Transitioning Recently Admitted Patients With Non-Insulin and Insulin-treated Type 2 Diabetes to Innovative Glucose Monitoring

1. Background and Rationale/Literature Review :

Suboptimal glycemic control after hospital discharge is common in patients with type 2 diabetes mellitus (T2DM). Continuous glucose monitors (CGMs) may improve time-in-range, reduce hypoglycemia, and inform therapy titrations compared with intermittent point-of-care (POC) capillary glucose checks. However, evidence specific to initiation at the transition from inpatient to outpatient care remains limited. This trial will evaluate whether CGM initiation prior to discharge improves glycemic control at 12 weeks versus usual care with POC monitoring.

Diabetes mellitus affects over 537 million people worldwide and substantially increases the risk of emergency department visits and hospital admissions. In the United States, individuals with diabetes account for over 17 million emergency department visits and 8 million hospital admissions annually. The hospital readmission rate for patients with diabetes ranges between 14% and 20%, nearly twice that of patients without diabetes. The 30-day all-cause readmission rate specifically has been reported at 10.8% in commercially insured and Medicare Advantage populations, with rates as high as 20-22% in certain cohorts. These readmissions carry significant clinical and financial implications, with 30% of hospitalized patients with diabetes requiring two or more days of hospital stays, accounting for over 50% of hospital costs for diabetes.

Notably, severe dysglycemic events-both hypoglycemia and hyperglycemia-account for 2.5% of readmissions, with hypoglycemia representing 61% of these dysglycemia-related readmissions.

Challenges in the Transition from Hospital to Home:

The transition from acute care to home presents substantial risks for all patients with diabetes. Multiple barriers to optimal glycemic control exist in the post-discharge period, including medication access and affordability, inadequate diabetes self-management education, lack of timely follow-up appointments, and clinical inertia in adjusting therapy. Hospitalized patients who have newly diagnosed or poorly controlled type 2 diabetes require initiation or intensification of their outpatient diabetes regimen, yet the cost of insulin and other diabetes medications can be prohibitive. Furthermore, the mismatch between caloric intake and insulin coverage, common during hospitalization, often persists after discharge.

Structured discharge planning is essential but frequently inadequate. The Agency for Healthcare Research and Quality recommends that discharge plans include, at minimum, medication reconciliation, structured discharge communication, and patient education.

Evidence for Continuous Glucose Monitoring in Type 2 Diabetes:

Continuous glucose monitoring (CGM) has emerged as a valuable tool for diabetes management, providing comprehensive glycemic assessment and enhanced prevention of hypoglycemia. Multiple meta-analyses demonstrate that CGM use in type 2 diabetes leads to modest but consistent HbA1c reductions of approximately 0.3-0.35% over 10-34 weeks compared to self-monitoring of blood glucose, with both real-time CGM and intermittently scanned CGM showing similar efficacy.

Beyond glycemic control, real-world observational data suggest that CGM may reduce acute complications and hospitalizations. A Veterans Affairs system analysis of 15,706 patients with type 2 diabetes on insulin showed that CGM initiation was associated with reduced risk of hyperglycemia-related hospitalizations (HR 0.87) and all-cause hospitalization (HR 0.89) over 12 months. A separate retrospective analysis of 74,679 adults with type 2 diabetes demonstrated sustained reductions in all-cause hospitalizations (-10.1% to -22.6%) and acute diabetes-related hospitalizations (-31.0% to -52.7%) at 6-12 months after transitioning from blood glucose monitoring to CGM.

Preliminary Evidence for CGM at Hospital Discharge:

Despite the growing evidence for CGM in outpatient settings, data specifically examining CGM initiation prior to hospital discharge remain limited. The American Diabetes Association guidelines note that recent studies have shown CGM may prevent emergency department visits and hospital admissions in people with type 1 and type 2 diabetes, suggesting it may be beneficial to initiate CGM just prior to discharge to facilitate follow-up and possibly prevent acute diabetes-related complications and readmission.

One pilot randomized trial provides preliminary evidence for this approach. Umpierrez et al. conducted a study of 100 insulin-treated patients with type 2 diabetes, comparing FreeStyle Libre 2 CGM versus capillary point-of-care glucose testing for up to 12 weeks after hospital discharge. \[19\] (Umpierrez et al, 2025) The CGM group showed trends toward improved glycemic control, including increased time in range (70-180 mg/dL), reduced time above range (>180 and >250 mg/dL), decreased hypoglycemia, and lower insulin requirements compared to standard monitoring. However, no differences were observed in emergency room visits or hospital readmissions between groups. This pilot study suggests potential glycemic benefits but was not powered to detect differences in clinical outcomes such as readmissions.

Knowledge Gaps and Study Rationale:

Substantial gaps remain in our understanding of CGM use as a discharge intervention for patients with type 2 diabetes. First, while the pilot trial demonstrated trends toward improved glycemic metrics, it was underpowered to detect clinically meaningful differences in readmissions, emergency department visits, or other patient-centered outcomes. Second, the patients that were studied were strictly insulin-dependent T2DM patients.

Given the high rates of readmission among patients with diabetes, the established association between dysglycemia and adverse outcomes, the challenges inherent in the transition from hospital to home, and the preliminary evidence suggesting CGM may improve glycemic control and reduce hospitalizations, a definitive randomized controlled trial is warranted. Such a trial would address whether CGM initiation prior to hospital discharge can reduce readmissions and improve glycemic control compared to standard care in patients with type 2 diabetes.
2. Objectives:

Primary Objective: To compare HbA1c at 12 weeks post-discharge between CGM and POC groups (intention-to-treat).

Secondary Objectives: To compare (a) diabetes-related hospitalizations within 12 weeks; (b) incidence of hypoglycemic events within 12 weeks as assessed during PCP or endocrinology clinic encounters. (c) Readmission for patients to the hospital for diabetes related complications

Other information to record: Adherence to CGM (process metric), clinic follow up rates
3. Trial Design:

Prospective, single-center, parallel-group, randomized controlled trial with 1:1 allocation to CGM vs POC. The trial will use an intention-to-treat (ITT) analysis framework.
4. Participants:

Inclusion Criteria-

Age ≥ 18 years Established or new diagnosis of type 2 diabetes mellitus (T2DM) HbA1C >/= 9% On insulin therapy and/or oral hypoglycemic agents at time of index hospitalization and/or Anticipated discharge to home within 48-72 hours Ability to provide informed consent and attend 12-week follow-up

Exclusion Criteria- Active psychiatric illness that interferes with consent or study adherence Cognitive disturbance/dementia impairing device use or self-care Housing instability (e.g., homelessness) Skilled nursing facility (SNF) admission or planned discharge to SNF Contraindication to CGM placement (e.g., severe skin allergy) Pre-existing CGM use Pregnancy ESRD Steroid use (Prednisone equivalent 10 mg/day or more) HgbA1C < 9%
5. Interventions:

CGM Arm: Initiate CGM prior to discharge. Provide standardized device training, written instructions, and 24/7 device support contact. Share data with outpatient clinicians per institutional pathway. No mandated therapeutic algorithm beyond standard-of-care clinic management. Diabetes Educator + Sweet Transitions to help with clinic follow up

POC Arm: Usual care with capillary glucose checks and standard discharge education. Frequency of POC monitoring and therapy adjustments per treating clinician. Diabetes Educator + Sweet Transitions to assist with clinic follow up
6. Outcomes:

Primary Outcome: HbA1c at 12 weeks (±2 weeks) post-discharge measured in a CLIA-certified laboratory.

Secondary Outcomes:
* Diabetes-related hospitalizations within 12 weeks (composite of ED visits leading to admission and inpatient stays with primary diabetes/metabolic indication).
* Hypoglycemic events within 12 weeks, as documented in PCP or endocrinology clinic notes or patient-reported and confirmed (glucose <70 mg/dL or requiring assistance).

Exploratory: time-in-range from CGM where available
7. Sample Size and Power:

The trial is designed to detect a clinically meaningful improvement in HbA1c at 12 weeks post-discharge, the primary endpoint. For planning purposes, we assumed a common standard deviation of 1.2% and evaluated a range of plausible between-group differences in mean HbA1c. Under a two-sided type I error rate of 0.05 and 80% power, approximately 92 participants per group are required to detect a 0.5 percentage point difference. To add a 15% attrition, the total number of participants will have to be 218 (109 in each arm).

Secondary clinical outcomes, such as diabetes-related hospitalization and hypoglycemia, will be evaluated to estimate effect sizes and generate preliminary evidence for future larger studies.
8. Randomization and Allocation Concealment:

Participants will be randomized 1:1 using a computer-generated sequence with variable block sizes. Allocation will be concealed using a secure electronic system accessible only to the study coordinator at the time of assignment.
9. Blinding:

Due to the nature of the intervention, participants and treating clinicians cannot be blinded. Outcome assessment for HbA1c and adjudication of hospitalizations/hypoglycemia will be performed by assessors blinded to group assignment.
10. Data Collection and Management:

Baseline data: demographics, diabetes duration, baseline HbA1c, current therapies, comorbidities. Follow-up data at \~2-4 and 12 weeks via clinic visits or telehealth. Data will be captured in REDCap with role-based access. Device data (CGM) will be imported when available; however, ITT analyses will not require CGM data completeness.
11. Statistical Analysis Plan:

All analyses will follow the intention-to-treat (ITT) principle. The primary endpoint is HbA1c at 12 weeks post-discharge and will be analyzed using linear regression with adjustment for baseline HbA1c (continuous) and baseline therapy (insulin vs non-insulin). The treatment effect will be reported as an adjusted mean difference with 95% confidence intervals using a two-sided α = 0.05. Model assumptions will be assessed, and robust methods will be applied if needed. Secondary outcomes, including hospitalizations/readmissions and hypoglycemic events, will be analyzed using logistic regression for binary outcomes and negative binomial or Poisson regression for count data; time-to-event methods may be used for hospitalization outcomes when appropriate. These analyses will be considered exploratory. Missing data will be handled using multiple imputation, with sensitivity analyses including complete-case analyses and conservative missing-data assumptions. A per-protocol analysis will be conducted to assess consistency with the primary ITT results. Exploratory subgroup analyses will evaluate treatment effect heterogeneity by baseline therapy and baseline HbA1c category using interaction terms. No interim efficacy analyses are planned. All analyses will be conducted using R 4.4.1.
12. Safety Monitoring and Adverse Events:

Adverse events related to CGM (e.g., skin reactions, insertion-site issues) will be recorded. Stopping rules for unexpected serious device-related events will be predefined.
13. Ethics and Regulatory Considerations:

The study will be conducted in accordance with the Declaration of Helsinki and Good Clinical Practice. Institutional Review Board approval has been obtained prior to initiation. Written informed consent will be obtained from all participants. The trial will be registered on ClinicalTrials.gov prior to first enrollment.
14. Operational Considerations:

Training: Diabetes educators will receive CGM training. Supplies and device logistics coordinated with device vendor. Standardized discharge instructions and follow-up scheduling workflows will be implemented.
15. Timeline:

Start-up (IRB, registration, training): 2-3 months; Enrollment: 9-12 months; Follow-up: 3 months after last enrollment; Data analysis and manuscript: 2-3 months.

• Continuous glucose monitor (Device), also known as CGM: Participants in the intervention arm will be educated and discharged from their hospitalization with a continuous glucose monitor and followed up in clinic.

Primary outcomes

• Glycemic control as measured by Hemoglobin A1c (HbA1c) [From enrollment to end of monitoring at 12 weeks]: A reduction in Hemoglobin A1C is being studied. With the current sample size, we expect a reduction in Hemoglobin A1C level by 0.5 as compared to the POC glucose monitoring arm to be significant after 12 weeks.

Secondary outcomes

• Hospital readmissions over 12 weeks [12 weeks]: Any cause readmission to the hospital in the 12 weeks after enrollment will be measured
• Diabetes related readmissions at 12 weeks [12 weeks]: A diabetes related readmission to the hospital in the 12 weeks after enrollment will be measured
• Hypoglycemic events as measured by low glucose reading and symptoms as described by patients at follow up [12 weeks]: Hypoglycemic events as measured by low glucose reading and symptoms as described by patients at follow up within 12 weeks

Sex: All, ages: 18 Years to any

Inclusion Criteria:

* Age ≥ 18 years
* Established or new diagnosis of type 2 diabetes mellitus (T2DM)
* HbA1C >/= 9%
* On insulin therapy and/or oral hypoglycemic agents at time of index hospitalization
* Anticipated discharge to home within 48-72 hours
* Ability to provide informed consent and attend 12-week follow-up

Exclusion Criteria:

* Active psychiatric illness that interferes with consent or study adherence
* Cognitive disturbance/dementia impairing device use or self-care
* Housing instability
* Skilled nursing facility (SNF) admission or planned discharge to SNF
* Contraindication to CGM placement (e.g., severe skin allergy)
* Pre-existing established CGM use
* Pregnancy
* ESRD
* Steroid use (Prednisone equivalent 10 mg/day or more)
* HgbA1C < 9%
* Hemoglobinopathy (where HgbA1C is not reliable)

• Umpierrez GE, Castro-Revoredo I, Moazzami B, Nayberg I, Zabala Z, Galindo RJ, Vellanki P, Peng L, Klonoff DC. Randomized Study Comparing Continuous Glucose Monitoring and Capillary Glucose Testing in Patients With Type 2 Diabetes After Hospital Discharge. Endocr Pract. 2025 Mar;31(3):286-291. doi: 10.1016/j.eprac.2024.11.018. Epub 2024 Dec 17. PMID 39694328

• Mercedes Falciglia, MD, Study director, University of Cincinnati
• Arunab H Mehta, MD, MEd, Principal investigator, University of Cincinnati
• Houman Varghai, MD, Principal investigator, University of Cincinnati