Assessing Sensitivity and Specificity of Surveillance Case Definitions for Zika Virus Disease

Angela Chow; Hanley Ho; Mar-Kyaw Win; Yee-Sin Leo


Emerging Infectious Diseases. 2017;23(4):677-679. 

In This Article

Abstract and Introduction


We evaluated performance of 5 case definitions for Zika virus disease surveillance in a human cohort during an outbreak in Singapore, August 26–September 5, 2016. Because laboratory tests are largely inaccessible, use of case definitions that include rash as a required clinical feature are useful in identifying this disease.


Zika virus infections in humans were first reported in Nigeria, Uganda, and Tanganyika (now Tanzania) in 1951–1952.[1,2] Until 2006, sporadic cases and small clusters of Zika virus infections were reported.[3] In 2007, the first major outbreak occurred on Yap Island, where ≈1/5 infected persons were symptomatic, predominantly with rash, fever, arthralgia, and conjunctivitis.[4] In a recent outbreak in Brazil in 2015, similar signs and symptoms predominated.[5] Rash (67%), fever (64%), arthralgia (29%), myalgia (24%), headache (22%), and conjunctivitis (21%) were the 6 most common signs and symptoms reported during January 1964–February 2016.[3]

Unlike dengue virus (a related flavivirus), Zika virus was not considered to be a major pathogen until recent reports of its association with Guillain-Barré syndrome and microcephaly.[6] Thus, there is little information on the performance of surveillance case definitions for detection of Zika virus disease.

Responding to the rapidly evolving Zika virus epidemic to guide surveillance for Zika virus disease, the US Centers for Disease Control and Prevention worked with the Council of State and Territorial Epidemiologists (CSTE) to approve an interim definition in February 2016 and a final case definition in June 2016 for noncongenital Zika virus disease as >1 of the following signs or symptoms: acute onset of fever, maculopapular rash, arthralgia, and conjunctivitis.[7] The interim case definition (February 2016) of the World Health Organization (WHO) for suspected Zika virus disease includes rash or fever and >1 of the following signs or symptoms: arthralgia, arthritis, and conjunctivitis (nonpurulent/hyperemic).[8] The case definition of the European Centre for Disease Prevention and Control (ECDC) includes rash and optional symptoms in the WHO definition plus myalgia.[9] The case definition of the Pan American Health Organization (PAHO) includes rash and ≥2 of the following signs or symptoms: fever, conjunctivitis (nonpurulent/hyperemic), arthralgia, myalgia, and periarticular edema.[10]

The first outbreak of Zika virus disease in Singapore occurred in August 2016.[11] Singapore is a densely populated tropical country to which dengue fever is endemic. With the identification of the first local case of Zika virus disease, the Singapore Ministry of Health (MOH) initiated active case finding.[12,13] The MOH recommended Zika virus screening for persons with fever and maculopapular rash, and 1 of the following: arthralgia, myalgia, headache, and nonpurulent conjunctivitis.

Clinical criteria for disease surveillance are a balancing act for satisfying 2 potentially conflicting needs: sensitivity and specificity. A more sensitive case definition will identify a larger proportion of true cases, but at the cost of finding a large number of cases from other causes. In comparison, a more specific case definition will provide a more accurate description of true cases, but at the expense of missing true cases.[14]