An Epidemiological Evaluation of the use of Microbiological Tools for Identifying Gonorrhoea Infection Networks

Abstract

We aimed to assess the utility of various techniques for identifying gonorrhoea infection networks. All residents of a non-metropolitan North Carolina county visiting a sexually transmitted disease (STD) clinic during a 17-month period were screened for gonorrhoea. Infection networks were estimated by serovar type combined with antibiotic resistance, arbitrarily primed polymerase chain reaction (AP-PCR), or temporal clustering. The residential addresses of infected patients were geocoded and mapped. Among 2 serovar types, the presence of distinguishing characteristics of a network, based on questionnaire data, was assessed with prevalence ratios and 95% confidence intervals (CIs) relative to those not in the network. Twenty-five serovar types were identified among 759 gonorrhoea infections. In one serovar, the networks further delineated by temporal clusters correlated with particular AP-PCR types. In most instances, however, different typing techniques painted different network pictures. No refined serovar network stood out as having a particular set of characteristics that could be used to shape intervention. Teasing out an individual infection network with unique characteristics will require the development and use of other microbiological tools.