Abstract
The use of abdominal radiographs contributes to increased healthcare costs, radiation exposure, and potentially to misdiagnoses. We evaluated the association between abdominal radiograph performance and emergency department (ED) revisits with important alternate diagnosis among children with constipation. Retrospective cohort study of children aged <18 years diagnosed with constipation at one of 23 eds from 2004 to 2015. the primary exposure was abdominal radiograph performance. the primary outcome was a 3-day ed revisit with a clinically important alternate diagnosis. rand university of california, los angeles methodology was used to define whether the revisit was related to the index visit and due to a clinically important condition other than constipation. regression analysis was performed to identify exposures independently related to the primary outcome. a total of 65.7% (185 439 282 225) of children with constipation had an index ed visit abdominal radiograph performed. three-day revisits occurred in 3.7% (10 566 282 225) of children, and 0.28% (784 282 225) returned with a clinically important alternate related diagnosis. appendicitis was the most common such revisit, accounting for 34.1% of all 3-day clinically important related revisits. children who had an abdominal radiograph performed were more likely to have a 3-day revisit with a clinically important alternate related diagnosis (0.33% vs 0.17%; difference 0.17%; 95% ci 0.13-0.20). following adjustment for covariates, abdominal radiograph performance was associated with a 3-day revisit with a clinically important alternate diagnosis (aor: 1.39; 95% ci 1.15-1.67). additional characteristics associated with the primary outcome included narcotic (aor: 2.63) and antiemetic (aor: 2.35) administration and underlying comorbidities (aor: 2.52). among children diagnosed with constipation, abdominal radiograph performance is associated with an increased risk of a revisit with a clinically important alternate related diagnosis.>
