Perceived risk factors and risk pathways of Rift Valley fever in cattle in Ijara district , Kenya

Dates: Received: 27 Feb. 2014 Accepted: 21 Aug. 2014 Published: 20 Nov. 2014 Ijara district in Kenya was one of the hotspots of Rift Valley fever (RVF) during the 2006/2007 outbreak, which led to human and animal deaths causing major economic losses. The main constraint for the control and prevention of RVF is inadequate knowledge of the risk factors for its occurrence and maintenance. This study was aimed at understanding the perceived risk factors and risk pathways of RVF in cattle in Ijara to enable the development of improved community-based disease surveillance, prediction, control and prevention. A cross-sectional study was carried out from September 2012 to June 2013. Thirty-one key informant interviews were conducted with relevant stakeholders to determine the local pastoralists’ understanding of risk factors and risk pathways of RVF in cattle in Ijara district. All the key informants perceived the presence of high numbers of mosquitoes and large numbers of cattle to be the most important risk factors contributing to the occurrence of RVF in cattle in Ijara. Key informants classified high rainfall as the most important (12/31) to an important (19/31) risk factor. The main risk pathways were infected mosquitoes that bite cattle whilst grazing and at watering points as well as close contact between domestic animals and wildlife. The likelihood of contamination of the environment as a result of poor handling of carcasses and aborted foetuses during RVF outbreaks was not considered an important pathway. There is therefore a need to conduct regular participatory community awareness sessions on handling of animal carcasses in terms of preparedness, prevention and control of any possible RVF epizootics. Additionally, monitoring of environmental conditions to detect enhanced rainfall and flooding should be prioritised for preparedness.


Introduction
Rift Valley fever (RVF) is a viral disease caused by Rift Valley fever virus (RVFV), a member of the genus Phlebovirus, family Bunyaviridae.It is a mosquito-borne viral zoonosis that periodically causes disease outbreaks in humans and livestock (Davies, Linthicum & James 1985).The virus is transmitted to humans through direct contact with tissues and blood of infected animals as well as bites from infected mosquitoes.The disease has been endemic in sub-Saharan Africa since 1912 (Peters & Linthicum 1994).Periodic outbreaks have been reported in many African countries, Saudi Arabia and Yemen (Al-Afaleg et al. 2003; Centers for Disease Control and Prevention [CDC] 2007 ;Gerdes 2004;Woods & Karpati 2002).
Outbreaks of RVF in Ijara district have been associated with unusually heavy rainfall that causes extensive flooding of basins and low-lying grassland depressions called dambos, triggering mass emergence of Aedes mosquitoes (Anyangu et al. 2010).In 1997In /1998In and 2006In /2007, massive outbreaks of RVF occurred in East Africa, both associated with El Niño events (CDC 2007;Woods & Karpati 2002), with an estimated 27 500 human cases and more than 600 deaths reported in 1997/1998 in Kenya alone.Historical outbreaks of RVF 'since the early 1950s have been associated with cyclical patterns of the El Niño/Southern Oscillation (ENSO) phenomenon, which results in elevated and widespread rainfall over the RVF-endemic areas of Africa' (Anyamba et al. 2010:43).The two outbreaks (in 1997/1998 and 2006/2007) were the most notable in terms of public health and socio-economic impact (Murithi et al. 2010).
Rift Valley fever virus has distinct endemic and epidemic cycles.During the endemic cycle, the virus persists in inter-epidemic periods through vertical transmission in Aedes mosquito eggs (Davies et al. 1985).Flooding of mosquito habitats can introduce RVFV into domestic animal populations by the emergence of vertically infected Aedes mosquitoes.Epidemic cycles are driven by the subsequent expansion of various Culex mosquito populations, which serve as excellent secondary vectors if immature mosquito habitats remain flooded long enough (Anyamba et al. http://www.ojvr.org2010).Apart from the mosquito transmission to domestic animals and humans during the epidemic cycle, aerosols and contact between infected animals and humans can transmit the virus to humans (Anyamba et al. 2010).The virus is amplified in people and animals.Flat topography, presence of water-retaining soil types and dense bush cover are important factors for flooding and/or mosquitoes breeding (Anyangu et al. 2010).
Risk pathway analysis involves investigation of the possibility of entry, release, exposure and eventual consequences of the disease.It helps establish the routes the disease follows for possible entry, establishment and spread (Breiman et al. 2010).Factors such as vectors, hosts, animal movement patterns and the role of wildlife are used to estimate the probability of occurrence of the disease.In the USA, risk pathway analysis for emerging and re-emerging diseases has been adopted in order to establish exact areas to target for better control and prevention (Kasari et al. 2008).Breiman et al. (2010) documented that RVFV can enter a new area through infected mosquitoes and their eggs, infected livestock, infected wildlife, infected humans and smuggling (terrorist activity).The virus can then spread through mosquito bites, environmental contamination by infected carcasses or aborted foetuses and movement of infected animals and humans, leading to massive losses of lives, livelihoods and trade.About 90% of the population of Ijara is directly dependent on livestock for food and as a source of income.During the last outbreak, a ban on livestock trade and imposition of quarantine resulted in severe economic losses of more than $9.3 million (Murithi et al. 2010).In Ijara district, cattle are driven over long distances towards the Tana River Delta or into Boni forest, passing through various ecosystems.
Ijara district 1 was one of the hotspots during the last RVF outbreaks in the arid and semi-arid North-Eastern province of Kenya (Nguku et al. 2010).The nomadic and semi-nomadic pastoralist communities maintain large livestock herds even in circumstances of limited pasture and water.Outbreaks of RVF have caused major disruptions to public health and the economic mainstay of this population.The movement of these viruses amongst animals and vectors with occasional involvement of human populations under the influence of environmental factors required further study.
The study used key informant interviews to investigate the perceived risk factors and risk pathways associated with RVF in cattle along the livestock movement corridors in Ijara.It was part of a bigger project whose overall objective was to bring about a better understanding of the environmental, biotic and socio-economic drivers of the emergence of RVF and other arboviruses and the viable control options in the 1.Ijara district is amongst the four districts (Lujis, Lagdera, Fafi and Ijara) of Garissa County after the promulgation of the new constitution in Kenya in 2010.Ijara district was carved out of Garissa district (the region of current Garissa County was originally called Garissa district) even before the new constitution.Many publications still use Garissa district to mean Garissa County including Ijara district.However, some publications differentiate between Ijara and Garissa district.In this publication, Garissa district refers to the district bordering Ijara district to the north (see Figure 1).
arid and semi-arid North-Eastern province of Kenya, with a focus on Ijara district, a major hotspot of the disease.The information from this study can be used to create awareness as well as to formulate prevention and mitigation measures for RVF.

Study area
The study was carried out in the arid and semi-arid region of North-Eastern Kenya, with Ijara district at the centre of the study.Ijara district is bordered by Garissa and Fafi districts to the north, Somalia to the east, Lamu district to the south and Tana River district to the west (Figure 1).It lies at approximately 33°E, 6°N and 43°E, 5°S and is devoid of mountains.It is characterised by low, undulating plains at an altitude ranging between 0 m.a.s.l. and 90 m.a.s.l.

Sampling method
A cross-sectional study of the perceived risk factors and risk pathways of RVF for cattle in Ijara was carried out between August 2012 and June 2013.Purposive sampling was used to select 31 individuals who were later interviewed in order to describe the perceived risk factors and risk pathways for RVF in cattle as understood by the locals (Chambers 1981(Chambers , 1994;;World Bank 2004).Targeted stakeholders for 31 key informant interviews included local leaders, veterinary officers, animal health assistants, community-based animal health workers, Kenya Wildlife Service personnel and local administrative officers.

Key informant interviews
The key informant interviews were used to identify and rank the RVF-associated risk factors and risk pathways as perceived by the local pastoralists in the study area.It is important to note that some key informants were technical experts who may have given their opinion rather than the local communities' perception.The key informants listed the risk factors and used informal ranking, which was later cross-checked by a more formal pair-wise and matrix ranking and scoring.The identified pathways were also qualitatively ranked as high or high importance (4), medium or important (3), low or less important (2) or negligible or not important (1) or high (3), moderate (2) and low (1) in the study area by the respondents.These qualitative rankings had no set criteria and depended on the key informants' responses.

Data analysis
At the end of the data collection process, all information gathered was ranked and qualitative measures of perceived importance of risk factors and pathways summarised to address the objectives of the study.

Qualitative ranking of perceived Rift Valley fever risk factors
According to the respondents, availability of vectors (31/31 = 4), large number of cattle (29/31 = 4; 2/31 = 3) and high rainfall (12/31 = 4; 19/31 = 3) are rated most important and/or important risk factors associated with RVF in Ijara (Table 1).There was varied perception on soil types, dambos, bushy vegetation, wildlife and flat topography as risk factors, ranging from 3 (important) to 1 (not important).All the respondents rated drought as not important as a risk factor associated with RVF.High temperature was also rated as a less important to not important risk factor.

Qualitative ranking of perceived Rift Valley fever entry pathways
As shown in

Qualitative ranking of perceived Rift Valley fever exposure or spread pathways
The perceived exposure or spread risk pathway were bites from infected mosquitoes at livestock watering points, around cattle bomas, in bushy environments coming in contact with cattle as well as the exposure to contaminated pasture and environment by infected aborted foetuses and fluids.Most of the respondents perceived spread of RVFV through mosquitoes' bite to be one of the possible forms of spread, whilst environmental contamination by infected aborted foetuses and fluids was categorised as a medium to low-risk pathway (Table 3 and Figure 3).

Qualitative ranking of perceived Rift Valley fever outbreak consequences
The most highly rated consequences are morbidity (30/31), quarantine (31/31) and ban on trade (31/31).Ratings for abortion, mortality, loss of appetite, reduced production and vaccination consequences ranged between high, medium and low for each consequence (Table 4).

Discussion
The informants rated availability of vectors, large number of cattle and high rainfall highly as perceived risk factors associated with RVF in Ijara.These findings were supported by those of Anyangu et al. (2010), who showed a strong association between severe infections of RVF and handling of a large number of animals, closeness to water sources and mosquitoes in the 2006/2007 outbreak.Respondents' low rating on the role of wildlife in RVF transmission does not correspond with the findings of Evans et al. (2008), who detected RVF antibodies in warthogs, gerenuk, waterbuck and buffalo.
The risk pathway analysis was based on three possible stages: assessment (entry, exposure and consequence), communication and management.There were basically three pathways for consideration for the above analysis in relation to the possible entry of the RVF virus into the study area, transmission and spread of the virus and release of the virus to neighbouring areas.
The entry risk pathways for RVF in Ijara district mentioned by the pastoralists were through infected mosquitoes, infected domestic animals, infected aborted foetuses and fluids and infected wild animals, which were corroborated by the study carried out by Breiman et al. (2010).The perceived exposure or spread risk pathways of infected mosquitoes at livestock watering points, around cattle bomas and in bushy environments coming into contact with cattle as well as the exposure to pasture and environment contaminated by infected aborted foetuses and fluids were indications of the communities' understanding of the risk factors associated with RVF.The likelihood of virus smuggling across the border with Somalia was rated as negligible.It is also important to note that the respondents considered aborted foetuses as a less important pathway in the entry and spread of RVF, a factor which does not correspond with the findings of Anyangu et al. (2010) that aborted foetuses were the single most important factors directly associated with severe RVF infections in humans during the 2006/2007 outbreak, compared to the presence of mosquitoes, water bodies and contact with livestock, which were jointly associated.Table 5 provides a summary of the risk pathway analysis.

Conclusion
Availability of the RVFV vectors (mosquitoes) and hosts (cattle) as well as high rainfall were the main risk factors understood by the locals to enable epidemics of RVF in Ijara.
The main perceived routes of entry, exposure and spread of RVF in Ijara were infected mosquitoes, infected domestic   and wild animals and environmental contamination from poor disposal of infected carcasses.However, the locals did not consider transmission from contaminated environment important.The close association between cattle and wildlife was not perceived to be amongst the pathways of RVF transmission in the study area.The areas of contact with wildlife were bushy grazing areas and watering points.
It is important to note that RVFV is endemic in Ijara district.As a result, the likelihood of its entry into Ijara may not be applicable since it is already present.This explains why the communities' perception about the source and/or spread of RVF is the same, namely the role played by mosquitoes.

Recommendations
Although an RVF decision support tool has been developed for the Horn of Africa (International Livestock Research Institute [ILRI]/Food and Agriculture Organization of the United Nations [FAO] 2009), it seems it is not actually implemented in Ijara.Furthermore, management of cattle and wildlife carcasses would cause a big loss to the pastoralists in Ijara in cases of RVF outbreak given their current perception of their being a low-risk factor.There is therefore a need to implement the RVF decision support tool, especially education and creation of awareness on proper

Endemic Cycle
Virus persists during dry season through vertical transmission in Aedes mosquito eggs

Epidemic Cycle
Flooding  ways of handling both domestic and wild animal carcasses to mitigate or prevent the possibility of RVFV transmission.Awareness should also include the likelihood of RVFV transmission between domestic and wild animals by limiting  wild-domestic animal interaction to a minimum.On the other hand, strategic vaccination, if carried out, can help protect the animals and reduce losses in case of an outbreak.
Continuous surveillance of RVF in Ijara in domestic animals, wildlife, humans and vectors as well as environmental monitoring of rainfall pattern should be carried out together with all the stakeholders to provide early warning signs that can help in preparedness, control and prevention of the occurrence of RVF.

Figure 2
Figure 2 is a map of Kenya showing the areas where RVF outbreaks occurred and their relationship with the study area.The study area (Figure1) falls in ecological Zone V-IV with a total forest cover of 2484 km².Boni forest, which is an indigenous open canopy forest that forms part of the northern Zanzibar-Inhamdare coastal Forest Mosaic, covers a major part of the study area.A section of the forest, the Boni National Reserve, is under the management of the Kenya Wildlife Service as a protected conservation area.The soil types are black cotton and alluvial types, temperatures range between 15 ºC and 38 ºC, bimodal rainfall ranges between 700 mm and 1000 mm per annum, and average relative humidity is 68 mm.

FIGURE 1 :
FIGURE 1: Map of study area showing the relationship between the study area, the surrounding districts and Somalia.
Source: Modified map from International Centre for Insect physiology and Ecology (ICIPE) Earth Observation Unit and Google

FIGURE 2 :
FIGURE 2: Map of Kenya showing Rift Valley fever endemicity and the relationship with the study area.

FIGURE 3 :
FIGURE 3: Endemic (on left) and epidemic (on right) life cycles of Rift Valley fever involving close association between heavy rainfall conditions, vector Aedes and Culex mosquitoes, domestic animals and humans.
Table2, the perceived entry risk pathways for RVF in Ijara district according to the key informants were infected mosquitoes, infected domestic animals, infected aborted foetuses and fluids and infected wild animals.All the respondents were in agreement that infected mosquitoes constituted the most likely entry pathway for RVF (31/31).

TABLE 1 :
Summarised results for qualitative ranking of Rift Valley fever risk factors in cattle as perceived by the pastoralists' key informants in Ijara, Kenya.
†, Numbers of key informants suggesting the ‡, qualitative measure.Score shows perceived degree of importance of Rift Valley fever risk factors: 4 = most important, 3 = important, 2 = less important, 1 = not important.

TABLE 2 :
Summarised results for qualitative ranking of Rift Valley fever entry risk pathway in cattle as perceived by the pastoralists' key informants in Ijara, Kenya.

TABLE 3 :
Summarised results for qualitative ranking of Rift Valley fever exposure risk pathway in cattle as perceived by the pastoralists' key informants in Ijara, Kenya.
†, Numbers of key informants suggesting the ‡, qualitative measure.Score shows perceived degree of importance of Rift Valley fever exposure risk pathway: 3 = high, 2 = medium, 1 = low.

TABLE 4 :
Summarised results for qualitative ranking of Rift Valley fever outbreak consequences in cattle as perceived by the pastoralists' key informants in Ijara, Kenya.

TABLE 5 :
A summary discussion table of risk pathway analysis for Rift Valley fever in cattle in Ijara, Kenya.