Objective: To describe the use of a pediatric observation unit (OU), including relapse rates for common pediatric illnesses, and to assess effectiveness of OU utilization.
Design: Retrospective, cohort of all emergency department (ED) visits, OU and inpatient unit (IU) admissions.
Setting: Tertiary care children's hospital.
Participants: All children evaluated in the ED and subsequently admitted to either the OU or IU over a 2-year period.
Main outcome measure: Rates with 95% confidence intervals (CI) for OU use and need for subsequent IU admission from OU, and odds ratios (OR) with 95% CI for use of the OU for specific pediatric disorders.
Results: During 10/1/96-9/30/98, there were 44,459 ED visits, 1798 (4.0%) OU admissions, and 3241 (7.3%) inpatient admissions (IA) from the ED. OU mean length of stay was 15.6 +/- 6.1 hours; mean age was 6 +/- 5.3 years with 31% under 2 years of age. Of the total admissions (IU and OU), diagnoses with high OU utilization were: asthma 274/575, 48%; croup 76/125, 61%; enteritis/dehydration 284/470, 60%; poisonings 82/118, 70%; and seizures 80/204, 39%. The likelihood of an OU admission for these illnesses versus IU (adjusted for subsequent need for IU admission) was: asthma OR 1.3 (1.1, 1.5), P < 0.005; croup OR 2.3 (1.6, 3.3), <0.001; enteritis/ dehydration OR 2.8 (2.1, 3.0), P < 0.001; poisonings OR 3.8 (2.5, 5.7), P < 0.001; and seizures OR 0.8 (0.6, 1.2), P = 0.28. For these diagnoses, OU admissions resulting in IU admission occurred for asthma 45/274, 16.4%; croup 7/76, 9.2%; enteritis/ dehydration 13/284, 4.6%; poisonings 3/82, 3.7%; and seizures 15/80, 18.8%, resulting in an overall need for further hospitalization to the IU for these diagnoses of 83/796, 10.4%, (95% CI 8.3, 12.6).
Conclusion: Admissions to the observation unit comprised over one third of all admissions from a pediatric ED. Certain pediatric illnesses appear to be well suited for admission to the observation unit, with low likelihood of the need for subsequent admission to the inpatient unit. Given the current trends in third-party payer reimbursements for short (<24 hours) admissions, observation unit use provides a more attractive alternative to inpatient admission for many pediatric patients.