Predictors of pediatric emergency patients discharged against medical advice.

Based on a retrospective 5-year medical record review, this study characterizes factors associated with patients discharged against medical advice (AMA) from a tertiary pediatric emergency department (ED) and compares rates of return to the ED and admission to the hospital with those of patients routinely discharged. Data from 94 patients discharged AMA are compared with those of 188 control patients. Pediatric patients at risk for discharge AMA are older than 15 years (odds ratio [OR], 3.561; 95% confidence interval [CI], 1.695-7.482), self-register independent of a parent (OR, 3.100; 95% CI, 1.818-152.770), arrive by ambulance (OR, 2.761; 95% CI, 1.267-6.018), involve a consultant (OR, 2.592; 95% CI, 1.507-4.458), and have a chief complaint of abdominal pain (OR, 3.095; 95% CI, 1.154-8.303). Negative predictors include urgent triage (OR, 0.155; 95% CI, 0.039-0.618), a chief complaint of upper respiratory tract illness or otitis media (OR, 0.229; 95% CI, 0.075-0.702), and discharge diagnoses of infection (adjusted OR, 0.053; 95% CI, 0.004-0.767), disease of the nervous system and sense organs (adjusted OR, 0.066; 95% CI, 0.005-0.898), respiratory illness (adjusted OR, 0.072; 95% CI, 0.007-0.718), and gastrointestinal disease (adjusted OR, 0.050; 95% CI, 0.006-0.419). Certain key elements of discharge AMA are well documented, including consequences of discharge AMA (74.5%) and instructions for care (54.3%). Other elements such as alternative therapies (1.1%) are poorly documented. Patients discharged AMA have a significantly higher return rate (24.5%) within 15 days compared with patients who have routine discharge (6.4%) (chi2=18.85, P<.001). Ninety-six percent of patients who return to the ED have the same chief complaint at both visits if discharged AMA compared with 50% of patients who are discharged routinely (P=.003), with 25% admission rates at the time of second visit for both types of discharges. Adolescents who register themselves are at increased risk for discharge AMA. Patients who are triaged as urgent or nonurgent or who have minor illnesses are likely to be dispositioned routinely. Patients discharged AMA are more likely to return to the ED with the same complaint than patients who are routinely discharged.

Risk factors for extremely long length-of-stay among pediatric emergency patients.

OBJECTIVE: Over time, we observed more visits in our pediatric emergency department with length-of-stay (LOS) of more than 10 hours, whereas our mean LOS was approximately 3 hours. We sought to characterize factors associated with this extremely long LOS. METHODS: Eighty-one visits with LOS more than 10 hours were identified from January 1, 2001, to June 30, 2003. In this retrospective study, we compared these cases with 405 randomly selected age-matched controls with LOS less than 10 hours (5 controls per case). RESULTS: The groups were similar for sex, visit month, arrival mode, and level of training of the supervising physician. Cases more frequently arrived during night shifts (30% vs 13%) and had laboratory tests (93% vs 32%), radiological studies (83% vs 34%), procedures (28% vs 15%), sedations (24% vs 4%), subspecialty consultations (84% vs 20%), chief complaints of abdominal pain (42% vs 6%) and diagnoses of appendicitis (10% vs 1%), and had a greater hospitalization rate (67 vs 19%). Although more cases involved white patients (57% vs 31%), race was not associated with LOS more than 10 hours in adjusted analysis. In multivariable analysis, longer waiting time (odds ratio [OR], 1.013; 95% confidence interval [CI], 1.007-1.019), night shift arrival (OR, 5.0; 95% CI, 1.9-12.8), higher triage acuity (lowest acuity: OR, 0.003; 95% CI, 0.0-0.286), radiology study other than radiographs (OR, 18.0; 95% CI, 7.5-43.1), and subspecialty consultation (OR, 7.6; 95% CI, 3.2-18.3) were associated with LOS more than 10 hours. CONCLUSIONS: In our pediatric emergency department, risk factors for LOS more than 10 hours included longer waiting time, night shift arrivals, high triage acuity, radiology studies, and subspecialty consultations. These factors may also be important considerations for quality improvement initiatives at other institutions.

Validation of the pulmonary score: an asthma severity score for children.

OBJECTIVES: In the absence of a validated "user-friendly" method of scoring asthma severity, the authors derived the pulmonary score (PS). The purpose of this study was to begin validation trials of the PS by comparing it with the peak expiratory flow rate (PEFR). METHODS: The study enrolled a convenience sample of children, aged 5-17 years, who came to the emergency department (ED) for treatment of an acute asthma exacerbation. The PEFR (best of three attempts) and the PS were measured before and after the first albuterol treatment by a physician and a nurse from a pool of 45 trained observers. The PS includes respiratory rate, wheezing, and retractions, each rated on a 0-3 scale. Decreasing PS and increasing PEFR indicate clinical improvement. Pre- and post-treatment PEFRs and PSs were compared using paired t-tests to establish construct validity. Correlation of pre- and post-treatment PSs with PEFRs was measured to establish criterion validity. RESULTS: Forty-six subjects completed the study. Mean percent predicted PEFR improved after treatment by 20.7% (p = 0.0001), and mean PS by 1.5 for nursing-obtained scores (p < 0.0001) and 1.9 for physician-obtained scores (p < 0.0001). Pre- and post-treatment PSs were significantly correlated with PEFRs. Correlations for the nursing-obtained scores were pre-treatment r = -0.57 (p = 0.0003) and post-treatment r = -0.67 (p = 0.0001), and for the physician-obtained scores were pre-treatment r = -0.44 (p = 0.003) and post-treatment r = -0.56 (p = 0.0001). The pre-treatment interrater reliability was 0.62 and the post-treatment was 0.53. CONCLUSIONS: These data support the construct and criterion validities of the PS as a measure of asthma severity among children in the ED. The PS is a practical substitute to estimate airway obstruction in children who are too young or too sick to obtain PEFRs.