Outbreak of yellow fever in central and southwestern Uganda, February–may 2016

Yellow fever is an acute viral hemorrhagic disease caused by the yellow fever virus, a single-stranded RNA virus that belongs to the genus Flavivirus [1, 2]. It is transmitted from humans to humans or from animals to humans by Aedes mosquitos [1]. The disease is endemic in tropical areas; in Africa, South America, and Central America [2, 3]. The Global Health Security Agenda (GHSA) encourages all member states to be able to rapidly detect and respond to outbreaks [4].

Yellow fever is considered to be a re-emerging disease due to increasing reports of its occurrence in different parts of the world in the recent years [5]. Worldwide, the number of yellow fever cases has increased over the past 20 years. This might be attributable to multiple factors, including declining population immunity to infection [6], increased human activities such as deforestation, urbanization [7], population movements [8], and climate change [9]. In 2013, the disease affected an estimated 130,000 people and caused about 78,000 deaths in Africa [10]. There is no specific treatment for yellow fever; only supportive treatment is available to manage symptoms. Without treatment, up to 50% of severely affected persons die. An effective yellow fever vaccine is available; a single dose of this vaccine provides life-time protection against the disease [3].

Lying in the so-called “yellow fever belt”, Uganda is one of the 32 African countries at risk of yellow fever transmission [2] . Since the identification of the first outbreak in 1941, several outbreaks have occurred in Uganda [2, 11], the largest of which affected 181 people and resulted in 45 deaths in northern Uganda in 2010 [12].

Since 2000, surveillance for yellow fever in Uganda has been conducted through the Integrated Disease Surveillance and Response (IDSR) strategy [13]. This strategy enables timely detection of and response to outbreaks to prevent further spread. On 26 March 2016, the IDSR focal person in Masaka District, southern Uganda, alerted the Public Health Emergence Operations Center (PHEOC) of the Ministry of Health (MoH) that within a one-month period, three men from the same extended family had died of a “strange disease” with bleeding symptoms. Fearing an outbreak of a viral hemorrhagic fever (VHF), the MoH immediately activated the VHF response plan, established an isolation unit at the Masaka Regional Referral Hospital, and initiated active case-finding. Six blood samples were collected from patients at the isolation unit and tested for Ebola Virus, Marburg Virus, Crimean-Congo Haemorrhagic Fever, and Rift Valley Fever in the Viral Special Pathogen Laboratory at the Uganda Virus Research Institute (UVRI). However, the samples tested negative for all the tested VHFs. Based on the clinical presentation of the patient and the initial laboratory results, the reserved samples were then sent to the Arbovirus laboratory for further testing including yellow fever testing. On 8 April 2016, three samples from Masaka District tested positive for yellow fever by both PCR and IgM antibody tests. On 9 April 2016, MoH declared a yellow fever outbreak and launched an outbreak response. After the declaration, another cluster of cases was reported in Rukungiri District, southwestern Uganda.

As part of the outbreak response, we conducted an epidemiologic investigation to determine the scope of the outbreak, identify risk factors for transmission, and recommend evidence-based measures for outbreak prevention and control.

From the retrospective review of health facility records, the outbreak started from February and ended in May 2016. The total number of people evaluated in this outbreak was 42. We identified 35 probable and 7 confirmed case-patients. Nine additional persons were evaluated and their blood samples were taken for testing; however, they did not meet the probable case definition and their blood samples were negative for yellow fever. Our final data analysis excluded these nine persons, and focused on the probable or confirmed cases.

Our epidemiologic investigation showed that onset of yellow fever during this outbreak was associated with agriculture activities in and near forests and swampy fields, and with presence of sylvatic monkeys in the agriculture fields. Environmental investigation showed presence of sylvatic monkeys, Aedes spp. and other species of mosquitoes involved in yellow fever transmission, and the breeding sites of the mosquitos around farmlands and homesteads of case-patients. The outbreak affected working-age men more than other demographic subgroups, and all case-persons were engaged in some kind of farming activities. Taken together, our investigation indicated that the outbreak probably was sylvatic in nature, and was spread by Aedes mosquitoes in the outbreak areas.

Introduction of yellow fever from sylvatic monkeys to the human population has caused many outbreaks in the past [1]. The areas where the current outbreak occurred are undergoing a massive deforestation in favor of agricultural activities [18], resulting in frequent close contact between monkeys and humans. Sylvatic monkeys are now part of the ecosystem in the outbreak areas, especially in Buwunga and Kebisoni Sub-counties, where monkeys were frequently present around homes and in agriculture fields. Anecdotally, we learned that some local residents keep monkeys as pets and consume monkey meat, although we were unable to substantiate these claims. These human-monkey interactions may greatly increase the risk of human exposure to yellow fever and other pathogens.

The stratified epidemic curve in our investigation appeared to suggest that this outbreak might have started in Masaka District, and moved to the other districts. Masaka District is located on a major highway, and the district town is a commercial center where people traveling by the highway would normally stop to shop and dine. These activities might have facilitated the spread of the disease to other districts. Alternately, the outbreaks could have occurred in different districts spontaneously and independently due to transmission of the virus from sylvatic monkeys to humans via mosquito bites. However, these hypotheses could not be differentiated due to lack of genetic sequencing of the viruses from different clusters. In Uganda, the period between March to May falls in the rainy season, during which the mosquito population would increase exponentially, which might partially explain why this outbreak occurred during these months.

In this investigation, we were able to find confirmed (severe) and probable forms of yellow fever infection. However when it comes to seeking healthcare services, as well as in disease outbreaks, it is those severely affected by the disease that either seek care or are detected by the responders in the community. It is therefore possible that the outbreak affected more than what is reported. Michael Johansson estimates that, a severe case of yellow fever may represent an additional three to twenty asymptomatic or mild infections. This therefore indicates that for any detected case, there is a considerable potential for further infections [19] as explained by the iceberg concept.

Early identification, early diagnosis, and prompt response are critical for the successful control of communicable disease outbreaks and for ensuring global health security. The previous yellow fever outbreak in Uganda occurred in 2010 [12]. The outbreak was identified and reported on 8 November 2010, but laboratory confirmation was not completed until 18 December, which was over 40 days after the outbreak was reported. By the end of January 2011, when the outbreak was contained, 181 cases had occurred (including 13 confirmed cases), and 58 persons had died. In comparison, the current outbreak was confirmed 12 days after the outbreak was reported. Consequently, the case count (42 vs. 181) and death toll (14 vs. 58) were reduced by more than 75%. During the six-year time period between 2010 and 2016, the MoH of Uganda under the auspices of the Global Health Security Agenda has substantially enhanced the capacity for outbreak response, the core components of which includes a robust Public Health Emergency Operations Center, enhanced laboratory testing capabilities, a nationwide network of specimen-referral system, and an advanced-level field epidemiology training program. During the current outbreak, all elements of this enhanced outbreak response system worked seamlessly together, which contributed to the shortened response time, reduced morbidity and mortality, and decreased risk of disease spread.

Vaccination is the most important measure for preventing and containing yellow fever outbreaks [3]. During our investigation, none of the study participants reported a history of vaccination against yellow fever. In Uganda, yellow fever vaccination is individually sought by international travellers as a requirement and a few other people who can afford the vaccination cost of $27. In addition, yellow fever vaccination is not part of the routine immunization. This suggests that the population immunity level is low in Uganda. This was a consequence of yellow fever vaccination not being part of routine immunization program in Uganda. As soon as this outbreak was confirmed, the MoH of Uganda mounted a vaccination campaign in and around the outbreak area from May 19, 2016. We believe that interventions such as aggressive case finding and case management, risk communication in affected communities, setting up toll free communication and referral of suspected case among others worked to end this outbreak and effectively prevented the spread of the disease to other areas. Since Uganda has experienced repeated yellow fever outbreaks in the past, adding the yellow fever vaccine to the routine immunization schedule in Uganda may be warranted.