Background: Reports of ocular syphilis (OS) have increased in North Carolina since 2014. Absence of routine surveillance for OS limits epidemiologic understanding of the condition. As part of syphilis surveillance, Disease Investigation Specialists (DIS) attempt to interview all residents with syphilis and document information in the NC electronic disease surveillance system (NCEDSS). Before February 2015, DIS documented reported ocular symptoms in text fields, but did not conduct systematic surveillance. In February 2015, DIS were instructed to ask about and standardize documentation of eye symptoms by checking a pre-existing eye/conjunctiva box (ECB) in NCEDSS. We evaluated whether this was an effective surveillance strategy to identify OS cases.
Methods: We identified syphilis cases during January 1, 2014–December 31, 2015 with the ECB checked and conducted a free-text search within NCEDSS to identify reported eye symptoms among syphilis cases without the ECB checked. We reviewed medical records for all identified patients to determine if they met the OS case definition—confirmed syphilis case and symptoms of ocular disease with no other defined etiology. We calculated sensitivity, specificity, and positive predictive value (PPV) of the ECB before and after February 2015.
Results: Of the 4234 syphilis cases reported during 2014–2015, 63 (1.5%) met the OS case definition (53 identified by ECB, 10 by free-text search). The OS prevalence, before and after February 1, 2015, was 1.1% (22/2025) and 1.8% (41/2209), respectively. Both periods had 100% specificity and PPV, but the ECB’s sensitivity for identifying OS cases increased from 59% before to 98% after February 1, 2015.
Conclusions: Routinizing use of a checkbox in NCEDSS to detect patients reporting eye symptoms was a successful and low-resource approach to identifying OS cases. By utilizing existing infrastructure for syphilis prevention and control, we were able to effectively and rapidly implement surveillance for this health threat.