Hospitalization of rural and urban infants during the first year of life

Abstract

Objectives: To examine hospitalizations and length of stay (LOS) for infants aged <1 year in rural and urban counties, hypothesizing that infants living in rural counties experience significantly different hospital use compared with urban infants.

Methods: Birth certificates for infants born in California hospitals between 1993 and 2005 and surviving to discharge were linked to hospital discharge records and death certificates during the first year of life, resulting in a study population of >6.4 million. Hospitalizations, cumulative LOS, readmission rates, and mortality were compared by using univariate and multivariable analysis for infants living in small rural, large rural, small urban, and large urban counties. Odds of hospitalization and cumulative LOS were also examined for common infant diagnoses.

Results: Infants living in increasingly rural counties experienced decreasing rates of hospitalization and decreasing number of hospitalized days during the first year of life. Infants living in small rural counties experienced 370 hospital days per 1000 infants compared with 474 hospital days per 1000 infants living in large urban counties. In multivariable analysis, infants in large urban counties experienced increased odds of hospitalization (odds ratio: 1.20 [95% confidence interval: 1.06-1.36]) and increased hospitalized days (incidence risk ratio: 1.17 [95% confidence interval: 1.06-1.29]) compared with infants in small rural counties. For most common diagnoses, urban residence was associated with either increased odds of hospitalization or increased cumulative LOS.

Conclusions: Infants living in rural California counties experienced decreased hospital utilization, including decreased hospitalization and decreased LOS, compared with infants living in urban counties.

FIGURE 1

Adjusted ORs for odds of hospitalization (A) and adjusted IRRs for cumulative LOS (B) according to county size. Results adjusted for gestational age, SGA, multiple gestations, gender, maternal age, race, parity, maternal education, insurance, neighborhood median income, and birth year, compared with small rural reference group after adjusting for clustering by residential county. ^aSignificant at 5% level after correcting for multiple comparisons through Bonferroni correction.

FIGURE 2

Adjusted ORs for odds of hospitalization and IRR for cumulative LOS of specific diagnoses in large urban compared with small rural areas. Results adjusted for gestational age, SGA, multiple gestations, gender, maternal age, race, parity, maternal education, insurance, neighborhood median income, and birth year, compared with small rural reference group after adjusting for clustering by residential county. ^aSignificant at 5% level after correcting for multiple comparisons through Bonferroni correction. Temp, temperature; Hyperbili, hyperbilirubinemia.