Early hospital discharge of infants born to group B streptococci-positive mothers: a decision analysis.

Objective: To compare the cost-effectiveness of an additional 24-hour inpatient observation for asymptomatic term neonates born to group B streptococcus (GBS)-colonised mothers with adequate intrapartum antibiotic prophylaxis (IAP) after an initial 24-hour in-hospital observation.

Design: Cost-effectiveness analysis from a societal perspective.

Setting: United States.

Population: Asymptomatic term neonates born to GBS-colonised mothers with IAP after an initial 24-hour in-hospital observation.

Methods: Monte Carlo simulation for a decision tree model incorporating the following chance events: development of GBS sepsis during the second 24 hours of life, development of GBS sepsis between 48 hours and 7 days of life, prompt versus delayed treatment for sepsis, neonatal mortality and long-term health sequelae.

Main Outcome Measures: Expected cost and quality-adjusted life years (QALYs), Incremental cost-effectiveness ratio (ICER).

Results: Delayed, versus early, hospital discharge results in similar mean expected QALYs, but substantially higher expected cost. The mean difference in QALY is 0.00016 (95% CI 0.00005-0.00040), whereas the mean difference in cost is $1170.96 (95% CI $750.13-1584.32). The ICER is estimated to be $9,771,520.87 per QALY (95% CI $2,573,139.89-24,407,017.82). The proportion of early-onset GBS that develops during the second 24 hours of life, the cost of 24 hours of inpatient observation, and the probability of long-term sequelae following prompt versus delayed treatment play important roles in determining the cost-effectiveness of delayed hospital discharge.

Conclusion: Cost-effectiveness analysis suggests that with adequate IAP, discharging asymptomatic term neonates to home after 24 hours is the preferred approach compared with 48 hours inpatient observation.