Concordance between emergency department and discharge diagnoses in patients admitted with right lower quadrant abdominal pain

Abstract

Study objectives: We compare emergency department (ED) diagnosis to discharge diagnosis in patients presenting with right lower quadrant pain and examine factors influencing agreement. Methods: We conducted a 7-month prospective observational study enrolling patients with right lower quadrant pain in an urban ED. Exclusion criteria were recent trauma or previous appendectomy. Patients discharged from the ED were followed up for 7 days (none required later admission). ED diagnosis, discharge diagnosis, demographic data, diagnostic studies, and inhospital management were recorded. Primary outcomes were agreement between ED diagnosis and discharge diagnosis (excluding discharged patients). Secondary outcome was ED diagnosis accuracy in appendicitis, controlling for age, race, sex, and insurance (sensitivity/specificity calculated with discharged patients). Data are reported with 95% confidence intervals (CIs). Results: Four hundred twenty-nine patients enrolled; 269 patients were admitted (63%). Agreement of ED diagnosis and discharge diagnosis was 80% (95% CI 75% to 85%). There was no difference in age groups (>65 versus ED diagnosis of appendicitis with CT scan (n=243) was as follows: accuracy 90% (95% CI 86% to 94%), sensitivity 91% (95% CI 87% to 95%), and specificity 90% (95% CI 86% to 94%). Patients not scanned were more likely to be triaged to the most critical category, to have a fever, and to have a WBC count outside the 95% CI for age-specific norms. For patients who ultimately required operative treatment (n=120), ED diagnosis had 90% agreement with discharge diagnosis (95% CI 85% to 95%). ED appendicitis diagnosis in those who required surgery was as follows: accuracy 94% (95% CI 89% to 99%), sensitivity 96% (95% CI 91% to 100%), and specificity 90% (95% CI 85% to 95%). Conclusion: ED diagnosis and discharge diagnosis have higher concordance in right lower quadrant pain in male patients and in patients requiring surgery. There may be less agreement in female patients, especially with obstetric and gynecologic pathology. Agreement does not appear to be affected by age, race, or insurance status.