2292-P: Characterizing the Clinical Presentation of Diabetic Ketoacidosis Hospitalizations in People with Diabetes Using Admitting Diagnoses

Introduction and Objective: Diabetic ketoacidosis (DKA) remains a life-threatening emergency for people with diabetes (PWD), yet its early signs and symptoms are often mistaken for benign illness (e.g. stomach flu) - reflecting ongoing gaps in clinical awareness and education. We characterize the initial clinical diagnoses of DKA hospitalizations versus the proportion diagnosed as DKA at admission. Methods: Using the CMS Limited Data Set (Jan 2017-Mar 2025), which includes Medicare and dually eligible Medicaid beneficiaries, we identified patients with type 1 diabetes (T1DM) or type 2 diabetes (T2DM), ICD-10 E10/E11, hospitalized with a primary diagnosis of DKA. We summarized the admitting diagnoses for these DKA hospitalizations - in particular, the proportion that were DKA versus other diagnoses, as well as the proportion that were deemed nonspecific from the list of the 30 most common diagnoses. The admitting diagnosis in CMS data represents the physician’s working diagnosis at the time of inpatient care admission. Results: Of 100,334 individuals with a DKA hospitalization in the study period, 45% (45,380/100,334) had a non-DKA admitting diagnosis. This was consistent by type of diabetes, 41% in T1DM and 42% in T2DM. Of the 45,380 individuals with non-DKA admitting diagnoses, 66% (30,158) had nonspecific admitting diagnoses (e.g. altered mental status, fatigue, weakness, GI symptoms), and 50% (15,173) were gastrointestinal (GI) symptoms (e.g. nausea, vomiting, abdominal pain, diarrhea). Conclusion: Nearly half of the PWD hospitalized for DKA had alternative diagnoses at admission, suggesting DKA is often difficult to identify early. The non-DKA admitting diagnoses were largely nonspecific, and were often GI-related. These findings highlight the need for tools and strategies that enable earlier and more accurate detection of DKA; delay of diagnosis and treatment may increase length of stay and mortality, especially in older adults. Disclosure E.M. Miller: Advisory Panel; Current; Abbott. Speaker's Bureau; Current; Bayer AG, Corcept Therapeutics, Eli Lilly and Company, Insulet Corporation, Novo Nordisk, Sanofi. C. Wysham: Advisory Panel; Current; Abbott Diabetes. Speaker's Bureau; Current; Abbott Diabetes. Research Support; Ended; AbbVie Inc., Bayer AG, Novo Nordisk. Speaker's Bureau; Current; Novo Nordisk. Advisory Panel; Current; Eli Lilly and Company. Research Support; Ended; Eli Lilly and Company. Speaker's Bureau; Current; Eli Lilly and Company. Stock/Shareholder; Current; Pendulum. Speaker's Bureau; Ended; MannKind Corporation. S. Mehta: Employee; Current; Abbott Diabetes. N. Virdi: Employee; Current; Abbott Diabetes. A. Kwist: Employee; Current; Abbott Diabetes. F. Levrat-Guillen: Employee; Current; Abbott Diabetes. Stock/Shareholder; Current; Abbott Diabetes.