CGM as medicine: Technology benefits extend to type 2 diabetes
Continuous glucose monitoring, first approved by the FDA for professional use in 1999, has become standard of care for people with type 1 diabetes, with benefits for improved glucose management, health outcomes and quality of life.
For people with type 2 diabetes, few efficacy trials of real-time CGM have been conducted, particularly in the primary care setting where most of these patients are seen. Medicare covers CGM for patients with insulin-treated type 2 diabetes who meet certain criteria, but the technology is not yet considered standard of care.
“CGM for the type 1 population is important because there is so much variability in the blood glucose, but they are also a much smaller population overall,” Athena Philis-Tsimikas, MD, a specialist in endocrinology, diabetes and metabolism and corporate vice president of the Scripps Whittier Diabetes Institute at Scripps Health, told Endocrine Today. “Approximately 95% of people with diabetes have type 2. A good portion of that population also uses insulin, can also experience variability in blood glucose, and do not have good control. ... An opportunity for these people to see glucose numbers in real time, and then adjust medications and lifestyle, can be life-changing.”
Recent data show that improvements in glucose management, time spent in the recommended glucose range, and reductions in hypoglycemia also extend to people with type 2 diabetes, including those using long-acting basal insulin therapy. Anecdotal evidence suggests CGM can also yield large benefits even for people with type 2 diabetes not on insulin therapy.
“The value of CGM for anyone with diabetes is it allows you to see the patterns of your glucose levels throughout the day and see what happens after, for example, breakfast, lunch or dinner,” Roy W. Beck, MD, PhD, president and medical director of the Jaeb Center for Health Research Foundation Inc., told Endocrine Today. “We see with people with type 2 who do not perform frequent blood glucose monitoring and then start wearing a CGM, for many of them, it is an eye-opener. Often, they have no idea that after a meal, their glucose was going up to 300 mg/dL. All they went by, typically, would be a fasting glucose in the morning.”
Among people with type 2 diabetes, the best outcomes have been observed for those with the highest baseline HbA1c, James R. Gavin III, MD, PhD, clinical professor of medicine at Emory University School of Medicine and chief medical officer of Healing Our Village, told Endocrine Today.
“Those are difficult patients to get on a better trajectory, and CGM has really been helpful for them,” said Gavin, who is also a Healio | Endocrine Today Co-editor. “Even intermittent use of CGM in type 2 patients has been shown to improve HbA1c values. Additionally, CGM use has been reported to enhance treatment satisfaction. When patients feel better, they feel more in control of what is happening with their diabetes treatment. They have more meaningful engagement and less fear of hypoglycemia.”
New primary care data
In a randomized controlled trial, the MOBILE study group analyzed data from 175 adults with type 2 diabetes receiving diabetes care in one of 15 primary care centers in the U.S. from July 2018 to July 2020. Participants, who had a mean age of 57 years and a mean baseline HbA1c of 9.1%, were prescribed one or two daily injections of long- or intermediate-acting basal insulin without prandial insulin, with or without noninsulin glucose-lowering medications. Participants were randomly assigned 2:1 real-time CGM (Dexcom G6; n = 116) or a traditional blood glucose meter (n = 59). Primary outcome was HbA1c at 8 months; secondary outcomes were CGM-measured time spent in the target glucose range (70-180 mg/dL), time spent in hyperglycemia (> 250 mg/dL) and mean glucose level at 8 months.
At 8 months, mean HbA1c levels decreased from 9.1% to 8% in the CGM group and from 9% to 8.4% in the blood glucose monitoring group, for an adjusted difference of 0.4 percentage points (95% CI, 0.8 to 0.1). Compared with participants assigned a glucose meter, those using CGM experienced a significantly better mean percentage of CGM-measured time in range (mean, 59% vs. 43%; adjusted difference, 15 percentage points; 95% CI, 8-23).
Results were similar for the mean percentage of time spent in hyperglycemia (mean, 11% vs. 27%; adjusted difference, –16 percentage points; 95% CI, –21 to –11).
“The interesting thing is there was no change in insulin dosing between the groups,” Tsimikas said. “That means there really were other things that were influencing the improvements in blood glucose. This is such a great tool to help the patients and the primary care physicians manage the disease.”
Richard M. Bergenstal, MD, executive director of the International Diabetes Center Park Nicollet and Health Partners in Minneapolis and an investigator in the MOBILE study, compared the effect of CGM to a diabetes drug.
“If we handed them CGM, and they had a 0.4% HbA1c reduction, that is about what a new diabetes drug does,” Bergenstal told Endocrine Today. “Using it as a lifestyle tool, it is like a drug. You are using CGM to make your own drugs work to their best ability. As providers, we have been dancing around insulin — every article says the biggest barrier with insulin therapy is hypoglycemia. ... Now we can say, ‘Where are they low, and how can we adjust?’”
In another analysis, also published in JAMA in June, researchers assessed data from 41,753 adults with insulin-treated diabetes (5,673 with type 1; 36,080 with type 2) from 2014 to 2019. Within the cohort, 3,806 adults initiated real-time CGM (91% with type 1 diabetes) and 37,947 were non-initiators (6% with type 1 diabetes).
Mean HbA1c declined among real-time CGM initiators from 8.17% to 7.76% and from 8.28% to 8.19% among non-initiators, for an adjusted difference-in-differences estimate of 0.4% (95% CI, –0.48 to –0.32). Hypoglycemia rates declined among real-time CGM initiators from 5.1% to 3% and increased among non-initiators from 1.9% to 2.3%, for a difference-in-differences estimate of 2.7% (95% CI, –4.4 to –1.1).
In a study presented at the American Diabetes Association Scientific Sessions in June, researchers analyzed data from the French nationwide reimbursement claims database on 33,203 residents with type 1 diabetes and 40,955 with type 2 diabetes who used a flash glucose monitoring system between August and December 2017. Data included the number of daily glucose test strips individuals used before the study period and reimbursement for ICD-10 codes for diabetic ketoacidosis from 1 year before to 1 year after first use of the FreeStyle Libre.
Overall, hospitalization rates for DKA declined by 47% for users with type 2 diabetes. Reductions were most dramatic for people who had rarely monitored their blood glucose levels previously — those who had used no test strips — at 51% for those with type 2 diabetes, and for those with type 2 diabetes who had used more than five daily test strips at 52%. Additionally, decreases in DKA hospitalization rates were observed among those using multiple daily injections and those using an insulin pump.
“These data have real implications for both the higher risk populations and primary care providers,” Gavin said. “What this says to the primary care providers is, this is a tool that you want to leverage more often even though there may be hurdles.”
Simplifying CGM adoption, use
Data from CGM reports can offer clinicians insight into a user’s day-to-day glucose profile to allow for more tailored treatment. Yet the main barrier to analyzing CGM reports in the clinic is the ability to easily obtain them, according to Amy B. Criego, MD, MS, department chair in pediatric endocrinology at Park Nicollet Clinic and a medical director at the International Diabetes Center in Minneapolis.
“I am not a primary care provider, but the pace of getting CGM data and the ability to integrate it into the electronic health record is critical for their efficiency every day,” Criego told Endocrine Today.
To improve access, the International Diabetes Center partnered with Abbott on a pilot initiative beginning in spring 2020 to make FreeStyle Libre CGM data available at the point of care in select clinics. The goal, Criego said, was for clinicians to place an order in the EHR for a patient with diabetes who agreed to share their CGM data. In real time, data would then be transferred from Abbott’s cloud-based system, LibreView, via an EHR platform, allowing providers to automatically view CGM data in patients’ lab results and diabetes flow sheet. Data metrics would include time in the recommended glucose range and visual alerts for out-of-range values.
The ambulatory glucose profile is also integrated into the EHR in a PDF format, allowing clinicians to track the patient’s glucose trends over time and adjust treatment regimens as needed, Criego said.
Currently, orders within the HealthPartners system clinics are available to directly integrate CGM data into the EHR once connection is established, Criego said. Data are obtained “within seconds” after the order is placed.
Tsimikas, whose institution adopted a remote patient monitoring program for CGM, said clinicians must be creative and consider new ways of facilitating engagement with this technology.
“Perhaps the primary care physician is not completely comfortable with the technology, but what if we had a virtual clinic with a specialist who knows how to do this?” Tsimikas said. “The provider can connect with them for one virtual visit. That is one way to integrate some virtual care to support both the primary care physician and the patient. And this tool can be accessed from anywhere. A formal study to evaluate that would be fascinating.”
Reaching underrepresented groups
Research conducted in the type 1 population, particularly among children, suggest race disparities persist in CGM uptake and continued use. Data published in Diabetes Care in November showed white children are twice as likely to initiate and four times as likely to persist with CGM at 1 year compared with Black and Hispanic children, regardless of insurance type. Similar data assessing CGM use among type 2 patients by race is scarce, though patterns are likely similar, Gavin said.
In the MOBILE study, 54% of participants were from underrepresented groups, demonstrating that device access matters for glucose control, Tsimikas said.
“[MOBILE] shows that when you provide that access, along with the education, support and recommendations around it, people from underrepresented groups did just as well,” Tsimikas said. “We should provide this type of tool and access to all populations.”
Recent changes in Medicare rules may help widen access for people with type 2 diabetes. In September 2020, CMS approved reimbursement for the FreeStyle Libre 2 (Abbott), expanding access of the integrated CGM system, or iCGM, to Medicare beneficiaries with diabetes. In July, a CMS policy change eliminated the four times per day testing requirement to qualify for a CGM device. The updated criteria went into effect July 18.
“For all diabetes technologies, we need to continue to score victories like the Medicare one, which lower barriers for access,” Gavin said. “Requirements for device access cannot be draconian.”
Gavin said diabetes stakeholders may also need to incorporate CGM into advocacy strategies with new approaches.
“For example, if patients are enrolled or engaged in diabetes self-management education and a clinician can validate that the person is doing the things they need to do but is still not improving, maybe under those circumstances, you may need to authorize 1-year access to CGM,” Gavin said. “Then, assess the person’s improvements. We must become more imaginative and supportive in the way we utilize technology, even if done on a trial basis to reach people in the most marginalized and vulnerable positions. ... We must be as creative in overcoming those gaps as we were intentional in generating those gaps.”
Cost, coverage questions
Researchers agree that payors want outcomes data that will justify the use of CGM by people with type 2 diabetes, lowering risks for both microvascular and macrovascular complications of diabetes.
“In the broader type 2 population, what payors are interested in is what is the bang for their buck?” Beck said. “If they spend money on CGM, will there be fewer ED visits? Fewer hospitalizations? Less use of insulin? Less medications? Now that we showed a benefit, we need to show how this could reduce health care costs in other ways.”
In one of the first studies to assess risk for diabetes outcomes associated with time in range, published in Diabetes Care in 2018, researchers assessed time in range among hospitalized patients with type 2 diabetes using masked CGM. Those in the highest quartile, defined as time in range of at least 86%, had the lowest incidence of mild, moderate or vision-threatening diabetic retinopathy.
In a study published in Diabetes Care in 2019, researchers evaluated associations of time in range with development or progression of retinopathy and microalbuminuria using the Diabetes Control and Complications Trial (DCCT) data set — calculating time in range from finger sticks — to validate use of the metric as an outcome measure. Researchers found time in range was “strongly associated” with the risk for microvascular complications and should be an acceptable endpoint for clinical trials.
In a retrospective, real-world study published in the Journal of the Endocrine Society in March, Bergenstal and colleagues analyzed IBM MarketScan Commercial Claims and Medicare Supplemental databases to assess the impact of flash CGM on diabetes-related events and hospitalizations among 2,463 people with type 2 diabetes prescribed short- or rapid-acting insulin therapy. Outcomes were changes in acute diabetes-related events and all-cause inpatient hospitalizations, occurring during the first 6 months after acquiring the flash CGM system compared with event rates during the 6 months prior to CGM initiation.
At 6 months, acute diabetes-related event rates decreased from 0.18 to 0.072 events per patient-year (HR = 0.39; 95% CI, 0.3-0.51). All-cause inpatient hospitalizations rates decreased from 0.42 to 0.283 events per patient-year (HR = 0.68; 95% CI, 0.59-0.78).
Bergenstal said the findings provide support for the use of flash CGM in type 2 diabetes patients treated with short- or rapid-acting insulin therapy to improve clinical outcomes and potentially reduce costs.
“Acute events, meaning emergency calls or ED visits, decreased by about 40% and hospital admissions decreased 32%,” Bergenstal said. “It is powerful data.”
Additionally, an analysis from the IQVIA Core Diabetes Model showed a potential cost savings of $6.7 billion to $9.7 billion over 10 years if time in range improved from the current average of 58% to 70% or 80% for people with type 1 and type 2 diabetes, when also factoring in the reduction in hypoglycemia.
“The big-time costs come in when someone has [a myocardial infarction] or stroke, or develops renal failure,” Beck said. “The problem is those are down-the-road costs, years away. An insurer’s perspective is often 3 to 5 years. It is a challenge to demonstrate CGM is a benefit to them when the benefit is unequivocally long term.”
Bergenstal said leveraging data from large CV outcomes trials might offer an answer.
“We have been clever, using data, for example, derived from DCCT. We used the best data we have to show that it looks like CGM falls in line with HbA1c,” Bergenstal said. “But the studies are short. Payors want 7-year data with CGM in type 2. The only way I see that happening is if [drug] companies add CGM data to their CV outcomes trials. If conducting a study with 10,000 participants and one drug for 6 years, assessing all of these adjudicated outcomes, and you threw in CGM twice a year, you would have your answer.”
In the meantime, Tsimikas said clinicians must do all they can to initiate CGM for any person with diabetes who is open to using it.
“CGM should not be restricted only to those who are already in the most advanced throes of the disease,” Tsimikas said. “People with type 2 diabetes need more access to these tools. If you can get it to them early, you might prevent complications.”
“Not everyone needs CGM, but many more people need CGM than are getting it,” Gavin said. “That is the gap we must fill. Most patients will derive meaningful clinical benefits. The more you know, the more comfortable you are in clinical decision-making.”
- Bergenstal RM, et al. J Endocr Soc. 2021;doi:10.1210/jendso/bvab013.
- Karter AJ, et al. JAMA. 2021;doi:10.1001/jama.2021.6530.
- Martens T, et al. JAMA. 2021;doi:10.1001/jama.2021.7444.
- Roussel R. 68-OR. Presented at: American Diabetes Association Scientific Sessions; June 12-16, 2020 (virtual meeting).
- Martens T, et al. JAMA. 2021;doi:10.1001/jama.2021.7444.
- For more information:
- Roy W. Beck, MD, PhD, can be reached at email@example.com.
- Richard Bergenstal, MD, can be reached at firstname.lastname@example.org.
- Amy B. Criego, MD, MS, can be reached at email@example.com.
- James R. Gavin III, MD, PhD, can be reached at firstname.lastname@example.org.
- Athena Philis-Tsimikas, MD, can be reached at email@example.com; Twitter: @athenatsimikas.