Assessment of a Day Hospital Management Program for Children With Type 1 Diabetes

IMPORTANCE: A shift in the setting of care delivery for children with a new diagnosis of type 1 diabetes led to a reorganization of treatment. OBJECTIVE: To determine whether a new diagnosis of pediatric diabetes can be successfully managed in a day hospital model. DESIGN, SETTING, AND PARTICIPANTS: This quality improvement study used retrospectively collected data on pediatric patients with a new diagnosis of diabetes who completed an inpatient program for education and insulin titration prospectively compared with patients completing a diabetes day hospital program. Baseline data were collected over 12 months (January-December 2015) and intervention data collected over 14 months (March 2016-May 2017). The study was conducted at a single institution and judged as a nonhuman participant project. The referral local base included a 100-mile radius. Patient inclusion was a new diagnosis of diabetes, age 5 years or older, and no biochemical evidence of diabetic ketoacidosis. Ninety-six patients completed the day hospital program and 192 patients completed an inpatient program. EXPOSURES: All patients received 2 consecutive days of insulin titration and education in either a day hospital or inpatient setting. MAIN OUTCOMES AND MEASURES: Primary outcomes included the mean length of stay, patient charge, and insurance denial/reimbursement rates. The hypothesis was that a day hospital program would be associated with a reduced length of stay, which would directly affect patient charges and insurance denials. RESULTS: Among the 96 day hospital patients, the mean (SD) age was 12.2 (4.7) years (range 5-20.3), with no patients experiencing diabetic ketoacidosis or hypernatremia. Among the 192 inpatient patients, the mean (SD) age was 9.4 (4.7) years (range, 1.6-20.1). The mean (SD) length of stay reduction in the day hospital was 46 (14.1) to 14 (5.1) hours. The mean day hospital patient charge was $2800, compared with a mean (SD) baseline carge of $24 103 ($9401). Within the first year, there was a cumulative reduction in patient charges of more than $2.1 million. CONCLUSIONS AND RELEVANCE: This study’s findings suggest that a diabetes day hospital setting was associated with reductions in length of stay and patient charges, with an increase in insurance reimbursements and a decrease in insurance denials. This study demonstrates an effective way to streamline new-onset diabetes education, which may reduce length of stay and patient charges. Reimbursement rates for patients with a new diagnosis of diabetes increased from 52% to 72% and reimbursement denial rates decreased from 80% to 0%.