Making a Good Thing Even Better: Expanding Access and Applicability of Automated Insulin Delivery Systems to Benefit All Youth With Type 1 Diabetes

Advanced diabetes technologies, such as insulin pumps, continuous glucose monitoring (CGM), and automated insulin delivery (AID), have revolutionized type 1 diabetes care resulting in significant improvement in glycemia and quality of life (1). For pediatric type 1 diabetes, this positive impact extends to the parents, guardians, caregivers, and family units as a whole (2). AID systems broadly facilitate the integration of real-time CGM data with automated algorithm-based insulin delivery, which results an in incremental improvement in diabetes-related short- and long-term outcomes. AID use is associated with decreased burden of diabetes management and a global improvement in diabetes outcomes. Despite the efficacy of diabetes technologies, disparities in their use have been increasingly recognized (3–6). In its most basic definition, a disparity is the inequitable access to an efficacious intervention. In pediatric diabetes, access and use of efficacious advanced diabetes technologies are key disparities (5–7).

In this issue of Diabetes Care, Zeng et al. (8) present a meta-analysis of over 1,300 participants pooled from 25 outpatient randomized control trials (RCTs) comparing AID with conventional therapy. A key strength, the authors chose to group all non-AID management (multiple daily insulin injections, continuous subcutaneous insulin infusion, and sensor-augmented pumps with/without low glucose suspension). By grouping insulin delivery methods with varying efficacy, the authors test the robustness of AID systems explicitly and make their findings applicable to the “real-world.” AID was associated with an improvement in glucose time in range (TIR) of 2.75 h and reduced time above range by ∼3 h without an increase in time below range. This meta-analysis also identified that dual-hormone AID systems outperformed single-hormone AID systems, conferring an additional 3 h TIR. Given that this meta-analysis is of outpatient RCTs, the authors stratified the data by supervision; supervision (camp or hotel setting with study staff) confers an additional improvement of 77 min TIR and 138 min time above range compared with unsupervised (home) settings. Benefits in time below range were only appreciated in supervised settings.

In this meta-analysis, the authors included outpatient RCTs and focused on CGM outcomes. In support of their findings, real-world studies representing the various U.S. Food and Drug Administration–approved AID systems show improvements in glycemia, psychosocial states, and family dynamics (9–13). The results of these studies expand the applicability of the evidence base and support the recommendation of AID systems to a broad group. In fact, the efficacy of AID systems is greatest for those with the highest HbA_1c or least amount of TIR (9). In addition to the decrease in burden with algorithm-driven insulin delivery, the significant improvement in glycemia (particularly overnight) is associated with improved sleep and psychosocial states for youth and their parents/guardians (12). Importantly, AID systems tend to be more forgiving of missed or late boluses, a frequent driver of hyperglycemia for youth, than non-AID systems (11).

For many of the studies included in this meta-analysis sociodemographic data were not collected, or minoritized youth were not included into these pivotal clinical trials. Of the 25 studies selected, a majority of studies were missing data on race, ethnicity, or ancestry (n = 17); parental education (n = 20); family income (n = 20); and insurance status (n = 21). For those studies where sociodemographic variables were reported, the majority of participants were non-Hispanic White (range 65–90%; >80% in 6 of 9 studies), college or graduate degree holders (69–94%), and privately insured (82–91%).

The lack of population- and disease-representative clinical research has adverse health outcomes for the understudied groups (14,15). For example, when historical CGM clinical trials underrecruited minoritized individuals, smaller and less-resourced pilot studies evaluated the efficacy of CGM in minoritized populations (16). It is worth considering whether there is a need to prove efficacy of AID or advanced diabetes technologies in minoritized populations. The idea that efficacious technology must be proven in less resourced groups has resulted in delayed adoption of equitable policies. Technology works effectively in minoritized populations (5,7,17)—it is more important that minoritized individuals are supported to minimize the impact of structural inequities (3). In this meta-analysis, the role of study supervision was critical to the optimization of AID. Study supervision is a proxy for the scaffolding and support of a research team to successfully transition youth to sustained AID use. Through appropriate education and scaffolding from clinical and research teams, CGM use can be equitable, efficacious, and sustained (3).

It is not surprising that AID outperform non-AID systems; the efficacy of AID is well established (1,2). However, equally well established are the multifactorial drivers of disparities in access to and use of diabetes technology (3–7). Three independent studies attributed an ∼0.3% HbA_1c difference between minoritized and nonminoritized groups, driven by advanced diabetes technology use (4,18,19). For many minoritized youth living with type 1 diabetes, the choice of whether to start technology use is made not by them but, rather, for them due to structural inequities. Structural drivers are best organized by the social ecological model where health outcomes for youth living with type 1 diabetes and their family are nestled within their interpersonal, institutional, community, and societal relationships (20) (Fig. 1). With AID, youth must initiate two separate and complementary devices, namely, a CGM and an insulin pump. Thus, a dual burden of device access and use exists for successfully incorporation of AID.

Given that the drivers of structural inequities are multifactorial, solutions to expand and stabilize access to AID are multifactorial as well. Mapping evidence-based solutions addressing disparities in CGM use onto AID use can offer a playbook for solutions to prevent and mitigate disparities in AID (Fig. 1). Offering CGM to all youth with new-onset diabetes resulted in similar improvements in HbA_1c for minoritized and nonminoritized populations (21). Similarly, approaching all eligible research participants for clinical research resulted in excellent recruitment and retention (22,23). Clinic protocols can ease clinician burden to prescribe technology through quality improvement initiatives (24). Insurers can expand coverage and streamline procedures to minimize interruptions in technology use (3,19). Device manufacturers can diversify their research, development, testing, and advertisement practices and develop inclusive and accessible (i.e., languages available and/or customization of readability for those with visual impairment) user interfaces. Allowing seamless communication of data between individuals living with type 1 diabetes, their families, and their clinical team reinstates ownership of the data back with the person generating it. Funding agencies can ensure adequate financial support to promote equitable research practices and fund diverse research teams. Finally, allocating a greater portion of a nation’s gross domestic product to social supports is associated with improved health outcomes (25). Including minoritized and historically excluded communities into all of the aforementioned levels with intentionality and authenticity can begin to repair historical injustices.

In contrasting current AID data with the historical perspective of CGM, a clear and familiar narrative emerges. CGM, then, and AID, now, have been and are emerging technologies that are efficacious, have broad applicability, and improve the multifaceted lives of youth living with type 1 diabetes and their families—revolutionizing diabetes care. Zeng et al. add to the compelling evidence base on the utility of AID. Clinical care centers, researchers, insurers, and device manufacturers have clear marching orders if we are to not repeat the inequities that led to worsening disparities in the last decade. The solutions to advance equity are multilayered, offering an opportunity for all within the diabetes world to engage and make a concerted effort to increase both CGM and insulin pump use, and thereby AID. Learning from lessons of the past, we have the opportunity to make AID uptake and use a model for health equity.