Greg Norman, PhD, is Director of Health Economics and Outcomes Research at Dexcom where he leads the generation of real-world evidence to support the clinical and health economic impact of continuous glucose monitors for people with diabetes. Katia Hannah, PhD, MPH, is the Senior Health Economics and Outcomes Research Specialist within Dexcom’s Global Access team. She is responsible for data management and the execution of real-world evidence studies.
The Health Economics and Outcomes Research team at Dexcom has been using the IHD Analytics platform for over a year to conduct a number of ad hoc analyses and sophisticated real-world studies. Three members of the team – Greg Norman, Katia Hannah, and Poorva Nemlekar – presented two retrospective real-world studies at the annual IHD User Conference on April 13th, 2022.
In this two-part blog series, we’ll hear from Greg, Katia, and Poorva on insights and learnings from their use of the IHD Analytics platform and get an in-depth look at two retrospective real-world studies that they conducted. In this first part, Greg and Katia introduce us to their team’s work and discuss one study titled Reduction in Diabetes-Related Hospitalization Rates After Real-Time Continuous Glucose Monitor (RTCGM) Initiation.
Q: Greg, as Director of the HEOR team at Dexcom, why is your diabetes research so important?
Greg: Diabetes is a highly prevalent chronic disease that affects more than 537 million people worldwide.1 The disease is also a major driver of mortality worldwide, with approximately 6.7 million adults aged 20-79 estimated to have died from the disease or its complications last year. Moreover, poorly controlled diabetes may cause serious acute and chronic complications that negatively impact patient quality of life and increase healthcare resource utilization and costs.
Our team is focused on studying real-time continuous glucose monitoring for diabetes patients to generate evidence to support the value of Dexcom’s real-time CGM from the patient, payer, and health care system perspectives.
Q: Can you give a little background on real-time continuous glucose monitoring?
Greg: Real-time continuous glucose monitors, or CGMs, measure the interstitial glucose level. So, they aren’t actually measuring blood glucose, but instead estimating blood glucose levels based on the glucose in the interstitial tissue. They display numerical and graphical information about current blood glucose levels and the direction and the velocity of change in glucose levels. The Dexcom G6 is our current model. The components of the system include a sensor worn on the abdomen, and a wireless transmitter attached to the sensor that sends data every 5 minutes to a display device which could be a phone, a smartwatch, or a dedicated receiver.
The American Diabetes Association recommends real-time continuous glucose monitoring for Type 1 diabetes (TD1) patients on intensive insulin therapy and Type 2 diabetes (T2D) patients on less intensive insulin therapy. However, there are currently no clinical guidelines for people with T2D on no insulin therapy, so that is definitely an area of research that we are trying to explore in our work.
Q: What value does real-world data analytics hold for CGMs?
Greg: Dexcom has conducted a number of clinical trials for CGMs which have shown good reductions in A1C, hypoglycemia, and what we call “time in range,” which is the sweet spot of where you want your glucose to be. However, these trials take time and money, so we want to compliment this clinical trial work with our real-world evidence studies.
To help with that, we started using the IHD Analytics platform about a year ago. My colleagues, Katia and Poorva, presented two of our CGM studies at the IHD User Conference in April. One study focused on healthcare utilization for patients with T1D and T2D on insulin using personal real-time CGMs. Another study focused on professional CGM use by patients with T2D not on insulin. This second study is particularly interesting because it is exploratory work for us and conducting the analysis helps us to understand the value of CGMs in this population where there is sparse clinical trial evidence.
Q: Why is glucose monitoring important for diabetes patients?
Katia: Diabetes disease management requires a multifaceted approach aimed at achieving optimal levels of glycemic control. The implications of inadequate glycemic control increase the risk of diabetes-related hospitalizations. Glucose monitoring is one of the most important components of optimizing glycemic control and CGMs are a useful tool for improving glycemic control by providing real-time glucose levels including trends and patterns.
Q: Can you give a little background on the study you presented?
Katia: “Reduction in Diabetes-Related Hospitalization Rates After Real-Time Continuous Glucose Monitor (RTCGM) Initiation” is a retrospective, observational analysis evaluating diabetes-related hospitalizations before and after CGM initiation. The study was specifically focused on analyzing the healthcare resource utilization of diabetes. We know that patients with diabetes accrue 2.3 times more in medical expenses than those without diabetes and in fact, diabetes-related hospitalizations account for 30% of medical expenses.2
Because of this, we are very interested in gaining insights about the disease that can help lower healthcare resource utilization. Studies have shown that inadequate glycemic control increases the risk of hospitalizations (and therefore resource utilization) for diabetes patients. For this study, we were interested in finding out how CGM use might be impacting diabetes-related hospitalizations and emergency room visits.
Q: What did your results show?
Katia: After real-time CGM initiation, there was a decrease in the number of emergency room visits in both Type 1 and Type 2 patients by 29% and 15% respectively. However, these decreases were not statistically significant.
There were statistically significant decreases in inpatient visits in both Type 1 and Type 2 patients by 54% and 48% respectively after device initiation. We also observed statistically significant reductions in the average hospital stay days after device initiation. We saw a reduction in Type 1 patients by .39 days and Type 2 patients by .88 days. Among patients with Type 2 who had a hospital stay after initiation of CGM, we saw a decrease of 1.22 days in the average length of stay.
Q: Why are these results significant?
Katia: This study helped us show that the Dexcom G6 CGM is associated with reduced diabetes-related hospitalizations after device initiation across both the Type 1 and Type 2 intensive insulin therapy patient populations. Studies like this one help support our clinical trial results by proving the value of the device in a real-world setting.
A decrease in healthcare resource utilization like we observed in this study also suggests a decrease in healthcare utilization related costs. Next, it would be interesting for our team to study the economic effect of CGM devices.
Q: How did Dexcom use these insights and what was the broader impact of this study?
Greg: You know, it's interesting how for FDA approval of a device like a CGM, it only requires accuracy and evidence of accuracy. But for payors, both Medicare and commercial in the US, they want to see evidence of efficacy and cost effectiveness. So that's where we can use this information. They want to see real-world evidence, and these studies conducted in IHD really allowed us to provide that. We were able to generate evidence for potential cost effectiveness and cost reductions for healthcare utilization, among other things.
Q: Do you find that you're able to run studies faster with IHD vs. other traditional statistical software like SAS or R?
Katia: Absolutely. We were able to easily get our code lists established, create cohorts, and understand the workings of IHD and how the data is mapped into it. Our work is much more streamlined for analyzing data, especially because we're able to copy projects instead of starting from scratch. In SAS, we would have to take our code and copy it over and make lots of changes, causing a huge bottleneck in the research process. It’s just much more straightforward and streamlined with IHD.
Q: You’re relatively new to the IHD community but you are already presenting scientific research, which is impressive. Can you describe how the onboarding experience was and how you integrated it in your organization so quickly?
Katia: I joined the Dexcom team about a year ago and hit the ground running with the IHD certification and onboarding process. Poorva and I went through that quickly. Then it was all about code lists and understanding what measures we needed for our study goals. Working with support has been wonderful. Coming from being proficient in SAS and other platforms, support and onboarding helped me bridge the gap between writing all of that code and managing the data myself to the seamless experience of IHD. There’s a learning curve, of course, but with the IHD support team, we're able to do a lot of troubleshooting and figuring out what we needed to do to get up and running to do these kinds of analyses.
1. Centers for Disease Control and Prevention. https://www.cdc.gov/diabetes/pdfs/data/statistics/national-diabetes-statistics-report.pdf. Published 2017. Accessed January 28, 2020.
2. American Diabetes Association. Diabetes Care. 41:917, 2018
In part two of this series which will be published next week, Greg Norman and Poorva Nemlekar discuss an exploratory study that their team conducted in IHD.
Contact us to learn how Panalgo's IHD Analytics platform can help your RWE teams scale their analytics capabilities and deliver richer insights, faster.