Reducing Legionella colonization in water systems with monochloramine.

Monochloramine disinfection of municipal water supplies is associated with decreased risk for Legionnaires' disease. We conducted a 2-year, prospective, environmental study to evaluate whether converting from chlorine to monochloramine for water disinfection would decrease Legionella colonization of hot water systems. Water and biofilm samples from 53 buildings were collected for Legionella culture during 6 intervals. Prevalence ratios (PRs) comparing Legionella colonization before and after monochloramine disinfection were adjusted for water system characteristics. Legionella colonized 60% of the hot water systems before monochloramine versus 4% after conversion (PR 0.07, 95% confidence interval 0.03-0.16). The median number of colonized sites per building decreased with monochloramine disinfection. Increased prevalence of Legionella colonization was associated with water heater temperatures <50 degrees C, buildings taller than 10 stories, and interruptions in water service. Increasing use of monochloramine in water supplies throughout the United States may reduce Legionella transmission and incidence of Legionnaires' disease.

Aggregated antibiograms and monitoring of drug-resistant Streptococcus pneumoniae.

Community-specific antimicrobial susceptibility data may help monitor trends among drug-resistant Streptococcus pneumoniae and guide empiric therapy. Because active, population-based surveillance for invasive pneumococcal disease is accurate but resource intensive, we compared the proportion of penicillin-nonsusceptible isolates obtained from existing antibiograms, a less expensive system, to that obtained from 1 year of active surveillance for Georgia, Tennessee, California, Minnesota, Oregon, Maryland, Connecticut, and New York. For all sites, proportions of penicillin-nonsusceptible isolates from antibiograms were within 10 percentage points (median 3.65) of those from invasive-only isolates obtained through active surveillance. Only 23% of antibiograms distinguished between isolates intermediate and resistant to penicillin; 63% and 57% included susceptibility results for erythromycin and extended-spectrum cephalosporins, respectively. Aggregating existing hospital antibiograms is a simple and relatively accurate way to estimate local prevalence of penicillin-nonsusceptible pneumococcus; however, antibiograms offer limited data on isolates with intermediate and high-level penicillin resistance and isolates resistant to other agents.