Diabetes Mellitus
Conditions
Keywords
Implementation, HbA1c, Multi-component Implementation, Continuous glucose monitoring
Brief summary
Continuous glucose monitoring (CGM) is a technology that helps individuals with diabetes track their sugar levels in real-time, leading to more in-range blood sugars, fewer episodes of dangerously low blood sugar, and improved quality of life. Despite these benefits, CGM is not widely used in primary care settings, where most people receive their diabetes care. The investigators aim to make CGM more accessible and equitably prescribed in primary care practices. The study team will support primary care to increase CGM use with a program called SPARK-CGM (Supporting Primary Care Adoption, Resources, and Knowledge for CGM) across a large network of primary care clinics at Montefiore Medical Center. This program will provide primary care providers (PCPs) with education, tools, and support to incorporate CGM into their routine care for people with diabetes. Investigators plan to test SPARK-CGM to evaluate whether it increases CGM prescriptions who are eligible to receive this technology.
Detailed description
Continuous glucose monitoring (CGM) has been shown to reduce HbA1c levels, hypoglycemia, and improve quality of life, representing a powerful tool to improve population-level outcomes in diabetes. The American Diabetes Association (ADA) recommends CGM for all individuals with any type of diabetes on insulin, but its widespread adoption into primary care remains low. SPARK-CGM (Supporting Primary care Adoption, Resources, and Knowledge for CGM) is a hybrid effectiveness-implementation trial evaluating a multifaceted practice transformation package designed to increase new continuous glucose monitoring (CGM) prescriptions among adult patients with diabetes treated wtih insulin. This includes: (1) strengthening clinic capacity by streamlining CGM prescribing pathways; (2) optimizing pharmacy and insurance authorization workflows to reduce administrative burden; (3) providing prescriber-focused training on CGM workflows, data access, and interpretation to support effective clinical use; (4) training nursing and support staff to facilitate device initiation, patient education, and data management; and (5) delivering regular performance feedback on prescribing rates across clinics and providers. The effectiveness component will be evaluated using a stepped-wedge cluster randomized trial in which primary care clinic clusters sequentially transition from usual care to the SPARK-CGM implementation intervention over the study period until all clinics receive the intervention. The implementation component will include mixed-methods evaluation using surveys and semi-structured interviews with patients, providers and clinic staff to assess implementation outcomes. Based on pre-implementation clinic counts, the stepped-wedge trial will include 14 non-pilot primary care clinics. The investigators expect to include up to approximately 20,000 adult patients with insulin-treated diabetes who may contribute data to the effectiveness analysis (Aim 1) during the study period . For implementation evaluation and semi-structured interviews (Aim 2), investigators plan to recruit up to 40 adults with insulin-treated diabetes (10 per cluster) and 15-20 clinicians or clinic staff participants (4-5 per cluster) in order to reach thematic saturation.
Interventions
OTHEREnhanced CGM Interventions
The study intervention involves creating a streamlined workflow for CGM prescribing that does not restrict the treatment options available to patients or clinicians. SPARK-CGM implementation strategy includes three core practice transformations: (1) building clinic infrastructure by establishing CGM prescription workflows, and by training clinic staff to place CGM devices at the point of care, (2) provider training on accessing and using CGM data effectively in practice, and (3) regular feedback on prescription rates across the network.
OTHERStandard of Care
Usual care
Sponsors
Montefiore Medical Center
American Diabetes Association
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
NONE
Intervention model description
This study uses type II hybrid effectiveness-implementation design combining (1) stepped-wedge cluster randomized trial (SW-CRT) evaluating effectiveness of SPARK-CGM and (2) post-intervention mixed methods implementation evaluation including interviews and surveys with patients, providers and clinic staff. Participating clinics will be randomized into four mutually exclusive cluster waves, stratified by clinic size, defined by the number of patients with insulin-treated diabetes, and by federally qualified health center status to ensure balance across waves. Clinics will sequentially transition from the usual care (control) condition to the intervention phase at three-month intervals until all clusters receive the intervention.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
Clinic level: * All adult Montefiore primary care sites * Clinician and clinic staff will be eligible if they provide direct patient care or are involved in CGM prescribing, authorization, onboarding, or education at participating primary care clinics. Eligible clinicians include physicians, nurse practitioners, physician assistants, and clinicians in training. Eligible clinic staff may include nurses, medical assistants, and other relevant administrative staff Patient level: * Age 18 years or older * Receive primary care at participating sites * Diagnosis of any diabetes mellitus * Treated with insulin therapy
Exclusion criteria
Clinic level: \- Sites participating in pilot phase of CGM initiative Patient level: \- Existing CGM prescription within 24 months before the study start
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Time to first CGM prescription by primary care provider | Up to 18 months following initiation of intervention | Time to first CGM prescription initiated by a primary care provider (PCP) will be defined as the date from a patient's first eligible primary care encounter during the study period to the date of the first CGM prescription, up to 18 months following intervention initiation, as recorded in the electronic health record. All CGM orders are captured in the EHR. Results will be summarized by study arm using descriptive statistics and analyzed using Cox proportional hazards models. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| CGM Utilization | Up to 18 months following initiation of intervention | A dichotomous (binary) measure of CGM usage obtained from device platforms and/or EHR. CGM utilization will be summarized by study arm using descriptive statistics. |
| HbA1c | Up to 18 months following initiation of intervention | Longitudinal repeated HbA1c measures over time will be obtained from EHR. HbA1c results will be summarized by study arm using descriptive statistics. |
| Hospitalizations | Up to 18 months following initiation of intervention | The number/percentage of patients who were admitted for inpatient hospitalization will be assessed based on data available from EHR from the initiation of the intervention to the end of the study, up to 18 months. |
| Emergency Department (ED) visits | Up to 18 months following initiation of intervention | The number/percentage of patients who visited an ED will be assessed based on data available from EHR from the initiation of the intervention to the end of the study, up to 18 months. |
| CGM prescription rate by race/ethnicity and payor | Up to 18 months following initiation of intervention | The number/percentage of participants with CGM will be evaluated by race/ethnicity and payor based on data available from EHR from the initiation of the intervention to the end of the study, up to 18 months. |
| Longitudinal glucose time in range (TIR) | Up to 18 months following initiation of intervention | TIR (i.e., percent time spent in glucose range 70-180 mg/dL) will be collected longitudinally among patients using CGM in the post-implementation trial phase. Repeated TIR measures will be obtained from device platforms and/or the EHR at baseline and throughout follow-up. Average TIR levels over time will be analyzed to assess within-participant change over time. |
| Longitudinal glucose time below range (TBR) | Up to 18 months following initiation of intervention | TBR (i.e., percent time spent in glucose range \<70 mg/dL) will be collected longitudinally among patients using CGM in the post-implementation trial phase. Repeated TBR measures will be obtained from device platforms and/or the EHR at baseline and throughout the follow-up. Average TBR levels over time will be analyzed to assess within-participant change over time. |
| Longitudinal glucose time above range (TAR) | Up to 18 months following initiation of intervention | TAR (i.e., percent time spent in glucose range \>180 mg/dL) will be collected longitudinally among patients using CGM in the post-implementation trial phase. Repeated TAR measures will be obtained from device platforms and/or the EHR at baseline and throughout the follow-up. Average TAR levels over time will be analyzed to assess within-participant change over time. |
| Longitudinal glucose management indicator (GMI) | Up to 18 months following initiation of intervention | GMI measures will be collected longitudinally among patients using CGM in the post-implementation trial phase. Repeated GMI measures will be obtained from device platforms and/or the electronic health record at baseline and throughout the follow-up. Average GMI levels over time will be analyzed to assess within-participant change over time. |
Countries
United States
Contacts
CONTACTRosina Antwi
PRINCIPAL_INVESTIGATORJovan Milosavljevic, MD
Montefiore Medical Center
Outcome results
None listed