Emergency Room Use among Low-income Children: Does Health Insurance Coverage Make a Difference

Suk-fong S Tang.  Department of Practice and Research, American Academy of Pediatrics, Elk Grove Village, IL.

 

Background: The cost-effectiveness of providing low-income children public health insurance coverage is often debated in terms of its effectiveness in channeling patients away from costly care sites. But despite studies showing higher use of physician offices by the insured and by new SCHIP enrollees, a direct relationship between coverage and reduced ER use has not been demonstrated.

 

Objective: This study compares physician office and ER utilization between insured low-income children who have consistent coverage versus those with insurance gaps during the year. It tests the hypothesis that privately insured and publicly covered low-income children with gaps in their coverage have fewer physician office visits but more ER admissions than those with full-year coverage.

 

Design/Methods: Using 1996-2000 Medical Expenditure Panel Survey monthly insurance coverage and utilization data, annual physician office visit and ER admission rates were compared between full-year and part-year insured poor and low-income children who were i) privately insured only, and ii) publicly-insured. SSI children were excluded to reduce variability in health. Those who had hospitalizations and enrolled in Medicaid mid-year were excluded to eliminate selection biases resulting from Medicaid enrollment following ER and hospital admissions. Instead of between-group difference, the direction of the difference was tested for statistical significance due to sample size limitations.

 

Results: For i) privately insured and ii) publicly covered children in 4 of 5 study years, those insured part-year had fewer physician office visits and more ER admissions than those covered full-year. On average, part-year insured children had 23% fewer office visits and 10% higher ER admissions than those full-year insured. With the hypothesized direction of the difference in utilization confirmed in 8 of 10 comparisons, the null hypothesis was rejected at p < .05.

 

Conclusions: While previous studies comparing the insured to the uninsured did not show reduced ER use as a benefit of health insurance coverage, the current study suggests that this benefit does exist. Focusing on low-income insured children with more or less stable coverage, this study likely reduced variability in a) ER use attributed to health status and b) families experience with accessing acute care in physician offices, and was able to demonstrate that ER use decreases as physician office visits increase with consistent health insurance coverage.