To the Editor in Chief:

We note the editorial by Dr Behrens (“Yellow fever recommendations for tourists to Kenya: a flawed risk assessment,” J Travel Med 2008; 15:285–286), in which he states that “… there is no current or foreseeable risk of infection (with yellow fever)” in Kenya and suggests that the risk of serious adverse events from yellow fever (YF) vaccine outweighs the benefits of vaccination. Dr Behrens raises an important point.

The World Health Organization, with a panel of experts including the authors of this letter, is currently assessing the risk of YF country by country. The panel concluded that western and northern Kenya remain endemic for YF.

YF is enzootic/endemic or is periodically introduced and persists for years in western Kenya. Two cases were described prior to the 1992 to 1993 outbreak in Kenya: in 1942 (Kitale) and in 1943 (Langata Forest Reserve, west of Nairobi). Serologic surveys in the 1940s showed evidence of infections in western Kenya.1 Following the 1992 to 1993 outbreak, 59 and 73 suspected cases were identified in 1994 and 1995, respectively, and cases occurred outside the epidemic zone.2

Kenya shares borders with Uganda, Sudan, and Ethiopia, where YF is endemic/epidemic. In 2003, a major epidemic occurred in the Equatoria Province in a bordering area of Sudan, and in 1966, an epidemic occurred in the Gamo Gofa Province of Ethiopia near the Kenyan border. Serologic studies in the Northern Frontier District of Kenya following the 1966 outbreak revealed recent YF transmission.3 Ten years later, YF transmission was found between monkeys in Ethiopia,4 emphasizing that the absence of human case reports does not imply the absence of sylvatic virus transmission.

Dr Behrens’ assumption that successful identification of a recent Rift Valley Fever (RVF) outbreak indicates that current surveillance for YF is effective and may not be accurate. RVF causes disease in domestic livestock, whereas no such sentinel of zoonotic YF exists in Africa. Human surveillance for YF in Kenya is passive, using a case definition of acute fever and jaundice (Dr Charles Nzioka, personal communication, Ministry of Health, Kenya, 2008). This surveillance system could easily miss mild or sporadic cases of YF, which might be the only indications of ongoing transmission.

The ecological setting in which YF is transmitted has not changed with the passage of time, and human case surveillance is insensitive and made less efficient by human vaccinations. For these reasons, we believe the evidence indicates that travel, particularly when there is extensive outdoor exposure to sylvatic Aedes vector mosquitoes in western and northern Kenya, engenders a risk of YF infection.


  1. Top of page
  2. References
  • 1
    Bonnel PH, Deutschman Z. La fièvre jaune en Afrique au cours des années récentes. Bull World Health Organ 1957; 11:325.
  • 2
    Sanders EJ, Borus P, Ademba G, et al. Sentinel surveillance for yellow fever in Kenya, 1993 to 1995. Emerg Infect Dis 1996; 2:236.
  • 3
    Henderson BE, Metselaar D, Cahill K, et al. Yellow fever immunity surveys in northern Uganda and Kenya and Eastern Somalia, 1966-67. Bull World Health Organ 1968; 38:229.
  • 4
    Wood OL, Lee VH. Yellow fever epidemic sites revisited—possible areas of yellow fever endemnicity in Ethiopia. Ethiop Med J 1975; 13:177179.