Background:Prevention of residual cases of neonatal group B streptococcus (GBS) early-onset disease (EOGBS) has become a goal in the past decade. This study is aimed at evaluating changes in the incidence of EOGBS over a 9-year period after the implementation of a screening-based approach and comparing 2 different protocols for managing healthy-appearing at-risk newborns (ARNs). Methods:A screening-based strategy was introduced in Emilia-Romagna (Italy) in 2003. A prospective, cohort study was conducted from 2003 to 2011; culture-proven EOGBS cases were analyzed in 2 periods: period 1 (2003 to 2008) and period 2 (2009 to 2011). ARNs (>35 weeks’ gestation) were managed according to 2 different protocols: laboratory testing plus observation (period 1) was replaced with expectant observation alone (period 2). Results:Ninety-one EOGBS cases were observed (incidence rate: 0.26/1000 live births). The incidence in full-term babies declined from 0.30 (period 1) to 0.14/1000 live births (period 2, P = 0.04). Recto-vaginal screening cultures in full-term mothers increased significantly from 10/45 (period 1) to 10/14 (period 2, P = 0.002). EOGBS was diagnosed earlier in ARNs than in not-at-risk newborns (mean age 5.5 versus 14.5 hours, P = 0.007). There were no differences in age at diagnosis irrespective of whether ARNs were managed with laboratory testing plus observation (mean 3.5 hours, period 1) or with expectant observation alone (mean 2.4 hours, period 2). Conclusions:When screening cultures were handled according to standard protocols, cases of EOGBS in full-term newborns simultaneously decreased. ARNs were diagnosed in a timely manner through both strategies. The clinical yield of laboratory testing was negligible. © 2013 Lippincott Williams & Wilkins, Inc.