You are on page 1of 21

Preventive Veterinary Medicine 78 (2007) 296316

www.elsevier.com/locate/prevetmed

Maasai perception of the impact and incidence


of malignant catarrhal fever (MCF) in
southern Kenya
Claire Bedelian a,*, David Nkedianye a,b, Mario Herrero a,b
a
International Livestock Research Institute (ILRI), P.O. Box 30709, Nairobi 00100, Kenya
b
Institute of Atmospheric and Environmental Sciences, The University of Edinburgh,
West Mains Road, Edinburgh EH9 3J6, Scotland, UK
Received 12 January 2006; received in revised form 19 October 2006; accepted 24 October 2006

Abstract

We investigated the perceived impact of malignant catarrhal fever (MCF) to pastoralists in Isinya
Division, a wildlife dispersal area of Nairobi National Park, and used a range of participatory
epidemiology methodologies. We compared the relative importance, incidence and impact of MCF
compared to other locally defined important diseases with a total of 158 respondents in 11 group
meetings and 21 household meetings in July 2004. Direct losses due to disease were investigated
through lowered prices as a result of the emergency sale of disease-infected animals.
Overall, Maasai in Isinya Division perceived east coast fever (ECF) to be the most important cattle
disease and to have the highest incidence. Anthrax was considered to have the largest impact. In areas
within or adjacent to the wildebeest calving zone, MCF was perceived to be the most important cattle
disease and also to have the largest impact. Outside the calving zone, MCF was considered the fourth-
most important disease with the fourth largest impact, and these were areas where wildebeest were
less common. MCF was also the fourth-most common disease, and across the Division incidence was
estimated at 5% in calves and 10% in adults. However, MCF incidence varied greatly throughout the
study area, from 3% to 12%, and the highest incidence risks were found in areas where wildebeest
came to calve. The percent drop in sale price per animal infected with MCF was estimated at 50% for
MCF for the year 20032004.
Forced avoidance movements away from wildebeest calves were reported to decrease livestock
production due to loss of access to prime grazing sites. As suggested by pastoralists in this study, the

* Corresponding author at: International Livestock Research Institute (ILRI), P.O. Box 30709, Nairobi 00100,
Kenya. Tel.: +254 20 4223000; fax: +254 20 4223001.
E-mail address: c.bedelian@cgiar.org (C. Bedelian).

0167-5877/$ see front matter # 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.prevetmed.2006.10.012
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 297

development of compensation schemes or incentives from wildlife would reduce the conflict between
livestock keeping and wildlife conservation.
# 2006 Elsevier B.V. All rights reserved.

Keywords: Malignant catarrhal fever; Maasai pastoralists; Participatory appraisal; Participatory epidemiology;
Disease impact; Disease incidence; Kenya

1. Introduction

In pastoral lands, conflicts can arise between livestock and wildlife due to competition
for key resources, predation and infection transmission. This can be most noticeably seen
in areas surrounding protected areas, where livestock and wildlife often mix freely.
Malignant catarrhal fever (MCF) is a wildlife-transmitted infection passed from wildebeest
to cattle following the wildebeest calving period. All age classes of cattle are susceptible to
MCF infection, although studies have shown a higher incidence in adults, and in particular,
in periparturient females (Barnard et al., 1994). MCF primarily affects the lymphoid organs
and the respiratory and alimentary tracts (Reid et al., 1984), and is caused by one or two
viruses belonging to the gammaherpesvirnae subfamily. MCF is characterized in cattle by
profuse nasal and ocular discharges, keraconjunctivitis with corneal opacity, and
enlargement of peripheral lymph nodes (Barnard et al., 1994). MCF is a fatal disease of
cattle, and with no available vaccine or treatment the only reliable method of control is to
separate wildebeest and cattle during the infective period. As a result, Maasai pastoralists
in Kenya and Tanzania are excluded from prime grazing sites in the wet season. This can
have serious consequences to herd productivity (Homewood et al., 1987) when a range of
pasture types can be critical in establishing body condition at the end of the dry season
(Swift, 1983).
MCF is of serious concern to pastoralists in Maasailand. In Kajiado District in Kenya,
MCF was consistently cited as one of the five most-important problems along with east
coast fever (ECF), contagious bovine pleuropneumonia (CBPP), foot-and-mouth disease
(FMD), and anthrax (Boone and Coughenour, 2001). In a study of Group Ranches in
Kajiado and Narok Districts, Kenya, MCF was ranked as the most important and third-most
important disease, respectively (Ngotho et al., 1999a). Maasai in Ngorongoro District,
Tanzania, ranked MCF as the second-most important disease after ECF; reasons were cattle
deaths and no available treatment (Cleaveland et al., 2001). However, MCF has not been
regarded by scientists as one of the top 20 diseases according to its impact on the poor in
Eastern, Central and Southern Africa (Perry et al., 2002). Maasai believe the problem of
MCF has been ignored by government and donors over many years and feel resentment at
the lack of research on methods for MCF control (Cleaveland et al., 2001).
In the Maasai pastoral areas of Kenya and Tanzania, incidence of MCF has been
estimated to vary from 1% to 21%. However, literature on the incidence of MCF is lacking,
with only one reported study in Kenya (Ngotho et al., 1999b), and three in Tanzania
(Plowright et al., 1975; Kalunda et al., 1982; Cleaveland et al., 2001) in the last 30 years.
Added to this, previous investigations have not always been very representative because
often they are only implemented in times of an outbreak (Kalunda et al., 1982). Incidence
298 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

rates can also be difficult to estimate because few cases are reported to Veterinary
Departments, because preventative and curative treatments are not available.
In the most recent study in Ngorongoro District, Tanzania, Cleaveland et al. (2001)
reported incidence rates of 5.66.2% in high-risk villages, and direct losses due to death,
emergency sales and slaughter, and lost milk production, ranging from 50,000 to
150,000 TSh (equivalent of 42125 at year of study) per case. In the Kenyan study, direct
losses resulted in a loss of herd value between 4.5% and 8.3% in low-incidence areas, and
between 13.6% and 24.2% in high-incidence areas (Ngotho et al., 1999b).
We used participatory appraisal (PA) to investigate the importance, incidence, impact
and cost (relative to other cattle diseases in the area) of MCF as perceived by Maasai
pastoralists in Isinya Division, Kajiado District, Kenya. Avoidance movements due to MCF
and the problems associated with it were also investigated. MCF has been shown to be a
disease of concern to pastoralists in other areas of Maasailand, but the impact on
pastoralists in Isinya Division has not been assessed. A lack of information on the
economic consequence of this disease (Bourn and Blench, 1999; Machange, 1988) has
undervalued it as a disease of importance. This research was carried out with the
International Livestock Research Institute (ILRI), Kenya, and was a result of a direct
request by the Maasai community in Isinya Division for an investigation into the problem
of MCF.

2. Isinya Division

Isinya Division lies south of Nairobi National Park (NNP) and includes the Kitengela
wildlife dispersal area. This is a Maasai pastoral area, which acts as a migratory corridor
and dispersal area for the parks migratory wildlife. NNP is fenced on its northern, eastern
and western boundaries, but only the Mbagathi River marks its southern boundary.
Wildlifelivestock conflicts arise as livestock mix freely with wildebeest and other
migratory species. Wildebeest leave the park at the onset of the rains and congregate on the
Kapiti plains in the south-eastern part of the ecosystem, where during the rains the grasses
are short, green and highly nutritious (Gichohi, 2000). Females use this area to calve, and
this is the area that highest wildebeest densities have been found (>83 wildebeest/km2)
(Gichohi, 2000).
Kitengela landowners have negative attitudes towards wildlife (Mwangi and Warinda,
1999). Although the land is privately owned, the Government owns the wildlife on the land
and Kenya Wildlife Service (KWS) is the official custodian. Damage to crops or livestock
by wildlife is not compensated by the government (Republic of Kenya, 1989), adding to
conflict because landowners feel they are being denied their basic security and property
rights. Incidences of retaliatory lion killings by Maasai have been reported after lions killed
cattle adjacent to the park (Mungai, 2003).
Maasai in this area are traditionally livestock keepers dependent upon their cattle, sheep
and goats. Livestockwildlife conflicts have been exacerbated in recent years because once
traditional pastoralists, Maasai are now increasingly growing crops and diversifying into
economic activities (Kristjanson et al., 2002). The close proximity to Nairobi has caused
much in-migration of non-Maasai tribes, and the development of Athi-River and Kitengela
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 299

townships has been encouraged by the location of an Export Processing Zone (an industrial
park for the manufacture of export goods) in Kitengela. Landowners in Isinya Division are
increasingly opting to fence their land.

3. Materials and methods

Before performing any work in the area, permission was sought to carry out the study.
Village chiefs were contacted, and it was explained to them that the aim of the research
would be to investigate the relative impact of cattle diseases. Some group meetings were
then organized through the chief of the village, whilst others were held more informally
without prior arrangement.
Group meetings allow a large number of views on the impacts of cattle diseases to be
incorporated into the study, but predominately comprise village elders and older men.
These people have much experience of livestock keeping and are therefore invaluable
informants. However, because it was important that a range of people were included, and
because not all members of the community can attend such meetings, the sampling of
households was included in the assessment. Key-informant meetings were also arranged to
gather more specific information on certain topics. These were held with the Divisional
Veterinary Officer (DVO) for Isinya Division, and with the Veterinary Officers at both
Isinya and Kitengela slaughterhouses.
The research team consisted of the first author and two Maasai researchers from the
study area. The latter had local knowledge of the area (hence were key in developing
contacts and arranging meetings) and were fluent in both Swahili and Maasai languages,
and were familiar with the Maasai cultural and social protocols. The first author had
previous training in PA, and the two other members of the research team were given 3 days
training in PA prior to the onset of the study. One person was responsible for facilitation of
the PA exercise, whilst the other was responsible for recording information. Discussions
were conducted predominantly in Maasai.
Isinya Division was split into three groups to reflect the different aspects that were
present there (Fig. 1). Study-group one was the area adjacent to NNP, which had the most
predation problems, and where high humanwildlife conflict is reported. Study-group two
was the area nearest to the towns, had major tarmac roads and represented closest access to
markets. Study-group three was the area furthest from NNP and towns, consisted of bigger
ranches and was at or adjacent to the main wildebeest calving zone. Here, it was thought
that the largest impact of MCF would be seen on the community, because cattle would be at
the greatest risk of infection from MCF infective wildebeest calves. These three grouped
areas allowed comparison by inspection of how the perceived impact of MCF varied
throughout the Division.
Ten Maasai informants compiled a list of Maasai households (stratified into the three
study-groups) which totaled 888. This was done as comprehensively as possible to avoid
the risk that certain households could have been left out. A sample of households was then
picked randomly from each study-group using a computer-generated random list. From
each study-group between five and nine households were visited, and also between three to
four group meetings where held. Households and group meetings were geo-referenced and
300 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

Fig. 1. Map showing the geo-referenced points of household and group meetings held in each study-group in
Isinya Division, Kenya, 2004.
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 301

these points added to a Geographical Information Systems (GIS) map using ArcView GIS.
3.2. (Fig. 1). All households and groups were sampled from within Isinya Division in
Kajiado District, and an adjoining area of Machakos District (Empakasi sublocation;
included in these data). Households and group meetings sampled in this sublocation have
been encompassed into Isinya Division for the purposes of this study.

3.1. Participatory appraisal methods

The methods chosen in this study were designed from previous participatory disease
investigations (Catley and Mariner, 2002; Cleaveland et al., 2001; Catley, 2003; Catley and
Admassu, 2003). Methods were tested before use to assess their suitability, and appropriate
adjustments were made. To quantify the impact and be able to compare different areas, we
used proportional piling and matrix scoring standardized to allow repetition across areas.
An advantage of these methods in disease investigations is that personal questions about
herd size need not be asked, and a comparative assessment of important diseases can be
investigated so as not to focus on one specific disease. (The latter can lead to over-estimated
costs in the hope of receiving subsequent treatment or vaccination; Catley, 2003.)
Qualitative methods of mapping and semi-structured questions were used to investigate the
seasonal movements of cattle and movements due to MCF.

3.2. Priority disease scoring

Participants were asked what are the five most-important diseases to the overall well
being of the family in the last year? Participants produced a list of diseases, and if more
than five, they were asked to choose the five most-important, and these were then written on
a card. All disease names were shown in Maasai. Participants were then asked to score each
disease relatively using 20 beans or stones. The more important a disease, the higher the
number of beans allocated to it. By scoring the diseases, rather than just simple ranking, it
was possible to see the relative weight given to each disease.

3.3. Disease incidence

Previous studies have reported difficulties in estimating disease incidence (Cleaveland


et al., 2001), because of the reluctance of livestock owners to reveal the size of their herd.
Catley (2003) also reported possible inaccuracies as a result of the exaggeration of cases
due to the hope of receiving free treatments or vaccinations. Therefore, to get a true
reflection of the problem of MCF in Isinya Division, we chose methods that did not ask
personal questions about herd size or just concentrate on MCF, but instead took a
comparative approach of locally defined problem diseases. This method was adapted from
Catley et al. (2002), and was carried out at household meetings only.
The five diseases named during priority scoring were used plus another category other
diseases. Cattle were categorized into two locally known categories: ilasho, calves (up to
about 1 year) and inkishu sapukin, adults (more than 1 year). These age categories were
confirmed with each household, and Maasai names were used for each disease name and
cattle age category, and written on a piece of card. For each age category, participants were
302 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

asked to divide 100 beans into sick cattle in the last year and healthy cattle in the last
year. Participants were then asked to subdivide the pile of sick cattle to show the relative
numbers of cattle suffering from each disease, plus other diseases. Finally, participants
were asked to further divide the piles of beans that represent incidence of each disease to
show the number of animals surviving relative to the number of animals dying. This
additional piling enabled calculation of herd mortality for different diseases. Probing
questions were asked by the research team to check and verify scores and follow-up points
of interest.
This method asked participants to consider the disease situation in their herd in the past
year. If people are not able to accurately recall past events there could be a possible source
of error here. However, because pastoralists have good knowledge of their livestock,
including histories of births and deaths, we hoped this error was not too large. Moreover,
because MCF is characterized by very high mortality risks, livestock owners are more
likely to remember incidences of animals being infected (and subsequently dying), than
compared with a disease with a lower mortality.

3.4. Impact of disease

Participants were first asked What are the benefits from cattle? The participants
usually produced a list of answers by individuals calling out, and when everyone had
agreed that was the complete list, each benefit was written on a card. With one randomly
selected group meeting in each study-group, the benefits mentioned were then scored in
terms of their relative importance using 20 beans (the more important got more beans).
(Time constraints restricted further repetition of this benefit scoring.)
We then placed the five most-important diseases from that proportion piling exercise
along the x-axis and the benefits along the y-axis to form a matrix. During group meetings,
the matrix was usually done on the ground using stones and Maasai sticks. In households, it
was usually done inside using beans on the floor or on a table. Participants were then asked
to compare how each disease affected each of the benefits. Ten bean/stones were given to
each cell in the matrix grid, and the respondents then took away up to 10 beans/stones from
each cell, to indicate relatively how much of the cell-specific benefit was lost when the cell-
specific disease was present. In placing the beans, there was usually much discussion
amongst participants before the decided number of beans was set down. Once all the
benefits had been worked through, an extra benefit was added to the matrix: low veterinary
costs (both preventive and therapeutic). A score of 10 represented very low treatment costs,
whereas a score of 0, very high costs.
To get an estimate of the financial cost attributed to a disease, participants were asked to
compare the prices of a healthy animal sold to the price if that animal was infected by a
particular disease and then sold. Three categories of animals were used: a breeding bull, an
adult cow and a replacement weaned. Again, open and probing questions were asked by the
research team to cross-check the prices reported and participants gave reasons for the drop
in price for a particular disease. This was also a useful way to cross-check matrix scores. If
information was contradictory, scores were adjusted where necessary.
When the scoring was finished, all the benefit scores for each disease were summed.
Participants were then asked to compare this situation to that of either 10 or 20 years ago to
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 303

show how the impact of each disease had changed. In summarizing the data, the reduction
in benefit scores were subtracted from 10 to get an impact score. These impact scores were
then weighted using the results of the proportion piling of benefits that was carried out once
per study-group. Weighted impact scores were summed per disease per household or group
meeting to get individual sum scores.

3.5. Mobility mapping and semi-structured questions

This method was used only with participants who had mentioned MCF as one of the
important diseases in the area. Participants were given paper and coloured pens and asked
to draw a map of the area (with their home as centre), the areas to which they move their
cattle in the dry and wet seasons, and any features they thought important (including roads,
rivers, villages and salting areas). Participants were asked to highlight areas associated with
disease, areas known for wildebeest, areas with a high risk to MCF, and to mark on the
direction of movements due to MCF avoidance (and the distances traveled).
One person was usually responsible for drawing the map with others commenting
throughout. In very large groups, many people drew maps so that more people were active.
Following the map some semi-structured questions were asked to obtain more specific
information about movements using the map as a focus point for discussion, and for views
on vaccine development and solutions to MCF. Semi-structured interviews take the form of
guided dialogues rather than interviews with a questionnaire (Waters-Bayer and Bayer,
1994). Open-ended questions were asked using a written checklist of points to discuss, and
new lines of questioning arose following comments on previous questions. We thus
allowed participants to raise issues of interest and importance to them, and we could
triangulate information and probe on any arising points of interest.
These methods avoided the possible bias of exaggerated responses, because we only
asked more-specific questions after the quantitative methods were finished. We cross-
checked results throughout the exercise by using a variety of methods (both quantitative
and qualitative). Scores from importance, incidence and impact scoring were compared
and probing questions used throughout to support the methods.

4. Results

In July 2004, 158 people (100 males and 58 females) participated in 11 group and 21
household meetings across Isinya Division. The number of people in the group meetings
ranged from 1 to about 25, and the number of people in the household interviews from 1 to 6.
The distribution of households and group meetings across the Division is shown in Fig. 1.

4.1. Important diseases and priority scoring

There was large variation between the study-groups in MCF being considered an
important disease (Table 1). However, all study-groups scored anthrax, ECF, FMD and
MCF as the four most-important diseases, and all recognized that MCF is a seasonal
problem. In study-group one, many participants reported that there were few wildebeest in
304 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

Table 1
The relative importance of different cattle diseases as scored by 158 Maasai pastoralists in Kenya, 2004
Study- No. of meetings Anthrax CBPP ECF FMD MCF Narok Lumpy skin Oloduab Ormiloc
group Kutukia
1 12 (group = 3, 49 13 100 41 15 3 1 12 4
HH = 9)
Rank 2 5 1 3 4 8 9 6 7
2 9 (group = 4, 31 6 63 55 18 0 0 3 3
HH = 5)
Rank 3 5 1 2 4 8.5 8.5 6.5 6.5
3 11 (group = 4, 30 0 69 33 77 0 6 5 0
HH = 7)
Rank 4 8 2 3 1 8 5 6 8
a
Narok Kutuki was reported in study-group one only as a disease spread by Vervet monkeys with similar signs
to MCF.
b
Olodua is the traditional Maasai word for rinderpest, however rinderpest is no longer present in Maasailand
and Olodua is now used by Maasai to describe an anaplasmosis disease with bile staining.
c
Ormilo is the traditional Maasai word for bovine cerebral theileriosis, however Maasai commonly refer to this
as heartwater.

the area, and therefore MCF was not a common disease. Nevertheless, those respondents
also reported that wildebeest may calve near their homes, so that cattle must be moved to
locations where lions are more common. In study-group three, many participants said that
wildebeest numbers had increased in the area, and rather than migrating back to the park,
they were seen all year round in the area. These are the areas where MCF was considered
most important. However, many participants in all study-groups explained that because of
increasing human population levels in the last few years, wildebeest numbers were
declining because they were increasingly losing their habitat.

Fig. 2. The relative incidence of the 7 most common diseases and healthy cattle for calves and adults during 2003
2004, as scored by 21 households in Isinya Division.
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 305

Table 2
Estimated median incidence of MCF in calves and adults as indicated by 21 households across Isinya Division for
the year 20032004
Study-group MCF
Calves (%) Adults (%)
1 3 4
2 5 4
3 8 12
Isinya Division 5 10

4.2. Estimated incidence of cattle diseases

Fig. 2 shows the healthy and disease estimates for calf and adult cattle across Isinya
Division. Sixty percent of calves and 50% of adults were estimated to have been sick during
20032004. ECF was the highest-incidence disease followed by FMD and anthrax. MCF
was the fourth-most common disease across Isinya Divison, with incidence estimated at
5% in calves and 10% in adults. Highest MCF incidence was found in study-group three
within the wildebeest calving zone, and MCF incidence decreased with increasing distance
from the calving zone (Table 2).

4.3. Variation in MCF

When asked about variation in incidence levels of MCF, participants responded that
incidence depended on the number of wildebeest calving in the area, which in turn
depended on rainfall and availability of grass. In study-group three, many participants
reported that an increase in the wildebeest population had increased the incidence of MCF,
and if there was available grass and water in the area, then incidence risks were higher. If
there was no rain, wildebeest did not come to the area as much and incidence of MCF was
low. However, participants in study-group one reported that in times of drought MCF was a
bigger problem, because wildebeest were more likely to stay near NNP where there were
permanent water sources. Across the three study-groups some participants also said that
variation depended on whether people moved or not, and how much people would chase
wildebeest away from their areas. Some participants in all study-groups thought that the
incidence of MCF during the study year was low compared to previous years.

4.4. Case-fatality risk

The risk of infected animals dying of each disease is shown in Fig. 3. Every household
estimated the case-fatality risk of MCF affected animals to be 100% for both calves and
adult cattle.

4.5. Perceived impact of cattle diseases

Income from sales was reported as the most-important benefit followed by milk
consumption (Table 3). Participants reported that income from the sale of cattle was used to
306 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

Fig. 3. Estimated case-fatality risk of the five most-commonly occurring diseases in Isinya Division during the
year 20032004.

pay for school fees for children, and could be used for buying treatment for other sick
cattle, or even for sick members of the family. Milk (including other dairy products such as
butter and ghee) remains a staple of the Maasai diet. In contrast, cattle were very rarely
slaughtered at home for meat consumption; goats or sheep would be slaughtered on most
occasions.
Table 4 shows the weighted matrix scores of the relative impact of the four most-
commonly mentioned diseases (anthrax, FMD, ECF and MCF) on the benefits derived
from cattle. Although ECF was scored as being the most-important disease in the first
exercise in study-groups one and two, they scored anthrax as having the biggest impact
(even though it is rare). When asked about this, participants recognized that although ECF
is the most-common disease in the area and present all year round, many benefits still could
be received from the ECF animal (milk and meat still could be consumed and animals sold
(in contrast to anthrax)). Almost all participants across the Division reported that they
would eat the meat from a MCF-infected animal. Most would also consume the milk of an
MCF-infected animal, although milk production was reduced due to the infection.
However, a few participants reported that milk from an infected cow was not fit for

Table 3
The relative importance of benefits received from cattle as indicated by Maasai pastoralists in three group
meetings in Kenya, 2004
Benefit Study-group 1 (%) Study-group 2 (%) Study-group 3 (%)
Cultivation 0 0 5
Dowry 0 5 10
Hides 5 5 5
Income from sales 35 35 30
Manure 15 10 15
Meat consumption 5 5 5
Milk consumption 30 20 20
Milk sales 10 20 10
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 307

Table 4
Weighted matrix scores of the relative impact on the benefits derived from cattle of the four most-commonly
mentioned diseases, Kenya 2004
Disease
Anthrax ECF FMD MCF
Study-group 1 2 3 1 2 3 1 2 3 1 2 3
No. of matrices completed 10 9 10 10 9 10 10 9 10 10 9 10
No of times disease used in matrix 10 9 8 10 9 10 9 9 10 5 7 9
Benefit
Cultivation 0 0 13 0 0 12 0 0 10 0 0 14
Dowry 0 3 8 0 4 5 0 3 2 0 3 6
Hides 18 29 22 19 27 34 26 35 39 10 29 30
Income from sales 224 221 171 179 168 168 140 130 144 119 130 180
Manure 93 53 77 78 54 90 81 48 92 53 42 104
Meat consumption 32 30 30 28 26 18 15 16 19 17 21 21
Milk consumption 222 156 154 186 140 162 180 136 152 117 118 164
Milk sales 79 166 67 74 156 81 75 158 78 43 134 78
Total weighted impact score 668 658 542 564 575 570 517 526 536 359 477 597
A higher score represents a larger impact.

consumption. Almost all participants said that when they realize an animal has MCF they
sell it immediately. An infected animal would only be consumed at home if the animal had
died before it could be taken to the slaughterhouse. Participants in study-group three
reported that an MCF animal was most likely to die in a household that lives far from the
slaughterhouse (although even here, there were middlemen who often brought the animal
and drove it to the slaughterhouse). The impact of MCF (Fig. 4) by study-group suggested
that the greatest impact of MCF is seen on pastoralists living within the wildebeest calving
zone.

Fig. 4. The relative impacts of the four most-commonly mentioned cattle diseases as indicated by 158 Maasai
pastoralists in Kenya, 2004.
308 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

Table 5
The relative scores of veterinary costs for the four most-common diseases as indicated by Maasai pastoralists in
Kenya, 2004

Scores shown are the medians, with minimum and maximum values.

4.6. Veterinary costs

ECF was consistently scored as the disease with the highest veterinary costs (Table 5),
because of the need for acaracides and antibiotics. MCF was scored as the disease with the
least costs, with all participants agreeing there was no preventive or curative treatment
available. Some participants said that sometimes, at the early stages of infection before
they realize an animal has MCF, they might mistakenly inject it thinking that it was ECF.

4.7. Changes in disease impact

When questioned about the changes in the impacts of these diseases over the last 1020
years, participants gave a variety of responses. In study-group one, most participants
agreed that the impact of MCF had gone down. Here, it was reported that there were few
wildebeest in the area, and many had not migrated back since the drought in 2000.
However, most participants expressed that the other diseases had gotten worse. One
household commented that whereas animals used to die of drought and famine, they were
now dying of disease. In study-group two, participants nearer the wildebeest calving zone
said that MCF was worse now because there were more wildebeest around. However,
participants west of the KitengelaNamanga road in Kisaju thought that the incidence of
MCF had either stayed the same or gone down in their area. Here, the impact of ECF had
got much worse, and was affecting more animals than it used to. There were large
variations in the presence or absence of wildebeest as perceived by participants over short
distances. In study-group three, many participants said that the impact of MCF was higher
now, and because of land fragmentation it was not always possible to avoid areas of MCF
risk. Some complained that because there was no longer any culling of wildlife there is
more risk of infection transmission. Some participants in study-groups one and two thought
that due to improved veterinary care, diseases have improved, although the costs of
treatment had increased. However, this was not the case for MCF. In study-group three,
participants believed that diseases were worse in the area despite increases in veterinary
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 309

care, and some complained that diseases had got worse because the government had
stopped running vaccination schemes.

4.8. Loss in sale price

Estimating losses due to the emergency sale of diseased animals was carried out for 15
household and community group meetings across the Division. Participants reported that
the price received for an MCF animal depended on how early the animal was sold after it
was known to be infected. If an animal is sold early on in the infection, and was still
relatively fit, then a better price is received than compared to when the owner waits longer
for the animal to recover, and then has to sell the animal in a poor condition (and knowing
that it is unlikely to recover). When they realized an animal has MCF, participants said they
would sell that animal immediately. ECF animals commonly received a higher price than
MCF because as many participants complained, in the case of MCF, the butcher or buyer
has an advantage, as he knows that the animal will die, and with no other options for the
owner, the price drops further.
Overall in Isinya Division it was reported that there is a 50% loss in price for all cattle
age categories infected with MCF when sold at the market (Table 6). The value lost varied
over the three study-groups with the largest losses estimated in study-group three (between
44% and 64% depending on the age category). The very high loss with breeding bulls can
partly be explained by the higher prices of these animals reported in this area.

4.9. Avoidance movements

These Maasai now move only short distances with their livestock. Most participants
reported only moving their animals away from their immediate area in times of drought or
for MCF avoidance. There were those who moved their cattle to different grazing areas to
avoid MCF but returned at night, and those who moved longer distances for a few months.
Participants throughout the Division reported moving their cattle before the wildebeest
gave birth (in January or February), and returning in July (when wildebeest calves were a few
months old). Many said that, in the case of heavy rain, they would return earlier because this
would wash away the disease. Participants said they moved to bushy areas to escape the
wildebeest and the threat of MCF, because these were areas unfavoured by wildebeest.
As with normal herding duties, it was generally younger herders who moved with the
cattle during MCF avoidance. However, some participants mentioned that at this time,

Table 6
The median percent drop in price per animal infected with MCF when sold, compared to that of a healthy animal as
indicated by Maasai pastoralists in Kenya, 2004
Study-group Breeding bull (%) Adult cow (%) Replacement weaned (%)
1 56 53 50
2 58 40 38
3 64 50 44
Isinya Division 50 50 50
310 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

more-experienced herders might be responsible, and that extra herders might be needed to
cope with the extra work involved. In the case of moving longer distances, many
participants reported that animals would be taken to a relatives house to be looked after
during the period of risk. In study-group one, participants reported traveling to Kitengela
and Athi-river, and moving up to distances of 20 km. In study-groups two and three, larger
distances traveled were reported, with many moving up to 50 km to Emarti in Central
Kajiado. However, others moved only a few kilometres to bushy areas.

4.10. Problems with movement

Lack of pasture, water and salt (and overgrazing of existing pasture) in the areas to
which people were forced to move to avoid MCF was cited as a big problem, and
participants complained about the extra cost of having to buy them. One household in
study-group three commented that from January to April wildebeest excluded their cattle
from 400 out of their 500 acres of land, putting them into competition with other herds for
grazing land. There was also increased conflict and cases of theft of animals reported
because people moved to areas with which they were not familiar.
There were risks of ECF and FMD because cattle were forced either to share resources or
move to tick-infested areas. The risk of predation also increased, and participants in study-
group one reported that their animals were more at risk from lion predation when moving (so
that it was necessary to hire extra herders). Many participants explained that the extra
workload during MCF avoidance meant that in addition to the costs of hiring extra labour,
household members were diverted from doing other household or farming activities.
Many participants also mentioned the problems of separating the family because
household members traveled with the cattle. This incurred extra costs such as financing the
running of two homes and also traveling to visit family or animals. Benefits from cattle
such as milk and manure were also lost if animals were away from the household for
extended periods. Milk productivity of the herd was reported by some participants to
decrease with increasing distances traveled. Most participants who practiced avoidance
behaviour nevertheless agreed that it was a beneficial strategy, because animals were less
likely to be infected with MCF.

4.11. Other control strategies

Participants across the Division reported chasing (sometimes with dogs) the wildebeest
away from their cattle, even though the wildebeest often return at night to give birth. Very few
reported using any traditional medicines, with participants saying there was not anything
available. Many participants (especially in study-group two) recognized the use of fencing as
a control measure, but this was often not possible or was unaffordable. A few participants also
reported killing wildebeest as a control strategy but they recognized that this was not legal.

4.12. Solution to MCF

Most participants replied that a vaccine was required to solve the problem of MCF. This
would allow wildebeest to graze alongside cattle, and the cattle would not be denied access
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 311

to key grazing sites. Most participants expressed that a vaccine would increase the number
of cattle owned and hence improve living standards. However, it was also said that
vaccination would not solve competition for resources between cattle and wildebeest.
Participants said that the wildebeest should either be taken back to NNP or shot, and/or that
individual households land or NNP should be fenced.
A few participants in study-group three had given up hope of a vaccine being found
and instead they wanted to see compensation schemes by the government to offset the
losses incurred. This was also commonly re-iterated in study-group one, where people
commented they were suffering also from predation losses. When asked who they
expected to solve the problem of MCF, most said that it was the responsibility of Kenya
Wildlife Service (KWS) and the Government (whom many blamed for not doing
anything to help them with the wildebeest on their private land). Participants also
expressed that it was the responsibility of veterinarians and scientists to come up with a
vaccine.

5. Discussion

Maasai pastoralists in Isinya Division ranked MCF overall as the fourth-most important
cattle disease and as the most important disease in the area where the wildebeest came to
calve. These results are very comparable to previous studies in Kenya and Tanzania which
show MCF as a disease of serious concern to pastoralists across Maasailand (Boone and
Coughenour, 2001; Ngotho et al., 1999a; Cleaveland et al., 2001).
Mean incidence of MCF across Isinya Division varied from 3% to 12%, also comparable
to previous studies where MCF incidence estimates in Kenya varied between 1% and 21%
(Ngotho et al., 1999b; Plowright et al., 1975) and between 3% and 12% in Tanzania
(Cleaveland et al., 2001). Incidence was found to be higher in adults than calves, also
similar to previous studies (Cleaveland et al., 2001) and literature (Barnard et al., 1994).
Incidence of MCF was associated with the density of wildebeest and distance from the
wildebeest calving zone. The large variation of MCF between the three study-groups
suggests incidence rates need to be carefully estimated on a localized scale. Even within
close distances incidence varied widely, and this combined with the fact that Maasai cattle
in Isinya Division do not move large distances means that incidence risks will vary widely
with location. Cleaveland et al. (2001) also highlighted that the extrapolation of incidence
data to wider populations should be treated with caution and is likely to exaggerate the
burden of disease.
Incidence risks presented here are only relevant for the year 20032004. Variations in
MCF incidence as reported by participants depended on wildebeest movements, which
depended on rainfall. Some participants recognized the year 20032004 as a low year for
MCF incidence compared to previous years. Many also recognized the year as quite a dry
year, and some herders were seen migrating with their cattle because of drought. If both of
these reports are taken into consideration, incidence of MCF should be lower in years with
less rainfall, agreeing with some study-group three participants who commented that if
there was little or no rainfall, MCF incidence was lower. This trend has also been reported
by Mirangi cited in George et al. (1999), who reported that Maasai pastoralists in Kajiado
312 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

District witnessed a 1015% increase in MCF incidence among their cattle during a heavy
rainfall season. However, the opposite trend has been described in Tanzania where it was
thought that an outbreak of MCF was exacerbated by severe drought preventing the
separation of cattle and wildebeest (Kalunda et al., 1982), and an extended dry period (with
unpredictable wildebeest movements and the extension of the migration into livestock
grazing areas) was thought to be the reason why there was high incidence of MCF
(Cleaveland et al., 2001).
The opinions of the study-group three participants could be explained by the fact that
during drought periods, wildebeest are forced from their traditional wet-season calving
grounds, and instead stay around permanent water sources (of which in study-group three
there are none), either near NNP or elsewhere in Kajiado. The south-eastern part of
Division is the driest area, and this may explain why the incidence of MCF will be low in
the calving zone in dry periods.
All participants estimated mortality risks of MCF affected animals to be 100% in calves
and adults. This is consistent with views of Maasai respondents in Ngorongoro District,
where there were no reports of any animals having recovered from MCF (Cleaveland et al.,
2001).
It has been suggested that because MCF provides meat fit for consumption it does not
represent a major economic problem. However, we found that the benefit of meat
consumption is not regarded as important to Maasai in Isinya Division (because they
seldom eat cattle meat). An animal infected with MCF will usually die within 45 days of
obvious clinical signs developing (Barnard et al., 1994), making it difficult for livestock
keepers in more remote areas to reach the market in time. Even if livestock owners are able
to get to a market to sell their diseased animal, the price is severely reduced because the
butcher can reduce the price given because he knows that the animal will die and, in an
epidemic, there can be a glut on the market. The impact of diseases such as MCF can
therefore be serious to pastoralists who rely on the market benefits from their livestock.
Such market-oriented groups often earn the most from livestock (Kristjanson et al.,
2002).
Impact scores from group meetings were higher than scores from household
meetings (data not shown). Individuals in a group meeting are probably more
encouraged to give a higher score due to the lively interaction of people within the
group, and these meetings tended to be more dynamic than the household meetings. In
their evaluation of bias in participatory methods in Kenya and India, the Livestock
Development Group (2003) also found differences between the narratives that were
obtained from groups and those from individual interviews. Group dynamics was
demonstrated to be a powerful factor in determining who was able to express their
views and how often. The most-influential people dominated group interactions, and
group dynamics hindered the expression of values or beliefs that were contrary to the
dominant voice. Challenges between group members did not occur or were quickly
squashed, and it was thought by LDG (2003) that even the more spontaneous group
meetings were viewed as formal communicative events in which the community
agenda could/should be conveyed. Such biases of working with groups need to be
recognized, and indicate that skilled facilitators are needed for the implementation of
PA methods.
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 313

5.1. Avoidance movement and control

Avoidance movements were reported to decrease livestock productivity. A computer


simulation of the Ngorongoro Conservation Area using the SAVANNA modelling system
(Boone et al., 2002), analysed the consequences of removing the threat of MCF and the
need for avoidance movement. As a result, cattle populations after 15 years increased by
18%. The effect this would have on the rangeland is difficult to predict. In Ngorongoro
District, there have been concerns that a rise in cattle numbers due to MCF control would
increase land-use pressure (Cleaveland et al., 2001). However, due to the increased demand
for cash to meet household needs such as school fees and medical care, Maasai are now
more likely to sell any excess animals (Cleaveland et al., 2001; Kristjanson et al., 2002).
Cattle populations have been declining in Isinya Division in the last 15 years, and this has
been attributed to increases in the sales of animals, increased incidence of disease, and also
frequent droughts (Kristjanson et al., 2002).
Field experiments in Narok and Kajiado Districts showed that incidence of MCF might
be reduced by avoiding wildebeest temporally as well as spatially (Rossiter et al., 1983).
When exposed to direct sunlight the MCF virus lost 99.6% of its infectivity within 25 min
at midday. Those authors suggested starting the daily grazing and watering of cattle 1 or
2 hours later than is the current practice, and to return an hour earlier in the evening. A few
herdsmen who practice late grazing support this view (Rossiter et al., 1983). Discussions
with herders at a community feedback meeting of the research findings revealed that this
was a practice also known to some herders in Isinya Division.
Maasai in Isinya Division are increasingly opting to fence their land to restrict wildlife
access and to prevent infection transmission. Fences are detrimental to migrations, and did
not prevent MCF transmission in South Africa (Barnard et al., 1989).
The development of compensation schemes for lost animals or financial incentives to
offset the losses due to MCF would reduce conflict between Maasai and wildlife in Isinya
Division, and help to alleviate the cost of disease. Maasai have to bear the costs of wildlife
conservation, which simultaneously hinders their own development. For the survival of
NNP and the ecosystems wildlife, conservationists need to work with pastoralists to create
ways to benefit both people and wildlife. A consolation scheme in Kitengela has been set
up to provide compensation to pastoralists whose livestock are killed by wildlife. Another
scheme, the Kitengela Wildlife Conservation Lease Programme is a mechanism that
provides financial incentives to landowners in the wildlife dispersal area, in return for
agreeing to leave their land open to wildlife. Landowners are paid 300 KSh/US$ 4 per acre
per year to not engage in quarrying, fencing, land subdivision and sale, or poaching
activities. The pilot project started in April 2000 with a total of 214 acres initially signed up
under the lease programme, and by 2003, 115 families were participating in the programme
with many on the waiting list (Chiemelu, 2004). Although currently constrained by
funding, the programme hopes to include landowners in the key parts of the migration
routes, and in the wildebeest calving zone (Chiemelu, 2004). As owner of the wildlife, the
Government of Kenya needs to support such compensation or wildlife-related financial
incentive schemes to offset the losses incurred by pastoralists. The lease programme has a
positive influence on landowners attitudes towards wildlife, mainly due to the direct
benefits they were getting in return for wildlife protection on their land (Nkedianye, 2004).
314 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

Large financial losses due to disease can have serious consequences to the socio-
economic well being of livestock keepers, because any reduction in livestock production
directly affects food security and income levels. The resulting higher dependencies on
crops can also have negative environmental impacts. Efforts to promote and improve
pastoral production systems via the provision of more reliable veterinary services might
increase the survival of animals and decrease mortality, providing pastoralists with
greater income from their cattle. This might in turn discourage the growth of crops (a
land use that is regarded as largely incompatible with wildlife), and reduce human
wildlife conflicts.
This study only concentrated on one area of Kajiado District, and it is recommended that
this research be extended to other areas in Kajiado District, in Narok District, and
elsewhere in Maasailand. Isinya Division is a unique area considering its close proximity to
Nairobi, large immigration of non-Maasai tribes, and its large industrial presence. The
challenges faced by Maasai in less developed areas are likely to differ, and as a result the
impacts due to disease need to be widely investigated.

6. Conclusion

 In the areas where wildebeest are common, the perceived impact of MCF by Maasai
pastoralists in Isinya Division was greater than in areas with few wildebeest.
 Incidence was highest in and around the wildebeest calving zone. Incidence risks were
comparable to those of other areas in Maasailand in Kenya and Tanzania.
 MCF was perceived to have the largest effects on income from animal sales and milk
consumption, and these were the most important benefits indicated by participants.
 Infection of animals with MCF reduced the sale value of animals by half, irrespective of
the age of the animal. This combined with a decrease in milk production and the costs of
moving herds away from infective areas creates a strong negative attitude towards
wildebeest by Maasai pastoralists.

Acknowledgements

This is an output of the project Better policy and management options for pastoral
lands: Assessing trade-offs between poverty alleviation and wildlife conservation, led by
the International Livestock Research Institute (ILRI), Nairobi, and funded by the Belgian
Government. The views expressed in this document are those of the authors and might not
represent the views of the donor. Financial assistance for the project also came from a
NERC half scholarship and a grant from the Edinburgh Small Projects Grant from the
University of Edinburgh Development Trust. We thank staff 8 at ILRI, including Fumi
Mizutani, Mohammed Said, Robin Reid, Shem Kifugo, Dan Omolo and Lucille Kirori.
Thanks are extended also to Sarah Cleaveland in Edinburgh, and Andy Catley for his
advice in choosing the participatory methods. We are very grateful to Nickson Ole Parmisa
and Joesph Ole Tuleto for their help in the field, and all the participants in Isinya Division
for their cooperation in the research.
C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316 315

References

Barnard, B.J.H., Van der Lugt, J.J., Mushi, E.Z., 1994. Malignant catarrhal fever. In: Coetzer, J.A.W., Thomson,
G.R., Tustin, R.C. (Eds.), Infectious Diseases of Livestock. Oxford University Press, New York.
Barnard, B.J.H., Van de Pypekamp, H.E., Griessel, M.D., 1989. Epizootology of wildebeest-derived malignant
catarrhal fever in an outbreak in the north-western Transvaal: indications of an intermediate host. Onder-
stepoort J. Vet. Res. 56, 135139.
Boone, R.B., Coughenour, M.B., Galvin, K.A., Elly, J.E., 2002. Addressing management questions for
Ngorongoro Conservation Area, Tanzania, using the SAVANNA modelling system. Afr. J. Ecol. 40, 138150.
Boone, R.B., Coughenour, M.B., 2001. A System for integrated management and assessment of east African
pastoral lands. Balancing food security, wildlife conservation, and ecosystem integrity. Final Report to the
Global Livestock Collaborative Research Support Program.
Bourn, D., Blench, R., 1999. Can Livestock and Wildlife Co-exist? Overseas Development Institute, London.
Catley, A., 2003. Validation of participatory appraisal for use in animal health information systems in Africa.
Doctor of Philosophy. The University of Edinburgh.
Catley, A., Admassu, B., 2003. Using participatory epidemiology to assess the impact of livestock diseases. FAO-
OIE-AU/IBAR-IAEA Consultative Group Meeting on Contagious Bovine Pleuropneumonia in Africa, 1214
November 2003. FAO Headquarters, Rome, Italy.
Catley, A., Irungu, P., Simiyu, K., Dadye, J., Mwakio, W., Kiragu, J., Nyamwaro, S.O., 2002. Participatory
investigations of bovine trypanosomiasis in Tana River District, Kenya. Med. Vet. Entomol. 16, 5566.
Catley, A., Mariner, J., 2002. Where there is no data: participatory approaches to veterinary epidemioloy in
pastoral areas of the Horn of Africa. IIED. Issue Paper No. 110.
Chiemelu, N., 2004. At Whose Expense? Policy Formation in Wildlife Conservation and Human Conflict in
Kenya. Luce Undergraduate Environmental Fellows Program. Watson Institute for International Studies.
Cleaveland, S., Kusiluka, L., Ole Kuwai, J., Bell, C., Kazwala, R., 2001. Assessing the Impact of Malignant
Catarrhal Fever in Ngorongoro District, Tanzania. A Study commissioned by the Animal Health Programme,
DFID.
George, K., Ireland, D., Shick, R., 1999. Assessing the Impact of Rainfall, Wildebeest Body Condition, Pregnancy
Rates, and Fetal Growth Rate on the Incidence of Malignant Catarrhal Fever (MCF) on the Athi Kapiti Plains.
The School for Field Studies, Centre for Wildlife Management Studies, Nairobi, Kenya.
Gichohi, H., 2000. Functional relationships between parks and agricultural areas in East Africa: the case of
Nairobi National Park. In: Prins, H.H.T., Grootenhuis, J.G., Thomas, T.D. (Eds.), Wildlife Conservation by
Sustainable Use. Conservation Biology Series. Kluwer Academic Publishers.
Homewood, K.H., Rodgers, W.A., Arhem, K., 1987. Ecology of pastoralism in Ngorongoro Conservation Area,
Tanzania. J. Agric. Sci. Camb. 108, 4772.
ILRI, 2004. Livelihoods Information Atlas for Kajiado District.
Kalunda, M., Shaka, S., Henrikson, K., Kalaye, J., Semugurka, W., 1982. An epizootic of bovine malignant
catarrhal fever in Northern Tanzania in 1976. Bull. Anim. Health Prod. Afr. 30, 229234.
Kristjanson, P., Radeny, M., Nkedianye, D., Kruska, R., Reid, R., Gichohi, H., Atieno, F., Sanford, R., 2002.
Valuing alternative land-use options in the Kitengela wildlife dispersal area of Kenya. ILRI Impact
Assessment Series 10. A joint ILRI (International Livestock Research Institute)/ACC (African Conservation
Centre) Report. ILRI, Nairobi, Kenya. 61 pp.
Livestock Development Group, 2003. Poverty and Participation: An Analysis of Bias in Participatory Methods.
The Livestock Development Group, School of Agriculture, Policy and Development, University of Reading.
Machange, J., 1988. Livestock/wildlife interactions. Technical Report No. 4. Ngorongoro Conservation and
Development Project.
Mungai, N., 2003. Nairobi National Park Threatened by Human Wildlife Conflict, http://www.sosrhino.org/news/
rhinonews031803.php.
Mwangi, A., Warinda, E., 1999. Socio-Economic Dimensions of Sustainable Wildlife Conservation in the
Kitengela Wildlife Dispersal Area. African Conservation Centre (ACC), Nairobi, Kenya, pp. 67.
Nkedianye, D.K., 2004. Testing the Attitudinal Impact of a Conservation Tool Outside a Protected Area: The Case
of the Kitengela Wildlife Conservation Lease Programme for Nairobi National Park. MAThesis, University of
Nairobi.
316 C. Bedelian et al. / Preventive Veterinary Medicine 78 (2007) 296316

Ngotho, R.N., Mcleod, A., Mirangi, P.K., 1999a. Perceptions and attitudes to malignant catarrhal fever (MCF) by
the Maasai people of Kajiado and Narok Districts in Kenya. In: Sutherland, J.A. (Ed.), Towards Increased Use
of Demand Driven Technology KARI/DFID NARP II End of Project Conference, KARI Headquaters,
Nairobi, Kenya, 2326 March 1999.
Ngotho, R.N., Mcleod, A., Mirangi, P.K., 1999b. Costs of malignant catarrhal fever (MCF) in group and private
ranches in Kajiado and Narok Districts of Kenya. In: Sutherland, J.A. (Ed.), Towards Increased Use of
Demand Driven Technology KARI/DFID NARP II End of Project Conference, KARI Headquaters, Nairobi,
Kenya, 2326 March 1999.
Perry, B.D., Randolp, T.F., McDermott, J.J., Sones, K.R., Thornton, P.K. (Eds.), 2002. Investing in Animal
Health Research to Alleviate Poverty. ILRI (International Livestock Research Institute), Nairobi, Kenya, p.
148.
Plowright, W., Herniman, A., Jesst, D., Kalunda, M., Rampton, C., 1975. Immunisation of cattle against the
herpesvirus of malignant catarrhal fever: failure of inactivated culture vaccines with adjuvant. Res. Vet. Sci.
19, 159166.
Republic of Kenya (Eds.), Kenya Population Census 1989, vol. 1. Ministry of Finance and Planning, Central
Bureau of Statistics, pp. 119.
Reid, H.W., Buxton, D., Berrie, E., Pow, I., Finlayson, J., 1984. Malignant catarrhal fever. Vet. Rec. 114, 582584.
Rossiter, P., Jessett, D., Karstad, L., 1983. Role of wildebeest fetal membranes and fluids in the transmission of
malignant catarrhal fever virus. Vet. Rec. 113, 150152.
Swift, J., 1983. The start of the rains. Research Memo. Institute for Development Studies, Sussex.
Waters-Bayer, A., Bayer, W. (Eds.), 1994. Planning with Pastoralists: PRA and More. A Review of Methods
Focused on Africa. GTZ.