Use of Syndromic Surveillance to Identify Areas at Risk for HCV and HIV Transmission through Injection Drug Use

Tuesday, June 21, 2016: 10:54 AM
Tikahtnu A, Dena'ina Convention Center
Lauren J. Stockman , Wisconsin Division of Public Health, Madison, WI
Ashley R. Bergeron , Wisconsin Division of Public Health, Madison, WI
BACKGROUND: Substantial evidence supports the link between paraphernalia used to inject drugs and transmission of the hepatitis C (HCV) and human immunodeficiency viruses (HIV).  National surveillance has revealed an emerging epidemic of HCV infection among young, non-urban people in communities mostly east of the Mississippi River.  In 2015, HIV infections spread rapidly within a network of people who injected drugs in rural Indiana.  Surveillance for indicators of injection drug use (IDU) may help identify areas at increased risk for HCV and HIV transmission. 

METHODS: Wisconsin Division of Public Health participates in the Centers for Disease Control and Prevention’s National Syndromic Surveillance Program, BioSense Platform.  BioSense is a public health surveillance system that collects information on individual emergency department (ED) visits, hospitalizations and ambulatory visits. During 2015, BioSense received data from more than 80% of Wisconsin hospitals. A word search algorithm was developed to search free-text chief complaint data from ED visits to find encounters related to IDU. Results of the algorithm were validated by comparing trends to heroin-related visits identified by specific ICD-9 codes (965.01 or e-code E850.0) in the Wisconsin ED dataset. Variation in the number of visits over time was evaluated at a county level using geographic information software, ArcGIS.

RESULTS: The BioSense query identified 4,339 visits from Wisconsin hospitals related to IDU from August 1, 2014 to October 31, 2015. Visits were related to opioid detoxification, opioid withdrawal symptoms, acute opioid overdoses, bacterial skin infections and non-specific syndromes where heroin use was noted by medical staff.  Monthly activity monitored by BioSense was available for analysis in real-time, up to eight months earlier than the Wisconsin ED dataset. BioSense-monitored activity visually correlated with the ED visits coded as heroin-associated by ICD-9 codes. The BioSense query identified an increase in IDU-related ED visits over the study time period in 29 of 72 Wisconsin counties.  

CONCLUSIONS: These findings suggest that the BioSense data system may be useful to identify trends and patterns in injection drug activity within Wisconsin and at a local level.  Wisconsin Division of Public Health can use BioSense to identify counties at risk for HCV and HIV transmission through IDU on a very timely basis and strengthen programs for IDU harm reduction in these areas.