Questionnaire Survey on Urban and Peri-urban Livestock Farming Practices and Disease Control in Kisumu Municipality, Kenya

INTRODUCTION The poverty rate as well as undernour-ishment is increasing in most sub-Saharan African countries 16. It is expected that by the year 2015, about half of the African population will be living in urban centres and poverty will have increasingly moved from rural to urban areas 14,16,17. With the declining purchasing power in the urban areas, many urban households have responded with diversification of income sources, the most notable one being urban agriculture


INTRODUCTION
The poverty rate as well as undernourishment is increasing in most sub-Saharan African countries 16 . It is expected that by the year 2015, about half of the African population will be living in urban centres and poverty will have increasingly moved from rural to urban areas 14,16,17 . With the declining purchasing power in the urban areas, many urban households have responded with diversification of income sources, the most notable one being urban agriculture 2 . Rapid expansion of urban farming has been noted in cities such as Dar es Salaam, where there was an increase from 18 % of the families engaged in agriculture in 1967 to 67 % in 1991, of which 74 % kept livestock 6 .
In most cities, the development of urban livestock production is unplanned and in densely populated neighbourhoods. There is a growing concern that it is creating health and environmental hazards 17,18 . Therefore, urban livestock keeping is often declared illegal, although enforcement of this has not been successful 7 . The local by-laws frequently date back over 100 years (Kisumu municipality by-laws on livestock were last reviewed in 1907) and forbid all agricultural activity within the boundaries of urban centres 7 . In spite of the illegality, urban livestock keeping is usually tolerated as long as it does not become a nuisance. Indeed, in some urban cities, for example Dar es Salaam and Kampala, authorities are encouraging the practice of Urban Agriculture (UA) in order to raise food-supply levels 4,6 .
The purpose of the current study was to characterise the urban livestock keeping practices and constraints in Kisumu municipality, Kenya. Such information is important in assessing the needs of the farmers, the importance of livestock in improving the household and urban livelihoods, and also to inform policy.

Study area
Kisumu, the 3rd-largest city in Kenya, is a regional capital and an administrative, commercial and industrial centre for the Lake Victoria basin. The population of the Municipality rose from 32 431 in 1969 to 322 734 in 1999 12,13 . Currently, the population is estimated at 500 000 people with a growth rate of 2.8 % per annum. About 60 % of Kisumu's population lives in slums and a further 60 % of the population residing in the municipality are involved in some form of urban agriculture and livestock keeping 17,18 . In fact, almost 80 % of Kisumu's municipal land area is rural in nature 7 .

Selection of study sites and farms
Field visits were undertaken between July and August 2007. The study sites and farms were purposively selected in collaboration with the government extension and administration officers. The visited sites were headed by a village/ ward chairperson and were selected on the basis of having a higher concentration of livestock keeping in urban and peri-urban areas. These sites included Nyamasaria, Nyalenda, Obunga, Manyatta, Mamboleo and Otonglo, located in Chiga, Nyalenda, Kanyakwar, Manyatta and Kolando sub-locations respectively. The Nyalenda, Obunga and Manyatta sites were within urban areas while Nyamasaria, Mamboleo and Otonglo were in a peri-urban area of Kisumu city. Nyalenda and Obunga were densely populated and are mainly regarded as slums 18 . At each study site, the farmers were randomly selected from a list prepared from the previous vaccination campaigns by the veterinary office in Kisumu. Of 150 listed farmers, 40 were randomly selected, of whom 34 were interviewed as the others were not available.

Questionnaire survey
A structured questionnaire was prepared and pre-tested before administration to the selected households. The questionnaire addressed the following: demogra-phy, size of land and ownership, livestock kept on the farm, production system, management practices, milk production, production constraints and common diseases at farm level. The questionnaire was administered to the household head or any other mature person (if the household head was not available) using the Kiswahili language, which is common in the area. For farmers who were unable to understand Kiswahili, the questionnaire was administered in the dholuo language by one of the extension officers.

Data entry and analysis
The responses to the questionnaires were entered into a computer spreadsheet, Microsft Excel ® (Microsoft Corporation, USA) followed by descriptive analysis undertaken using Statview ® Version 5.0.1 (SAS Institute Inc., 1995-1998, Cary, NC, USA).
The farming activities involving the HH were categorised as either livestock (41 %), or both crops and livestock (59 %). The HH also involved themselves with other activities including shop/kiosk selling (18 %), livestock trading (12 %), building houses (masonry and carpentry) (15 %), fishing (6 %), formal employment (15 %). Eleven household heads (32 %) did not engage in other enterprises and were purely involved in livestock keeping. All the land was privately owned and in the majority of cases (56 %) the farm size (plot or allotment size) was less than an acre. Those whose farm size ranged between 1 and 2 acres were 9/34 (26 %), while 6/34 (18 %) farmers had a farm size of more than 6 acres. The targeted farmers were those keeping mainly cattle but in descending order the number of livestock kept by the farmers included cattle, chicken, goats, ducks, sheep and pigs (Table 1). Most (98 %) of the cattle were of the indigenous type, with farmers indicating that the exotic animals (2 %, Holstein and Ayreshire breeds) were capital-and labour-intensive and therefore undesirable to keep.
Most (94 %) of the farmers had been keeping livestock on their farms for a period of more than 5 years. The reasons for keeping the livestock (cattle, sheep and goats) included: source of income (97 %), domestic consumption (59 %) and cultural considerations (funerals, dowry etc.) (29 %). Cattle and small ruminants were mainly owned by the male spouse (97 %). The ownership of pigs was by the male spouse (67 %) and son (33 %), while that of poultry was by both the male (59 %) and female (41 %) spouse.
Most farmers (94 %) grazed their rumi-nants in the free open spaces in the municipality. These open spaces included unutilised government/municipal lands, roadsides, undeveloped private lands (mainly plots) and rubbish dumping sites. A small proportion (6 %) of farmers undertook zero-grazing where exotic animals were kept in stalls and fed on fodders and commercial feeds. The reasons given by farmers for the preference for a specific farming system included increased productivity (68 %), availability (27 %) and reduction in disease challenge (6 %). The source of labour for grazing and milking the cattle was mainly hired help (76 %) rather than family members (24 %). Housing (night boma for ruminants) that was surrounded by timber was provided by 76 % of the farmers ( Table 2). Only the zero-grazing shed had concrete floors, while the rest of the night-bomas had floors of natural earth. The majority (59 %) of the farmers did not  supplement their ruminants with feeds other than natural grazing ( Table 2). The rest of the farmers supplemented the animals with left-overs from the markets (18 %), brewers waste (machicha in Kiswahili), commercial feed (9 %) and self-grown Napier grass (6 %). The water sources for the livestock included local rivers (59 %), local wells (41 %), tap water (35 %), floodwater/pans (18 %) and lake water (12 %) ( Table 2). In total, 33 farmers were milking their animals, and primarily used the milk for domestic consumption (91 %). The average volume of milk produced per cow per day was 2.7 , while the average volume of milk consumed at household level was 2 (range = 0. [3][4][5]. Seventeen (50 %) farmers sold milk to hotels and neighbours. The mean selling price milk was US$ 0.4 per litre (range = 0.14-0.6) and the average quantity sold was 10.1 (range = 0.6-56) daily.

Constraints to livestock production
In descending order, the production constraints mentioned by farmers included: diseases (100 %), poor fertility/ low genetic potential (68 %), lack of feed (56 %), conflict with local authorities and neighbours (53 %), inadequate veterinary services (41 %), stock theft (18 %), lack of water (9 %) and lack of labour (3 %). Most of the farmers obtained replacement cattle either from other farmers (79 %) or local markets (21 %). The purchased cattle were walked to and from the local markets, which were close to the farms (5-10 km). The marketing constraints were categorised as: low prices (35 %), inadequate markets (32 %) and presence of middle men (29 %) who often dictated the prices to the farmers. Other marketing constraints included lack of credit (9 %) and poor infrastructure (3 %).
The farmers mentioned that the most common diseases in cattle were lumpy skin disease (LSD) (71 % of the farmers), diarrhoea (65 %), helminthosis (62 %), foot-and-mouth disease (FMD) (32 %), and respirator y conditions (32 %) ( Table 3). The farmers identified these diseases using either local names or by clinical signs, whose real name was interpreted by veterinary extension officers. The disease vectors which were mentioned by farmers included tsetse flies (12 %) and ticks (24 %).
Only 3 pig farmers were interviewed in the study, owing to the decimation of the pig population by a recent (November 2006 -February 2007) African swine fever epidemic in Kisumu municipality and neighbouring districts 9 . The common diseases mentioned by farmers were African swine fever (100 %), respiratory distress manifesting mainly as coughing (100 %), mange (67 %), worms (33 %), diarrhoea (33 %) and FMD (33 %). The farmers mentioned that the most common diseases in chickens were Newcastle disease (100 % of the farmers) and respiratory distress, which mainly manifested as coughing (40 %).
The farmers sought advice on management and treatment of the livestock from private veterinarians (56 %), government veterinary officers (50 %) and Agroveterinary shops (6 %). The distance to the nearest veterinary services was categorised as 3-5 km (91 % of the farmers) and less than 2 km (9 %). Twenty-seven percent (27 %) treated the animals on their own, mainly using stock remedies sourced from chemists and Agro-veterinary shops. Fifteen percent (15 %) used unidentified ethno-veterinary products. Sixteen (47 %) farmers were able to describe the cost of treatment for their livestock. The mean cost of treatment per year for these 16 farmers was US$ 31.2 (range: 0-171.4).

DISCUSSION
Kisumu city experiences exceptionally high food poverty (53 %), which is defined as the percentage of the population that is unable to afford or have reasonable access to food which provides a healthy diet 3,17 . In Kenya, other cities and towns that have lower food poverty include  Nairobi (8 %), Mombasa (30 %) and Nakuru (30 %) 17 . Most of the food consumed in Kisumu city comes from neighbouring provinces, which is in contrast to other cities in Africa (e.g. Dar es Salaam and Nairobi) where urban and peri-urban agriculture has contributed to urban food security 17 . In contrast to peri-urban areas where land sizes are larger, the majority of the urban farmers, especially in slums, had small land sizes (plots) and some claimed to be 'squatters', which could complicate planning interventions. This is similar to a study undertaken in Kumasi city in Ghana, where 85 % of the livestock enterprises were on small plot-sized parcels of land 11 . The majority (94 %) of the farmers in the current study grazed their animals in open land spaces, especially government and municipal lands. The presence of unfenced open land often serves as encouragement for urban livestock keeping. Grazing of livestock by the roadsides and in rubbish dumps is not only a nuisance, but also poses a risk of transmission of zoonotic diseases. As in other African societies, most of the families in the current study were male-headed 6,8 . In the current study, most farmers were more than 45 years old and had been keeping livestock for a period of more than 5 years. The majority of the farmers had an educational level ranging from lower primary to tertiary, which might make it easier to educate them on improved livestock production. The types of animals kept by the farmers included ruminants, pigs, chicken, ducks, dogs and cats. A similar range of animals has been reported in other urban studies 6,8 . Ownership of the animals was mainly due to their economic value and the results of this study are similar to others in Africa where men mainly owned cattle, sheep and goats, while women mainly owned the backyard poultry 6,11 .
Most of the farmers kept indigenous cattle, which is in contrast to Nairobi and Nakuru in Kenya, where small-scale dairy farming with exotic breeds (mainly Holstein and Ayreshire) is common 5 . In other African cities like Maseru in Lesotho, 94 urban producers contributed 40 % of the city's overall milk production 2 . In the current study, only 6 % of the farmers undertook zero-grazing, and this lack of intensive farming could be associated with lack of farmer education and transfer of technology. Exotic livestock is also severely affected by diseases which could deter farmers from keeping them 11 . On the other hand, indigenous livestock are often hardy and require low inputs 10,11,15 . Several market constraints could also be hindering livestock trading, including low prices and the presence of middlemen. As suggested elsewhere, it would be important to determine the feasibility of micro-financing programmes in reducing poverty amongst the local people 17 .
The main constraints to livestock production were animal diseases, poor fertility, lack of feed, conflict with local authorities and inadequate veterinary services. These constraints have been cited by other studies conducted in cities in developing countries 11,17 . Since the area is tropical, animals in Kisumu city are bound to suffer from a variety of parasitic diseases. The poor fertility can be addressed through better management and crossbreeding with improved breeds. Conflicts with local authorities and inhabitants are a common problem amongst livestock farmers in the cities 5 . This arises from the fact that livestock keeping in the cities is illegal, although the government veterinary officers nevertheless attend to livestock when they are sick, in accordance with the control of notifiable and transboundary diseases.
Farmers indicated that the main diseases affecting cattle are LSD, diarrhoea, helminthosis, FMD and respiratory conditions. The LSD response was probably influenced by recollection of a recent outbreak of LSD in the area (Dr Makori, pers. comm. 2008). The local veterinarians indicated that LSD and FMD are endemic in the area and the spread of these diseases through illegal animal movements to local markets and abattoirs was suspected. The consumption of rubbish predisposes to sub-lethal poisoning and bacterial infections such as salmonellosis, all of which may lead to the reported diarrhoea. The farmers indicated that the main diseases affecting sheep and goats included diarrhoea, worms/flukes, unthriftiness and respiratory conditions. The high frequency (64-68 %) of farmers who indicated diarrhoea, helminthosis as well as unthriftiness ('thin sheep and goat disease') shows that the diseases should be further investigated.
The farmers interviewed mainly relied on both private (56 %) and government veterinary officers (50 %) for advice and treatment. In Kumasi, Ghana, 39 % of livestock keepers reported that they used the government veterinary service for animal health problems but they also carried out many treatments themselves 11 . In the current study, most of the extension service providers were located at a distance of 3-5 km from each farm and thus farmers would be able to walk to their premises and seek their services. This scenario is in contrast to that in Busia in rural Kenya, where farmers have to walk long distances to reach the veterinary ser-vices 2 . The extension services in Kenya were privatised in late 1980s as part of the World Bank structural adjustment programmes 10 . This led to a negative impact on veterinary service extension as most resource-poor farmers were unable to pay for the services of either government or private veterinarians 10 . With the high level of poverty amongst the urban livestock farmers (average income in people living in slums in Kisumu is US$30-40 per month 17 ), free or subsidised veterinary extension services could improve the control of major livestock diseases and thus improve food supply in the cities.