Yu Bai, MD, Health Policy & Administration, Pennsylvania State University, 104 Henderson Building, University Park, PA 16802, 814-238-2735, yzb103@psu.edu

Background: There are approximately 200,000 childhood asthma hospitalizations annually in the US, and many are considered potentially avoidable with high quality outpatient care. Health insurance is expected to play an important role in improving access to outpatient health care services, and may therefore be associated with the need for hospitalization and length of hospital stay among children with asthma. Objective: To evaluate the effects of insurance status on length of stay among children hospitalized with asthma. Methods: Hospitalization data for Pennsylvania children (0-19 yrs) in 2001 whose primary diagnosis was asthma (ICD9-CM 493) was merged with Census 2000 data on rural/urban residence and median household income of residential zipcode. A multivariate linear regression model was used to evaluate the association between insurance status (private insurance/ Medicaid/uninsured) and length of stay, controlling for age, gender, race/ethnicity (nonHispanic white/nonHispanic black/Hispanic/other or unknown), rural/urban residence, income, and urgency of child’s condition at admission (emergency/urgent/elective). Results: Among the 7734 childhood asthma hospitalizations in 2001, 45 percent were covered by private insurance, 53 percent were covered by Medicaid, and less than 3 percent were uninsured. Privately insured children were disproportionately likely to be white and to live in areas with higher estimated median income (p<0.01). Sixty nine percent of black children and 65 percent of Hispanic children had Medicaid coverage. Among children with Medicaid coverage, 82 percent were admitted in emergency medical condition, compared to 68 percent of privately insured children and 72 percent of uninsured children (p<0001). Children with Medicaid coverage also had longer lengths of stay than children with private insurance and uninsured children (2.3 vs 2.1 and 1.8, respectively; p<0.01), even after controlling for potentially confounding variables. Regression results also revealed that, net of other factors, rural children had significantly longer lengths of stay than urban children (p<0.01), and estimated income was inversely associated with length of stay (p=.0025). Conclusion: Children with Medicaid coverage were more likely to have longer hospital stays than other children, and were more likely to be admitted in poorer clinical condition. These findings suggest that among Pennsylvania children with asthma, Medicaid coverage does not provide the same access to appropriate outpatient health care as private insurance. Policymakers should focus on identifying and removing barriers to appropriate outpatient care for children with asthma who are enrolled in Medicaid.

Learning Objectives: