Prior to initiation of Libre use, about a quarter of each group was using 0 fingerstick test strips per day, about 19% of the type 1 diabetes group and 28% of the type 2 diabetes group were using 1-3 strips per day, and about half of both groups were using 4 or more strips per day.
Compared with the year prior to the date of first reimbursement for the Libre, hospitalization rates for DKA during the first year of Libre use fell by 52% in the type 1 diabetes group, from 5.46 to 2.59 per 100 patient-years, and by 47% in the type 2 diabetes group, from 1.70 to 0.90 per 100 patient-years.
The impact of Libre on DKA hospitalizations was most dramatic among those not using any test strips prior to Libre use, with a 60% reduction for the type 1 diabetes group (8.31 to 3.31 per 100 patient-years) and a 51% reduction in the type 2 diabetes group (2.51 to 1.23 per 100 patient-years).
But interestingly, the next-biggest impact was among those who had been using more than 5 test strips per day, with drops of 59% among those with type 1 diabetes (5.55 to 2.26 per 100 patient-years) and 52% in the type 2 diabetes group (1.88 to 0.90 per 100 patient-years).
This finding is important for the United States, Argento said, because some insurers including Medicare require that the patient performs at least 4 fingerstick glucose measurements per day to qualify for reimbursement for the Libre or any CGM system.
"I think that speaks to the importance of not requiring that patients first show they're frequently doing self-blood glucose monitoring before they can get these devices," he observed.
The large benefit in the high strip use group is interesting too, Argento said. "It's a different group of people. They're more engaged in their care...This U-shaped curve they showed is fascinating."
Reductions in DKA hospitalizations were also similar between patients using insulin pumps and those using multiple daily injections of insulin, Roussel reported.
"It is plausible that use of the FreeStyle Libre system allowed people to detect and limit persistent hyperglycemia, and subsequently ketoacidosis," Roussel said.
"This analysis has significant implications for patient-centered clinical care in diabetes and also for long-term health economic outcomes in the treatment of diabetes at a national level."
All-Cause Hospitalizations Drop 30% With Libre in Type 2 Diabetes
Richard M. Bergenstal, MD, executive director of the International Diabetes Center at Park Nicollet, Minneapolis, Minnesota, presented the US results, obtained from the IBM Watson Health MarketScan, a database of commercial and Medicare supplemental insurance claims for over 30 million Americans.
The study population included 2463 patients with type 2 diabetes using basal-bolus daily insulin injections but who had not previously used Libre or any other CGM, and for whom data were available 6 months prior to and after Libre initiation.
Compared with 6 months prior to Libre use, the number of acute diabetes-related events — including hyperglycemia, hypoglycemia, DKA, hypoglycemic coma, and hyperosmolarity — in the subsequent 6 months dropped by 60%, from 0.180 to 0.072 events per patient-year (P < .001).
Similarly significant reductions were seen between males and females, and among those aged ≥ 50 years or < 50 years.
All-cause hospitalizations also significantly dropped by 33% (P < 0.001), from 0.420 to 0.283 events per patient-year. Among diagnostic codes for the hospitalizations, circulatory system causes remained number one during both time periods, with little change from pre-Libre to during Libre use.
However, "endocrine, nutritional, and metabolism system" codes dropped from the second position pre-Libre (6.4 events/100 patient-years) down to the fifth position (2.6 events/100 patient-years).
And, Bergenstal noted, other major diagnostic categories that also dropped included respiratory (3.5 to 2.1 events/100 patient-years), kidney and urinary tract (3.3 to 1.7 events/100 patient-years), and hepatobiliary system and pancreas (2.4 to 1.4 events/100 patient-years).
"We're seeing a resurgence of certain types of complications, but all of these were reduced in the 6 months after Libre," Bergenstal pointed out.
And, pertinent to the current COVID-19 situation, "infectious and parasitic disease and disorders" dropped as well, from 4.8 to 2.8 per 100 patient-years.
Argento commented: "The fact that infections went down speaks to something that is important right now. Hyperglycemia impairs immune function chronically, but also acutely...so patients who become ill and their blood glucose deteriorates rapidly are much more likely to have a poor outcome regardless of infection. There are data for COVID-19 now."
"These findings provide compelling support for use of [Libre] to improve both clinical outcomes and potentially reduce costs in this patient population," Bergenstal concluded.
ADA 2020 Scientific Sessions. Presented June 13, 2020. Abstracts 68-OR, 69-OR.
www red DiabetologNytt