Objective To compare 2 approaches in the management of neonates at risk for group B Streptococcus early-onset sepsis: laboratory tests plus standardized physical examination and standardized physical examination alone. Study design Prospective, sequential study over 2 consecutive 12-month periods, carried out in the maternity hospitals of the region Friuli-Venezia Giulia (north-eastern Italy). All term infants were included (7628 in the first period, 7611 in the second). In the first period, complete blood count and blood culture were required for all infants at risk, followed by a 48-hour period of observation with a standardized physical examination. In the second period, only standardized physical examination was performed. Study outcomes were: (1) number of neonates treated with antibiotics; and (2) time between onset of signs of possible sepsis and beginning of treatment. Results There was no difference between the 2 periods in the rate of maternal colonization (19.7% vs 19.8%, P = .8), or in other risk factors. The interval between onset of signs of sepsis and starting of antibiotics was not different in the 2 periods. Significantly fewer infants were treated with antibiotics in the second period (0.5% vs 1.2%, P < .001). Conclusions Laboratory tests together with standardized physical examination seem to offer no advantage over standardized physical examination alone; the latter was associated with fewer antibiotic treatments. Our results are in agreement with the Center for Disease Control and Prevention's 2010 recommendations.