Group B streptococcal (GBS) infections are the leading cause of bacterial disease and death among newborns in the United States and an important cause of morbidity among peripartum women and nonpregnant adults with chronic medical conditions. Disease in infants usually presents as sepsis, pneumonia, or meningitis but also may include cellulitis or osteomyelitis (1). In 1990, GBS infections caused an estimated 7600 serious illnesses and 310 deaths among US infants aged less than or equal to 90 days; infections among infants aged less than 7 days (ie, early-onset disease) accounted for approximately 80% of these illnesses (2). To determine the incidence of GBS disease during 1993-1995, CDC conducted surveillance for this disease in an aggregate population of 12.5 million persons with 190,000 annual live-born infants. This report summarizes the findings of surveillance in this population, which indicate that a statistically significant decline in the incidence of early-onset GBS disease occurred in some surveillance areas.

Surveillance was conducted in the three-county San Francisco Bay area, California; four urban counties in Tennessee; the eight-county metropolitan area of Atlanta, Georgia; and the entire state of Maryland. At biweekly intervals, surveillance personnel requested standardized reports of cases of invasive GBS disease from contacts in each laboratory that served acute-care hospitals within specified surveillance areas. Periodic audits of all laboratories were conducted to validate completeness of reporting. A case of invasive GBS disease was defined as isolation of group B streptococci from a normally sterile site (eg, blood or cerebrospinal fluid) from a resident of an area under surveillance. Cases were categorized as early-onset and late-onset (illness onset at age 7-90 days). To calculate the incidence of neonatal GBS disease for the surveillance areas, the number of live-born infants for 1993-1995 was obtained from the respective state health departments or from CDC's National Center for Health Statistics. Race-specific data are presented only for blacks and whites because numbers for other racial/ethnic groups were too small for meaningful analysis.