BACKGROUND: Starting in December 2015, Angola experienced an unprecedented outbreak of yellow fever (YF). Early in the outbreak, many suspected cases were reported, but true burden of YF was unclear as clinical diagnosis relied on nonspecific symptoms of fever and jaundice. Accurate YF diagnosis is challenging in settings such as Angola, where endemic diseases have similar clinical presentation (e.g., malaria) and there is limited laboratory capacity. To more fully understand true YF burden in Angola, we evaluated clinically-diagnosed YF deaths to assess feasibility of excluding patients positive for other infections from the surveillance case definition.
METHODS: In March 2016, we abstracted charts for clinically-diagnosed YF deaths from 5 tertiary hospitals in Luanda Province to determine if the case met World Health Organization’s (WHO) suspected YF case definition of fever and jaundice <14 days of symptom onset. We cross-referenced infectious disease testing results in hospital and Ministry of Health laboratory databases.
RESULTS: Of all 101 clinically-diagnosed YF deaths, 86 (85%) met WHO’s suspected YF case definition. Only 9 of 86 patients (10%) had YF testing; 3 (33%) were confirmed positive. Forty-eight patients were tested for malaria by smear or rapid diagnostic test; 26 (54%) were positive for malaria, and 2 were also Widal positive for typhoid. Two of 3 laboratory-confirmed YF patients were also positive for malaria.
CONCLUSIONS: Applying a more specific case definition excluding deaths laboratory-positive for other infections would have excluded confirmed YF cases from consideration as suspected cases. YF deaths with coinfections would also be missed. To capture all possible YF cases to guide outbreak response, we recommended continued use of existing WHO case definition, improved diagnostic capacity and increased YF testing.