The use of clindamycin and intravenous immunoglobulin (IVIG) in treatment of invasive group A streptococcal (iGAS) infection, and the need for prophylactic antibiotics in close contacts, remains contentious. Controlled trials are unlikely to be conducted, so prospective, observational studies provide the best data to inform practice. We conducted population-based, prospective, active surveillance of iGAS infections throughout the state of Victoria, Australia (population 4.9 million), from March 2002 through August 2004. Eighty-four cases of severe iGAS infection (streptococcal toxic shock syndrome, necrotizing fasciitis, septic shock, or GAS cellulitis with shock) were identified. Clindamycin-treated patients had more severe disease than clindamycin-untreated patients but lower mortality (15% vs 39%; odds ratio [OR], 0.28; 95% confidence interval [CI], .10-.80). Among those who received concurrent IVIG, the fatality rate was lower still (7%). The adjusted point estimate of the OR for mortality was lower in clindamycin-treated patients (0.31; 95% CI, .09-1.12) and clindamycin plus IVIG-treated patients (0.12; 95% CI, .01-1.29) compared with clindamycin-untreated patients. Three confirmed cases of iGAS infection occurred in household contacts of index cases. The incidence rate of iGAS disease in contacts was 2011 (95% CI, 413-5929) times higher than the population incidence in Victoria. Our data suggest that clindamycin treatment of patients with severe iGAS infections substantially reduces mortality and that this effect may be enhanced by concurrent treatment with IVIG. The dramatically increased risk of iGAS disease among household contacts within 1 month of the index case highlights a potential role for antibiotic prophylaxis.