Using an algorithm to recommend insulin dosing, both in and out of the hospital, was shown to improve glycemic control and aid population health management.
The march toward value-based care is occurring on many fronts, with diabetes quality metrics front and center. Providers from giant hospital systems to primary care practices are looking for ways to find savings–and make patients healthier.
In this mix is Glytec, a 10-year-old company founded on the idea that taking guesswork and time out of insulin dosing, first in hospital settings and then outside them, could dramatically reduce episodes of hypo- and hyperglycemia. This would provide savings and lead to better glycated hemoglobin (A1C) levels for patients with previously uncontrolled diabetes.1
Glytec presented a series of studies at the 76th Scientific Sessions of the American Diabetes Association that took place June 10-14, 2016, in New Orleans, Louisiana, and study authors spoke with The American Journal of Managed Care during conference about what the findings mean for value-based care and population health.
The company seeks to manage diabetes through cloud-based systems, using FDA-cleared algorithms, known as Glucommander, that take into account a patient’s weight, age, and other clinical indicators. A big step has involved integrating the algorithms with Epic, the electronic health record (EHR) platform used by many health care systems.
The integration has meant that Glucommander has been “very easily accepted” by the nursing staff, said Jagdeesh Ullal, MD, an endocrinologist with Sentara Healthcare, a 12-hospital system in Virginia and North Carolina that adopted the technology in 2013. Glytec previously reported that since the change, Sentara reported an in-hospital hypoglycemia rate of 0.83% for intravenous (IV) and subcutaneous insulin dosing.2 (Ullal was also a co-author on an economic analysis of the cost of hospital-associated hypoglycemia.3)
Outpatient Dosing. Bruce Bode, MD, FACE, a specialist with Atlanta Diabetes Associates and a clinical associate professor at Emory University, was an author on a separate study involving Glytec’s outpatient technology: the Glucommander can allow providers to remotely manage patients with type 1 and type 2 diabetes by using the algorithm to generate dosing recommendations.4
Only the provider uses the algorithm, not the patient, Bode explained; information on dosing is then forwarded to the patient via a HIPAA-compliant text or email. “That is why the FDA allowed us to do this,” Bode said.
In the study, 31 patients were treated for 12 weeks using Glucommander Outpatient, which made personalized dosing recommendations based on patients’ self-management blood glucose data. The mean A1C at baseline was 10.4% and the mean A1C after 12 weeks was 7.7%. Of the 7940 blood glucose tests recorded, mild hypoglycemia was found in 1.6% of the blood glucose tests (40-70 mg/dL), and no episodes of major hypoglycemia were reported ( < 40 mg/dL). In patient satisfaction surveys, 96% of the respondents said they would recommend the system to a family member or friend.4
In an interview, Bode explained that there are great opportunities for remote monitoring and dosing adjustments with staff such as certified diabetes educators. “One CDE could manage 1000 patients,” he said. With shortages at the primary care level to manage this population, the technology could greatly improve diabetes management for the difficult “between visit” periods, when patients often struggle.
Opportunity for Savings. Ullal discussed a study led by Joseph A. Aloi, MD, of section chief of endocrinology and metabolism at Wake Forest Baptist Health. While a landmark study in AJMC in 2011 calculated the cost of an inpatient admission due to hypoglycemia at $17,564,5 Aloi, Ullal and their co-authors took a different tact–they looked at the cost of events that occur once a patient is already admitted to the hospital, as well as the cost of recurrent events, in light of rising costs since 2011.3
The study evaluated data from 44,000 patients admitted to an academic medical center during a 1-year period with a diagnosis of diabetes who were prescribed insulin; in other words, the population that might be managed by Glytec’s cloud-based technology. They looked at length of stay (LOS) and overall cost for those on insulin who had a significant hypoglycemic event vs those who did not, defining such an event as those who had a blood glucose of 50 mg/dL or less. Of the study population, 387 met the criteria.
Those with a significant event had a LOS of 12.3 days, compared with 4.3 days for the rest. Costs for those with an event were $83,000 compared with $29,700 for those without. Each incremental significant event increased costs by about $7000: 2 events were $83,000, 3 events were $103,000, 4 were $110,000 and more than 4 were $117,000.
Reducing Nursing Time. Valerie Garrett, MD, who practices at Mission Health in Asheville, North Carolina, led a retrospective analysis that compared insulin management with Glucommander for 219 patients after cardiovascular surgery with 1755 patients whose insulin had been managed with a different infusion algorithm, EndoTool.6 The capillary blood glucose (CBG) target was adjusted from < 120 mg/dL with EndoTool to < 140 mg/dL with Glucommander. Incident of hyperglycemia (BG > 180 mg/dL) was reduced, from 6% with EndoTool to 3.9% with Glucommander. There was no severe hypoglycemia with Glucommander and less mild hypoglycemia with this algorithm than with EndoTool.
In an interview, Garrett said the integration of Glucommander into the EHR led to a streamlining of the process for the staff, which contributed directly the finding that nursing time from capillary blood glucose due to IV insulin titration was cut nearly in half (47%), from 19.9 minutes to 10.5 minutes. Less hypoglycemia for patients and less nursing time for the provider are important results, she said. “From the perspective of managed care, it all matters,” Garrett said.
1.Â Â Â Â Â Glytec website. Our Story. Â https://www.glytecsystems.com/company/about. Accessed June 20, 2016.
2.Â Â Â Â Â Sentara Healthcare ushers in a new standard of care in glucose management [press release]. Greenville, SC: Business Wire; May 7, 2014. http://www.businesswire.com/news/home/20140507005373/en/CORRECTING-REPLACING-Sentara-Healthcare-Ushers-Standard-Care.
3.Â Â Â Â Â Aloi J, Ullal J, McFarland R, Henderson A. The frequency of hypoglycemia in hospitalized patients treated with insulin increases overall costs associated with hospitalization and length of stay. Diabetes. 2016; 65 (suppl1): 152-LB.
4.Â Â Â Â Â Clarke JG, Bode BW. Glucommander Outpatient, a cloud-based insulin management solution adjusted insulin doses and achieved 2.7% drop in A1C percentage points. Diabetes. 2016; 65 (suppl1): 84-LB.
5.Â Â Â Â Â Quilliam BJ, Simeone JC, Ozbay B, Kogut SJ. The incidence and cost of hypoglycemia in type 2 diabetes. Am J Manag Care. 2011;17(10):673-680)
6.Â Â Â Â Â Garrett V, Fornoff L, Becker J, Booth R, Henderson A, McFarland R. A comparison of glycemic outcomes for two computerized insulin infusion algorithms in CV surgery patients. Diabetes. 2016; 65 (suppl1): 101-LB.