<p><strong>OBJECTIVE: </strong>To assess the timing of pediatric asthma rehospitalization, variation in rate of rehospitalization across hospitals, and factors associated with rehospitalization at different intervals.</p>

<p><strong>STUDY DESIGN: </strong>Retrospective cohort analysis of 44,204 hospitalizations for children with asthma within 42 children's hospitals between July 2008 and June 2011. The main outcome measures were rehospitalization for asthma within 7, 15, 30, 60, 180, and 365 days of an index asthma admission.</p>

<p><strong>RESULTS: </strong>The rate of asthma rehospitalization ranged from 0.5% (n = 208) at 7 days to 17.2% (n = 7603) at 365 days. Black patients and patients with public insurance had higher odds of rehospitalization at 60 days and beyond (P ≤ .01 for both). Adolescents (12- to 18-year-old), patients with a diagnosis of a complex chronic condition, and patients with a prior year asthma admission had higher odds of rehospitalization at every time interval (P ≤ .001 for all). Significant hospital variation in case-mix adjusted rates of rehospitalization existed at each time interval (P ≤ .01 for all). Rates at 365 days were ≤ 10.9% for the top 10% of hospitals; if all hospitals achieved this rate, 36.6% of rehospitalizations might have been avoided.</p>

<p><strong>CONCLUSIONS: </strong>Significant variation in asthma rehospitalization rates exists across children's hospitals from 7 to 365 days after an index admission. Racial/ethnic and economic disparities emerge at 60 days. By 1 year, rehospitalizations account for 1 in 6 hospitalizations. Assessing asthma rehospitalizations at longer intervals may augment our current understanding of and approach to post-hospitalization care improvement.</p>

<p><strong>OBJECTIVE: </strong>To assess the utility of imputing race/ethnicity using U.S. Census race/ethnicity, residential address, and surname information compared to standard missing data methods in a pediatric cohort.</p>

<p><strong>DATA SOURCES/STUDY SETTING: </strong>Electronic health record data from 30 pediatric practices with known race/ethnicity.</p>

<p><strong>STUDY DESIGN: </strong>In a simulation experiment, we constructed dichotomous and continuous outcomes with pre-specified associations with known race/ethnicity. Bias was introduced by nonrandomly setting race/ethnicity to missing. We compared typical methods for handling missing race/ethnicity (multiple imputation alone with clinical factors, complete case analysis, indicator variables) to multiple imputation incorporating surname and address information.</p>

<p><strong>PRINCIPAL FINDINGS: </strong>Imputation using U.S. Census information reduced bias for both continuous and dichotomous outcomes.</p>

<p><strong>CONCLUSIONS: </strong>The new method reduces bias when race/ethnicity is partially, nonrandomly missing.</p>