Abstract
Introduction
In recent years, the diabetes community has focused on documenting the presence of inequities in diabetes outcomes and management, focusing particularly in the diabetes technology space. These real-world data repeatedly indicate that glycemia improves with the addition of continuous glucose monitoring, insulin pumps, and hybrid closed-loop systems; these results also demonstrate the persistence of inequities. While these findings are consistent, it is important to acknowledge that published studies do not represent the majority of people living with diabetes, and that understanding the context of real-world evidence is nuanced. Last year's article on real-world data highlighted studies from the United States (1 –3) and Europe (4) that shed light on the socioeconomic and racial/ethnic disparities in technology uptake (5). Identifying and describing these disparities is the first step in advocating for those who are “missing” from the diabetes technology landscape and for which areas need further study for comprehensive real-world evidence. Further elucidating patterns of disparities, delving into causes, and discussion of novel ways to measure disparities are the extension of this work.
This year's article on real-world evidence takes new approaches to examining these contexts, progressing through three themes. First, it aims to broaden the representation of published studies to new countries and regions and spotlights the populations represented in large benchmarking studies. This is important because it is insufficient to assume that all real-world evidence is applicable or representative of all people with diabetes. Second, our article highlights new populations using diabetes technology, to help understand the broad utility of devices for global health. Finally, our article highlights the application of varied methodologies to address disparities in diabetes technology utilization. We highlight four methodological approaches utilized to address disparities with the goal of encouraging the diabetes community to move beyond only documenting disparities by race/ethnicity and socioeconomic status alone and lay a roadmap to action.
This article includes original research articles retrieved from PubMed that were published between July 2021 and June 2022 and contain search terms related to diabetes technologies, including insulin pump, hybrid closed loop, HCL, continuous glucose monitor, CGM, intermittently scanned CGM, isCGM, real-time continuous glucose monitoring, and rtCGM. Important context terms included “disparities”, “real-world use,” “barriers,” “discontinuation,” “practical,” and “clinical care.” Over 300 article titles were reviewed for pertinence and possible inclusion in this article. Of these, 54 abstracts were reviewed in detail, and 17 were selected for inclusion in this article.
Key Articles Reviewed
Marigliano M, Eckert AJ, Guness PK, Herbst A, Smart CE, Witsch M, Maffeis C, SWEET Study Group
Gerhardsson P, Schwandt A, Witsch M, Kordonouri O, Svensson J, Forsander G, Battelino T, Veeze H, Danne T on Behalf of the SWEET Study Group
Choe J, Won SH, Choe Y, Park SH, Lee YJ, Lee J, Lee YA, Lim HH, Yoo JH, Lee SY, Kim EY, Shin CH, Kim JH
De Silva JD, Lepore G, Battelino T, Arrieta A, Castañeda J, Grossman B, Shin J, Cohen O
Hohendorff J, Gumprecht J, Mysliwiec M, Zozulinska-Ziolkiewicz D, Malecki MT
Munshi M, Slyne C, Davis D, Michals A, Sifre K, Dewar R, Atakov-Castillo A, Toschi E
Toschi E, Atakov-Castillo A, Slyne C, Munshi M
Karter AJ, Parker MM, Moffet HH, Gilliam LK, Dlott R
Carlson AL, Daniel TD, DeSantis A, Jabbour S, Karslioglu French E, Kruger D, Miller E, Ozer K, Elliott T
Miller E, Kerr MSD, Roberts GJ, Nabutovsky Y, Wright E
Auzanneau M, Rosenbauer J, Maier W, von Sengbusch S, Hamann J, Kapellen T, Freckmann G, Schmidt S, Lilienthal E, Holl RW
Fallon C, Jones E, Oliver N, Reddy M, Avari P
Johnson SR, Holmes-Walker DJ, Chee M, Earnest A, Jones TW on behalf of the CGM Advisory Committee and Working Party and the ADDN Study Group
Schmitt J, Fogle K, Scott ML, Iyer P
Odugbesan O, Addala A, Nelson G, Hopkins R, Cossen K, Schmitt J, Indyk J, Jones NY, Agarwal S, Rompicherla S, Ebekozien O
Tremblay ES, Ruiz J, Dykeman B, Maldonado M, Garvey K
Mencher SR, Weinzimer SA, Nally LM, Van Name M, Nunez-Smith M, Sadler LS
REAL-WORLD OUTCOMES ACROSS POPULATIONS AND REGION
Association of the Use of Diabetes Technology with HbA1c and BMI-SDS in an International Cohort of Children and Adolescents with Type 1 Diabetes: The SWEET Project Experience
Marigliano M1, Eckert AJ2,3, Guness PK4, Herbst A5, Smart CE6, Witsch M7, Maffeis C1, SWEET Study Group
1Regional Center for Pediatric Diabetes, University of Verona, University City Hospital, Verona, Italy; 2Institute of Epidemiology and MedicalBiometry, ZIBMT, University of Ulm, Ulm, Germany; 3German Center for Diabetes Research (DZD), Munich-Neuherberg, Germany; 4Nongovernment Organization, Quatres Bornes, Mauritius; 5Department of Pediatric and Adolescent Medicine, Hospital Leverkusen GmbH, Leverkusen, Germany; 6Department of Paediatric Endocrinology, John Hunter Children's Hospital, New Lambton Heights, Australia; 7Pediatric Diabetology, Centre Hospitalier de Luxembourg, Luxembourg, Luxembourg
Objective
Two devices used in diabetes management are insulin pumps (for continuous subcutaneous insulin infusion [CSII]) and glucose sensors (for continuous glucose monitoring [CGM]). The aim of this study was to determine whether the use of at least one of these devices was associated with metabolic control (HbA1c) and body adiposity (body mass index standard deviation score [BMI-SDS]) in children and adolescents with type 1 diabetes who have never used these devices before.
Subjects and Methods
A total of 4643 T1D patients (aged 2–18 years, T1D ≥1 year, without celiac disease, no CSII or CGM before 2016) participating in the SWEET prospective multicenter diabetes registry were enrolled. Data were collected at two points (2016 and 2019). Metabolic control was assessed by glycated hemoglobin (HbA1c) and body adiposity by BMI-SDS (according to WHO standards). Patients were categorized by treatment modality (multiple daily injections [MDIs] or CSII) and the use or not of CGM. Linear regression models, adjusted for age, gender, duration of diabetes and region, were applied to assess differences in HbA1c and BMI-SDS among patient groups.
Results
The proportion of patients using MDI with CGM and CSII with CGM significantly increased from 2016 to 2019 (7.2% to 25.7%, 7.8% to 27.8% respectively; P<.001). Linear regression models showed a significantly lower HbA1c in groups that switched from MDIs to CSII with or without CGM (P<.001), but a higher BMI-SDS for those who switched from MDIs without CGM to CSII with CGM (P<.05) and those who switched from MDI without CGM to CSII without CGM (P<.01).
Conclusions
Switching from MDI to CSII was significantly associated with improvement in glycemic control but increased BMI-SDS over time. Diabetes technology may improve glucose control in youths with T1D, although further strategies to prevent excess fat accumulation are needed.
The SWEET Project 10-Year Benchmarking in 19 Countries Worldwide Is Associated with Improved HbA1c and Increased Use of Diabetes Technology in Youth with Type 1 Diabetes
Gerhardsson P1, Schwandt A2,3, Witsch M4, Kordonouri O5, Svensson J6,7, Forsander G8, Battelino T9, Veeze H10, Danne T5,11 on Behalf of the SWEET Study Group
1Department of Epidemiology, Institute of Applied Economics and Health Research, Copenhagen, Denmark; 2Institute of Epidemiology and Medical Biometry, ZIBMT, Ulm University, Ulm, Germany; 3German Center for Diabetes Research (DZD), Munich-Neuherberg, Germany; 4Department of Pediatrics DCCP, Center Hospitalier de Luxembourg, Luxembourg; 5Children's Hospital AUF DER BULT, Hannover Medical School, Hannover, Germany; 6Department of Pediatrics and Adolescents, Copenhagen University Hospital, Herlev and Gentofte, Herlev, Denmark; 7Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark; 8Department of Pediatrics, Institute for Clinical Sciences, Sahlgrenska Academy, University of Gothenburg and Region Västra Götaland, Sahlgrenska University Hospital, Queen Silvia Children's Hospital, Gothenburg, Sweden; 9UMC-University Children's Hospital and Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia; 10Diabeter, Diabetes Center for Pediatric and Adolescent Diabetes Care and Research, Rotterdam, Netherlands; 11SWEET e.V., Hannoversche Kinderheilanstalt, Hannover, Germany
This manuscript is also discussed in DIA-2023-2508, page S-118.
Objective
The international SWEET registry (trial registration: NCT04427189) was initiated in 2008 to improve outcomes in pediatric diabetes. A 10-year follow-up allowed researchers to study time trends of key quality indicators in youth with type 1 diabetes (T1D) who were seen in at least one of 22 centers from Europe, Australia, Canada, and India.
Methods
Aggregated data per person with T1D <25 years of age were compared between the 2008–2010 and 2016–2018 time periods. Hierarchic linear and logistic regression models were applied. Models were adjusted for gender, age, and diabetes duration groups.
Results
The first and second time periods included 4930 and 13,654 persons, respectively (51% versus 52% male, median age 11.3 [IQR, 7.9–14.5] vs 13.3 [IQR, 9.7–16.4] years); median T1D duration was 2.9 (IQR, 0.8–6.4) years for the first time period and 4.2 (IQR, 1.4–7.7) years for the second period. The adjusted hemoglobin A1C (HbA1c) improved significantly (P<.0001) from 68 (95% CI, 66–70) mmol/mol to 63 (60–65) mmol/mol or 8.4% (95% CI, 8.2%–8.6%) to 7.9% (95% CI, 7.6%–8.1%). Across all age groups, HbA1c was significantly lower in pump and sensor users. Severe hypoglycemia declined from 3.8% (95% CI, 2.9%–5.0%) to 2.4% (95% CI, 1.9%–3.1%) (P<0.0001), whereas diabetic ketoacidosis events increased significantly with injection therapy only. Body mass index-standard deviation score also showed significant improvements, from 0.55 (95% CI, 0.46–0.64) in the first period to 0.42 (95% CI, 0.33–0.51) in the second period (P<.0001). Over time, the increase in pump use from 34% to 44% preceded the increase in HbA1c target achievement (<53 mmol/mol) from 21% to 34%.
Conclusions
Twice yearly benchmarking within the SWEET registry was associated with significantly improved HbA1c and increased pump and sensor use over a 10-year period in young persons with T1D.
Temporal Trends for Diabetes Management and Glycemic Control Between 2010 and 2019 in Korean Children and Adolescents with Type 1 Diabetes
Choe J1, Won SH2, Choe Y3, Park SH3, Lee YJ3,4, Lee J5, Lee YA3,4, Lim HH6, Yoo JH7, Lee SY4,8, Kim EY9, Shin CH3,4, Kim JH1,4
1Department of Pediatrics, Seoul National University Bundang Hospital, Seongnam, South Korea; 2Medical Research Collaborating Center, Seoul National University Bundang Hospital, Seongnam, South Korea; 3Department of Pediatrics, Seoul National University Children's Hospital, Seoul, South Korea; 4Department of Pediatrics, Seoul National University College of Medicine, Seoul, South Korea; 5Department of Pediatrics, Inje University Ilsan Paik Hospital, Goyang, South Korea; 6Department of Pediatrics, Chungnam National University Hospital, Daejeon, South Korea; 7Department of Pediatrics, Dong-A University Hospital, Busan, South Korea; 8Department of Pediatrics, SMG-SNU Boramae Medical Center, Seoul, South Korea; 9Department of Pediatrics, Chosun University Hospital, Gwangju, South Korea
Purpose
There is increasing use of modern devices in the management of patients with type 1 diabetes (T1D). We investigated temporal trends for diabetes management and outcomes in Korean pediatric T1D patients over 10 years.
Methods
We retrospectively collected the data from 752 participants (311 [41.4%] boys) diagnosed with T1D and aged ≤18 years, with ≥1 year of follow-up between 2010 and 2019 in any of the seven study hospitals in Korea.
Results
Over the 10-year study period, use of continuous glucose monitoring (CGM) increased from 1.4% to 39.3%. From 2010 to 2019, use of multiple daily insulin injections (MDIs) increased from 63.9% to 77.0%, and continuous subcutaneous insulin infusion (CSII) increased from 2.1% to 14.0%. However, during the same time period, conventional insulin therapy (CIT) decreased from 33.9% to 9.0%. Mean glycated hemoglobin (HbA1c) decreased from 8.56% to 8.01% (P<.001) and was lower in younger patients, boys, and CGM users (P<.001). MDI and CSII users had lower mean HbA1c levels than CIT users (P=0.003). Regarding the acute complications of T1D, CGM use was associated with lower incidences of diabetic ketoacidosis (P=.015), and CSII users were likely to experience less severe hypoglycemia (P=.008).
Conclusions
The use of CSII and CGM increased ∼7- and 30-fold, respectively, over the 10-year study period. The glycemic control of pediatric T1D patients in Korea improved from 2010 to 2019, probably because of increased use of T1D technologies.
Comments
Several articles were published this year related to outcomes in large and small regional populations (6 –11). The three highlighted here present benchmarking data, with the SWEET projects using registry data from 19 countries around the world (10,11) and a new report from Korea (9), a country often not represented in larger registry studies, prompting notable mention. All three articles report on an increase in CGM use and insulin pump use over time, as well as glycemic improvement in the respective cohorts over the same time period. In the SWEET registry, CGM use increased most dramatically over a decade, from negligible percentages to over 50% of the main cohort (11) and from 1.4% to 39.3% in the Korean cohort (9). Insulin pump use also increased across all studies, a finding that was associated with better glycemic outcomes and reduction in HbA1c.
Marigliano and colleagues (10) uniquely report an increase in body adiposity in SWEET registry youth from 2016 to 2019 who switched from multiple daily injections to insulin pump with or without CGM use (P<.05 and P<.01, respectively). This effect was not seen in any other technology group. This finding is important because excess body weight is directly associated with cardiovascular risk for people with diabetes (12,13). While associations between technology use and glycemic outcomes are important contributions to our body of knowledge, this report reminds us that clinical context is important, and uncovering the effect of technology on other diabetes risk factors presents a more holistic picture of long-term health. This should be a topic of further study and a model for how to use expanded clinical characteristics in benchmarking analyses.
Real-World Performance of the Minimed 780G System: First Report of Outcomes from 4120 Users
Da Silva JD1, Lepore G2, Battelino T3, Arrieta A4, Castañeda J4, Grossman B5, Shin J5, Cohen O1
1Medtronic International Trading Sàrl, Tolochenaz, Switzerland; 2Unit of Endocrine Diseases and Diabetology, ASST Papa Giovanni XXIII, Bergamo, Italy; 3University Children's Hospital, University Medical Centre Ljubljana, and Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia; 4Medtronic Bakken Research Center, Maastricht, The Netherlands; 5Medtronic, Northridge, CA
Background
The MiniMed 780G system includes an advanced hybrid closed-loop (AHCL) algorithm that provides both automated basal and correction bolus insulin delivery. The preliminary performance of the system in real-world settings was evaluated.
Methods
Data uploaded from August 2020 to March 2021 by individuals living in Belgium, Finland, Italy, the Netherlands, Qatar, South Africa, Sweden, Switzerland, and the United Kingdom were aggregated and retrospectively analyzed to determine the mean glucose management indicator (GMI); percentage of time spent within (TIR; 70–180 mg/dL), below (TBR; <70 mg/dL), and above (TAR; >180 mg/dL) target glycemic ranges; system use; and insulin consumption in users having ≥10 days of sensor glucose (SG) data after initiating AHCL. The impact of initiating AHCL was evaluated in a subgroup of users also having ≥10 days of SG data before AHCL initiation.
Results
Users (N=4120) were observed for a mean of 54±32 days. During this time, they spent a mean of 94.1%±11.4% of the time in AHCL and achieved a mean GMI of 6.8%±0.3%, TIR of 76.2%±9.1%, TBR of 2.5%±2.1%, and TAR of 21.3%±9.4%, after initiating AHCL. There were 77.3% and 79.0% of users who achieved a TIR >70% and a GMI of <7.0%, respectively. Users for whom comparison with pre-AHCL was possible (N=812) reduced their GMI by 0.4%±0.4% (P=.005) and increased their TIR by 12.1%±10.5% (P<.0001), post-AHCL initiation. More users achieved the glycemic treatment goals of GMI <7.0% (37.6% vs 75.2%, P<.0001) and TIR >70% (34.6% vs 74.9%, P<.0001) than they did before AHCL initiation.
Conclusion
Most MiniMed 780G system users achieved TIR >70% and GMI <7% while minimizing hypoglycemia in a real-world condition.
Intermittently Scanned Continuous Glucose Monitoring Data of Polish Patients from Real-Life Conditions: More Scanning and Better Glycemic Control Compared to Worldwide Data
Hohendorff J1, Gumprecht J2, Mysliwiec M3, Zozulinska-Ziolkiewicz D4, Malecki MT1
1Department of Metabolic Diseases, Jagiellonian University Medical College, Krakow, Poland; 2Department of Internal Medicine, Diabetology and Nephrology, Medical University of Silesia, Katowice, Poland; 3Department of Pediatrics, Diabetology and Endocrinology, Medical University of Gdansk, Poland; 4Department of Internal Medicine and Diabetology, Poznan University of Medical Science, Poland
Background
Randomized trials and observational studies have shown that the use of FreeStyle Libre intermittently scanned continuous glucose monitoring system (isCGMS) is associated with improved glycemic indices and quality of life.
Materials and Methods
In this retrospective, real-world data analysis, we described country-specific glucometrics among isCGMS users from Poland and compared them with international data. The analyzed time period for the Polish data ranged between August 2016 and August 2020, and the analyzed time period for the international data ranged from September 2014 to August 2020.
Results
Data from the Polish population were collected from 10,679 readers and 92,627 sensors with 113 million automatically recorded glucose readings. The worldwide database included information from 981,876 readers and 11,179,229 sensors with 13.1 billion glucose readings. On average, isCGMS users of from Poland performed substantially more scans/day (21.2±14.2 vs 13.2±10.7), achieved lower eHbA1c (7.0%±1.2% vs 7.5%±1.5%), and spent more time-in-range (TIR) (64.2%±17.3% vs 58.1%±20.3%) and less time-above-range (TAR) (29.7%±18.0% vs 36.6%±21.3%) (P<.0001 for all comparisons). Moreover, they were more likely to achieve TIR >70% (36.3% vs 28.8%), but spent more time-below-range (TBR) (4.7% vs 3.6%). Our results confirmed that analyzed glucometrics improve as the scan rate frequency increases. However, at a similar scanning frequency to the comparative group, users from Poland achieved lower eHbA1c, higher TIR, lower TAR, but higher TBR.
Conclusions
We report more scanning and better glycemic control in isCGMS users in Poland than in users worldwide. The cause of this observation remains unknown. Our data also show that in real-life practice, a large number of patients may be willing to perform scanning more frequently than is usually assumed.
Comments
These two articles analyze de-identified, large datasets obtained from Minimed 780G hybrid closed-loop downloads (14) and Libre isCGM. The strength of these types of analyses is that device settings and user behavior (15) can be analyzed. The high TIR results with 780G reported by DeSilva and colleagues were impressive considering the majority of users did not tune the device with optimal settings: only 50% of the users had the algorithm target set at the recommended 100 mg/dL (16). In the paper by Hohendorff et al., higher TIR correlated to more frequent isCGM scans, both in the Polish cohort and in the international cohort.
It is important, however, to pay attention to the inherent limitations to de-identified dataset analyses. While these results illuminate ideal behaviors and tuning for diabetes devices, they do not give us information on who might expect similar results, especially in light of socioeconomic and racial/ethnic disparities in access and differences in clinical populations of device users by country or region. The user age, duration of diabetes, and gender are all unknown. Hohendorff and colleagues make the point that the Polish cohort seemed to achieve better glycemia than the international cohort with the same number of isCGM scans; however, without understanding the clinical characteristics of the cohort, this cannot be meaningfully interpreted. It is thus important that we extrapolate the correct conclusions from de-identified datasets; it is not that these glycemic outcomes are achievable by everyone, or that one country is better at diabetes care than another, but rather that behavior and device tuning are what matter. We can all learn best practices with diabetes devices to achieve optimal outcomes, and likely these lessons can be translated across all populations.
EXPANDING DIABETES TECHNOLOGY TO NOVEL POPULATIONS
Use of Technology in Older Adults with Type 1 Diabetes: Clinical Characteristics and Glycemic Metrics
Munshi M1,2,3, Slyne C1, Davis D1, Michals A1, Sifre K1, Dewar R1, Atakov-Castillo A1, Toschi E1,2,3
1Joslin Diabetes Center, Clinical Research, Boston, MA; 2Beth Israel Deaconess Medical Center, Department of Medicine, Boston, MA; 3Harvard Medical School, Boston, MA
Background
The use of diabetes-related technology, both for insulin administration and glucose monitoring, has shown benefits in older adults with type 1 diabetes (T1D). However, the characteristics of older adults with T1D and their use of technology in real-world situations are not well documented.
Methods
Older adults (aged ≥65 years) with T1D who were using insulin pump or multiple daily injections (MDIs) for insulin administration and continuous glucose monitoring (CGM) or glucometer (blood glucose monitoring [BGM]) for glucose monitoring were evaluated. Participants wore CGM devices for 2 weeks, completed surveys, and underwent laboratory evaluation.
Results
We evaluated 165 older adults with T1D (mean age 70±10 years, diabetes duration 40±17 years, and A1c 7.4%±0.9% [57±10 mmol/mol]). For insulin administration, 63 (38%) were using MDIs, while 102 (62%) were using pump. Compared to MDI users, pump users were less likely to have cognitive dysfunction (49% vs 65%, P=.04) and had lower scores on the hypoglycemia fear survey (P=.03). For glucose monitoring, 95 (58%) used CGM, while 70 (42%) used BGM. Compared to BGM users, CGM users were more likely to report impaired awareness of hypoglycemia (IAH) (P=.01), and had lower A1c (P=.02). Participants who used any technology (pump or CGM) had lower A1c (pump, P=0.04; CGM, P=.006), less hypoglycemia ≤54 mg/dL (pump, P=.0006; CGM, P<.0001) and <70 mg/dL (pump, P=.0002; CGM, P=.0001), and lower glycemic variability (pump, P=0.0001; CGM, P<.0001), while reporting more IAH (pump, P=.04; CGM, P=.006) and diabetes distress (pump, P=0.02; CGM, P<.004).
Conclusion
Older adults with T1D who use newer diabetes-related technology had better glycemic control, lower hypoglycemia risk, and fewer glycemic excursions. However, they were more likely to report IAH and diabetes-related distress.
Closed-Loop Insulin Therapy in Older Adults with Type 1 Diabetes: Real-World Data
Toschi E1,2, Atakov-Castillo A1, Slyne C1, Munshi M1,2,3
1Joslin Diabetes Center, Boston, MA; 2Harvard Medical School, Boston, MA; 3Beth Israel Deaconess Medical Center Boston, MA
Objective
To assess the impact of initiating closed-loop control (CLC) on glycemic metrics in older adults with type 1 diabetes (T1D) in the real world.
Methods
Retrospective analysis of electronic health records from a single tertiary diabetes center of older adults prescribed CLC between January 2020 and December 2020.
Results
A total of 48 patients (mean age 70±4 years, T1D duration 42±14 years) were prescribed CLC, and 39/48 started on the CLC. Among the CLC starters, 97.5% and 95% were prior pump and continuous glucose monitoring (CGM) users, respectively. CGM metrics showed an increase in time-in-range (mean±SD 62±14% to 76±9%; P<.001) and a reduction in both time spent at <70 mg/dL (median[IQR] 2[1-3]% to 1[1-2]%; P=0.03), and time spent >180 mg/dL (mean±SD 30±11% to 20±9%; P<0.001) at 3 months.
Conclusion
In these real-world data, most of the older patients with T1D initiating CLC were prior pump or CGM users. Initiation of CLC improved glycemic control and reduced time spent in hypoglycemia with respect to the levels during prior therapy.
Comments
With the rising incidence of type 1 diabetes across multiple age groups and improvements in medical care that result in better long-term outcomes, there are more older adults living with type 1 diabetes than ever before. Data from the Type 1 Diabetes Exchange and the Diabetes Patienten Verlaufsdokumentation (DPV) registries indicate the majority of older adults with type 1 diabetes are not meeting recommended A1c targets, even with use of newer diabetes technologies (17 –19). Increased susceptibility to hypoglycemia, complicated medical diagnoses, visual and hearing impairments, and cognitive changes all present potentially significant obstacles to the attainment and maintenance of optimal diabetes management. These two articles (21,22) report the use of insulin pumps, glucose sensors, and automated insulin delivery systems in this important but understudied population.
In a cohort of 165 adults over 60 years old (mean age 70 years) with type 1 diabetes, Munshi and colleagues (20) found that use of diabetes technology, most notably CGM (more than pumps), was associated with lower average blood glucose levels, higher time in range, less hypoglycemia, and lower glucose variability. Impaired awareness of hypoglycemia and increased diabetes distress were also seen with the diabetes technology. In this real-world study, it is likely that diabetes-related distress and impaired hypoglycemia awareness were the drivers for choosing to utilize diabetes technologies, not the result, although more robust clinical trials are needed to truly understand this relationship. In a related study by the same group, Toschi et al. (21) found that initiating hybrid closed-loop systems in older adults (mean age 70 years) was associated with significant improvements in glycemic metrics, including time in range, hypoglycemia, hyperglycemia, and glucose variability. It important to note that over 80% of the study population initiated on the Control-IQ system were already using CGM and CSII prior to starting hybrid closed-loop systems; furthermore, impairments in manual dexterity or visual acuity and discomfort with trying a new system were cited as reasons for not starting or continuing closed-loop control in 13% of the population. Diabetes technologies, including automated insulin delivery systems, are clearly beneficial in older adults: designing systems that are easy to operate for those with visual and manual impairments and successfully introducing and maintaining their use in older adults with little experience or comfort with technology are challenges we must meet.
Continuous Glucose Monitor Use Prevents Glycemic Deterioration in Insulin-Treated Patients with Type 2 Diabetes
Karter AJ1, Parker MM1, Moffet HH1, Gilliam LK2, Dlott R3
1Kaiser Permanente—Division of Research, Oakland, CA; 2Kaiser Permanente Northern California, South San Francisco, CA; 3The Permanente Medical Group, Martinez, CA
Abstract
Background
Continuous glucose monitoring (CGM) is indicated in poorly controlled insulin-treated patients with type 2 diabetes (T2D) to improve glycemic control and reduce the risk of hypoglycemia, but the benefits of CGM for lower risk patients have not been well studied.
Methods
Among 17,422 insulin-treated patients with T2D, hemoglobin A1c (HbA1c) <8%, and no recent severe hypoglycemia (based on emergency room visits or hospitalizations), CGM initiation occurred in 149 patients (17,273 noninitiators served as reference). Changes in HbA1c and severe hypoglycemia rates for the 12 months before and 12 months after CGM initiation were calculated.
Results
CGM initiation was associated with decreased HbA1c (−0.06%), whereas noninitiation was associated with increased HbA1c (+0.32%); a weighted adjusted difference-in-difference model of change in HbA1c yielded a net benefit of −0.30% (95% CI, −0.50% to −0.10%; P=.004). No significant differences were observed for severe hypoglycemia.
Conclusions
CGM may be useful in preventing glycemic deterioration in well-controlled patients with insulin-treated T2D.
Flash Glucose Monitoring in Type 2 Diabetes Managed with Basal Insulin in the USA: A Retrospective Real-World Chart Review Study and Meta-analysis
Carlson AL1, Daniel TD2, DeSantis A3, Jabbour S4, Karslioglu French E5, Kruger D6, Miller E7, Ozer K8, Elliott T9
1International Diabetes Centre IDC, HealthPartners Institute, Minneapolis, MN; 2Diabetes Wellness Clinic, Pearland, TX; 3Department of Family Medicine, The Charlotte-Mecklenburg Hospital Authority d/b/a Atrium Health, Charlotte, NC; 4Diabetes Research Center, Thomas Jefferson University, Philadelphia, PA; 5Department of Endocrinology, University of Pittsburgh Medical Center, Pittsburgh, PA; 6Division of Endocrinology, Henry Ford Health System, Detroit, MI; 7Diabetes and Obesity Care LLC, Bend, OR; 8Texas Diabetes & Endocrinology, Round Rock, TX; 9BC Diabetes, Vancouver, British Columbia, Canada
Introduction
Evidence supporting the use of continuous glucose monitoring in type 2 diabetes treated with basal insulin is limited. The purpose of this real-world study is to evaluate the impact of flash glucose monitoring on the glycated hemoglobin (HbA1c) levels of adults with type 2 diabetes who are managed with insulin.
Research Design and Methods
Medical records were reviewed for adult individuals with type 2 diabetes using basal insulin for ≥1 year with or without additional antihyperglycemic medication; participants were required to have HbA1c between 8.0% and 12.0% for ≥90 days before they began using FreeStyle Libre flash glucose monitoring, and they were also required to have an HbA1c measurement recorded between 90 and 194 days after device use began. Exclusion criteria included utilization of bolus insulin. Meta-analysis data are from the current study (United States) and a similar Canadian cohort.
Results
Medical record analysis (n=100) from eight US study sites showed significant HbA1c decrease of (mean ± SD) 1.4% ±1.3%, from 9.4%±1.0% at baseline to 8.0%±1.2% after device use began (P<.0001). Meta-analysis of medical records from US and Canadian sites (n=191) showed HbA1c significantly decreased by (mean ± SE) 1.1%±0.14%, from baseline 9.2%±1.0% to 8.1%±1.1% (P<.0001), with moderate to high heterogeneity between sites (Q=43.9, I2=74.9, P<.0001) explained by differences in baseline HbA1c between sites. The HbA1c improvement in both groups was observed by age group, body mass index, duration of insulin use, and sex at birth.
Conclusions
In a real-world retrospective US study and a meta-analysis of a larger US and Canadian cohort, HbA1c significantly reduced in basal insulin–treated type 2 diabetes without bolus insulin initiation and following the commencement of flash glucose monitoring.
Flash CGM Associated with Event Reduction in Nonintensive Diabetes Therapy
Miller E1, Kerr MSD2, Roberts GJ2, Nabutovsky Y3, Wright E4
1Diabetes and Obesity Care, Bend, OR; 2Abbott, Sylmar, CA; 3Abbott, Santa Clara, CA; 4Charlotte AHEC, Charlotte, NC
Objectives
We evaluated the effects of acquiring a flash continuous glucose monitoring (CGM) system in the population with type 2 diabetes (T2D) treated with basal or noninsulin therapy.
Study Design
This was a retrospective database analysis of the IBM MarketScan Commercial Claims and Medicare Supplemental databases that assessed rates of acute diabetes-related events (ADEs) and all-cause inpatient hospitalizations (ACHs) in a large population with T2D treated with basal insulin therapy or noninsulin medications. ADE and ACH rates 6 months prior to and 6 months after CGM acquisition were compared.
Methods
Inclusion criteria for analysis were as follows: diagnosis of T2D; age 18 years or older; treatment with long-acting, neutral protamine Hagedorn or premixed insulin or noninsulin therapy; naive to CGM; and acquisition of their flash CGM system between October 2017 and March 2019. Patients served as their own controls. Event rates were compared by using weighted Cox regression with Andersen-Gill extension for repeat events.
Results
A cohort of 10,282 adults with T2D (mean [SD] age, 53.1 [9.6] years; 51.9% male) who met inclusion criteria were assessed. ADE rates decreased from 0.076 to 0.052 events per patient-year (HR, 0.68; 95% CI, 0.58–0.80; P<.001). ACH rates decreased from 0.177 to 0.151 events per patient-year (HR, 0.85; 95% CI, 0.77–0.94; P=.002).
Conclusions
After patients acquired flash CGM systems, significant reductions were seen in their outpatient and inpatient ADEs and ACHs. These data strongly demonstrate that patients with T2D treated with basal insulin therapy or noninsulin therapy who start using flash CGM improve their clinical outcomes and potentially incur lower costs.
Comments
Recent epidemiological studies have estimated the worldwide number of people with type 2 diabetes at over 460 million, corresponding to over 6% of the total global population (22). This number is likely an underestimate, due to undiagnosed cases and the rise in prevalence during the years since these data were first published. Diabetes technologies, such as insulin pumps, are being increasingly utilized in the management of insulin-requiring type 2 diabetes, and recent randomized controlled trials have demonstrated the effectiveness of continuous glucose monitoring for both intensive (23) and nonintensive (24) regimens.
It is clear now from the three real-world studies highlighted in this section (25 –27) that CGM has a place in the routine management of type 2 diabetes, not only in the more obvious role in improving outcomes in patients not meeting recommended targets or in those using intensive regimens who are at greatest risk for hypoglycemia, but as a standard component of treatment for intensive, nonintensive, and even noninsulin-requiring therapies. Increasing the availability and reimbursement for CGM to the broadest communities of people with type 2 diabetes provides many clinical advantages, including improved safety and convenience for patients, earlier identification of the need to intensify individual treatment regimens, and easy and effective incorporation into telehealth. Furthermore, as seen in the study by Miller et al. (27), despite the incremental increase in cost of CGM, large-scale use is likely to have positive economic impact on health-care utilization costs by reducing emergency and inpatient care. More “out-of-the-box” use of diabetes technologies holds the promise of improving outcomes in other underexplored forms of diabetes as well, as evidenced by recent reports of closed-loop devices in patients with cystic fibrosis–related diabetes (28,29).
METHODOLOGICAL APPROACHES TO ADDRESS DIABETES TECHNOLOGY DISPARITIES
In this section, we highlight the role of population level data, policy changes, quality improvement initiatives, and qualitative data as four methodological approaches to address disparities. The studies highlighted in this section were selected as they 1) offer a practical and actionable roadmap to address diabetes and 2) are varied in methodology so they may be applicable to a broad audience in the diabetes community.
Heterogeneity of Access to Diabetes Technology Depending on Area Deprivation and Demographics Between 2016 and 2019 in Germany
Auzanneau M1,2, Rosenbauer J2,3, Maier W2,4, von Sengbusch S5, Hamann J6, Kapellen T7, Freckmann G8, Schmidt S9, Lilienthal E10, Holl RW1,2
1Institute of Epidemiology and Medical Biometry, ZIBMT, University of Ulm, Ulm, Baden-Württemberg, Germany; 2German Center for Diabetes Research (DZD), Neuherberg, Germany; 3Institute for Biometrics and Epidemiology, German Diabetes Center, Leibniz Center for Diabetes Research at Heinrich Heine University, Düsseldorf, North Rhine-Westphalia, Germany; 4Institute of Health Economics and Health Care Management, Helmholtz Zentrum München, German Research Center for Environmental Health (GmbH), Neuherberg, Bayern, Germany; 5Department of Pediatrics and Adolescent Medicine, Division of Pediatric Endocrinology and Diabetes, University of Lübeck, Lübeck, SchleswigHolstein, Germany; 6Department of Pediatrics, St. Marien Hospital Landshut, Landshut, Germany; 7 Department of Women and Child Health, Hospital for Children and Adolescents, University of Leipzig, Leipzig, Sachsen, Germany; 8Institut für Diabetes-Technologie, Forschungs- und Entwicklungsgesellschaft mbH an der Universität Ulm, BadenWürttemberg, Ulm, Germany; 9Department of Paediatrics and Adolescent Medicine, Klinikum Dritter Orden, Munich, Bavaria, Germany; 10Department of Pediatrics, St. Josef-Hospital, Ruhr-University Bochum, Bochum, Nordrhein-Westfalen, Germany
Background
Although technology is increasingly being used to manage type 1 diabetes, ethnic and socioeconomic disparities continue to be reported. This study examined the evolution of pump therapy use and of continuous glucose monitoring in Germany in the context of demographics and area deprivation.
Method
We investigated the use of insulin pump and CGM between 2016 and 2019 in 37,798 patients with type 1 diabetes aged <26 years from the German Prospective Follow-up Registry (Diabetes Patienten Verlaufsdokumentation [DPV]). Associations with federal state, area-deprivation quintile (German Index of Multiple Deprivation 2010 on district level), gender, and migration background were investigated over time by using multiple logistic regression.
Results
Between 2016 and 2019, the regional distribution of insulin pump use did not change substantially, and the association with area deprivation remained nonlinear and statistically nonsignificant. The effect of area deprivation on CGM use decreased continuously and disappeared in 2019 (OR [95% CI] Q1 vs Q5: 1.85 [1.63–2.10] in 2016; 0.97 [0.88–1.08] in 2019). The effect of migration background on the use of either technology decreased over the years but remained significant in 2019. Female participants had constantly higher odds of using an insulin pump than did male participants (OR, 1.25 [1.18–1.31] in 2019), whereas no gender difference was identified for CGM use.
Conclusions
Although disparities decreased in Germany, access to diabetes technology still depended on migration background in 2019, and gender differences in pump use remain. As technological advances are made, further research is needed to understand the reasons for these persistent disparities.
The Impact of Socio-Economic Deprivation on Access to Diabetes Technology in Adults with Type 1 Diabetes
Fallon C1, Jones E1,2, Oliver N1,2, Reddy M1,2, Avari P1,2
1Department of Metabolism, Digestion and Reproduction, Imperial College London, London, UK; 2Department of Diabetes and Endocrinology, Imperial College Healthcare NHS Trust, London, UK
Background
As technology advances, there is growing concern over the inequalities in care provision and diabetes outcomes in places of greater deprivation. The purpose of this study was to evaluate the relationship between socioeconomic status and access to diabetes technology and whether that relationship affects outcomes in patients with type 1 diabetes.
Methods
This is a retrospective observational analysis of adults attending a tertiary center comprising three urban hospitals in the UK. Socioeconomic deprivation was assessed by the English Indices of Deprivation 2019. Data analysis was performed using one-way analysis of variance (ANOVA) and chi-squared tests.
Results
In total, 1631 adults aged 44±15 years and 758 (47%) women were included, with 391 (24%) using continuous subcutaneous insulin infusion, 312 (19%) using real-time continuous glucose monitoring, and 558 (34%) using intermittently scanned continuous glucose monitoring. The highest use of diabetes technology was in the least deprived quintile, and the lowest was in the most deprived quintile (67% vs 45%; P<.001). HbA1c outcomes were available in 400 participants; no association with deprivation was observed (P=.872). Participation in structured education was almost twice as high in the least deprived than in the most deprived quintiles (23% vs 43%; P<0.001). Adults whose ethnicity was White or mixed were more likely to use technology than those whose ethnicity was Black (60% vs 40%; P<.001).
Conclusions
Adults living in the most deprived quintile used less technology. Irrespective of socioeconomic status or ethnicity, glycemia was positively affected in all groups. It is imperative that health disparities are further addressed.
Comments
These two studies underscore the utility of “big data” to identify system-level disparities with a specific focus on area deprivation. While prior publications using data from Germany's Diabetes Patienten Verlaufsdokumentation (DPV) initiative have demonstrated the relationship between area-level deprivation and diabetes management and outcomes (1,30), this study by Auzanneau et al. independently implicated ancestry, the closest proxy to race/ethnicity in Germany, in the utilization of diabetes technology (31). The authors present a roadmap on how to evaluate for race/ethnicity, and the resultant inequitable care provision, through the utilization of proxy variables. These and similar learnings from the DPV initiative underscore the utility of a nationally representative database and real-world data. For countries that do not have a national registry, Fallon et al. present a strategy evaluating data from three urban multicenter hospitals to explore the role of area deprivation on technology use and HbA1c (32). These data are consistent with prior reports of socioeconomic disparities in diabetes outcomes (1,30).
Taken together, these two studies demonstrate the power of population and large-scale data to identify systemic inequities in the utilization of diabetes technologies. Notably, these two studies do not simply focus on the redemonstration of known disparities. Rather, these studies give context to disparities in a new geographic area (such as the United Kingdom) or new sociodemographic factors (such as ancestry as a proxy for race/ethnicity) in which disparities have not been previously evaluated. Population-level real-world data are requisite to identify disparities including the most important areas of action, access and utilization of diabetes technology.
Universal Subsidized Continuous Glucose Monitoring Funding for Young People with Type 1 Diabetes: Uptake and Outcomes over 2 Years, a Population-Based Study
Johnson SR1,2, Holmes-Walker DJ3,4, Chee M5, Earnest A6, Jones TW7,8 on behalf of the CGM Advisory Committee and Working Party and the ADDN Study Group
1Department of Endocrinology and Diabetes, Queensland Children's Hospital, Brisbane, Queensland, Australia; 2Faculty of Medicine, University of Queensland, Herston, Queensland, Australia; 3Department of Diabetes and Endocrinology, Westmead Hospital, Sydney, New South Wales, Australia; 4Faculty of Medicine and Health, University of Sydney, Sydney, New South Wales, Australia; 5JDRF Australia, St Leonard's, New South Wales, Australia; 6Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia; 7Perth Children's Hospital, Nedlands, Western Australia, Australia; 8Telethon Kids Institute, Nedlands, Western Australia, Australia
This manuscript is also discussed in DIA-2023-2502, page S-15 and DIA-2023-2508, page S-118.
Objective
The use of continuous glucose monitoring (CGM) is increasing among patients with type 1 diabetes. However, the models for funding CGM differ throughout the world. In this study, researchers determined the uptake rate and glycemic outcomes that occurred after a national health policy went into effect in Australia to subsidize CGM funding for people under the age of 21 years who have type 1 diabetes.
Research Design and Methods
Longitudinal data from 12 months before the subsidy began until 24 months after were analyzed. Measures and outcomes included age, diabetes duration, HbA1c, episodes of diabetic ketoacidosis and severe hypoglycemia, insulin regimen, CGM uptake, and percentage CGM use. Two data sources were used: the Australasian Diabetes Database Network (ADDN) registry, which is a prospective diabetes database, and the National Diabetes Service Scheme (NDSS) registry, which includes almost all individuals with type 1 diabetes in Australia.
Results
CGM uptake increased from 5% before to 79% 2 years after the subsidy began. After CGM introduction, the odds ratio (OR) of achieving the HbA1c target of <7.0% improved at 12 months (OR, 2.5; P<.001) and was maintained at 24 months (OR, 2.3; P<.001). The OR for an HbA1c >9.0% decreased to 0.34 (P<.001) at 24 months. Of the CGM users, 65% used CGM >75% of the time and had a lower HbA1c at 24 months than those whose usage was <25% (7.8±1.3% vs 8.6±1.8%; P<.001). Diabetic ketoacidosis was also reduced in this group (incidence rate ratio 0.49; 95% CI, 0.33–0.74; P<.001).
Conclusions
After the national subsidy began, CGM use was high and associated with sustained improvement in glycemic control. This information will inform economic analyses and future policy and serve as a model of evaluation diabetes technologies.
Comments
Johnson et al. present the impact of advocacy and policy change to address disparities and offer a roadmap to evaluate the impact of policy change on technology utilization and glycemia (33). In this analysis, a policy change increasing diabetes technology coverage in Australia resulted in a significant increase of CGM use 2 years after improved coverage. As expected, a greater number of individuals used CGM and had an associated improvements in HbA1c and reductions of DKA rates.
Real-world data that evaluate the impact of policy changes on the outcome of technology utilization and glycemia are critical to evaluate the efficacy of the policy change, encourage similar policy changes in other care delivery models, and remind the clinicians of the changing nature of many payer coverage policies. The dynamic and changing nature of payers' coverage policies on diabetes technology is a commonly cited barrier for both individuals with diabetes and their providers (34) and has recently been implicated as one potential cause for insurance-mediated provider implicit bias (35,36). In addition to evaluating policy efficacy, real-world analyses on outcomes after a policy change can identify persistent gaps in equitable care; ameliorating these gaps can form the foundation for next steps.
Improving Equitable Access to Continuous Glucose Monitors for Alabama's Children with Type 1 Diabetes: A Quality Improvement Project
Schmitt J1, Fogle K2, Scott ML1, Iyer P1
1University of Alabama at Birmingham Department of Pediatrics, Birmingham, AL; 2University of Alabama at Birmingham Heersink School of Medicine, Birmingham, AL
Background
Continuous glucose monitors (CGMs) can reduce the burden of self-monitoring glucose values in children and adults with type 1 diabetes (T1D), are associated with improved glycemic control, and are associated with reduced fear of hypoglycemia. Unfortunately, disparities in access to CGM exist, and rates of CGM access in Alabama in 2019 were below national (US) averages. We aimed to increase CGM access and reduce disparities in access by race, insurance status, and high-risk diabetes status.
Methods
Stakeholder input identified barriers to CGM access and defined the existing process. Process changes were implemented and studied for effect. Data were collected from electronic health records to track rates of CGM access in patients aged 2 years and above with T1D for >3 months.
Results
For the eligible population, rates of CGM access increased from a baseline of 50% to 82%. Rates for CGM access in patients with high-risk T1D increased from 34% to 85%. Disparity in CGM access between non-Hispanic White and non-Hispanic Black patients decreased from 18% to 6%. Disparity in CGM access between privately insured and Medicaid-insured patients decreased from 38% to 12%.
Conclusions
Targeted quality improvement projects using stakeholder input can increase access to diabetes technology while reducing disparities. As technology advances, concerted efforts are needed to ensure equitable access to evolving therapies for all patients with T1D.
Implicit Racial-Ethnic and Insurance-Mediated Bias to Recommending Diabetes Technology: Insights from T1D Exchange Multicenter Pediatric and Adult Diabetes Provider Cohort
Odugbesan O1, Addala A2, Nelson G3, Hopkins R4, Cossen K5, Schmitt J6, Indyk J7, Jones NY8, Agarwal S9, Rompicherla S1, Ebekozien O1
1T1D Exchange, QI & Population Health Department, Boston, MA; 2Stanford University, Division of Pediatric Endocrinology & Diabetes, Lucile Packard Children's Hospital, Stanford, CA; 3Le Bonheur Children's Hospital, Pediatric Endocrinology, Memphis, TN; 4SUNY Upstate Medical Center, Division of Endocrinology and Metabolism, Syracuse, NY; 5Children's Healthcare of Atlanta, Division of Pediatric Endocrinology, Atlanta, GA; 6The University of Alabama Pediatric Endocrinology and Diabetes at Birmingham Hospital, Birmingham, AL; 7Nationwide Children Hospital, Division of Endocrinology, Columbus, OH; 8Cincinnati Children's Hospital, Division of Endocrinology, Cincinnati, OH; 9Yeshiva University Albert Einstein College of Medicine, Division of Endocrinology, Bronx, NY
Background
Despite documented benefits of diabetes technology in managing type 1 diabetes, inequities persist in the use of these devices. Provider bias may be a driver of inequities, but the evidence is limited. Therefore, we aimed to examine the role of race/ethnicity and insurance-mediated implicit provider bias in recommending diabetes technology.
Method
We recruited 109 adult and pediatric diabetes providers across seven US endocrinology centers to complete an implicit bias assessment composed of a clinical vignette and ranking exercise. Providers were randomized to receive clinical vignettes with differing insurance and patient names as proxy for racial-ethnic identity. Bias was identified if providers 1) recommended more technology for patients with an English name (racial/ethnic bias) or private insurance (insurance bias) or 2) race/ethnicity or insurance was ranked high (racial/ethnic bias and insurance bias, respectively) in recommending diabetes technology. Provider characteristics were analyzed using descriptive statistics and multivariate logistic regression.
Results
Insurance-mediated implicit bias was common in our cohort (n=66, 61%). Providers who were identified to have insurance-mediated bias had more years in practice (9.3±9 years vs 5.3±5.3 years, P=.006). Racial-ethnic-mediated implicit bias was also observed in our study (n=37, 34%). Compared with those without racial/ethnic bias, providers with racial/ethnic bias were more likely to state that they could recognize their own implicit bias (89% vs 61%, P=.001).
Conclusion
Implicit provider bias in recommending diabetes technology was observed to be based on insurance and race/ethnicity in our pediatric and adult diabetes provider cohort. These data raise the need to address providers' implicit bias in diabetes care.
Comments
These two publications from the United States demonstrate the utility of quality improvement methodology to address disparities in diabetes technology use. Schmitt et al. present a single-site quality improvement initiative in which stakeholders, including diabetes educators, clinic nurses, administrative assistants, physicians, nurse practitioners, patients, and family members each identified factors that impacted CGM access at the site (37). The authors then outline the framework to address these stakeholder-identified barriers to institutional policy change resulting in a decrease but not complete elimination of racial/ethnic disparities in technology use. The purpose in highlighting this manuscript is to offer a template for a single-institution intervention to advance equitable access to diabetes technology.
For institutions that are part of a multisite registry, Odugbesan et al. present a strategy to tackle a specific driver of diabetes disparities, namely implicit provider bias (35). The T1D Exchange Quality Improvement initiative has developed a framework to address multiple systemic barriers to equitable diabetes care (38). This publication focused on the preassessment of implicit provider bias by insurance status and race/ethnicity. These data redemonstrate that a provider's implicit bias against public insurance increases with the numbers of years in practice and that awareness of explicit racial/ethnic bias alone is not sufficient to protect providers against implicit racial/ethnic bias.
Taken together, these two studies offer a roadmap for institutional changes via a quality improvement framework to address disparities in diabetes care and management.
Hispanic Caregivers' Experience of Pediatric Type 1 Diabetes: A Qualitative Study
Tremblay ES1, Ruiz J2, Dykeman B3, Maldonado M4, Garvey K1
1Division of Endocrinology, Department of Pediatrics, Boston Children's Hospital, Boston, MA; 2Department of Pediatrics, Boston Children's Hospital, Boston, MA; 3Institutional Centers for Clinical and Translational Research, Boston Children's Hospital, Boston, MA; 4Children's Hospital Primary Care Center, Social Work, Boston Children's Hospital, Boston, MA
Objective
It is widely recognized that type 1 diabetes (T1D) outcomes are worse among Hispanic children; however, little is published about the perspectives of these patients and their caregivers. Our intent was to characterize the lived experience of Hispanic caregivers of children with T1D, focusing on the role of language and culture and their perspectives on current medical care and alternative care models. We studied Hispanic caregivers of patients aged 2 to 17 years with T1D of greater than 6 months' duration.
Research Design and Methods
We completed semistructured interviews and focus groups of a purposive sample of 20 members of our population of interest. We developed a codebook and completed multidisciplinary consensus coding, then conducted iterative thematic analysis using qualitative software and discussion to generate themes.
Results
We gathered data from 20 Hispanic caregivers of T1D patients (11.37±3.00 years old, 4.80±2.84 years since diagnosis). Of the caregivers, 85% were female, 80% preferred Spanish, and 15% were college educated. Our analysis yielded four themes across the participants: 1) culturally based nutrition challenges, 2) social isolation and lack of support for T1D care, 3) hesitancy to fully embrace diabetes technology, and 4) deferential views of care experience and providers. The commonality among these themes was support for Hispanic group-based models of care tailored to address these concerns.
Conclusions
The unique concerns among Hispanic caregivers of children with T1D suggest the importance of culturally tailored interventions to improve care. With successful implementation, such interventions could diminish widening disparities in health-care outcomes.
Technology Utilization in Black Adolescents with Type 1 Diabetes: Exploring the Decision-Making Process
Mencher SR1, Weinzimer SA1,2, Nally LM1, Van Name M1, Nunez-Smith M3, Sadler LS2,4
1Department of Pediatrics, Division of Endocrinology, Yale University School of Medicine, New Haven, CT; 2School of Nursing, Yale University, Orange, CT; 3Equity Research and Innovation Center, Yale University School of Medicine, New Haven, CT; 4Yale Child Study Center, New Haven, CT
Background
Significant disparities in diabetes device (DD) use exist for Black adolescents with type 1 diabetes (T1D) and merit further exploration. We sought to describe how Black adolescents with T1D and their parents make decisions about using DDs and understand personal, familial, and cultural beliefs that may influence use.
Materials and Methods
Nineteen Black adolescents with T1D and 17 parents participated in individual qualitative semistructured interviews. Adolescents were purposively sampled for a range of socioeconomic and clinical demographics. Interview data were recorded, transcribed, coded for thematic analysis, analyzed separately for parents and adolescents, and then compared across groups. Data collection continued until thematic saturation was achieved.
Results
Adolescents and parents reported similar themes related to 1) the intersectionality of multiple identities: T1D experience of Black adolescents; 2) decision to use DDs: complexities of T1D management and easing the burden; and 3) reasons for differential uptake of DDs in Black adolescents. Adolescents reported lacking peers with T1D “who look like me,” leading to stigmatization exacerbated by device visibility and alarms. Cultural and familial traditions as well as individual factors were described as both facilitators and barriers in DD use. Lack of familiarity with T1D, limited exposure to DDs, and mistrust of the medical community, both historically and currently, were brought up as reasons for inequities in DD use.
Conclusions
Understanding the decision-making process surrounding DDs in one sample of Black adolescents and their parents is critical to guide further research to improve equity in DD use and glycemic outcomes.
Comments
These two qualitative studies focused on the experience of Black and Latinx/Hispanic families in the United States with type 1 diabetes with respect to diabetes devices (39,30). The utility of qualitative studies in addressing disparities is clearly delineated by comparing these two studies. Black and Latinx/Hispanic youth have the lowest diabetes technology uptake and poorest glycemic outcomes in the United States (41) and these disparities are additionally compounded by socioeconomic status and social determinates of health (1,34).
Tremblay et al. (40) present qualitative data from families who are Latinx/Hispanic, whereas Mencher et al. (39) present data from families who are Black. When evaluating these two studies, both groups discussed the importance of cultural traditions and burden of social isolation. However, the groups differed in other areas. Black families additionally focused on mistrust and intersectionality, whereas Latinx/Hispanic families focused on nutritional support and device hesitancy. While these two studies have underlying methodological differences in study design, these data demonstrate that different racial/ethnic groups have different priorities. These findings underscore importance evaluating the lived experiences of underrepresented groups in formulating and prioritizing barriers to equitable diabetes care.
Conclusion
One common message that connects these disparate themes is that in order to successfully apply diabetes technologies to broader populations of people living with diabetes, we need to better understand their particular characteristics and circumstances. Design factors that have been effective and appealing in younger people may in fact present obstacles to older individuals; the lived experiences of marginalized individuals may influence their approach to diabetes technologies in ways that are not relevant to other groups. It is apparent that there is not only an ethical justification for expanding access of advanced technologies to previously unselected populations, but also measurable societal and financial justifications. It is hoped that in coming years, the idea of real-world studies will indeed include the whole real world of those living with diabetes.
Footnotes
Author Disclosure Statement
Laurel H. Messer reports receiving consulting and speaking honoraria from Tandem Diabetes, Capillary Biomedical, Eli Lilly, and Dexcom. Her research team receives grants from Medtronic, Insulet, Beta Bionics, Dexcom, Eli Lilly, and Tandem. Ananta Addala has no disclosures or conflicts of interests. Stuart A. Weinzimer reports receiving speaking honoraria from Abbott Diabetes and consulting honoraria from Zealand Pharma. His research team has received research support (to his institution) from Abbott Diabetes and Zealand Pharma.