Implementing a Continuous Glucose Monitoring Hospital Discharge Program: Strategies and Best Practices

Introduction

Diabetes mellitus is a common condition among hospitalized patients, affecting approximately 30% of all individuals admitted to United States hospitals. ¹ Continuous glucose monitoring (CGM) systems are now considered the standard of care for monitoring glucose for individuals with diabetes on insulin therapy. Moreover, the American Diabetes Association’s 2025 Standards of Care recommend that all people with diabetes be offered access to CGM. ² This technology offers real time insight into glycemic trends which help avoid short term complications of diabetes such as hypoglycemia or hyperglycemia and long term complications such as retinal, renal and cardiovascular disease. ³ In addition, CGM can help individuals identify how lifestyle behaviors, such as dietary choices or activity, can influence glycemic control. ⁴ For these reasons, CGM use has been associated with significant improvements in Hemoglobin A1c (HbA1c), reduced diabetes distress, and increased engagement in diabetes self-management. ⁴ With these potential benefits, CGM initiation should be considered at or soon after diabetes diagnosis. Given that many people with diabetes are initially diagnosed incidentally during a hospitalization, the transition period prior to discharge represents a unique opportunity to introduce CGM. ⁵

Initiating CGM at hospital discharge is an attractive option, as it offers an opportunity to educate patients about diabetes, reinforce proper device use, compare CGM values with capillary glucose readings, and review glycemic trends under provider supervision. ⁶ At the time of discharge, CGM may offer improved glycemic control and better detection of hypoglycemic events compared to traditional blood glucose monitoring (BGM).^7-9 In preliminary studies, CGM demonstrated an approximately 20% higher detection rate of hypoglycemia and severe hypoglycemia, as well as a 13% higher median time in range, compared to traditional capillary blood glucose testing.^7,8 Patients discharged from the hospital with CGM also reported 30% higher scores on diabetes-related quality-of-life satisfaction surveys and 41% higher diabetes management scores compared to those without CGM. ¹⁰

Although implementing a CGM hospital discharge program holds promise for improving diabetes care, several challenges continue to hinder its widespread adoption.^11,12 These challenges include therapeutic inertia, appropriate patient selection, insurance coverage for devices, limited hospital resources, and a lack of clear guidance on effective implementation strategies.^11,12

In this article, we present explicit guidance and a protocol for implementing CGM at the time of hospital discharge. We believe this protocol offers hospitals and providers actionable steps to facilitate CGM initiation and establish a foundation for successful outpatient follow-up. The protocol was developed by the Armstrong Institute Diabetes Clinical Community at Johns Hopkins Medicine and is based on the current practices of Endocrine Hospitalist teams at three community hospitals: Johns Hopkins Howard County General Hospital in Columbia, MD; Suburban Hospital in Bethesda, MD; and Sibley Memorial Hospital in Washington, DC^13,14 (Figure 1).

OPEN IN VIEWER

Figure 1.

CGM hospital discharge workflow.

CGM Hospital Discharge Program

The authors propose the following protocol for optimal initiation of CGM on discharge.

The first step in the process is multidisciplinary coordination. Key stakeholders should include the legal team, CGM suppliers, diabetes care and education specialists, clinical staff, and the inpatient glycemic management team. All team members should have a basic understanding of CGM devices and their use. Educational resources for clinical staff are available through the device manufacturers.^15,16 Each company also has representatives available for outreach.^17,18 Continuous glucose monitors can be obtained either by the hospital purchasing the CGMs directly, or by obtaining free sensor samples from manufacturers. If the CGMs are to be obtained as free samples from manufacturers (i.e., via a manufacturer CGM hospital discharge program), the authors recommend contacting the hospital’s legal team to obtain authorization for their use.^19,20 Importantly, if samples from multiple companies are used, this should not be seen as favoritism or a conflict of interest—unless a member of the team has a direct business relationship with a specific company, in which case disclosure would be necessary.

The use of actual CGMs at the time of discharge offers a seamless and integral opportunity for patient education. It allows clinical staff to teach patients directly how to insert the sensor subcutaneously and pair the device with their smartphone or sensor to receive glucose data. Training videos alone are often insufficient, particularly for overwhelmed, ill or patients who are uncomfortable with the new technology. After the initial education, the prescriber would have to submit a prescription, so the patient would then purchase further CGMs either through their pharmacy or durable medical equipment company.

The authors strongly recommend offering CGMs from competing manufacturers, to ensure the most appropriate device is selected based on patient preference and insurance coverage. Arrangements with competitor representatives should be made to secure educational supplies for inpatients. This approach is expected to improve patient safety, enhance the quality of care, and potentially reduce hospital readmissions by ensuring patients are properly trained on CGM use before discharge.^12,21 If complimentary CGM samples are to be used, consider using the following sample letter to the hospital’s legal counsel (Figure 2).

OPEN IN VIEWER

Figure 2.

Sample letter to the legal counsel justifying the use of complimentary samples for the CGM discharge program.

The second step of the process is patient selection. All individuals with diabetes are potential candidates, and selection criteria should aim to be as inclusive as possible. However, because initiating CGM on every patient with diabetes at discharge may not be realistic, the following characteristics should be considered when prioritizing patients. High risk features for future readmission include uncontrolled diabetes (HbA1c ≥9%), newly starting insulin on discharge, or significant adjustments made to the insulin treatment regimen, admission for glycemic emergencies (e.g., diabetic ketoacidosis, hyperosmolar hyperglycemic state, severe hypoglycemia) and recent readmission.

Other selection considerations include type 1 or type 2 diabetes on multiple daily injections or continuous subcutaneous insulin or, financial means to pay for outpatient CGM (i.e., health insurance likely to cover CGM devices or patients are agreeable to purchase CGM if not covered by health insurance (Table 1). ²² Special patient populations to consider include pregnant individuals and those with hypoglycemia unawareness as CGM has been shown to reduce fetal diabetes complications and decrease incidence of hypoglycemia in the latter. ²² Review of CGM use in the geriatric population has also shown improved glycemic control with support from health care professionals. ²³

OPEN IN VIEWER

Table 1.

Medicare Criteria for CGM Coverage.

Criteria	Medicare CGM coverage
1	Diagnosis of diabetes
2	Require insulin or history of “problematic” hypoglycemia
3	The patient or caregiver has been trained in use of CGM
4	CGM is prescribed according to FDA indications

All Criteria must be met for CGM to be covered by medicare. ¹⁷

Patients also need access to a smartphone that supports the desired CGM application, which can be used as the reader/receiver. Importantly, certain Apple or Android smartphone models do not have the capability to pair with a CGM. This requirement may not be necessary if a separate standalone CGM reader/receiver is available for the patient. Continuous glucose monitor manufacturers have programs in place to provide free CGM starter kits for patients nearing discharge that include sensors and readers/receivers if needed.^19,24

Patients should also ideally be discharging home. Patients being discharged to post-acute care facilities (e.g. rehabilitation centers) may not be good candidates, as nursing home facilities may not have the means in place to support the use of a CGM within their facility. The ability to respond to CGM alarms and alerts should also be evaluated prior to initiation. Ideal candidates should demonstrate receptiveness to using CGM or have strong social support systems in place to facilitate its use.

Once a candidate has been identified and expresses interest in CGM, the third step in the protocol is patient education. The CGM education process should ideally begin as early as possible. It is important to be mindful of timing as selected patients should not be critically ill and should not have planned imaging studies that may interfere with the device (per manufacturer specific restrictions) at the time of education. At Johns Hopkins Medicine community hospitals, an order is placed in the electronic health record by a provider (e.g., Endocrine Hospitalist), and most CGM sensors are placed by an inpatient certified diabetes care and education specialist (CDCES).

The implementation of a structured CGM protocol with defined training and competency requirements is essential to ensure quality and safety. CDCES bring a unique skill set to this effort, providing both direct patient education and supporting skill development for nursing staff and other licensed clinical providers. As inpatient CGM technology continues to evolve, nurse councils and educators will play a critical role in staff training—particularly when a CDCES is not available for bedside CGM placement or patient/caregiver instruction.

Training and competency requirements may vary based on hospital resources and priorities, especially given that formal inpatient diabetes programs are not uniformly present across institutions. Staff training should include objective education on CGM benefits and limitations, as well as hands-on instruction in CGM setup, sensor insertion, device pairing or activation, and application use. Education should emphasize how CGM differs from standard capillary glucose testing, the interpretation of CGM data, and sensor calibration (if applicable). Staff must also be prepared to address common challenges such as skin irritation, poor adhesion, or device malfunction—issues that require critical thinking and troubleshooting.

Patient education should include an assessment of clinical appropriateness, patient and caregiver readiness, access to follow-up, and availability of necessary resources. Patients must be instructed on setting up their receiver or smartphone, understanding trends and alerts, and anticipating sensor changes to ensure continuity of care. Although CGM manufacturers provide video tutorials and customer service, patients may still face barriers due to limited digital literacy or technology access.

To maintain neutrality, staff must present device options objectively and avoid endorsing one product over another. All education provided—whether to patients or staff—should be documented clearly in the medical record (see Figure 3). Increasing the number of staff trained in CGM use and education will expand patient access and better support glycemic control at discharge.

OPEN IN VIEWER

Figure 3.

Sample documentation of CGM education.

Patients should be informed that CGM devices are not yet U.S. Food and Drug Administration (FDA)-approved for inpatient use (although approval in the near future is anticipated). ²⁵ Most hospitals will continue to rely on point-of-care glucose checks for treatment decisions while the patient remains admitted. This distinction must be clearly communicated to set appropriate expectations. A signed consent form for CGM use during hospitalization may be required depending on local legal regulation.

Notably, while CGM education may appear resource-intensive, staff training to support patient use typically takes less than 30 minutes. At Johns Hopkins, discharge planning begins with the patient’s initial evaluation, and CGMs are generally placed 24 to 48 hours prior to discharge. This proactive approach helps ensure a smooth transition without increasing the patient’s length of stay. ²⁶

The fourth step of the process is to identify the provider who will follow up on the CGM data after discharge. The selection of clinical staff responsible for reviewing CGM data should be individualized based on the patient’s needs and the resources of the treating hospital. Robust communication between the discharging team and the receiving outpatient team is essential to identifying the most appropriate clinician to assume follow-up responsibilities. While the patient’s primary care provider is often the default choice, some primary care clinicians may not yet be equipped to manage CGM data due to unfamiliarity with the technology or lack of support infrastructure. In such cases, referral to an endocrinologist or CDCES may be warranted. In addition, most patients should be referred to a Diabetes Self-Management Training program, if available, for continued support.

Prior to discharge the inpatient prescriber must submit a CGM prescription. Insurance coverage and cost are often barriers to CGM initiation. Prior authorization or additional documentation may be necessary to successfully submit a prescription. Tools such as the Parachute Health platform can be used to facilitate communication with durable medical equipment suppliers and streamline device fulfillment. ²⁷

At our institution, we utilize an order that enables us to send patients a secure link to connect their new CGM with their outpatient provider. This link is delivered via email or the inpatient portal, facilitating seamless data sharing post-discharge. In addition, both major CGM manufacturers offer the option to share CGM data using a clinic code or practice ID, which can be provided to the patient to ensure their glucose data is linked directly to the outpatient clinic for ongoing monitoring and management.^28,29 The patient enters clinic code/practice ID within the CGM application to start sharing tracing. Third party cloud services utilize similar means of CGM data sharing and can be considered based on provider preferences. ³⁰ Continuous glucose monitor data should ideally be connected prior to the outpatient clinic appointment, however this can also be done in clinic within minutes should it be required.

At Johns Hopkins Health System, we prioritize referring patients to the endocrinology or diabetes care program within the same institution where inpatient care occurred whenever possible. This facilitates continuity of care by allowing outpatient providers to access the integrated health record and review inpatient management decisions, including those related to CGM initiation. When internal referral is not feasible, we ensure a smooth transition by sending the relevant CGM-related documentation—including provider notes and the discharge summary—to the outpatient clinician. In addition, our inpatient team frequently contacts the outpatient provider directly to communicate the CGM plan and request continued follow-up.

The final step of the process is consistent outpatient follow up. Timely outpatient appointments allow for CGM data review, therapeutic adjustments, and reinforcement of patient education. These follow-up visits are essential for addressing any knowledge gaps that may have arisen during the inpatient education phase. In our practice, timely outpatient follow-up—typically scheduled within 2 to 4 weeks—helps reinforce key concepts and support ongoing management. This appointment may be with an endocrinologist, primary care physician, or diabetes educator, depending on the patient’s needs and care plan.

The responsibilities of the outpatient care team are as follows. If circumstances allow for CGM sensor placement prior to discharge, the educating staff has the opportunity to help the patient accumulate initial glucose readings. This enables real-time education on how to interpret CGM data and respond appropriately to different scenarios and glycemic trends. Outpatient education should build on the education begun on the inpatient side. At every visit, providers should affirm and reinforce patient understanding of CGM and its use.

Gaps in patient understanding of CGM should be actively sought out and addressed. Continuous glucose monitor readings should be reviewed, trends should be discussed and education regarding best response to trends should be communicated. Limitations of CGM should be reinforced and adjunctive blood glucose checks should be encouraged for routine validation of CGM readings, especially when CGM readings may not be consistent with patient symptoms.

The importance of consistent follow up needs to be impressed upon the patient. Virtual visits and remote monitoring where possible can help make consistent follow up more accessible for the patient. When patients are admitted, outpatient providers should feel empowered to communicate with the inpatient care team and coordinate transition to outpatient care as a patient nears discharge.

Discussion

The purpose of this article is to provide actionable steps that providers and hospitals can use to address key challenges and successfully implement a CGM discharge program. Despite existing barriers, the authors believe—based on preliminary studies—that such an initiative has the potential to improve patient outcomes.^7-9,12,31 Folk et al ⁷ and Umpierrez et al ⁹ showed that use of CGM increased time in range (glucose 70-180 mg/dl) and improved detection of hypoglycemia. In another study, Crawford et al ³¹ demonstrated a 32 to 34% increase in time in range among medically complex patients who initiated CGM at discharge and received provider-driven interventions based on CGM data over the 30 days following discharge. O’Connor et al evaluated CGM initiation following discharge from the emergency department and found that 90% of patients and providers considered it useful. ³² In addition, patients who received CGM experienced a greater absolute reduction in HbA1c compared to those without CGM. ³² While these studies provide valuable insight into the benefits of initiating CGM at discharge, most are short pilot studies with either incomplete reporting, limited methodological detail, or designs that are difficult to generalize and adapt for widespread implementation. Efforts to provide guidance in discharging patients with diabetes has been attempted by our team and others.^11,12 Tian et al ¹² provides a comprehensive literature review discussing the impact of initiating CGM at discharge, the associated challenges, and potential strategies to overcome them. The broad categories of barriers defined by Tian et al ¹² were regulatory, behavioral, technical, logistical, staffing and systemic. The current regulatory limitation of CGM is the lack of FDA approval for inpatient use.^12,25 As a result, hospitalized patients who have never used CGM, lose an opportunity to learn how to fully utilize CGM under provider supervision, an experience which would be invaluable for further self-management after discharge. ¹² Dexcom, Inc currently has FDA breakthrough device designation for inpatient use which will likely expedite CGM use in hospitals and change this situation.^12,33 The behavioral barrier refers to learning to use CGM for the first time in the hospital, which is often a stressful environment. ¹² This barrier could be addressed by transitional care programs, such as the CGM discharge program we propose, and post discharge virtual visits. ¹² The technical barrier refers to the requirement of a compatible smartphone, technological literacy, sensor design limitations and the need to remove CGM during imaging. ¹² This barrier should be decreased as smartphone access increases and application interfaces become more user friendly. ¹² The logistic barrier describes access to health care, direct and indirect cost of CGM and diabetes management, and the challenges of obtaining insurance coverage for CGM. ¹² Addressing this barrier is limited largely by social determinants of health. Enhanced comfort with prescribing and interpreting CGM in the primary care setting could increase access to this technology in the most vulnerable communities. Furthermore experience and education around CGM prescribing would inform the process of navigating insurance barriers that may include prior authorizations and formulary preferences. The staffing barrier describes the limited staff familiarity with CGM available to teach patients how to use CGM. ¹² This barrier could be alleviated by providing teaching to existing staff around the use of CGM via in person or online courses. ¹² The systemic barrier refers to the difficulty integrating the large amount of data that CGM can provide into current physician workflows. ¹² Additional training for primary care providers and non-diabetologists is essential for reviewing and interpreting CGM data to address this barrier. ¹²

Our team has also developed strategies to not only support the successful discharge of patients with diabetes, but to also address the associated challenges. ¹¹ While prior efforts have been informative, they do not offer specific, actionable steps for the targeted implementation of a CGM discharge program—an area that is the sole focus of our work.

The Endocrine Hospitalist teams at the three community hospitals we serve recognize the advantage of having dedicated inpatient diabetes educators and institutional resources to support the CGM-at-discharge program we describe. These resources enable us to conduct multiple training sessions as needed to ensure staff are well-prepared to support patients initiating CGM. However, we acknowledge that not all hospitals have similar infrastructure or personnel, and implementation of such a program may present significant logistical and operational challenges in resource-limited settings.

The authors also acknowledge that initiating CGM at hospital discharge presents several challenges that may hinder implementation, even when resources are available. For example, impaired patient cognition during recovery from acute illness may limit a patient’s ability to engage with new technology. ²¹ The authors acknowledge that CGM is not appropriate or feasible for all patients. In our practice, there have been instances where a patient was initially identified as a suitable candidate, but during the placement process, it became clear that CGM was not appropriate. In such cases, CGM initiation was appropriately aborted. In addition, CGM devices may not be compatible with certain imaging studies, such as MRI or CT scans; as a result, some sensors may need to be removed during hospitalization and subsequently replaced, creating further logistical and clinical hurdles. ³⁴ Although preliminary studies are promising, further research is needed to confirm the costs and long-term benefits of initiating CGM at the time of hospital discharge.^7,8,10,31,32

The estimated annual cost of diabetes in the United States is $412.9 billion, with health care expenditures for individuals with diabetes approximately 2.6 times higher than for those without the condition. ³⁵ Despite its higher upfront cost compared to traditional glucose testing, CGM is projected to save individuals with diabetes thousands of dollars annually by preventing hospitalizations due to diabetic ketoacidosis and hypoglycemia. In addition, CGM use has been associated with reductions in long-term complications—retinal by 17%, renal by 20%, and cardiovascular by 2.4%. ³⁶

In addition to indirect cost savings, direct cost savings can be estimated from CGM prevention of hypoglycemia in the hospitalized setting. Continuous glucose monitor has been shown in large multicenter research clinical trials to significantly decrease the rate of incidence of hypoglycemia and time spent below glucose levels of 70 mg/dL compared to traditional capillary glucose check. ³⁷ This study by Spanakis et al ³⁷ showed that patients using CGM experienced 47% fewer hypoglycemic episodes, and 63% fewer severe hypoglycemic episodes compared to capillary glucose check. Hypoglycemia is associated with associated with longer length of stay in hospitals and nearly a two-fold increase in hospital cost and length of stay in the case of severe hypoglycemia. ³⁸ A Retrospective study by Quilliam et al ³⁹ estimated that the cost of hypoglycemia $394 for outpatient hypoglycemia, $1387 for an ED visit and, $17 594 for inpatient admission. As preliminary studies have shown that starting CGM before discharge decreases hypoglycemia, direct cost savings proportional to the number of patients in whom CGM is started can be expected.^7-9 However, more studies with larger patient populations will be needed for better estimation of cost savings.

Although often intended to improve patient care, introducing CGM to patients nearing discharge raises ethical considerations. Recommending CGM at this vulnerable time may inadvertently promote a specific product and manufacturer, influencing both patient and provider preferences. While this bias is difficult to avoid once CGM is initiated, it’s important to recognize that CGM is now considered standard of care in outpatient diabetes management and, in some cases, may be more cost-effective than capillary BGM.^40-42 In contrast to interventions that reduce individual costs while increasing systemic health care spending, CGM may actually lower overall health care costs by reducing diabetes-related hospitalizations and long-term complications. Given these potential benefits, the authors believe that initiating CGM in patients nearing discharge remains a worthwhile consideration.^36-39,42

Conclusion

Initiating CGM at hospital discharge represents a critical yet underexplored opportunity to improve glycemic outcomes and enhance the transition of care for patients with diabetes. Although challenges remain—including therapeutic inertia, logistical barriers, and the current lack of formal inpatient FDA approval—early experiences suggest that CGM initiation at discharge is feasible, beneficial, and well-received by patients and providers alike. Our proposed protocol offers a structured, actionable approach to overcoming common obstacles and laying the groundwork for broader adoption of CGM in the hospital-to-home transition. As CGM technology continues to evolve and outpatient use expands, systematic efforts to initiate CGM at discharge may help close gaps in diabetes management, reduce readmissions, and ultimately improve long-term outcomes. Future studies are needed to validate the cost-effectiveness and clinical impact of CGM initiation at discharge on a larger scale.