I Danel1, C Johnson, C Berg, L Flowers, H Atrash. 1. Centers for Disease Control and Prevention, Division of Reproductive Health, Atlanta, Georgia 30341, USA. ixd1@cdc.gov
Abstract
OBJECTIVES: To determine the independent association of selected maternal and hospital characteristics with length of maternal hospital stay for uncomplicated vaginal deliveries. METHOD: Linear regression analysis using National Hospital Discharge Survey data from 1988 to 1995. Independent variables were year, maternal age and race, method of payment, and hospital ownership, size, and geographic location. The outcome measure was length of maternal hospital stay for uncomplicated vaginal deliveries. RESULTS: Length of stay was independently associated with year, geographic region, payment method, and hospital size. From 1988 to 1995, the mean length of stay fell from 2.1 to 1.5 days. The rate of decrease was similar for all regions, methods of payment, and hospital size. Women in the West had a shorter mean length of stay (1.5 days) than women in the Northeast (2.2 days). The difference by method of payment was smaller. Length of stay was shortest for women without insurance (1.8 days) and longest for women covered by Blue Cross (2.1 days). Maternal age and race and type of hospital ownership were not independently associated with the length of stay. CONCLUSIONS: Significant variations existed in the length of time women are hospitalized for normal childbirth. These variations are primarily associated with where a woman lives and whether she is insured. Given the current public debate on the impact of shortened hospital stays, these variations need to be explored and their effects on maternal and infant well-being clarified.
OBJECTIVES: To determine the independent association of selected maternal and hospital characteristics with length of maternal hospital stay for uncomplicated vaginal deliveries. METHOD: Linear regression analysis using National Hospital Discharge Survey data from 1988 to 1995. Independent variables were year, maternal age and race, method of payment, and hospital ownership, size, and geographic location. The outcome measure was length of maternal hospital stay for uncomplicated vaginal deliveries. RESULTS: Length of stay was independently associated with year, geographic region, payment method, and hospital size. From 1988 to 1995, the mean length of stay fell from 2.1 to 1.5 days. The rate of decrease was similar for all regions, methods of payment, and hospital size. Women in the West had a shorter mean length of stay (1.5 days) than women in the Northeast (2.2 days). The difference by method of payment was smaller. Length of stay was shortest for women without insurance (1.8 days) and longest for women covered by Blue Cross (2.1 days). Maternal age and race and type of hospital ownership were not independently associated with the length of stay. CONCLUSIONS: Significant variations existed in the length of time women are hospitalized for normal childbirth. These variations are primarily associated with where a woman lives and whether she is insured. Given the current public debate on the impact of shortened hospital stays, these variations need to be explored and their effects on maternal and infant well-being clarified.