Staphylococcus aureus Colonization and Infection Among Drug Users: Identification of Hidden Networks

摘要

Objectives. We combined social-network analysis and molecular epidemiology to investigate Staphylococcus aureus among drug users. Methods. From 2003 through 2005, we recruited adult drug users in Brooklyn, New York. Of 501 individuals recruited, 485 participated. Participants were screened for HIV infection and S. aureus carriage, and they answered a questionnaire assessing risk factors for S. aureus . Participants were asked to nominate up to 10 members of their social networks, and they were invited to recruit nominees to participate. Results. We identified 89 sociocentric risk networks, 1 of which contained 327 (67%) members. One third of participants were either colonized (20%) or infected (19%) with S. aureus . Overall strain similarity was unusually high, suggesting spread within and across networks. In multivariate analysis, 7 health-related and drug-use variables remained independently associated with infection. Moreover, 27% of nominees were not drug users. Conclusions. We found a large, linked, hidden network among participants, with no discernible clustering of closely related strains. Our results suggest that once a pathogen is introduced into a sociocentric network of active drug users, an identifiable community S. aureus reservoir is likely created, with significant linkages to the general population. There has been a dramatic increase in the number of community-based Staphylococcus aureus infections during the past decade. 1 – 3 This increase in infections extends well beyond outbreak settings (e.g., jails, daycare facilities), suggesting that reservoirs of these strains probably already exist in the community. 4 – 6 Despite considerable research, knowledge of how these epidemic strains of S. aureus spread and become established in different settings, including community settings, remains limited. 4 , 7 , 8 Understanding these transmission pathways is increasingly important to the prevention and control of S. aureus infections, particularly those that are methicillin-resistant. 2 , 4 , 9 Traditional epidemiological survey methods often used in health care settings have identified a number of factors, such as nasal colonization, as risks for subsequent S. aureus infection. It is unclear whether these same health care–associated risk factors are relevant for infections that originate in the community. 10 – 13 Community-based groups at high risk for S. aureus infection, including HIV-infected individuals and injection drug users, are often hidden and are therefore difficult to access by means of these methods. Difficulty in accessing certain subgroups of the population has highlighted the need for alternative research strategies. An approach that integrates microbiology with social-network methodologies at the community level has been used effectively to study infection transmission dynamics for pathogens as varied as HIV, sexually transmitted infections, hepatitis C virus, tuberculosis, and meningococcal disease. 14 – 24 When these social-network methods are used to understand infectious disease transmission in community settings, research finds that the degree and type of contact between infected and uninfected individuals play a significant role in disease prevalence. Therefore, it is logical to use social-network techniques in community settings to explore the prevalence and distribution of S. aureus colonization and the diversity of strains. On the basis of previous research with infectious pathogens, we hypothesized that a significant minority of this population would be either colonized or infected with S. aureus and that related strains of S. aureus would cluster within network components. 4 , 7 , 10 – 15 The current study integrated social-network methodologies with molecular epidemiology to investigate S. aureus carriage and infection among a large sample of active drug users in a well-defined community. We used this combined approach to examine: (1) prevalence of S. aureus colonization and recent drug use–associated infection, (2) risk factors associated with colonization and recent infection, and (3) diversity and distribution of S. aureus strains within and across networks in a high-risk population of drug users in central Brooklyn, New York.