded or assessed. Future efforts to assess the impact of AID and CGM on this patient population would need to facilitate reporting of data when a CGM was unavailable. Additional limitations include lack of access to ED and hospital data for external facilities because sources consisted of patient self-report and chart review.
In closing, this short intensive intervention demonstrated improvement in glycemic control and increased use of diabetes technologies in a population where technology adoption has been challenging. Future studies could pursue larger cohorts of youth over a longer duration of time.
We would like to thank patients and their families for their participation. We would also like to thank Olga Gupta, MD, for her support and feedback on the article.
Lisa Rasbach: Writing - review & editing, Writing - original draft, Visualization, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Ginna Purrington: Writing - review & editing, Writing - original draft, Project administration, Methodology, Investigation, Data curation, Conceptualization. Deanna Adkins: Writing - review & editing, Writing - original draft, Visualization, Supervision, Methodology, Investigation, Conceptualization. Robert Benjamin: Writing - review & editing, Writing - original draft, Visualization, Supervision, Methodology, Investigation, Funding acquisition, Conceptualization.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
This quality improvement initiative was possible because of the generosity of the Innovation Health Grant 2933476
This quality improvement initiative was determined exempt by Duke University Institutional Review Board.
Because this initiative was determined exempt by the institutional review board, the requirement for an informed consent was waived.
Lisa Rasbach https://orcid.org/0009-0007-5139-6585
The data that support the findings of this study are available from the corresponding author, LR, upon reasonable request.
Lawrence JM, Divers J, Isom S, et al. Trends in prevalence of type 1 and type 2 diabetes in children and adolescents in the US, 2001-2017. JAMA. 2021;326(8):717-727. doi:10.1001/jama.2021.11165
Foster NC, Beck RW, Miller KM, et al. State of type 1 diabetes management and outcomes from the T1D exchange in 2016-2018. Diabetes Technol Ther. 2019;21(2):66-72. doi:10.1089/dia.2018.0384
Sherr JL, Heinemann L, Fleming GA, et al. Automated insulin delivery: benefits, challenges, and recommendations. A consensus report of the Joint Diabetes Technology Working Group of the European Association for the Study of Diabetes and the American Diabetes Association. Diabetes Care. 2022;45(12):3058-3074. doi:10.2337/dci22-0018
Farrington C. Psychosocial impacts of hybrid closed-loop systems in the management of diabetes: a review. Diabet Med. 2018;35(4):436-449. doi:10.1111/dme.13567
Lipman TH, Hawkes CP. Racial and socioeconomic disparities in pediatric type 1 diabetes: time for a paradigm shift in approach. Diabetes Care. 2021;44(1):14-16. doi:10.2337/dci20-0048
Agarwal S, Simmonds I, Myers AK. The use of diabetes technology to address inequity in health outcomes: limitations and opportunities. Curr Diab Rep. 2022;22(7):275-281. doi:10.1007/s11892-022-01470-3
Ebekozien O, Fantasia K, Farrokhi F, Sabharwal A, Kerr D. Technology and health inequities in diabetes care: how do we widen access to underserved populations and utilize technology to improve outcomes for all? Diabetes Obes Metab. 2024;26(suppl 1):3-13. doi:10.1111/dom.15470
Burckhardt MA, Addala A, de Bock M. Editorial: equity in type 1 diabetes technology and beyond: where are we in 2022? Front Endocrinol (Lausanne). 2024;15:1400240. doi:10.3389/fendo.2024.1400240
Keenan ME, Berlin KS, Cook JL, et al. Predictors of HbA1c trajectories in predominantly Black adolescents with type 1 diabetes. J Pediatr Psychol. 2021;46(3):241-250. doi:10.1093/jpepsy/jsaa124
Hill-Briggs F, Adler NE, Berkowitz SA, et al. Social determinants of health and diabetes: a scientific review. Diabetes Care. 2020;44(1):258-279. doi:10.2337/dci20-0053
Lipman TH, Willi SM, Lai CW, Smith JA, Patil O, Hawkes CP. Insulin pump use in children with type 1 diabetes: over a decade of disparities. J Pediatr Nurs. 2020;55:110-115. doi:10.1016/j.pedn.2020.08.007
Agarwal S, Crespo-Ramos G, Leung SL, et al. Solutions to address inequity in diabetes technology use in type 1 diabetes: results from multidisciplinary stakeholder co-creation workshops. Diabetes Technol Ther. 2022;24(6):381-389. doi:10.1089/dia.2021.0496
Thomas LV, Wedel KR, Christopher JE. Access to transportation and health care visits for Medicaid enrollees with diabetes. J Rural Health. 2018;34(2):162-172. doi:10.1111/jrh.12239
Dhediya R, Chadha M, Bhattacharya AD, Godbole S, Godbole S. Role of telemedicine in diabetes management. J Diabetes Sci Technol. 2023;17(3):775-781. doi:10.1177/19322968221081133
American Diabetes Association Professional Practice Committee. 1. Improving care and promoting health in populations: standards of care in diabetes-2025. Diabetes Care. 2025;48(suppl 1):S14-S26. doi:10.2337/dc25-S001
Scalzo P. From the Association of Diabetes Care & Education Specialists: the role of the diabetes care and education specialist as a champion of technology integration. Sci Diabetes Self Manag Care. 2021;47(2):120-123. doi:10.1177/0145721721995478
Rasbach L, Purrington G, Adkins D, Benjamin R. Innovations: using telemedicine to improve care of predominantly non- White youth with suboptimal insulin dependent diabetes control. Diabetes Res Clin Pract. 2024;210:111613. doi:10.1016/j .diabres.2024.111613
Mejia-Otero JD, Adhikari S, White PC. Risk factors for hospitalization in youth with type 1 diabetes: development and validation of a multivariable prediction model. Pediatr Diabetes. 2020;21(7):1268-1276. doi:10.1111/pedi.13090
Puri I, Hollingshead CM, Tadi P. Quality Improvement. StatPearls; 2025.
American Diabetes Association Professional Practice Committee. 2. Diagnosis and classification of diabetes: standards of care in diabetes-2025. Diabetes Care. 2025;48(suppl 1):S27-S49. doi:10.2337/dc25-S002
Varni JW, Delamater AM, Hood KK, et al.PedsQL 3.2 diabetes module for children, adolescents, and young adults: reliability and validity in type 1 diabetes. Diabetes Care. 2018;41(10):2064-2071. doi:10.2337/dc17-2707
Anderson RM, Fitzgerald JT, Gruppen LD, Funnell MM, Oh MS. The Diabetes Empowerment Scale-Short Form (DES-SF). Diabetes Care. 2003;26(5):1641-1642. doi:10.2337/diacare.26.5.1641-a
Polonsky WH, Fisher L, Earles J, et al. Assessing psychosocial distress in diabetes: development of the Diabetes Distress Scale. Diabetes Care. 2005;28(3):626-631. doi:10.2337/diacare.28.3.626
From Division of Pediatric Endocrinology and Diabetes, Duke University Medical Center, Durham, North Carolina (Dr Rasbach, Ms Purrington, Dr Adkins, Dr Benjamin).
Corresponding Author: Lisa Rasbach, Division of Pediatric Endocrinology and Diabetes, Duke University Medical Center, 3000 Erwin Road, Durham, NC 27705, USA. Email: lisa.rasbach@duke.edu
