ABSTRACT
BACKGROUND: Older adults frequently experience nonspecific clinical features. However, there is limited evidence on how often admission diagnoses for hospitalized older patients are incorrect, potentially leading to treatment delays. OBJECTIVES: To determine the consistency between hospital admission and discharge diagnoses, and identify factors associated with diagnostic discrepancies in older adults. DESIGN AND SETTING: Population-based cohort study in the United States. We included adults aged ≥ 18 years who were admitted from emergency departments (EDs) to hospitals, identified using the 2005-2010 National Hospital Ambulatory Medical Survey, a nationally representative survey. METHODS: Three admission diagnoses and the principal discharge diagnosis were captured and classified as discrepant if they involved considerably different conditions within the same organ system, or different organ systems altogether. RESULTS: Each year, 12 million adults were hospitalized following ED visits in the United States; 45% were aged ≥ 65 years. These patients' mean age was 79 years and 58% were women. Diagnostic discrepancies between admission and discharge were more common among adults ≥ 65 years (12.5 versus 8.3%; P < 0.001). Certain admission diagnoses had particularly high rates of diagnostic discrepancies: 26-27% of patients presenting with mental disorders or with endocrine and metabolic diseases had substantial diagnostic discrepancies between admission and discharge. Substantial diagnostic discrepancy was independently associated with longer hospitalization and higher in-hospital mortality. CONCLUSION: One out of eight older adults hospitalized from EDs was discharged with a principal diagnosis differing considerably from the admission diagnosis. Given that missed or delayed diagnoses are a critical safety problem, clinicians should be vigilant and frequently cogitate alternative diagnostic possibilities