Abstract
Background: Continuous glucose monitoring (CGM) for patients with type 1 and type 2 diabetes is associated with improved clinical, behavioral, and psychosocial patient health outcomes and is part of the American Diabetes Association’s Standards of Medical Care. CGM prescription often takes place in endocrinology practices, yet 50% of adults with type 1 diabetes and 90% of all people with type 2 diabetes receive their diabetes care in primary care settings. This study examined primary care clinicians’ perceptions of barriers and resources needed to support CGM use in primary care.
Methods: This qualitative study used semistructured interviews with primary care clinicians to understand barriers to CGM and resources needed to prescribe. Participants were recruited through practice-based research networks. Rapid qualitative analysis was used to summarize themes from interview findings.
Results: We conducted interviews with 55 primary care clinicians across 21 states. Participants described CGM benefits for patients with varying levels of diabetes self-management and engagement. Major barriers to prescribing included lack of insurance coverage for CGM costs to patients, and time constraints. Participants identified resources needed to foster CGM prescribing, for example, clinician education, support staff, and EHR compatibility.
Conclusion: Primary care clinicians face several challenges to prescribing CGM, but they are interested in learning more to help them offer it to their patients. This study reinforces the ongoing need for improved clinician education on CGM technology and continued expansion of insurance coverage for people with both type 1 and type 2 diabetes.
- Continuous Glucose Monitoring
- Primary Care Physicians
- Primary Health Care
- Qualitative Research
- Type 1 Diabetes Mellitus
- Type 2 Diabetes Mellitus
- Wearable Technology
Introduction
Continuous glucose monitoring (CGM) technology has potential to improve health for the estimated 13% of the US population with diabetes,1 especially because most patients with diabetes do not achieve recommended glycemic targets.2,3 CGM is associated with reduced glycohemoglobin (A1c), hypoglycemia, and weight; increased treatment satisfaction; and improved psychosocial outcomes (eg, reduced diabetes distress).4⇓⇓–7 In primary care settings, CGM is associated with greater A1c reductions compared with usual care.8,9 CGM reduces or replaces capillary glucose checks and is approved for therapeutic decision making.10 The American Diabetes Association (ADA) Standards of Medical Care in Diabetes recommend CGM for patients with type 1 or type 2 diabetes whose treatment includes insulin. ADA Standards further note that even periodic use of CGM can be helpful for diabetes management among those for whom full-time CGM is not desired or available.11 Despite the benefits and recommendations, most CGM prescription occurs in endocrinologists’ offices, to which access is limited or nonexistent in much of the US.12 CGM prescribing is limited in primary care settings, though the majority of adults with diabetes in the US see a primary care clinician for managing their diabetes.13,14 Although CGM prescription in primary care is accelerating, this is happening in a somewhat different patient profile than in endocrinology.15 The issues facing primary care specifically need to be understood, rather than extrapolating the endocrinology experience for application to primary care.
Barriers to CGM use in primary care include insufficient clinician knowledge about CGM technology, inadequate insurance coverage, lack of diabetes resources such as Diabetes Care and Education Specialists (DCES)16 and other clinical staff with knowledge about diabetes and related devices,17,18 and burdensome prior authorization processes.19 Evidence from outside of primary care describes other challenges. Cost can be prohibitive for patients, depending on insurance.20,21 As with the introduction of many new treatments, learning about and implementing CGM requires time and effort that many physicians cannot spare, given competing priorities.20,22 It is also challenging to incorporate CGM into practice due to lack of a widespread, standard approach to interpreting data22 despite the existence of standardized formats for reporting data.11 Many primary care and other physicians, DCESs, and other members of the health care team need more education on CGM.20 Specific detail about barriers and training needs is necessary to inform strategies for primary care practices to make CGM use part of diabetes care. In this study, we examined primary care clinicians’ perceptions of barriers and resources needed to support CGM use in primary care.
Methods
This qualitative study consisted of semistructured interviews with primary care clinicians. The Colorado Multiple Institutional Review Board approved this study as exempt from review.
Participants and Recruitment
Participants were recruited from survey respondents in a larger study.23 Survey recruitment was conducted in collaboration with 2 practice-based research networks (PBRNs), the American Academy of Family Physicians National Research Network (AAFP NRN) and the Meta-network Learning and Research Center (Meta-LARC), a consortium of regional PBRNs. We used maximum variation sampling, a form of purposive sampling24 to select interviewees who could provide input on the different barriers and resource needs that different groups of clinicians might have. Sampling characteristics included varying levels of experience prescribing CGM, intention to prescribe CGM in the future, and distance from the nearest endocrinologist. We recruited potential interview participants via e-mail with a maximum of 3 contact attempts at 1-week intervals. A $50 gift card was offered for participation.
Data Collection
Interviews were conducted using a semistructured interview guide (Table 1), designed to explore additional details about and explanations for survey findings from our larger study on CGM use in primary care. The guide included open-ended questions to collect participants’ input on barriers to and facilitators for prescribing CGM, desired resources to help them feel more comfortable prescribing CGM at all or more often, overall perceptions of CGM and potential impacts of CGM. A PhD-level AAFP NRN researcher (MF) and a master’s-level AAFP NRN researcher (BM) with qualitative training and experience conducted interviews from the AAFP NRN sample. Two master’s-level research assistants (MW, TH) with qualitative training and experience conducted interviews from the Meta-LARC sample. Interviews were conducted from August through December 2020 via videoconference or telephone, were audio recorded, and lasted 15 to 35 minutes. Interviewers took detailed notes during and immediately after interviews. Recruitment continued until no new concepts emerged. See Table 1.
Guide for Interviews Conducted with US Primary Care Physicians, NPs and PAs (n = 55), 2020
Data Analysis
Rapid qualitative analysis was employed to facilitate prompt translation of interview findings into educational resources to support CGM prescription in primary care (resources not described here). This rapid approach involves summarizing each interview using a highly structured form with domains aligning with the study aim. Summary forms are then compiled into a matrix with rows for each interview and columns for each interview domain, which is then reviewed to identify themes within and across domains.25,26 Domains for this analysis included participant characteristics, a description of a typical and a CGM-focused diabetes visit, barriers, facilitators, resources needed, the “ideal candidate” for CGM, and perceptions of CGM. We then used inductive analysis to identify the most salient themes.
Two qualitative analysts (MW, TH) completed interview summary forms using interview notes supplemented by audio recordings. Details from transcripts were summarized and categorized in the summary form according to their corresponding study domain. Initially, each analyst independently completed summary forms for the same 2 interviews, then analysts met to compare the categorization of interview content and level of detail. An additional interview was then selected for both analysts to review. Each analyst independently completed a summary form for this additional interview to assess consistency in detail and categorization of topics across both analysts. On the analysts’ reaching consistency in categorization of topics from interviews into summary forms, the remaining interviews were divided evenly between the analysts. Each took responsibility for reviewing their assigned transcripts and completing the corresponding summary form. One analyst then compiled all completed summary forms into a matrix. Both independently reviewed the matrix to identify themes within domains and patterns across cases, then met 5 times to discuss and compare emergent themes; they periodically presented themes to the larger study team for interpretation and discussion. Finally, the analysis team summarized the themes in a report shared with the research team.
Results
We conducted interviews with 55 clinicians across 21 states. Most participants were attending, faculty, or community physicians (78%); residents (7%) and advanced practice clinicians (APCs) including nurse practitioners and physician assistants (15%) also participated. Most specialized in Family Medicine (90%). Participants represented a wide variety of practice settings, including private practice (24%), Federally Qualified Health Centers (FQHCs) (22%), and hospital-owned practices (20%). Just more than one-third (36%) had experience prescribing CGM. See Table 2 for complete participant characteristics.
Interview Participant Characteristics (n = 55 US Primary Care Physicians, NPs and PAs), 2020
Participants had positive perceptions of CGM benefits. Several noted that CGM can help explain why glucose values vary and thereby inform changes to the treatment plan. Many participants were hopeful for broader adoption of CGM.
Participants identified clinician and patient factors that hindered CGM prescription, such as the interaction between insurance coverage, costs, and time. They also discussed the need for resources and education to support prescribing and the need for diabetes resources, staff, and access to subspecialists in primary care settings.
Insurance-Related Roadblocks, Costs to Patients, and Time Constraints Make CGM Prescribing Difficult
Insurance-related factors and costs were the most frequently cited barriers to prescribing CGM. Participants reported that many insurance providers limited CGM coverage to a specific subset of patients or did not cover it at all. These challenges were noted by the majority of participants from all practice settings. Both private and public insurance coverage of CGM varied, though a few participants noted that coverage was improving among some payers.
Even when encountering patients whose circumstances appeared to prescribers that they should clearly qualify for CGM authorization under their insurance, the process and the time required to obtain that authorization presented challenges. As 1 participant stated, “I’ve never been able to successfully get it going for someone. There have been a couple that I've tried for and was not able to get it. Mostly due to loss to follow up and a pretty high burden of documentation that was required.” They later added, “It is sort of a waste of time and energy to keep trying something when you know that it is just going to get declined.” (Physician, community health center, participant ID #755) Participants characterized documentation required by some insurance providers to determine CGM coverage as excessive. Participants indicated that certain requirements, such as performing multiple capillary glucose checks for an extended period or being insulin-treated, were prohibitive for some patients. One physician stated:
“I believe Medicaid in [my state] covers it, but the patient has to have documented that they're checking their blood sugars four times a day for at least a month…. I have patients who simply can't accomplish that and so I can't really get it covered for them. It's unfortunate because that's part of the whole point is that if you can get past this barrier, then they don't have to be checking it like that manually every time. But people have complicated lives. Sometimes it's just really hard to get patients to actually do that.” (Physician, hospital-owned clinic, participant ID #771)
Time to complete prior authorizations was a barrier for many participants. One physician described the issue as follows: “You spend more energy getting it through the insurance companies, which takes an enormous amount of my time.” (Physician, academic medical center, participant ID #115) Some who had few patients for whom they thought that CGM was appropriate did not feel that the time investment for learning about CGM and obtaining insurance approval was worthwhile. However, participants with dedicated time to complete prior authorizations and related paperwork described their experience as less burdensome.
Participants indicated that out-of-pocket costs for CGM could be prohibitive, particularly among low-income patient populations. They attributed this to limited or complete lack of insurance coverage for CGM or to high deductibles even when coverage was in place. As 1 participant noted, “I am working in sort of a poor area so people cannot afford these things themselves… nobody here really has the extra income.” (NP, private practice, participant ID #713) Another participant similarly described how financial challenges prevent CGM use for their patients: “Every single visit, it is not just a matter of you prescribed insulin. They cannot afford it this month or, they are homeless this month. It is all that stuff that gets in the way of doing what the textbook says ought to happen.” (Physician, hospital-owned clinic, participant ID #76) Some participants were apprehensive to prescribe CGM if they perceived it as too costly for the patient.
Resources and Education Are Needed to Address Gaps in Clinician Knowledge and Experience
Many participants described gaps in their own knowledge and experience as barriers to prescribing CGM. Specific education needs included how to operate and troubleshoot the devices, differences across brands, and how to prescribe CGM. One physician summarized this:
“I think a lot of it is just knowing the process. Some sort of checklist or education around the steps to the process… I still feel like it's a new enough thing that probably some physician education like patient selection, who to recommend it for, when do you think about it? What insurance will pay for it? That kind of thing would be helpful.” (Physician, FQHC, participant ID #737)
Participants also wanted to better understand CGM accuracy compared with traditional capillary measurements and how to interpret and act on the data. As one physician stated, “I think we also as providers would need a little bit more training. I can probably come up with a plan, but I think a lot of other providers know how best to react to an A1C level, than to these periodic ups and downs. It just makes for a trickier management plan.” (Physician, academic medical center, participant ID #112) Participants wanted to ensure they were recommending CGM to the appropriate patients and using it properly to inform treatment. Less often, participants raised concerns about whether CGM made a difference in improving patient health outcomes and wanted further evidence of its benefits.
Participants who had CGM knowledge and experience gained it through clinical rotations, colleagues who prescribed and managed CGM, family members or friends who used CGM, patients who shared their own or loved ones’ experiences with CGM, previous professional experience (eg, as a DCES or pharmacist), conferences, and journal articles. Some participants expressed that some patients initiated discussion about CGM after learning about it from advertisements, family members, and friends.
Participants described resources that could help address knowledge gaps and increase their likelihood of prescribing CGM (Table 3 and Table 4) and suggestions to format educational resources most effectively. Participants described trainings conducted as short videos, self-paced online sessions, and other asynchronous opportunities with continuing medical education credits as most useful. Some requested live options. For hands-on learning, several participants wanted a CGM device in their office for patient demonstrations and noted that having a device on-hand would increase their comfort with prescribing.
Resources to Support Patient Education on CGM Suggested by US Primary Care Physicians, NPs and PAs (n = 55), 2020
Educational Topics of Interest for Practice Teams Identified in Interviews with Primary Care Physicians, NPs and PAs (n = 55), 2020
Primary Care Practice Environments Need Staff and Access to Diabetes Resources and Subspecialists to Increase Availability of CGM to Patients
Participants felt that there was not enough time during appointments to thoroughly educate patients about CGM. They wanted other professionals (eg, DCES and care managers) on staff to alleviate these time-related barriers. However, some noted that such staff members did not always have sufficient training or knowledge about CGM to optimally educate and support patients. One participant described the importance of other care team members to support patients’ use of CGM: “Diabetes nurse educators, they are key. I cannot imagine doing what I do for the management of our diabetic patients without them, because they have the time to invest, they have the time to answer all the questions. We just do not have that time in today's medical world.” (Physician, academic medical center, participant ID #189) Another physician elaborated on time as a challenge:
“We have hundreds and hundreds of patients. So being able to spend the time with the patient to explain how it works and how to do it, getting them in for close enough follow up, me having enough time in my schedule to figure out how to do it and plan things, because I don't get any administrative time in my job… That is challenging.” (Physician, FQHC, participant ID #741)
Partially due to limited knowledge, experience, and time, participants cited the need for additional staff and access to subspecialists to start or feel more confident offering CGM. They frequently cited DCESs, ideally on-site, as resources needed for ordering CGM, on-boarding patients with CGM devices, and providing patient education. Many participants highlighted the benefits of having clinical pharmacists with CGM experience and administrative staff to support insurance authorization, navigation, and billing. Some felt more comfortable prescribing if they or their patient could consult with an endocrinologist if needed. Participants wanted a specific contact person from CGM manufacturers and insurance companies to help clarify questions and concerns about the devices and insurance approval processes, respectively.
Electronic Health Record (EHR) Systems Lack Compatibility with CGM Data
Many participants wanted CGM devices to integrate with their clinics’ EHR system. They hoped integration would enable seamless transfer of patients’ data from CGM into their records and patient portal, and potentially enable clinicians to more easily prescribe CGM. One participant describes the current challenges: “It is awkward to download and get that real time feedback. Something we've been seeing, especially remotely, how do I get these reports into the chart in such a way that it is easily shared?” (Physician, hospital-owned clinic, participant ID #803) Another stated, “We use Epic and I would love to get that integrated. I think that would be really helpful to see. I know that there's a way that I can log on to a provider account for the [CGM], but it is yet another system, another log on, another thing for me to learn.” (Physician, hospital-owned clinic, participant ID #496) Participants who were able to upload and receive data from patients’ CGM devices noted that it was particularly helpful for virtual visits, even when not integrated into the EHR.
Additional Considerations
Seven of 55 participants (13%) noted in survey responses that they were not at all likely to prescribe CGM in the future (Table 2). In interviews, multiple reasons for this were mentioned, such as regularly referring patients to specialists for CGM, cost of CGM, technological requirements, and questioning the appropriateness of CGM for type 2 diabetes. However, there was no clear consensus for any single reason, with each mentioned by only 1 or 2 interviewees.
Discussion
CGM has potential to improve glucose management for patients with diabetes, yet access is often limited to those under endocrinology care.12 Our study sought to understand barriers facing primary care clinicians in offering CGM to patients and identify resources to address these barriers. We found that primary care clinicians have favorable attitudes toward CGM but would like additional resources and education to mitigate barriers. Collaborating with 2 large practice-based research networks allowed us to study different types of practices across the country to assess perceptions of CGM, barriers to and facilitators for CGM use, and resources needed to support CGM use in primary care settings.
Consistent with previous studies,17,20,21 insurance coverage, requirements and costs related to CGM were among the greatest barriers to clinicians’ willingness to offer CGM. The use of discount programs and point-of-care sample CGM devices to introduce patients to the technology is a promising strategy to help alleviate some of the costs and insurance challenges.27 Insurance coverage varies widely, particularly across public payers. Some states, such as Colorado and Oregon (from which our sample drew heavily), had limited or no coverage for CGM under the state Medicaid plans at the time of the study. Large-scale policy measures to broaden eligibility requirements and simplify prior authorization processes could increase the use of CGM in primary care and thereby increase achievement of glycemic targets. Since this study was conducted, Medicare eliminated the need for 4 times daily fingerstick glucoses in 2022. In April 2023, Medicare expanded CGM coverage to all people with diabetes taking insulin (including basal only) and those not on insulin at all but with problematic hypoglycemia. Future studies are needed to determine if these changes result in greater uptake of CGM in primary care settings.28
Primary care physicians and APCs most commonly cited consultation on insurance issues and a training workshop on CGM as resources that would support them in prescribing CGM.23 This study expands on those quantitative survey findings to describe and provide examples of how insurance requirements hinder primary care physicians’ ability to offer CGM, and the topics and format of training needed. Lack of knowledge about CGM was a challenge for primary care clinicians in 2013,17 and our study confirms that it remains so today. Study findings provide specific suggestions for the topics and format of resources and education that primary care physicians and APCs want and further build on calls for more education about CGM for the entire practice team.20 Clinicians vary in their level of readiness to promote CGM,29 and some have negative attitudes toward CGM.22,30 Clinicians’ exposure to and knowledge of CGM is associated with favorability toward future prescribing.23 Targeting specific clinician or practice types for training and whose patient populations have the greatest need could be beneficial, particularly if the practices lack support staff and other resources. Providing opportunities for hands-on learning and short, self-paced training modules to work within their busy schedules could be helpful. One such training that was informed by this and a previous study23 is the AAFP’s Transformation in Practice Series module Continuous Glucose Monitoring (CGM): Enhancing Diabetes Care, Workflows, Education, and Payment.31 This module was free and available to AAFP members and nonmembers alike, as is its update from 2024.
Insurance coverage and costs to patients were described as barriers to CGM use and clinicians’ interest in prescribing. However, it was unclear if these issues were reported by patients or assumed by clinicians. When appropriate, based on standards of care and/or clinical discretion, it is important that clinicians introduce therapy options for patients and allow patients to determine whether they might like to pursue them rather than having the clinician choose for them. This is as true of CGM as it is of any other therapy option. When clinicians have the tools and knowledge to explain CGM and the associated costs, patients can make informed decisions in determining the emotional, logistic, and financial costs and benefits of using CGM.
More information is needed to understand if and how factors such as geographical differences (eg, rural or urban setting, distance from nearest endocrinologist), organization type, or insurance status influence CGM uptake, alone or in interaction with other barriers. Ongoing research building on this work incorporates broader perspectives of health economists and other implementation scientists to better understand CGM use in primary care.
Limitations
This study gathered input from clinicians regarding their own perceptions as well as patient perceptions and did not include input directly from patients, so certain themes may not accurately represent patient perceptions. In addition, interview and survey recruitment language included a description of the study and language about CGM in primary care. Clinicians who agreed to participate may have been influenced by pre-existing interest in, experience with, or favorability toward CGM. Lastly, this article presents findings from interviews that took place during the COVID-19 pandemic. Due to the economic climate, this may have resulted in themes related to barriers due to loss of patients’ incomes and insurance coverage. Furthermore, the timing of Meta-LARC network interview recruitment was particularly affected by COVID-19, with multiple participants citing COVID-related priorities in their community and clinics making them less available for interviews.
Conclusion
Mitigating or eliminating barriers such as insurance authorization, patient out-of-pocket costs, time, staffing needs, and subspecialist support could lead to more CGM use in primary care. These challenges are not unique to the adoption of CGM in primary care, and future research should test whether strategies to address similar barriers in management of other chronic diseases or the uptake of other health care technologies could be effective to increase CGM access. Despite these challenges, primary care clinicians are interested in learning more to help them offer CGM to their patients. This study reinforces the ongoing need for improved and accessible clinician education on CGM technology, continued expansion of insurance coverage for CGM, and a streamlined process for obtaining insurance authorization – barriers that have persisted for more than a decade despite the progress of CGM over that same period into recommended standard care in many instances. Importantly, this article describes preferences of primary care physicians and APCs that should inform future training and support resources about CGM aimed at a primary care audience.
Acknowledgments
This study would not have been possible without the individual clinicians from the participating practice-based research networks.
Notes
This article was externally peer reviewed.
Conflict of interest: Drs. T. Oser and S. Oser have received Advisory Board Consulting fees (through the University of Colorado) from Dexcom, Ascensia, and Blue Circle Health and research grants (through the University of Colorado) from NINR, NIDDK, the Helmsley Charitable Trust, Abbott Diabetes, Dexcom, and Insulet. They report no conflicting or competing interests in connection with these collaborations. They do not own stocks in any device or pharmaceutical company. All other authors report no conflicts of interest.
Funding: This study was supported by a research grant from the Leona M. and Harry B. Helmsley Charitable Trust.
To see this article online, please go to: http://jabfm.org/content/37/4/671.full.
- Received for publication February 2, 2024.
- Revision received March 1, 2024.
- Accepted for publication March 11, 2024.
