We conducted a cross-sectional analysis of 2004-2008 hospital discharge information from the Healthcare Cost and Utilization Project Nationwide Inpatient Sample (HCUP-NIS). The HCUP-NIS, a 20% stratified sample of acute hospital admissions, is the largest all-payer inpatient database in the United States.3 Sampling weights are used to generate national estimates. We selected all patients 100 years or older included in the HCUP-NIS from 2004 through 2008. Admissions missing mortality data (<0.01%) were excluded. We calculated the 10 most common principal discharge diagnoses and grouped them based on International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes (eTable). Comorbidity was measured using the age-independent Charlson Comorbidity Index, which incorporates 19 medical conditions, including dementia, cardiovascular disease, cerebrovascular disease, pulmonary disease, renal disease, diabetes mellitus, and malignant disease, and classified as mild (Charlson score, 0-1), moderate (Charlson score, 2-4), and severe (Charlson score, ≥5).4 We calculated rates of utilization, all-cause mortality, and disease-specific mortality. Data analysis and treatment were performed using SAS statistical software (version 9.2; SAS Institute Inc). This study was exempted by the institutional review board of Yale University because HCUP-NIS is a public database with no personal identifying information.