Risk Factors for Hospital Readmission following Implementation of a Diabetic Ketoacidosis (DKA) Power Plan

Abstract

Patients admitted to the hospital with DKA are at high-risk for readmission. We have previously demonstrated that implementation of a DKA PowerPlan (PP) that guides IV insulin dosing following acute management and transition to SC insulin reduces the frequency of recurrent DKA and rebound hyperglycemia during the index hospitalization. The purpose of this study was to determine whether the DKA PP reduces all cause and diabetes-related (recurrent DKA, hyperglycemia, hypoglycemia) hospital readmissions at 30 days and 1 year following an index hospitalization, as well to examine the contribution of high risk co-morbidities (psychiatric illness and/or history of alcohol/polysubstance abuse) associated with readmission risk in this group of patients. Retrospective chart review was performed for patients admitted with primary diagnosis of DKA prior to (pre-PP, n=60) and following DKA PP implementation (Post-PP, n=60). The groups were similar in age (Pre- vs. Post-PP: 38 ± 13.5 vs. 40 ± 16.6), BMI (26 ± 6.7 vs. 27 ± 7 kg/m2), gender (%Male: 38% vs. 48%), HbA1c (11.7 ± 2.4 vs. 11.3 ± 2.9%), admission BG (560 vs. 615 mg/dl) and anion gap (AG) (27.6 vs. 27.4), but differed in the prevalence of comorbidities which were higher in the pre-PP group (58 vs. 40%, p = 0.045). All cause readmissions were similar at 30 days (Pre vs. Post-PP: 1.4 ± 0.8 vs. 1.2 ± 0.4) but were lower at 1 year in the post-PP group (5.2 ± 4.7 vs. 3.5 ± 2.7, p = 0.07). Patients with co-morbidities in the post-PP group had lower all-cause (p = 0.023) but not diabetes related readmissions at 1 year, compared to patients with co-morbidities in the pre-PP group. In summary, these findings suggest that a DKA PP may be associated with a reduction in all cause hospital readmission in high risk patients with underlying psychiatric or substance abuse disorders. These results emphasize the need to identify and address patient specific factors that contribute to readmission in this high risk group. Disclosure N. Karajgikar: None. A. Donihi: None. R.A. Salata: None. R. Codario: None. R. Acharya: None. P. Manroa: None. M.T. Korytkowski: Advisory Panel; Self; Novo Nordisk Inc.. Other Relationship; Self; JAEB Center For Health Research.