Estelle M Everett1,2,3, Timothy P Copeland4, Tannaz Moin1,2,3,5, Lauren E Wisk2,4. 1. Division of Endocrinology, Diabetes, & Metabolism, Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, CA, USA. 2. Division of General Internal Medicine & Health Services Research, Department of Medicine, David Geffen School of Medicine, University of California, Los Angeles, CA, USA. 3. Veteran Affairs Greater Los Angeles Healthcare System, Los Angeles, CA, USA. 4. Department of Health Policy and Management, Fielding School of Public Health, University of California, Los Angeles, CA, USA. 5. HSR&D Center for the Study of Healthcare Innovation, Implementation & Policy, Veteran Affairs Greater Los Angeles Healthcare System, Los Angeles, CA, USA.
Abstract
BACKGROUND: Insulin pump use in type 1 diabetes management has significantly increased in recent years, but we have few data on its impact on inpatient admissions for acute diabetes complications. METHODS: We used the 2006, 2009, 2012, and 2019 Kids' Inpatient Database to identify all-cause type 1 diabetes hospital admissions in those with and without documented insulin pump use and insulin pump failure. We described differences in (1) prevalence of acute diabetes complications, (2) severity of illness during hospitalization and disposition after discharge, and (3) length of stay (LOS) and inpatient costs. RESULTS: We identified 228 474 all-cause admissions. Insulin pump use was documented in 7% of admissions, of which 20% were due to pump failure. The prevalence of diabetic ketoacidosis (DKA) was 47% in pump nonusers, 39% in pump users, and 60% in those with pump failure. Admissions for hyperglycemia without DKA, hypoglycemia, sepsis, and soft tissue infections were rare and similar across all groups. Admissions with pump failure had a higher proportion of admissions classified as major severity of illness (14.7%) but had the lowest LOS (1.60 days, 95% CI 1.55-1.65) and healthcare costs ($13 078, 95% CI $12 549-$13 608). CONCLUSIONS: Despite the increased prevalence of insulin pump in the United States, a minority of pediatric admissions documented insulin pump use, which may represent undercoding. DKA admission rates were lower among insulin pump users compared to pump nonusers. Improved accuracy in coding practices and other approaches to identify insulin pump users in administrative data are needed, as are interventions to mitigate risk for DKA.
BACKGROUND: Insulin pump use in type 1 diabetes management has significantly increased in recent years, but we have few data on its impact on inpatient admissions for acute diabetes complications. METHODS: We used the 2006, 2009, 2012, and 2019 Kids' Inpatient Database to identify all-cause type 1 diabetes hospital admissions in those with and without documented insulin pump use and insulin pump failure. We described differences in (1) prevalence of acute diabetes complications, (2) severity of illness during hospitalization and disposition after discharge, and (3) length of stay (LOS) and inpatient costs. RESULTS: We identified 228 474 all-cause admissions. Insulin pump use was documented in 7% of admissions, of which 20% were due to pump failure. The prevalence of diabetic ketoacidosis (DKA) was 47% in pump nonusers, 39% in pump users, and 60% in those with pump failure. Admissions for hyperglycemia without DKA, hypoglycemia, sepsis, and soft tissue infections were rare and similar across all groups. Admissions with pump failure had a higher proportion of admissions classified as major severity of illness (14.7%) but had the lowest LOS (1.60 days, 95% CI 1.55-1.65) and healthcare costs ($13 078, 95% CI $12 549-$13 608). CONCLUSIONS: Despite the increased prevalence of insulin pump in the United States, a minority of pediatric admissions documented insulin pump use, which may represent undercoding. DKA admission rates were lower among insulin pump users compared to pump nonusers. Improved accuracy in coding practices and other approaches to identify insulin pump users in administrative data are needed, as are interventions to mitigate risk for DKA.
Authors: P Ross; A R Gray; J Milburn; I M Kumarasamy; F Wu; S Farrand; J Armishaw; E Wiltshire; J Rayns; P Tomlinson; B J Wheeler Journal: Acta Diabetol Date: 2016-09-01 Impact factor: 4.280
Authors: Steven M Willi; Kellee M Miller; Linda A DiMeglio; Georgeanna J Klingensmith; Jill H Simmons; William V Tamborlane; Kristen J Nadeau; Julie M Kittelsrud; Peter Huckfeldt; Roy W Beck; Terri H Lipman Journal: Pediatrics Date: 2015-03 Impact factor: 7.124
Authors: Sari Hoshina; Gregers S Andersen; Marit E Jørgensen; Martin Ridderstråle; Dorte Vistisen; Henrik U Andersen Journal: Diabetes Technol Ther Date: 2018-02-13 Impact factor: 6.118
Authors: Beate Karges; Anke Schwandt; Bettina Heidtmann; Olga Kordonouri; Elisabeth Binder; Ulrike Schierloh; Claudia Boettcher; Thomas Kapellen; Joachim Rosenbauer; Reinhard W Holl Journal: JAMA Date: 2017-10-10 Impact factor: 56.272
Authors: Joseph E O'Reilly; Anita Jeyam; Thomas M Caparrotta; Joseph Mellor; Andreas Hohn; Paul M McKeigue; Stuart J McGurnaghan; Luke A K Blackbourn; Rory McCrimmon; Sarah H Wild; John R Petrie; John A McKnight; Brian Kennon; John Chalmers; Sam Phillip; Graham Leese; Robert S Lindsay; Naveed Sattar; Fraser W Gibb; Helen M Colhoun Journal: Diabetes Care Date: 2021-07-08 Impact factor: 19.112
Authors: Stephen R Benoit; Yan Zhang; Linda S Geiss; Edward W Gregg; Ann Albright Journal: MMWR Morb Mortal Wkly Rep Date: 2018-03-30 Impact factor: 17.586
Authors: Kelly R Bergmann; Amanda Nickel; Matt Hall; Gretchen Cutler; M Jennifer Abuzzahab; Brianna Bretscher; Shea Lammers; Dave Watson; Gabrielle Z Hester Journal: JAMA Netw Open Date: 2022-05-02