Continuous Glucose Monitoring: Facilitating the Initiation and Use for Patients with Type2 Diabetes Mellitus in a Rural Primary Clinic

Type 2 diabetes mellitus continues to be one of the most significant health problems worldwide. It can lead to several health complications, such as amputations, vision loss, neuropathy, kidney disease, and cardiovascular disease. These complications can be delayed or even avoided by maintaining glycemic stability through effective self- and medical management. Continuous glucose monitors (CGM) help both patients and providers manage glycemic trends and maintain stability, which can decrease the risk for complications and improve patient outcomes. By increasing rural health care providers' awareness of the processes required for successful CGM initiation, more patients will experience long-term benefits and a reduction in devastating complications. A quality improvement project was implemented to increase rural providers' awareness of the processes required for successful CGM initiation for patients with type 2 diabetes mellitus and insulin dependence. A workflow algorithm was developed and used to accomplish this task in one small, rural Utah clinic. An educational presentation of the workflow algorithm was first introduced to clinic staff, which consisted of office personnel, a physician, and a diabetes educator, before implementation. Pre-project data of CGM initiation were collected and compared to post-project CGM initiation data. Usability, feasibility, and satisfaction were analyzed in a post-survey provided to both clinic staff and qualified patients who participated. Post-intervention, the project analysis found there was an 89.6% increase (17.4% vs 33%) in CGM initiation with the use of the workflow algorithm. Clinic staff reported 100% (n=5) in the ease of use and satisfaction with the algorithm, and 80% (n=4) reported plans for continued use. An unexpected benefit from the use of the algorithm was found when the clinic staff learned that the initiation of CGM and the interpretation of CGM data could be billed, which is something they had not done previously. Patients reported overall satisfaction (n=22, 88%) with the personal care provided, continuity of care, and their improved self-management of diabetes. Three patients reported "dissatisfied" or "very dissatisfied" (n=3, 12%) in improved self-management and overall satisfaction. The CGM workflow algorithm appears to be a valuable resource to help facilitate CGM initiation in a small, rural Utah clinic. Replication of this quality improvement project is needed to determine if the use of this workflow algorithm in other rural clinics would be beneficial and reflect similar results.

1 Continuous Glucose Monitoring: Facilitating the Initiation and Use for Patients with Type 2 Diabetes Mellitus in a Rural Primary Clinic Laura M. Dotson College of Nursing, University of Utah NURS 7703: DNP Scholarly Project III April 17, 2022 2 Abstract Background: Type 2 diabetes mellitus continues to be one of the most significant health problems worldwide. It can lead to several health complications, such as amputations, vision loss, neuropathy, kidney disease, and cardiovascular disease. These complications can be delayed or even avoided by maintaining glycemic stability through effective self- and medical management. Continuous glucose monitors (CGM) help both patients and providers manage glycemic trends and maintain stability, which can decrease the risk for complications and improve patient outcomes. By increasing rural health care providers’ awareness of the processes required for successful CGM initiation, more patients will experience long-term benefits and a reduction in devastating complications. Methods: A quality improvement project was implemented to increase rural providers’ awareness of the processes required for successful CGM initiation for patients with type 2 diabetes mellitus and insulin dependence. A workflow algorithm was developed and used to accomplish this task in one small, rural Utah clinic. An educational presentation of the workflow algorithm was first introduced to clinic staff, which consisted of office personnel, a physician, and a diabetes educator, before implementation. Pre-project data of CGM initiation were collected and compared to post-project CGM initiation data. Usability, feasibility, and satisfaction were analyzed in a post-survey provided to both clinic staff and qualified patients who participated. Results: Post-intervention, the project analysis found there was an 89.6% increase (17.4% vs 33%) in CGM initiation with the use of the workflow algorithm. Clinic staff reported 100% (n=5) in the ease of use and satisfaction with the algorithm, and 80% (n=4) reported plans for continued use. An unexpected benefit from the use of the algorithm was found when the clinic staff learned that the initiation of CGM and the interpretation of CGM data could be billed, 3 which is something they had not done previously. Patients reported overall satisfaction (n=22, 88%) with the personal care provided, continuity of care, and their improved self-management of diabetes. Three patients reported “dissatisfied” or “very dissatisfied” (n=3, 12%) in improved self-management and overall satisfaction. Conclusions: The CGM workflow algorithm appears to be a valuable resource to help facilitate CGM initiation in a small, rural Utah clinic. Replication of this quality improvement project is needed to determine if the use of this workflow algorithm in other rural clinics would be beneficial and reflect similar results. 4 Continuous Glucose Monitoring: Facilitating the Initiation and Use for Patients with Type 2 Diabetes Mellitus in a Rural Primary Clinic Problem Description Type 2 diabetes mellitus (T2DM) is a significant health problem worldwide, with most individuals never reaching or maintaining glycemic control (Mannucci et al., 2017). Glycemic variability and the progression of T2DM are associated with several complications such as amputations, vision loss, neuropathy, kidney disease, and cardiovascular disease (Lambrinou et al., 2019). Research has shown that critical components to achieving glycemic stability are diet, exercise, stress management, and appropriate medication management (Carlson et al., 2017). Practicing a structured regimen of assessing blood glucose several times a day is necessary to determine how effective both clinical and self-management strategies are working (Mannucci et al., 2017). Unfortunately, many individuals report not following a structured regimen to selfmonitor their blood glucose, and common reasons include cost, inconvenience, pain, frustration, and discouragement (Mannucci et al., 2017). Social determinants of health in rural communities can further complicate an individual's goal of reaching glycemic stability through self-management of T2DM (Bennett, 2018). Examples of these social determinants include the level of financial stability, medical insurance, access to healthy foods, access to exercise opportunities, transportation, and level of education. Studies have shown that the prevalence of T2DM is higher among individuals with lower economic status (Isaacs et al., 2021). While lower socioeconomic status directly correlates with other unhealthy behaviors (e.g., smoking, sedentary lifestyle, excessive alcohol consumption, poor nutrition), it is also related to suboptimal quality healthcare (Isaacs et al., 2021). 5 Continuous glucose monitoring (CGM) devices have been shown to help clinicians, and individuals with T2DM identify significant glycemic trends, which help determine appropriate treatment measures and manage variables appropriately and effectively (Cox et al., 2020). Continuous glucose monitoring is an innovative alternative to self-monitoring blood glucose (SMBG), and eliminates the inconvenience and pain associated with multiple finger sticks a day. Although this approach to diabetes management continues to grow in primary care, rural clinicians are reluctant to prescribe CGM due to unfamiliarity of the devices and their affordability, usability, and feasibility (Hirsch & Miller, 2021). A clinic workflow algorithm designed to help facilitate the integration and use of CGM in a rural clinic setting would help alleviate the reluctance of some providers to prescribe CGM. Available Knowledge Numerous studies provide evidence that CGM paired with self-management strategies of individuals with type 1 (T1) and insulin-dependent type 2 (T2) diabetes mellitus improve glycemic stability and quality of life (Hirsch & Miller, 2021; Bergenstal et al, 2018; Cappon et al, 2019). The American Diabetes Association (ADA) has included CGM and its effectiveness in treating individuals with T1 and T2 who are on intensive insulin regimens in the standards of medical care (ADA, 2019). Significant evidence has also demonstrated the correlation between CGM and the reductions in hemoglobin HbA1c and diabetes-related complications leading to hospitalizations (Charleer et al., 2018). Individuals who use insulin to manage T1 or T2 diabetes must adjust treatment based on glucose readings. However, multiple blood glucose readings a day are required to do this effectively and can be costly and inconvenient if using the traditional glucometer and finger sticks (Cox et al., 2020). CGM allows the user to continuously evaluate the effects of self- 6 management behaviors and medication adjustments on glucose levels. Additionally, most CGM devices will notify the user of glucose trends leading to dangerously low or high levels (Wysham & Kruger, 2021). The ability to evaluate glycemic trends continuously has been shown to improve confidence in self-management decisions, improve outcomes, and decrease diabetes distress in these individuals (Charleer et al., 2020). Rationale Numerous studies demonstrate that continuous glucose monitoring (CGM) improves outcomes for patients with diabetes mellitus who use insulin (Isaacs et al., 2021, Lafell et al, 2020; Cox et al, 2018). With this current knowledge, there still seems to be a lag in CGM utilization in rural and underserved regions (Isaacs, 2021; Cox et al, 2016). This lag may be linked to recent evidence that social determinants of health, such as gender, race, education level, and lower socioeconomic status are correlated with CGM underutilization by primary care providers (Isaacs et al., 2021). Providers may also be wary of prescribing these devices because they are unfamiliar with their limitations, strengths, and potential (Wysham & Kruger, 2021). The first assumption of this project was that a structured workflow algorithm for CGM initiation would increase the likelihood of providers prescribing CGM for patients regardless of their socioeconomic status, educational level, or history of self-management behaviors. A second assumption was that patients with T2DM who use insulin would increase competency in selfmanagement, confidence in their choices, and motivation to continue these behaviors regularly. The Plan-Study-Do-Act (PDSA) theoretical framework, commonly used by healthcare organizations to improve processes and outcomes, was used to guide this project (Institute for Healthcare Improvement, 2020). It was used to assess process improvement with the implementation of the CGM workflow algorithm in a rural primary care clinic. The PDSA model 7 addresses problems in a current process and is used to create process improvement by focusing on three critical questions (Brega et al., 2015): 1. What are we trying to accomplish? 2. How will we know that a change is an improvement? 3. What change can we make that will result in an improvement? With the implementation of the PDSA cycle for this project, several meetings with staff and the provider were needed to identify problems so that improvements could be made. The Plan phase of the PDSA cycle included the identification of a need for a structured workflow to increase the number of CGM initiations in the clinic. The Do phase included creating the workflow algorithm and implementing it. The Study phase analyzed the data to determine if the use of the algorithm increased CGM initiation. Finally, the Act phase was used to determine if any changes needed to be made. Specific Aims This project aimed to facilitate the use of CGM in a rural primary practice to support people with T2DM in the self-management of their diabetes. There were five objectives to achieve this purpose. The first objective was to identify individuals with T2DM who use insulin at a rural primary care practice. This process would also include the collection of demographic information, such as gender, race, and insurance type (see Table 1). The second objective was to develop of a clinic workflow algorithm to assist providers with the initiation and use of CGM for patients with T2DM and to provide CGM education (see Appendix A). The third objective was to implement the CGM workflow algorithm, starting with the introduction and education of the clinic staff and providers. The fourth objective was to evaluate the adoption of the algorithm and assess the feasibility, usability, and satisfaction of the CGM project (see Appendix B). The final 8 objective was to assess if any gaps are evident in demographics, insurance status, and education level of patients with T2DM and the initiation of CGM. Methods Context This quality improvement project was introduced to a small, rural clinic in southeastern Utah that primarily treats adolescents and adults. The clinic serves populations in two counties, with approximately 30,000 residents spread out over more than 6,000 square miles (United States Census Bureau, 2019). The providers and staff in this clinic consist of one family practice medical doctor, one medical assistant, and two office personnel. Other participants in this quality improvement project were new or established patients who have T2DM, are over 18 years of age, and use insulin daily. Before implementing this project, the provider has initiated CGM for patients without a structured protocol for follow-through. Intervention The first phase in this quality improvement project was to identify individuals who qualify to participate. Qualified individuals were defined as being over 18 years of age, having a current diagnosis of T2DM, and taking insulin multiple times a day. Individuals who had a current CGM prescription were excluded. The electronic medical record (EMR) was utilized to retrieve these data points, resulting in 282 qualified patients. Concurrently, each qualifying participant's demographic data, including gender, race, and insurance status, were gathered for the purpose of determining any gaps existing in CGM initiation. Other information retrieved from the EMR was the number of established patients with T2DM who had been prescribed a CGM between June 1, 2021, and September 1, 2021, before the implementation of the CGM workflow algorithm in the clinic. 9 The second phase included the development of a clinic workflow algorithm to guide the process of CGM initiation within the clinic for qualified individuals identified in the first step. The algorithm was developed based upon materials identified in a literature review on CGM usage in primary care. Additionally, professional input from the diabetes educator, the physician who is the content expert, and the faculty project chair, were utilized in its review and revision. The workflow algorithm starts with the identification of eligible patients for the initial appointment. It then walks through the initiation, use, and education of the CGM with the final step consisting of medical management, patient consultation, and referral to the diabetes educator (see Appendix A). This quality improvement project was introduced to the office personnel, medical assistant, and physician at a monthly in-service meeting demonstrating the use of the workflow algorithm. The purpose of the workflow algorithm is to help guide staff and providers in the initiation and use of CGM in this rural clinic to support individuals with T2DM in their selfmanagement. Several revisions to the algorithm were made based upon feedback from the staff during the first PDSA cycle that helped improve the implementation process, including the addition of order forms required by Medicare, Medicaid, and private insurers. The start and end dates for the project were agreed on as a group to be from October 13, 2021, through December 31, 2021. Study of the Intervention(s) The approach used to assess the impact of this quality improvement project was to first collect retrospective data from the EMR on the number of qualified patients with T2DM who had a CGM initiated in the clinic. These data were collected from the last three months before the launching of the project. Next, a weekly EMR review was done to analyze the use of the 10 workflow algorithm and the number of CGMs initiated in the office during the project. Finally, at the close of the three-month project, a comparison of both pre and post-data was made to determine the impact of the project. Before implementing this project, there was not a specific clinic workflow or process used to initiate CGMs for patients with T2DM. For this reason, it is reasonable to conclude that any increase in CGM initiation may have been the result of the use of the CGM workflow algorithm implemented for this project. Finally, the demographics, including gender, race, and insurance status, of the patients who qualified for CGM initiation were analyzed and summarized in Table 1. This is useful in determining if any gaps existed in CGM initiation. The utilization of two post-implementation surveys determined the usability, feasibility, and satisfaction of the project. The first survey administered to the clinic staff and the provider contained both closed and open-ended questions (see Appendix B). The last question asked for feedback and suggestions for improving the algorithm. The second survey was distributed to all participating patients and evaluated their satisfaction of care during the project (see Appendix C). Measures This quality improvement project was performed over five months. The first two months were spent collecting the total number of qualified patients in the clinic, the number of these patients seen and the number of CGM’s initiated to complete a retrospective analysis for the three months before the actual implementation of the workflow algorithm. The number of patients who were prescribed a CGM during these three months and the patients’ demographics were documented using a Microsoft Excel spreadsheet. Once the project was implemented in the clinic, three more months of data were collected from the clinic EMR. In addition, any new 11 patients who received a CGM who were not on the original list of qualified participants were added weekly where indicated. The pre- and post-data of the quality improvement project were used to compare the rate of CGM initiation and determine the effectiveness of the workflow algorithm and staff trainings. Both quantitative and qualitative analysis of an eight question post-survey were used to evaluate the feasibility, usability, and satisfaction of the workflow algorithm using both closedand open-ended questions (see Appendix B). This survey was distributed to the office staff, Medical Assistant (MA), the provider, and the diabetes educator associated with the clinic. Feasibility was assessed through the first question, which asked participants how often they used the algorithm and which offered the following responses to choose from: "not at all," "sometimes," "about half the time," "most of the time," and "always." Two follow-up, openended questions asked participants to describe what facilitated and impeded the use of the workflow algorithm. Finally, the level of satisfaction in using the algorithm was evaluated with four questions relating to its ease of use, administrative support, workload requirement, and functionality. These were scored on a five-point Likert scale ranging from "very satisfied" to "very dissatisfied." One open-ended response concluded the survey by asking for feedback on the algorithm and suggestions for improvement. A separate four-question satisfaction survey was provided to patients who participated (See Appendix C). These questions were scored on the same five-point Likert scale used in the clinic staff and provider survey. The level of patients’ satisfaction was evaluated based on four criteria: patient's perception of personalized care, continuity of care, improved self-management of diabetes, and overall satisfaction. Analysis 12 A summary of demographic data, comparison of means, and a category count were described using frequency distributions. A chi-square test was used to analyze the percentage change in the number of CGMs implemented during the project using the workflow algorithm. Finally, survey data were collected from Likert scale questions and open-ended questions to determine any barriers and facilitators of the project. Data were categorized, organized, and summarized. Ethical Considerations The University of Utah Institutional Review Board determined this study to be exempt from human subject review and approved the study as a quality improvement project. An ethical consideration was anonymity. All participation in this project and post-survey was voluntary, and all participant identities were left anonymous. There were no conflicts of interest identified during this quality improvement project. Results During the three-month period prior to the implementation of the workflow algorithm, from June 1, 2021, to August 1, 2021, the rural clinic had a CGM initiation rate for qualified patients of 17.4% (16/92). Three months after implementing the workflow algorithm, the CGM initiation rate for qualified patients was 33% (38/115), an 89.6% increase. A chi-square test of independence showed a significant increase in CGM initiation rates pre to post, X2 (1, N = 207) = 6.49, p = .0108 (see Figure 1). The feasibility, usability, and satisfaction of the use of the CGM workflow algorithm were determined from the survey responses from the clinical staff and patients who participated in the algorithm implementation (see Appendix B). Of the clinic staff, there were 5 (100%) that completed the survey, including a provider, a medical assistant, a receptionist, an office manager, 13 and a diabetes educator. All of the clinic staff (n=5, 100%) reported that the workflow algorithm was both “reasonable” and “easy to use.” Most of the clinic staff (n=4, 80%) reported that they intend to continue to use the algorithm. No additional feedback on the algorithm or suggestions for improvement was provided. Barriers and facilitators to using the workflow algorithm were addressed with clinic staff in two open-ended questions. Barriers included patient frustration with device failure (n=2, 40%) and delay in prior authorization (n=1, 20%). Facilitators to the use of the algorithm were identified in a comment made by the provider: “The best part of this algorithm is that it educated us in how we can get paid for what we have already been doing.” In addition, all the clinic staff agreed that the use of the algorithm increased their awareness of effective processes for successful CGM initiation (n=5, 100%). All of the clinic staff reported “very satisfied” or “satisfied” with the ease of use, administrative support, workload requirement, and functionality of the workflow algorithm (n=5, 100%). Of the 38 patients who received a CGM during the implementation of the workflow algorithm, there were 25 (66%) who completed the patient satisfaction survey. The patients reported “very satisfied” or “satisfied’ in the personal care provided, continuity of care, improved self-management of their diabetes, and overall satisfaction (n=22, 88%), and three patients reported dissatisfied or very dissatisfied (n=3, 12%) in improved self-management and overall satisfaction. Of the 115 patients seen in this clinic, the majority were male (63%), Caucasian (87%), and older than 65 years of age (80%). These patients also tended to use Medicare (73%) as their 14 primary insurance. Relationship between the sociodemographic characteristics and outcome, were not explored due to low diversity of patients receiving a CGM. Contextual Elements Contextual elements identified with this quality improvement project included manual data collection and inconsistencies in the documentation of CGM initiation during the three months before project implementation. Data extraction from the clinic EMR was done manually and became quite a tedious and time-consuming task. The office manager was able to provide structured data, such as demographics, diagnosis of T2DM, and insurance status. but the rest of the data, such as insulin use and CGM initiation, had to be collected through manual review of each patient’s EMR, which could have led to missed or incorrect data due to human error. In addition, documentation of CGM initiation was not consistently done by the provider during the three months prior to implementation of the algorithm. The provider frequently initiated CGM by using samples and did not document it, which may have resulted in inaccurate pre-project numbers. Although the data collection process and missing documentation in the EMR may have affected the precision of the data collected during the first phase of this project, this phase did lead to an unexpected benefit to the clinic. During one of the PDSA cycles, the clinic office manager was asked if we could track the number of CGM initiations by the CPT code used for billing. It was identified that the clinic staff were unaware of the CPT codes that could be used for both the initiation of a CGM and interpretation of CGM data. At that point, the enthusiasm and support for the workflow algorithm and this QI project by all clinic staff increased. 15 Another element that was identified influencing data collection was the current COVID19 pandemic. This may have influenced the number of established patients with T2DM coming to their regular scheduled check-ups with the provider. Missing Data The impact of the implementation of the CGM workflow algorithm may have been affected by the number of patients seen at different times of the year. The data collected during the three months before implementation was done during the summer months of June, July, and August. These months are reported by the staff to have lower numbers of patients seen overall in the clinic. Compared to the months of October, November and December, when the project was implemented. December is a month when the number of patients seen is especially high in comparison. Most likely due to the clinic trying to meet end-of-quarter goals and patients avoiding the first of the year deductibles. The education level of qualified patients was not collected as part of the demographic data used to determine any relationship to CGM initiation. This data field was found to be left blank on the majority of qualified patients' EMR, so it was not included in the final summary. Discussion Summary This project aimed to facilitate the use of CGM in rural primary practice to support people with T2DM in the self-management of their diabetes. Before the implementation of the workflow algorithm, the rate of CGM initiation in a rural clinic was 17.4% (16/92); after three months of using the workflow algorithm, the rate of CGM initiation was 33% (38/115), an 89.6% increase in initiation rates. The rate increase may suggest that the workflow algorithm is efficacious in improving CGM initiation in this rural clinic. The post-survey results from clinic 16 staff indicated that the workflow algorithm increased awareness of effective processes for successful CGM initiation, resulting, therefore, in increased rates. The majority of patients reported that they were “very satisfied” or “satisfied’ in the following: personal care provided, continuity of care, improved self-management of diabetes, and overall satisfaction with the CGM initiation process (n=22, 88%). This finding demonstrates that the patient's perception of improved self-management of T2DM was achieved. Interpretation Numerous studies demonstrate that continuous glucose monitoring (CGM) improves outcomes for patients with diabetes mellitus who use insulin (Lafell et al, 2020; Cox et al, 2016; Cox et al, 2018). It has also been demonstrated in one study that there seems to be a lag in CGM utilization in rural and underserved regions linked to lower socioeconomic status and other social determinants of health (Isaacs, 2021). With that current knowledge, it is hypothesized that rural providers may be wary of prescribing these devices because they are unfamiliar with their limitations, strengths, and potential (Wysham & Kruger, 2021). Continuous glucose monitor initiation was increased in a rural clinic after the implementation of a CGM workflow algorithm with this quality improvement project (17.4% vs 33%). This finding is supportive of the assumption that with education and increased awareness of processes required for successful CGM initiation, providers in rural clinics are more apt to initiate CGMs for qualified patients. The impact of this project was evident in the benefits to the rural clinic. As a result of using the workflow algorithm, the clinic staff learned that there were specific CPT codes for both the initiation of a CGM and the interpretation of CGM data allowing the clinic to bill for these services. With this new knowledge, the enthusiasm and support for the workflow algorithm and 17 this QI project by all clinic staff increased. It is reasonable to assume that the workflow algorithm, or components of it, will be sustainable and its use will continue in this clinic. The impact this project had on self-management behaviors, improvement in glycemic control, and decreased diabetes distress for patients with T2DM was not formally measured for this project due to the short implementation timeframe. Patient demographics, such as gender, race, and insurance status, did not appear to impact CGM initiation in this rural clinic, but it is not reasonable to conclude that these social determinants of health did not influence CGM initiation. Relationship between the sociodemographic characteristics and outcome, were not explored due to the low diversity of patients receiving a CGM. An assumption could be made that due to the initiation of the workflow algorithm, all qualified patients were offered CGM initiation regardless of insurance status. The education level of patients was not included in the data collected as originally planned due to poor documentation of this demographic in the EMR. Limitations This project has some limitations. First, the project was only initiated in one small rural Utah clinic with only one provider and three staff members. This small sample prevents any evidence of the workflow algorithm’s usability, feasibility, and satisfaction in other clinics. Another limitation was the low diversity of the target population. This clinic specializes in the primary care of older adults, which limited the generalizability of insurance status. The race of the target population was also found to be 86.9% (100/115) white and non-Hispanic. This limited a reliable analysis of gaps between CGM initiation and these social determinants of health. Conclusions 18 Innovations in the treatment and management of patients with T2DM, including continuous glucose monitoring (CGM) devices, have great potential in the prevention and reduction of many of the health complications associated with poor glycemic stability (Charleer et al., 2020). Many rural clinics and providers, like those in the participating clinic, may have trepidations to CGM use due to their unfamiliarity with the devices and the processes needed for successful initiation. The implementation of this project’s workflow algorithm resulted in an 89.6% increase in the rate of CGM initiation, a statistically significant improvement. To further verify these results, this project should be continued in other rural clinics. The accessibility of this project’s materials and data can help guide other providers and clinics implementing a similar workflow algorithm. This project would also benefit from examining the impact the algorithm has on patient outcomes. 19 References American Diabetes Association (ADA) (2019). Lifestyle management: Standards of medical care in diabetes. Diabetes Care, 42(1), 46–60. DOI 10.2337/dc19-S005 Bennett, N.M. (2018). Social Determinants of Health: Improve health outcomes beyond the clinic walls. American Medical Association: Steps Forward. Retrieved from https://edhub.ama-assn.org/. Bergenstal, R. M., Beck, R. W., Close, K. L., Grunberger, G., Sacks, D. B., Kowalski, A., Brown, A. S., Heinemann, L., Aleppo, G., Ryan, D. B., Riddlesworth, T. D., & Cefalu, W. T. (2018). Glucose management indicator (GMI): A new term for estimating A1C from continuous glucose monitoring. Diabetes Care, 41, 2275–2280. DOI 10.2337/dc18-1581 Brega, A. G., Barnard, J., Mabachi, N. M., Weiss, B. D., DeWalt, D. A., Brach, C., Cifuentes, M., Albright, K., & West, D. R. (2015). Plan-Do-Study-Act (PDSA) directions and examples. Health literacy universal precautions toolkit (2nd ed.). Agency for Healthcare Research and Quality. https://www.ahrq.gov/health-%09literacy/improve/precautions/tool2b.html Cappon G., Vettoretti M, Sparacino G., & Facchinetti A. (2019). Continuous glucose monitoring sensors for diabetes management: A review of technologies and applications. Diabetes Metab Journal, 43(4), 383–397. DOI 10.4093/dmj.2019.0121 Carlson, A. L., Mullen, D. M., & Bergenstal, R. M. (2017). Clinical use of continuous glucose monitoring in adults with type 2 diabetes. Diabetes Technology & Therapeutics, 19(2), 4–11. DOI 10.1089/dia.2017.0024. Charleer, S. Block C. D., Van Huffel, L., Broos, B., Fieuws, S., Nobels, F., Mathieu, C., & Gillard, P. (2020). Quality of life and glucose control after 1 year of nationwide reimbursement of intermittently scanned continuous glucose monitoring in adults living 20 with type 1 diabetes (FUTURE): A prospective observation real-world cohort study. Diabetes Care, 43, 389–397. DOI 10.2337/dc19-1610 Cox, D. J., Taylor, A. G, Moncrief, M., Diamond, A., Yancy, W. S. Jr., Hegde, S., McCall, A. L. (2016). Continuous glucose monitoring in the self-management of type 2 diabetes: A paradigm shift. Diabetes Care, 39, 71–73. DOI 10.2337/dc15-2836 Cox D. J., Basu R., & McCall A. (2018). Continuous glucose monitoring functions as a negative feedback loop for individuals with type 2 diabetes. Annals of Diabetes Metabolic Disorders Control, 1(3), 121-123. Cox, D. J., Banton, T. Moncrief, M., Conaway, M., Diamond, A., & McCall, A. L. (2020). Minimizing glucose excursions (GEM) with continuous glucose monitorinig in type 2 diabetes: A randomized clinical trial. Journal of the Endocrine Society. 4(11), 1–7. DOI 10.1210/jendso/bvaa118 Hirsch, I.B., Miller, E. (2021). Integrating continuous glucose monitoring into clinical practices and patient’ lives. Diabetes Technology & Therapeutics, 23 (3), 72-80. DOI 10.1089/dia.2021.0233 Isaacs, D., Bellini, N. J., Biba, U., Cai, A., & Close, K. L. (2021). Health care disparities in use of continuous glucose monitoring. Diabetes Technology & Therapeutics, 23(3), 81–87. DOI 10.1089/dia.2021.0268 Institute for Healthcare Improvement (IHI) (2020). Science of Improvement: How to Improve. Institute for Healthcare Improvement. Retrieved from http://www.ihi.org/resources/Pages/HowtoImprove/ScienceofImprovementHowtoImprov e.aspx 21 Janapala, N. R., Jayaraj, J. S., Fathima, N., Kashif, T., Usman, N., Dasari, A., Jahan, N., & Sachmechi, I. (2019). Continuous glucose monitoring versus self-monitoring of blood glucose in type 2 diabetes mellitus: A systematic review with meta-analysis. Cureus, 11(9), e5634. DOI 10.7759/cureus.5634 Laffel, L. M., Kanapka, L. G., Beck, R. W., Bergamo, K., Clements, M. A., Criego, A., DeSalvo, D. J., Goland, R., Hood, K., Liljenquist, D., Messer, L. H., Monzavi, R., Mouse, T. J., Prahalad, P., Sherr, J., Simmons, J. H., Wadwa, R. P., Weinstock, R. S., Willi, S. M., & Miller, K. M. (2020). Effect of continuous glucose monitoring on glycemic control in adolescents and young adults with type 1 diabetes: A randomized clinical trial. JAMA, 323(23), 2388–2396. DOI 10.1001/jama.2020.6940 Lambrinou, E., Hansen, T. B., Beulens, J. (2019). Lifestyle factors, self-management and patient empowerment in diabetes care. European Journal of Preventive Cardiology, 26(25); 5563. Mannucci, E., Antenore, A., Giorgino, F., & Scavini, M. (2018). Effects of structured versus unstructured self-monitoring of blood glucose on glucose control in patients with noninsulin-treated type 2 diabetes: A meta-analysis of randomized controlled trials. Journal of Diabetes Science and Technology, 12(1), 183–189. DOI 10.1177/1932296817719290 United States Census Bureau (2019). Quick Facts: Carbon County, Utah; Emery County, Utah. https://www.census.gov/quickfacts/fact/table/emerycountyutah,carboncountyutah,US/PS T045219 Wysham, C.H., & Kruger, D.F. (2021). Practical considerations for initiating and utilizing flash continuous glucose monitoring in clinical practice. Journal of the Endocrine Society, 5(9), 1–9. DOI 10.1210/jendso/bvab064 22 Table 1 Demographics of Qualified Participants Characteristic N=115 % CGM Initiated n=38 % Gender Male 72 63 20 53 Female 43 37 18 47 Caucasian 100 87 29 76 African American 1 0.8 1 3 Native American 2 0.2 1 3 Hispanic or Latino 12 10 7 18 Medicare 84 73 20 53 Medicaid 8 7 2 5.3 Uninsured 5 4 1 2.6 Private 18 12 15 39 Race Insurance status 23 Figure 1 Continuous Glucose Monitor Initiation Rates 77 76 80 # of Qaulified Patients 70 60 38 50 40 16 30 20 10 0 Pre-Project Post-Project Axis Title No CGM Initiated CGM Initiated 24 Appendix A CGM Workflow Algorithm 25 Appendix B Post-Survey Questions: Clinic Staff Q1 How often did you use the workflow algorithm? About half the time Never (1) Sometimes (2) (3) o o o Most of the time (4) o Always (5) o Q2 What barriers did you encounter with it's use? ____________________________________________________________ Q3 What facilitated the use of the algorithm? ________________________________________________________________ Q4 Were you satisfied with the ease of use of the algorithm? Extremely dissatisfied (1) o Somewhat dissatisfied (2) Neither satisfied nor dissatisfied (3) o o Somewhat satisfied (4) o Extremely satisfied (5) o Q5 Did you feel that you had administrative support when needed? Extremely dissatisfied (1) o Somewhat dissatisfied (2) Neither satisfied nor dissatisfied (3) o o Somewhat satisfied (4) o Extremely satisfied (5) o Q6 Was the workload requirement appropriate? Extremely dissatisfied (1) o Somewhat dissatisfied (2) Neither satisfied nor dissatisfied (3) o o Somewhat satisfied (4) o Extremely satisfied (5) o Q7 Were you satisfied with the functionality of the algorithm? Extremely dissatisfied (1) o Somewhat dissatisfied (2) o Neither satisfied nor dissatisfied (3) o Somewhat satisfied (4) o Extremely satisfied (5) o Q8 Please provide feedback on the algorithm and any suggestions for improvement ________________________________________ 26 Appendix C Post-Survey Questions: Patient Q1 Were you satisfied with the level of care received during this project? Extremely dissatisfied (1) Click to write Statement 1 (1) Somewhat dissatisfied (2) o Neither satisfied nor dissatisfied (3) o o Somewhat satisfied (4) o Extremely satisfied (5) o Q2 Did you feel this increased the level of continuity of care? Extremely dissatisfied (1) Click to write Statement 1 (1) Somewhat dissatisfied (2) o Neither satisfied nor dissatisfied (3) o o Somewhat satisfied (4) o Extremely satisfied (5) o Q3 Did you feel this project improved your self-management of diabetes? Extremely dissatisfied (1) Click to write Statement 1 (1) Somewhat dissatisfied (2) o o Neither satisfied nor dissatisfied (3) o Somewhat satisfied (4) o Extremely satisfied (5) o Q4 What is your overall satisfaction of this project? Extremely dissatisfied (1) Click to write Statement 1 (1) o Somewhat dissatisfied (2) o Neither satisfied nor dissatisfied (3) o Somewhat satisfied (4) o Extremely satisfied (5) o