Diffusion and Disparities: Rural Uptake of Continuous Glucose Monitors

Type 1 diabetes is a chronic disease that imposes a significant burden on pediatric patients and their families. Continuous glucose monitors (CGMs) are a new technology that can reduce the burden of painful finger sticks, reduce the risk of hypoglycemia, and improve glycemic control. Despite the advent of CGM and other diabetes technologies, nationwide reports show that the majority of pediatric patients with type 1 diabetes do not achieve their hemoglobin A_1c goal (1) and that patients from marginalized groups have worse outcomes (2–4).

In this issue of Diabetes Care, Tilden et al. (5) present a retrospective cohort study exploring CGM use in pediatric patients with type 1 diabetes between January 2018 and December 2021. Through the electronic health record system of the Vanderbilt Pediatric Diabetes Program, the team compared the odds of CGM interpretation during a clinic visit for children residing in urban areas, small rural towns, and isolated rural towns. The CGM interpretation billing code in the electronic health record served as a proxy for CGM usage. Living in a more rural area was associated with significantly lower odds of CGM use even after adjustment for sex, race/ethnicity, hemoglobin A_1c, visit year, and insurance type. Over the 4 years of the study, the proportion of visits including CGM interpretation increased for all patient types but the gap persisted between the patients living in the most rural and the urban areas. The authors also found that patients with public insurance, non-White race/ethnicity, and residence in areas with a higher Neighborhood Deprivation Index had significantly lower odds of CGM use in comparison with their counterparts.

This study is a contemporary study of diffusion of innovation similar to those dating back to the 1960s. Diffusion of innovation often follows an S-shaped curve, with a slow early adoption period, then a middle period with rapid adoption, and then a late period with flattening out and approach of full diffusion (6,7). In studies of health care innovations investigators have frequently found this diffusion to be uneven across patient socioeconomic status, racial/ethnic background, and geographic location of practices for a wide range of different types of innovations, such as the spread of digital mammography (8), statins (9), implantable cardioverter defibrillators (10), and laparoscopic surgery (11). In these studies, early adopters of new innovations tend to be of higher socioeconomic status, White, and located in urban areas. Uneven diffusion of new innovations can contribute to and exacerbate health care disparities. Figure 1 shows a schema adapted from Rogers (7) of the S-shaped curve and our hypothesis of how the temporal trend of urban versus rural adoption may occur.

In the diabetes technology literature, the findings from this study are consistent with prior research that has shown disparities in the prescription and use of CGMs by race, ethnicity, socioeconomic status, and type of insurance. This study is unique, however, in that the disparities in adoption of CGM all occurred within a single health care system. Presumably, provider education regarding new technologies, provider billing practices, and access to other diabetes resources would have been more equal across practice sites. The finding of rural/urban differences in CGM adoption and use within this single system suggests that patient and family factors may play a more prominent role in these disparities. Some literature has suggested that patients living in rural areas may have more negative attitudes toward technology and thus slower uptake of technologic interventions (12). However, disparities also exist in health metrics for rural communities for “low-tech” interventions such as breastfeeding rates and well-child visits (13,14), suggesting that CGM disparities are not solely related to technology.

Provider implicit biases and the provision of health care that is discordant with patient values can lead to racial and ethnic health care disparities (15). Despite the study setting in a single health care system, these factors could affect provider prescribing habits between rural and urban patients. There also may be differences in clinic sites and the availability of staff to help with paperwork or patient education.

Cost can drive inequities as well. While many commercial insurance plans cover CGMs, state Medicaid policies have varying levels of coverage. As CGM sensors typically need to be replaced every 7–14 days, any copays can become financially burdensome. In a study where CGMs were fully subsidized with zero copay, there were no significant differences in CGM prescription or adherence by patient race/ethnicity (16). CGMs can be covered as durable medical equipment or as a pharmacy benefit, but Medicaid-contracted pharmacies may not always be easily accessible in rural areas (17).

This study, alongside the abundant literature on disparities in diabetes care and technology uptake, reminds us that providing truly equitable care requires extra consideration for those who have been historically marginalized. At the provider level, it is important to provide care that is cognizant of the values and unique experiences of marginalized patients and to recognize biases that lead to differences in prescription of CGMs. Clinics and health care systems should also provide adequate psychological support, patient education, and financial assistance to make CGM adoption easier for underserved patients. For a health care system like the one in this study, differences in CGM adoption by clinic site might provide a direction for future implementation efforts that acknowledges the different needs of practice sites. And at a policy level, more comprehensive coverage by insurance plans can improve continued use of CGMs.

In future research investigators should explore more fully the reasons for disparities in CGM use among rural populations to create targeted interventions to improve outcomes. Participation of community health workers, who are individuals with a close tie to a particular community, can be an effective way to reach marginalized communities and has been shown to improve health outcomes among rural populations (18). Group education sessions and psychosocial counseling to reduce diabetes distress among marginalized groups can also potentially improve CGM use (19).