Children who live in Boston experience twice the risk of hospitalization for all medical conditions as do children living Rochester, NY. Children in New Haven, CT, sustain an intermediate risk. This proposal will identify several key factors underlying this variation and will clarify the health implications of hospitalization in these regions. The study will meet these objective through reviewing inpatients charts for all children residing in these three geographic areas and hospitalized for one of five conditions during 1988. These conditions include asthma/lower respiratory disease, meningitis, head trauma, abdominal pain/appendicitis and toxic ingestions. Data will be obtained to assess admission severity of illness, appropriateness of hospital use, attachment to primary care, the process leading to hospitalization and the cosioeconomic status of the child. In addition, outcome data will include severity at 72 hours, frequency of specified adverse events, and frequency of condition specific morbidities. Differences in the process of inpatient care will be documented as will be difference in hospitalization costs across the three regions. Severity will be determined via MedisGroups; appropriateness will be measured via the Pediatric Appropriateness Evaluation Protocol. Analyses using log-linear models will focus on whether the distribution of severity differs across these regions, how patient and health care factors influence severity, and whether these influences vary by region. By identifying factors underlying variation, and by clarifying health implications of hospitalization per se, this study will inform strategies for child health practices and policies directed towards improving outcomes and reducing costs.
