Description, socio-economic characteristics, disease managements and mortality dynamics in smallholder's dairy production system in coastal humid region of Tanga, Tanzania

E S Swai, E D Karimuribo*, L Schoonman**, N P French***, J Fitzpatrick**** D Kambarage* and M J Bryant*****

Veterinary Investigation Centre, P O Box 1068, Arusha, Tanzania
emasw@yahoo.co.uk
*Faculty of Veterinary Medicine, Sokoine University of Agriculture, P O Box 3019, Morogoro Tanzania
**Tanga Dairy Development Trust, P O Box 1474, Tanga, Tanzania
***Department of Veterinary Clinical Science and Animal Husbandry University of Liverpool, UK
****The Moredun Research Institute, UK
*****Department of Agriculture, University of Reading, UK

Abstract

This retrospective cross-sectional study aimed to provide descriptive information on socio-economics, disease management practices and mortality dynamics in an intensive dairying area of coastal humid region of Tanzania. In this study, an interview response rate of 100% was recorded.

The farm demographic information revealed farm size to be 8.0 hectares (ha) for rural (R), 4.2 ha for peri-urban (P) and 2.7 ha for urban (U) with a median of 3-4 dairy cattle per farm. Eighty seven percent of the farmers were males and dairying was ranked third as a major source of household income. The study identifies and listed difficult milk marketing, feeds and diseases as the most challenging farm constraints. The use of permanent labour was revealed to be common feature in all farm classes (rural, peri-urban and urban) and the tendency of using family labour was noted more in rural than urban and peri-urban located smallholder farms.

The study revealed that most farms were using acaricides that require on-farm dilution, and that most farms were under-dosing with acaricides and using acaricide at irregular or inconsistent intervals. Most farms seem to be aware of udder disease (mastitis) as reflected by high pre- milking udder preparation responses, but milking techniques and post- milking udder management were inadequately carried out.

During the study, 894 animals that were alive at some stage in 1998 were followed for a total of 648 animal years and the overall mortality rate was 8.5 animals per 100 year-risk. Of the total deaths, 56% were animals less than one year old and 40% of these were attributed to tick born diseases (TBDs - theileriosis and anaplasmosis). Animal less than 12 months of age and male stock had higher death rates than older animals and females.

The results of the study therefore show that for a smallholder dairy enterprise to be adopted on a wide scale, and for it to be sustainable, there is a need for the effective implementation of policies on disease control, proper farm planning and milk marketing, adequate input supply to the farmers as well as strengthening the extension service to enhance the production of dairying in the Tanga region of Tanzania.

Key words: Dairy cattle, demographic profiles, mortality, smallholders, TBDs, Tanga, Tanzania

Introduction

A major challenge facing several nations of the world today, including Tanzania is the need to feed the ever increasing populations (Winrock International 1992; Kurwijila 1994). This fast growing human population coupled with malnutrition and the low income of the people strongly suggests the need for further establishment, expansion and promotion of dairy farming in Tanzania. As a result, the per capita milk consumption has remained 22 litres less than the African average for the last two decade (Kurwijila 1994; Sansoucy 1995). In order to increase the milk supply to major urban centres, market-oriented intra-urban, peri-urban and rural dairy production systems have recently emerged.

Smallholder dairying gained momentum during the early 1980's, immediately after agriculture and livestock policy reform (Ministry of Agriculture and Livestock Development 1984) whereby strong emphasis was directed to smallholder producers (rural, peri-urban and urban) in favour of the then under performing large-scale farms. Under mixed, peri-urban dairy production systems, animals are normally semi- or zero-grazed and depend on farm-established or collected roadside pastures, crop residues and locally-obtained compounded concentrates. Marginal capital investment (Leslie et al 1999) and involvement of the farmer in other farm enterprises are common features (Msanga 1997). Due to small capital investment and less attention of farmers on key management issues such as nutrition and disease preventive measures, return on investment in relation to other alternate farm activities has been low (Leslie et al 1999). Low milk yields, poor growth rate for young stock, high morbidity and animal losses through mortalities (Msanga 1997; Tanga Dairy Development Programme (TDDP) 1994) have further compromised the overall farm output in these systems.

The recent rapid appraisal study in Tanga and Iringa regions of Tanzania, revealed that over 70% of the formal marketed milk in Tanzania came from smallholder dairy farms (Leslie et al 1999). However, most of the increased production in the smallholder sector has been due to increased use of land and livestock resources rather than from higher individual-cow productivity (Walshe et al 1991). In Tanzania, approximately 80% of the smallholder dairy cattle are stall-fed for the greater part of the year (ILCA 1979). Such intensification will require improved management and increased resources per cow. An important step in evaluating potential development alternatives is to identify the major constraints and opportunities for increased productivity on smallholder dairy farms. To this end, a structured cross-sectional study of randomly selected smallholder dairy farms in the coastal humid region of Tanzania was initiated. The objective of the study was to identify and quantify potential production constraints. In this paper we describe smallholder dairy unit profiles and quantify mortality events in a typical Tanzanian smallholder dairy setting.

Materials and Methods

Study area/location

This cross-sectional field study was conducted in Tanga Region, located between Latitude 4^o 21' and 6^o 14' S, and Longitude 36^o 11' and 38^o 26' E, in northeast Tanzania. The study area comprises the humid coastal, hinterland and highland zones of the region at an altitude of between 200 and 2000 metres (m) above sea level. Characteristics of the region are described in detail by Swai (1999) but briefly the study sites comprise five administrative districts (Tanga, Muheza, Pangani, Lushoto and Korogwe), with the land classified into five 'agro-ecological zone' (AEZ) basing on elevation, soil types, rainfall (amount and pattern), water retention ability and crops grown (National Coconut Development Council (NCDC) 1981). The location of each farm in each administrative district was also classified as rural, urban or peri-urban. Urban farms were those occurring within towns, peri-urban farms were those peripheral to towns but within a 15 kilometres (km) of the town centre and rural farms were those occurring at 15 or more km from a town centre. Such location of farm by administrative district could influence farm productivity (i.e. availability of supplement feeds, breeding service), due to ease of access to suppliers. Farm locations (by administrative district and farm class) were considered as explanatory (independent) variables during data analysis.

Climate and topography

The climate of the region is tropical heterogeneous with a bimodal rainfall pattern (the short rains falling from September to November and the long rains from March to May). The mean annual rainfall varies from 500 to 1400 mm/year depending on altitude. The topography of the area ranges from coastal plain to inland upland slopes. Soils are sandy in the coastal belt, clay to loamy in the hinterland, and leached mineral laterite in the highlands. The relative humidity of the day ranges from 70% on the coast to 40-50% in the highlands for most of the year. Temperatures range typically from a low of 15^oC in June to August and a high of 35 ^oC in December to March. Most areas within the region receive between 2300 to 3100 hours of sunshine a year.

Study farms and husbandry practices

Farms for the study were identified from the TDDP database. Two hundred farms from a sampling frame of 3001 small holders, distributed in five out of the six districts found in Tanga Region, were randomly selected for the study that began in October 1998. Dairy stock kept included crosses of Bos taurus (Friesian, Ayrshire, Jersey, Simmental) and Bos indicus (Tanzania Shorthorn Zebu, Boran, Sahiwal) at levels of Bos taurus genes that varies typically from 50 to 85%.

Data collection

A pre-tested, structured questionnaire (PSQ) administered on each farm at each visit was used to collect information on animal and farm level management practices. Data collected by the questionnaire included details of the type of animal housing (cowshed with and without roof, Kraal), whether or not animals had access to minerals, whether the animal had grazed or been zero grazed, whether the cattle owner had attended any dairy husbandry training, contact rate with extension agent, herd size, feeding regime, breeding regime, age of the animal, breed, filial generation (classified as F1, F2 and F3 basing on the level of Bos taurus genes and breeding records), source of animals (home-bred or brought-in) and tick control practices. All the information collected related to farm (land size, labour profile) and animal events occurring in, or relevant to 1998. This involved detailed tracing of all animals on the farm, and examination of any written records, so that all ages of the cattle, calving dates, date of deaths and other movements of cattle on and off the farms agreed chronologically. The study was conducted during the period of January-April 1999.

Data analysis

Descriptive statistics for the animal and farm level explanatory variables examined in the study were developed using Epi-Info version 6.04d (1996), Centre for Disease Control, Atlanta, USA. The proportions of categorical variables were screened for significant association between outcome variable and explanatory variables using a 2x2 contingency table. The level of significance was set at 0.05. Graphical results were developed using the Excel soft ware program (Microsoft Inc., USA)

Mortality rate estimates

Mortality rate was estimated using the following formula :(Martin et al 1987):

Animal days at risk are the total number of days the study animals were present during the year under study. An animal's number of days present during the study was calculated as the difference between its date of exit (or end of December 1998) and its date of entry (or start of 1998). For mortality estimates, the farm was the primary sampling unit. The study population was all dairy stock that was alive at any time during 1998. Using Cox proportional -hazard (Cox 1972) modelling (with farm as a random effect), the relationship between mortality and animal level variables of age and sex were estimated using S-plus 2000 (Math Soft Inc. 2000).

Results and discussion

Smallholder unit profiles

All 200 surveyed farms (comprising 698 animals) that represented 6.6% of all farms in the sampling frame were visited and interviewed. (a 100% voluntary response rate).

The average mean land size was 8.0 ha for rural, 4.2 ha for peri-urban and 5.0 ha for urban, with a range of 0 - 38 ha. The reserved land for dairy enterprises ranged from 0 to 6 ha and mean values for the farm classes were 1.2 ha for rural, 0.6 ha for peri-urban and 0.4 ha for urban (Table 1). The average number of dairy cattle per farm was between 3 and 4. A similar number of dairy animals was reported in smallholder dairy production in Kiambu district of Kenya (Gitau et al 1994a). Many farms had access to water (tap, pond or borehole) at a mean radius of 243 m for rural, 122 m for peri-urban and 33 m for urban farming communities.

Farm constraints

Generally milk marketing, feed availability and diseases were ranked as the most important farm constraints in all farm classes (Table 2).  Feed availability was ranked first and was significantly (P< 0.001) higher in peri-urban than in rural and urban areas. This observation was in agreement with the findings of Msangi (1998), Leslie et al (1999) and Rutamu (1999), who reported animal feeds as a major farm constraint for zero grazed dairy cattle.

Position of dairying and the source of household income

During individual interviews , farmers were asked which enterprise was the most important source of household income (Figure 1). Rural located farmers ranked crop farming (24%) first, closely followed by dairying (12%). Off-farm employment (19%) and trading (5%) were ranked first in urban located farms. Overall, dairying was ranked third as the most important source of household income in the study area. Many studies (Van Munster 1997; Leslie et al 1999) complemented dairying for its immense contribution as a source of income for rural areas. Regular flow of cash, milk for household consumption and for collateral or security being the most cited reasons (Swai et al 1992).

Labour profile in smallholder unit

Overall, the study survey identified four major forms of labour engagement, arranged in the order of importance: Permanent labour (47%), Family labour (26%), seasonal labour (23.5%) and casual (3.5 %). Proportionally, high use of family labour was noted as one move toward peri-urban and rural located farms (P = 0.002). Because of off-farm activities (in many urban/peri urban farms), the tendency was to recruit more permanent labour. Seasonal labour was a common feature in most of the rural located households (Figure 2). Dairy farming by its nature is a labour demanding activity that greatly influences industry performance.

Distributions of farm management variables

Details of the farm management factors investigated are summarised in Table 3.

Dairy farming experience ranged from more than 5 years for 38.5% and less than five years for 61.5% of the farms surveyed. Most of the cattle owners were males (87%). Over half (61.5%) of farmers had attended some form of training on dairy husbandry or disease control courses within the last three years prior to the start of the study. In addition, 94.5% of farms had an access to extension agents contact more than four times in a year. This possibly explains the value tied up in dairy cattle at household level.

Cowsheds with roofs (90.5%) and concrete floors (51%) were distinct on many farms, and peri-urban located farms had a significantly higher number of concrete-constructed floors compared to urban and rural located farms (P < 0.0048).

Over two thirds (78.5%) of animal feeds came from farm established fodder units or as post harvested crop residues. Most of the respondents (95%) fed their animals on maize bran and a combination of other locally available and compounded protein rich concentrates as shown in Table 3. In the area of study, availability of some concentrates is seasonal due to various weather related factors like rains and drought (Msangi 1998). Restricted suckling (71%) and bucket feeding (29%) were the commonest form of pre-weaning calf rearing system practised. Many farmers feel that mastitis can be reduced by practising restricted suckling. The mean weaning age was 130 days (range 0 - 360 days).

Disease management practices at farm level

Detailed disease preventive measures at farm level are summarised in Table 4

Most farms (98%) claimed to practice some form of tick control. Over 50% used acaricide for tick control either once or twice per week and there was no variation (P = 0.07) between wet and dry seasons. The products commonly used were Chlorfenvinphos-based acaricide Stelladone (58.2%), Superdip (7.9%) and Pour on or Synthetic pyrethroids based compound - Ectopor and Spot on. Acaricide application was mostly carried out by the attendants or relatives (53%) and to a lesser extent by the farm owner (47%). Hand spraying was the commonest method used and a clear variation (P< 0.006) between farm class and area was noticed. Synthetic pyrethroids were often used in rural located and tetse tetse fly infested districts i.e. Pangani. There were striking differences in the concentrations used on different farms. 37 farms (33.6%), which used Stelladone, and 13 farms (87%), which used Superdip, were using over-strength concentrations. On the other hand, 73 farms (66.4%), which used Stelladone, and 2 farms (13%), which used Superdip, used these acaricides at concentrations less than those recommended by the manufacturers. No farms used the correct concentration. Over- or under-dosing pose by far detrimental risk to dairy cattle, if tick control by acaricide has to continue for a longer period in the coming years. Over-dosing is a waste of money and detrimental to the environment and possibly to the farmers' health too. Under-dosing animals with acaricides can expose ticks to sub-lethal strengths of acaricide leading to the possible danger of increasing tick resistance to acaricides (De Castro 1997; Wellcome Eastern Africa Ltd 1980). Consumption of animal products derived from animals treated with acaricides or prolonged use and contact, has been reported to be associated with possible hazards for human health (Keating 1987). Prophylactic use of anthelminthics was practised by 71% of farmers. Many farms practised routine pre-milking udder preparation. Washing with water was practised by 96.2% and 61% used a wet cloth to clean the udders and 72.4% used teat lubricants during milking. In addition, at milking 53.5% checked the udder for abnormalities and 59.4% checked the milk. No farm used teat dip and the most common milking technique was stripping.

Animal dynamics

Movement of animals on and off study farms

Information gathered from the study farms included detailed tracing of all animals that stayed at some stage in the study farm during the year 1998. 191 animals left the study area due to various reasons including 105 (55%) sold for slaughter, 14 (7 %) sold for breeding or paying back credit, 55 (29%) animals died and 17 (9%) left for other reasons including gifts, farmers leaving study area or leaving farming altogether. During the same period, 249 animals entered the study area as a result of births (201; 82.7%) or purchases for breeding (48; 19.7%). At the end of the study, data were available for 894 animals that were alive at some stage during 1998. These animals contributed 236483 animal days (equivalent to 648 animal years) to the study during the year 1998.

Mortality rate estimates

Of the 894 dairy cattle that were reported to have been alive at some stage in 1998, 55 were reported to have died between January and December 1998 including two stillbirths. Of these, 32 (58.2%) were males and 23(41.8%) were females. The overall estimated mortality rates by administrative district are shown in Table 5. Pangani district was associated with high mortality rates compared to other districts.

The major cause and monthly pattern of deaths due to all causes and East coast fever (ECF) are shown in Figure 3 and Figure 4.

Tick borne diseases (TBDs), specifically ECF, were reported to be the major cause of deaths. Deaths (all causes) were reported to occur in all months of 1998. Death due to ECF was reported to occur in all months of the year except for January. Though not statistically significant, peak ECF mortality rates seemed to occur between August to December. The critical period of fodder shortage for zero grazed animals in Tanga lies between November to March. Grazing is more evident in these months. This might explain the higher risk in December.

Table 6 summarises the mortality rate by age category on smallholder dairy cattle. Thirty-one (56.3%) out of 55 reported deaths were of young stock less than 12 months old. Mortality in animals in the one to two years age category represented 20% of total mortality whereas mortality in animals over 2 years old represented 23.7% of total mortality. Nearly half (40%) of the recorded mortality in young stock less than 12 months old were related to tick borne diseases mainly ECF and anaplasmosis (Figure 3). After allowing for confounding with breed, higher relative risk was observed for young stock (less than one year), and males, than older and female animals. Age and sex of the dairy cattle remain the most significant explanatory variables in the final hazard regression model, with young (<12 months old) and male animals being five and three times more likely to die than older and female dairy stock. (Relative risk [RR] = 5.03, P < 0.001 for young animals and 3.66, P < 0.001 for males stock).

Survival of calves is important for replacement of slaughtered cattle in a beef herd and for expansion of dairy herds but their loss can have large effects on overall farm productivity. Besides TBDs, mainly ECF and Anaplasmosis, non-infectious related deaths (i.e. strangulation, snake bite, poison) were reported and identified as the other causes of deaths - highlighting the need for improvement in basic husbandry. Male stock had a higher mortality than female cattle, reflecting the relative value attached to female stock, either as future replacement stock or for commercial reasons. Feeding male calves with milk was not economically justifiable in the specialist dairying farming system. Gitau et al (1994b) and French et al (2001) recorded similar findings in central Kenya and in communal areas of Zimbabwe, respectively. Despite the fact that studied animals were at risk of a variety of causes of mortality, mortality due to tick borne disease was reported to account for over one third of all deaths. Although this observation may be confounded by those practising tick control being those most at risk of mortality due to TBDs, it still suggests that acaricide application on many smallholder farms is not effective enough to prevent cattle from being exposed to ticks (Table 4). Similar observations have been reported in smallholder dairy farms of coastal regions of Kenya (Maloo et al 2001). The under-dosing concentration levels of the popular acaricides used (Table 4) could possibly explain for this anomaly. Alternatively, the nature of the study that involved interviewing one farmer once may not have been enough to probe and gather detailed information on actual daily tick management practice, rather than what the farmer wanted the interviewer to hear.

Conclusion

• Generally, the farming activities of most smallholder dairy farmers in Tanga dairying potential areas are varied and most of the people involved were males. Median farm size (both hectares and animal numbers) is small for urban and peri-urban located farms. The use of family labour was high particularly in rural located farms. For this reason, many farmers have adopted zero grazing systems. Milk marketing, disease risks, inherent household-level factors (i.e. small size farms and distances from markets), feeds and diseases are all constraints but their individual importance may vary amongst the farm classes. Farmers have adopted, to some extent, 'newer' practices such as tick control, use of pre-milking udder preparations and de-worming procedures. However, despite of being aware of the consequences of disease (especially tick-borne disease and mastitis), yet correct methods of acaricide application, milking techniques (five finger squeezing as opposed to stripping) and post milking udder preparation appearedto be poorly understood.

• Young stock group (less than 12 months) and male calves were more likely to die than females or older age group. Therefore, for smallholder dairy enterprises to be adopted on a wide scale, and for them to be sustainable, there is a need for the effective implementation of policies on disease control, farm planning, milk marketing, adequate input supply to the farmers as well as strengthening the extension services to enhance the production of dairying in the Tanga region of Tanzania.

Acknowledgements

The authors are very grateful to the farmers, extension staff who gave their time for this research. We also thank the Director, Directorate of Veterinary Service, for permission to publish this work. This publication is an output from a research project funded by the UK Department for International Development (DFID) for the benefit of developing countries. The views expressed are not necessarily those of DFID (Animal Health Programme: Project No. R7271)

References

Cox D R 1972 Regression models and life table. Journal of Research Statistics Society. Biometrika 34: 187-202.

De Castro J J 1997 Sustainable tick and tick borne disease control in livestock improvement in developing countries. Veterinary Parasitology. 7: 77-97.

Epi-info 1996 Epi-info Centre for disease control, version 6.04d, Atlanta, USA and Geneva, Switzerland.

French N P, Tyrer J and Hirst W M 2001 Smallholder dairy farming in the Chikwakwa communal land, Zimbabwe: birth, death and demographic trends. Preventive veterinary Medicine 48: 101-112.

Gitau G K, O'callaghan C J, McDermott J J, Omore A S, Odima P A, Mulei C M and Kilungo J K1994a Description of smallholder dairy farms in Kiambu District. Preventive veterinary Medicine 21: 155-156.

Gitau G K, McDermott J.J, Waltner-toews D, Lissemore K D, Osumo J M and Muriuki D 1994b Factors influencing calf morbidity and mortality in smallholder dairy farms in Kiambu District, Kenya. Preventive veterinary Medicine 21: 167-177.

International Livestock Centre for Africa 1979 Economic Trends: Dairy Products.ILCA Bull.No.4, ILCA, Addis Ababa, Ethiopia

Keating M T 1987 Diaxathion (delnav) residues in milk in Kenya. Tropical Animal Health and Production 19: 147-152.

Kurwijila L R 1994 Dairy processing and Marketing in Tanzania. Lesson and future challenges. Tanzania veterinary Journal, Volume 18: (Supplement 4): 192-208

Leslie J, Swai E S, Karimuribo E and Bell C 1999 Tanga and Southern Highland Dairy Development programmes: Socio- Economic Aspects and Farmer perception of Dairy cattle keeping and Animal diseases, DFID/NRRD -Animal Health Research Programme.

Maloo S H, Thorpe W, Kioo G, Ngumi P, Rowland G J and Perry B D 2001 Seroprevalence of vector transmitted infections of small-holder dairy cattle in coastal Kenya. Preventive veterinary Medicine 52 (No1): 1-16.

Martin S W, Meek A H and Willeberg P 1987 Veterinary Epidemiology, Principles and methods, IOWA State University Press, Ames, IOWA, USA.

Math soft 2000 S-plus 2000 Guide to Statistics, MathSoft, Inc, Seattle, Washington,USA.

Ministry of Agriculture and Livestock Development 1984 T anzania National food strategy Volume II, Dar es Salaam, Tanzania.

Msanga N 1997 Productivity of cross breed cattle and calf rearing System in smallholder dairying in Northern Coastal Tanzania. PhD Thesis, University of Reading, UK.

Msangi B S 1998 The effect of feeding on the performance of crossbred dairy cattle in the smallholder dairy system along the coast of Tanzania. PhD Upgrade Report, Department of Agriculture, University of Reading, UK.

National coconut development council 1981 TIRDEP/GTZ Tanga Regional water master plan map, Tanga, Tanzania.

Rutamu 1999 Improving the performance of growing heifers in coastal areas of Tanzania. MSc Thesis, Swedish University of Agricultural Sciences, Sweden.

Sansoucy R 1995 Livestock - a driving force for food security and sustainable development. World Animal Review 84/85: 5-17. http://www.fao.org/ag/AGa/AGAP/FRG/FEEDback/War/V8180b/v8180b07.htm

Swai E S Minja F N and Zylstra L 1992 Dairy Development in Tanga, Tanzania. Proceeding of a workshop on Future of Livestock Industries in East and Southern Africa held in Kadoma Motel Ranch, Zimbabwe, July 1992 (Editors) J.A., Kategile and S. Mubi. ILCA pg 157 - 164.

Swai E S 1999 Tick borne diseases: Production and animal health constraints in smallholder dairying systems in Tanga region of Tanzania. PhD Upgrade Report, Dept. of Agriculture, University of Reading, UK.

Tanga dairy development programme 1994 Annual Progress Report, Tanga, Tanzania.

Van Munster B 1997 A cow is a sign of Development, Report of the TDDP social economic survey, Tanga, Tanzania.

Walshe M J, Grindle J, Nell A and Bachmann M 1991 Dairy development in sub-Saharan Africa: a study of issues and options. World Bank Technical Paper No. 135. World Bank, Washington, USA.

Wellcome Eastern Africa limited 1980 The control of cattle ticks in East and Central Africa. Kabete, Kenya.

Winrock International 1992 Assessment of Animal Agriculture in sub-Saharan African. Winrock International Institute for Agriculture Development, Arkansas, USA. pp. 125.