Sexually transmitted infections (STIs) spread along sexual networks whose structural characteristics promote transmission that routine surveillance may not capture. Cases who have partners from multiple localities may operate as spatial network bridges, thereby facilitating geographical dissemination. We investigated how surveillance, sexual networks, and spatial bridges relate to each other for syphilis outbreaks in rural counties of North Carolina.
We selected from the state health department’s surveillance database cases diagnosed with primary, secondary, or early latent syphilis during October 1998 to December 2002 and who resided in central and southeastern North Carolina, along with their sex partners and their social contacts irrespective of infection status. We applied matching algorithms to eliminate duplicate names and create a unique roster of partnerships from which networks were compiled and graphed. Network members were differentiated by disease status and county of residence.
In the county most affected by the outbreak, densely connected networks indicative of STI outbreaks were consistent with increased incidence and a large case load. In other counties, the case loads were low with fluctuating incidence, but network structures suggested the presence of outbreaks. In a county with stable, low incidence and a high number of cases, the networks were sparse and dendritic, indicative of endemic spread. Outbreak counties exhibited densely connected networks within well-defined geographic boundaries and low connectivity between counties; spatial bridges did not seem to facilitate transmission.
Simple visualization of sexual networks can provide key information to identify communities most in need of resources for outbreak investigation and disease control.
From the aSchool of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC; bGillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC; cDalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada; and dQuintus-ential Solutions, Colorado Springs, CO.
Submitted 26 May 2011; accepted 25 November 2011; posted 21 September 2012.
Supported by a grant from the National Institutes of Health grant number 1R01AI067913-01A1.
Correspondence: Irene A. Doherty, 130 Mason Farm Road, CB 7030, Chapel Hill NC, 27599-7030. E-mail: email@example.com.