©Haramaya University, 2020 ISSN 1993-8195 (Online), ISSN 1992-0407(Print) East African Journal of Sciences (2020) Volume 14 (2) 151-162 Licensed under a Creative Commons *Corresponding Author. E-mail: mitikuguya@yahoo.com Attribution-NonCommercial 4.0 International License. Producing, Processing, Marketing and Hygiene of Cow Milk in the Supply Chain of Girar Jarso District of Oromia Regional State, Ethiopia Alemnesh Yirda1, Mitiku Eshetu2*, and Firew Kassa3 1Department of Animal Sciences, Selale University, P.O. Box 245, Fitche, Ethiopia 2School of Animal and Range Sciences, Haramaya University, P.O. Box 138, Dire Dawa, Ethiopia 3Holetta Agricultural Research Center, P.O. Box 31, Holetta, Ethiopia Abstract Background: Production and productivity of dairy is very low in Ethiopia. This problem is exacerbated by high contamination with microorganisms and other contaminants during production, procurement, processing and distribution. To tackle the problem, understanding the production, processing and microbial load of raw milk and measuring its hygiene quality is necessary. Objectives: The study was conducted with the objective of assessing milk production, handlings, constraints of milk production and marketing, and its hygiene in urban and peri urban areas of Girar Jarso district of Oromia Regional State, Ethiopia. Materials and Methods: A total of 150 respondents were interviewed using pretested questionnaire to collect data on dairy cattle management, milk production, hygienic conditions, milk production constraints and marketing. Moreover, 60 milk samples were collected and analyzed for mean aerobic mesophilic bacterial count (AMBC), total coliform count (TCC), and spore forming bacterial count (SFBC). Results: The major feed resources were communal grazing land, crop residues, grass hay, concentrate feeds and non-conventional feed such as atella. The mean estimated daily milk yield/day/cow was 12.15 ± 0.26 and 2.69 ± 0.04 liters for crossbred and local cows, respectively. Average lactation lengths of local and crossbred dairy cows were 6.58 ± 0.22 and 9.19 ± 0.11 months, respectively. Shortage of feed, lack of clean water, appropriate utensils and adequate markets during fasting season were the major constraints to dairy production in the study area. The mean AMBC, TCC and SFBC for milk samples collected from producers at farm gates were 6.42 ± 0.07, 4.49 ± 0.09 and 2.59 ± 0.05 log10 cfu ml-1, respectively. Conclusion: It is concluded that dairy productivity in the study area is low and of poor quality as a result of different constraints and therefore good dairy husbandry and hygienic milk handling practices should be promoted to improve milk productivity and milk quality in the study area. Keywords: Feed; Hygienic handling; Microbial quality; Milk yield; Production constraints. 1. Introduction Dairy production is an important component of livestock farming in Ethiopia (Azage et al., 2013). Ethiopia is endowed with diverse topographic and climatic conditions favorable for dairy production that support the use of improved, high milk yielding dairy breeds, and offer relatively disease-free environments for dairy production (Berhanu, 2012). Cattle milk constitutes the larger proportion of the milk produced nationally (83%) (Pongruru and Nagalla, 2016). However, this potential has been hampered by different challenges such as lack of improved breeds, poor performance of local breeds, and shortage of feed in terms of quality and quantity (Pongruru and Nagalla, 2016). Microorganisms may contaminate milk at various stages including production, procurement, processing and distributions. There is a steady challenge to those involved in milk production to prevent or minimize the entry and subsequent growth of microorganisms in milk (O'Connor, 1994). Therefore, an understanding of the microbial load of raw milk is important to measure its hygienic quality as high microbial load and presence of harmful pathogenic microorganisms in the milk samples are evidences of unhygienic milk production practices (Abrahamsen et al., 2007). The intention of flourishing quality control is not routinely employed at individual farm level, and there is scarcity of data pertaining to the level of spoilage microorganisms and pathogens in commercially available raw cow’s milk. There is a steady challenge to those involved in milk production to prevent or minimize the entry and subsequent growth of microorganisms in milk (O'Connor, 1994). Teshome et al. (2014) reported an average total coliform count (TCC) of 4.99 ± 0.081log cfu ml-1 for milk marketed in Shashemene town. Similarly, Amistu et al. (2015) also reported 5.42 ± 1.735 to 5.78 ± 0.985 log 10 cfu ml-1 in Alemnesh et al. East African Journal of Sciences Volume 14 (2) 151-162 152 special zone of Oromia and Asaminew et al. (2011) 4.84 log10 cfu ml-1 for milk samples collected from Bahir Dar milk shed. According to Quality and Standards Authority of Ethiopia (Ethiopian Standards, 2008), TCC of good quality raw milk should not exceed 3 log10 cfu ml-1. The presence of high TCC in milk indicates unsanitary conditions of milk production, processing and storage. There was no formal quality control system in place to monitor and control the quality of milk produced and sold in the Girar Jarso district. Therefore, studying the production practice, quality of cow milk along the milk market chain is important. Therefore, the objective of this study was to assess milk production and handling practices, marketing system, production constraints and hygienic quality of raw cow milk along the milk market chain in urban and peri-urban areas of Girar Jarso District of Oromia Regional State, Ethiopia. 2. Materials and Methods 2.1. Description of the Study Area The study was conducted in urban and peri urban areas of Girar Jarso District which is one of the thirteen districts of North Shewa Zone of Oromia Regional State, Ethiopia. The district is geographically located at 09°45’121’’N latitude and 038°46’728’’E longitude and at an altitude of 2,677 meters above sea level. The district is located at a distance of about 112 km from Addis Ababa. The mean maximum and minimum temperatures of the area are 22.13 °C and 10.26 °C, respectively and average long-term annual rainfall is 1000 mm (NSMoLF, 2016, unpublished). 2.2. Design of the Study This study had two parts (survey and laboratory experiment). For this study, urban (Fitche town) and peri-urban (rural Kebeles adjacent to Fitche town) of Girar Jarso district were purposively selected based on their milk production and marketing potential. 2.3. Survey study A two-stage sampling technique was used for this study. In the first stage, urban and peri urban areas of the district were selected purposively based on milk production potential and participation in marketing. In the second stage, two kebeles from urban and two kebeles from peri-urban areas of the district were randomly selected. The list of all milk producers was obtained from Agriculture and Rural Development Office of the District. Then the respondents were selected proportionally using random sampling techniques. In addition, 30 milk collectors (15 from urban and 15 from peri-urban areas) were selected to assess milk hygiene across the milk market chain. 2.4. Milk Sampling Techniques A total of 60 samples of raw cow milk (250 ml) were randomly collected based on the lottery method from the previously surveyed dairy farmers at farm gate and milk collection centers during January to March 2016. Morning milk was taken from the containers of each of the producer and bulk milk samples were collected from collection centers. The cow milk samples were collected aseptically using sterile bottles and immediately kept in an ice box and transported to Dairy Technology and Microbiology Laboratory of Holetta Agricultural Research Center for analysis. The milk samples were kept in a refrigerator at 4 °C upon arrival. The samples were analyzed within 24 hours as described by American Public Health Association (APHA, 1992) and all laboratory analyses were conducted in duplicates. 2.5. Microbiological Analysis The microbiological analysis was done through enumeration of major microorganisms namely total aerobic mesophilic bacterial count (AMBC), total coliform count (CC), and spore-forming bacterial count (SFBC). To determine AMBC, 1 ml milk sample was diluted in 9 ml sterile peptone water (Oxoid, CM0009) and serial dilutions were made in sterile peptone water diluents until the expected level of 30- 300 count was obtained. One of the milk samples from a chosen dilution was placed on the sterile plate. Then, plate count agar media (Oxoid, CM0325) of 15−20 ml was poured on to the plate and thoroughly mixed with the sample and allowed to solidify for 15 minutes. Then the plates were incubated for 48 ± 2 hours at 35 °C in an inverted position. Finally, colonies were counted manually (FDA, 2003). Total coliform Count was determined using sterile violet red bile agar (VRBA) (Oxoid, CM0107). One ml of raw milk sample was added into a sterile test tube containing 9 ml of sterile peptone water (Oxoid, CM0009). After thoroughly mixing, the sample was serially diluted up to 10-9 and duplicate samples (each with 1 ml) were pour plated using sterile 15−20 ml VRBA. After gently mixing, the resulting plates were allowed to solidify and then incubated at 32 ± 1 oC for 24 hours (Murphy, 1996). Following incubation, typical dark red or purplish red or pink colonies appearing on the plates, measuring 0.5 mm or more in diameter on un-crowded plates and with bile precipitation around them were counted as coliforms (FDA, 2003). To determine SFBC, milk samples were first heat treated in a water bath (Chifton, UK) at 80 oC for 10 minutes. Appropriate dilutions of the milk samples (1 ml) were plated on duplicate solid plate count agar (Oxoid, CM0325) media. Then, colonies were counted after 3 days of incubation at 30 oC (Roberts and Greenwood, 2003). Alemnesh et al. ‘Producing, Processing, Marketing and Hygiene of Cow Milk 153 2.6. Data Analysis The survey data were analyzed using descriptive statistics (mean and percentage) of SPSS (Statistical Package for Social Sciences) software, version 20 (SPSS, 2011). Microbiological data were subjected to analysis of variance (ANOVA), SAS procedure, version 9.0 (SAS, 2009). Tukey's Studentized Range (HSD) test was employed to detect mean differences among sample sources. The numbers of microorganisms (colony forming units) per milliliter of milk samples were expressed using the following mathematical formula (FDA, 2003): Where, N = Number of colony forming units per milliliter of milk ∑C = Sum of all colonies counted on plates n1 = Number of plates in the first dilution counted n2 = Number of plates in the second dilution counted d = Dilution factor of lowest dilution used Microbial count data were first transformed to logarithmic values (log10) before statistical analysis. The log10 transformed values were analyzed using the General Linear Model (GLM) procedure of analysis of SAS software. Yij = µ + Li+ Sj + eij; where, Yij = the dependent variables; µ =overall mean; Li = location effect (peri- urban and urban); Sj = collection sites (farm gate and collection center) and eij = random error. 3. Results and Discussion 3.1. Household and Farming Practices in the Study Area The majority of the respondents (80%) in the study area were married and male-headed households (Table 1). The educational levels of household heads varied between urban and peri-urban areas. About 60% of the household heads in urban areas completed high school education and above, whereas the majority of the household heads (56.7%) in peri-urban areas never went to school. Low level of education may have a direct impact on milk production, quality and safety of milk and milk products. Education is perceived as one of the prerequisites for the development of market oriented dairy farming and understanding determinants of market channel choices among smallholder dairy farmers (Zewdie, 2010). Table 1. Marital and educational status and farming system of respondents in the study area. Variable Category Urban (N = 60) Peri urban (N = 60) Overall mean (N = 120) N % N % N % Marital status Single 7 11.7 1 1.7 8 6.7 Married 47 78.3 50 83.3 97 80.8 Divorced 5 8.3 8 13.3 13 10.8 Widowed 1 1.7 1 1.7 2 1.7 Educational status Illiterate 12 20.0 34 56.7 46 38.3 Read and write 3 5.0 12 20.0 15 12.5 Elementary school 0 0 4 6.7 4 3.3 Junior school 9 15.0 5 8.3 14 11.7 High school 20 33.3 3 5.0 23 19.2 Above high school 16 26.7 2 3.3 18 15.0 Farming system Livestock only 51 85.0 6 10.0 57 47.5 Mixed crop-livestock 9 15.0 54 90.0 63 52.5 Note: N = Number of respondents. Mixed crop-livestock farming system was found to be the major practiced farming system as reported by 90% of the respondents in peri-urban area of the study district (Table 1). On the contrary, intensive livestock rearing was the sole farming activity in urban areas, which could be attributed to shortage of land. Among the livestock species, cattle are the most important component of the mixed crop-livestock farming system. 3.2. Dairy Cattle Management 3.2.1. Feed resource and feeding Communal grazing land (64%), crop residues (of barley, teff, wheat and oat straw) (90.8%), grass hay (100%), concentrate feeds (64%) and non-conventional feed (atella) were the major feed resources of dairy cattle in the study area (Figure 1). In line with this result, Kibru et al. (2015) reported that communal gazing, private grazing and stall feeding were major feeding system in Aleta Chuko district, Southern Ethiopia. About 82% of dairy producers in urban area of the district were using purchased concentrate feed as supplement. However, only 47% of the respondents were using purchased supplement feeds in the peri- urban areas. Hay making was the most commonly used means of feed preservation technique in the study area. This was used to mitigate livestock feed shortage during dry Alemnesh et al. East African Journal of Sciences Volume 14 (2) 151-162 154 periods of the year and to avoid wastage of feed in times of surplus production during rainy season. In urban areas, grazing land was hardly available except in the backyards and some open communal fields. As a result, the majority of the cattle are kept indoor and fed on purchased hay, crop residues, concentrate and non- conventional feeds (Figure 1). Figure 1. The major feed resources available in the study area (%). 3.2.2. Dairy cattle breeding About 38% of the respondents were used artificial insemination (AI) to breed dairy cows. However, nearly 24% and 20% of the respondents depended on natural mating system using genetically improved bulls and combination of both methods, respectively (Table 2). In the absence of AI and improved bull services, some of the farmers were compelled to use (17.2%) local bulls. Both AI and veterinary services were delivered by the District Livestock Agency in both urban and peri urban areas. The AI service was delivered at the cost of 12.00 Birr per service and respondents reported that the price was affordable. Similarly, a study conducted by Kibru et al. (2015) indicated that the majority of the farmers (91%) practiced natural mating system using local bulls available in the area while some of the farmers used both natural and artificial mating system (7%) and only 2% of them used AI in Aleta Chuko District of Southern Ethiopia. The present finding agreed with the findings of Zewdie (2010) who reported that AI, crossbred and local bulls were the most commonly used methods to breed dairy cows in the central highlands of Ethiopia. Table 2. Breeding system used and available services in the study areas. Variables Urban (N = 60) Peri urban (N = 60) Overall mean (N = 120) N % N % N % Breeding system AI 33 55 13 21.7 46 38.3 Local breed bull 2 3.3 19 31.7 21 17.6 Crossbred bull 13 21.7 16 26.6 29 24.1 AI and crossbred bull 12 20 12 20 24 20 Available breeding Service providers Government (AI) 39 65 14 23.3 53 44.1 Own bull 0 0 17 28.3 17 14.1 Neighbors’ bull 21 35 29 48.4 50 41.8 Note: AI = Artificial insemination and N = Number of respondents. 3.3. Family Labor Division for Dairy Production Members of the households have a range of responsibilities for different dairy farm operations. Milking, cleaning of milk containers and barns, animal health management, milk processing and marketing, heat detection and feeding of dairy animals was the major dairy farm activities identified. Wives were highly engaged in milking and cleaning of milk containers and barns while husbands were mainly responsible for marketing of milk, feeding, heat detection and animal health management (Table 3) in the urban area of the Alemnesh et al. ‘Producing, Processing, Marketing and Hygiene of Cow Milk 155 district. Children assist in herding activities after and before school times. In the peri-urban area of the district, husbands were most often involved in feeding, watering, health management, milking cows and milk marketing. Money earned from sale of milk was exclusively controlled by the husband. Kibru et al. (2015) reported that milking, milk processing, barn cleaning and sale of dairy products were mainly performed by wives while live animal marketing and stall feeding were performed by husband in Aleta Chuko district, southern Ethiopia. Table 3. Family labor division for milk production related activities in the study area. Activities Location Responsibility sharing among family members (%) Men Women Children Hired Labor Feeding and watering of dairy animals Urban 8.3 31.7 33.3 26.7 Peri urban 28.3 8.4 53.3 10.0 Overall mean 18.4 20.0 43.3 18.3 Barn cleaning Urban 0.0 45.0 36.7 18.3 Peri urban 1.7 66.7 20.0 11.6 Overall mean 0.9 55.8 28.3 15.0 Cleaning of milk container Urban 15.0 38.3 25.0 21.7 Peri urban 0.0 71.7 21.7 6.6 Overall mean 7.5 55.0 23.4 14.1 Milking of cow Urban 20.0 38.4 8.3 33.3 Peri urban 6.7 70.0 5.0 18.3 Overall mean 13.4 54.2 6.6 25.8 Milk marketing Urban 41.7 15.0 40.0 3.3 Peri urban 40.0 16.6 36.7 6.7 Overall mean 40.9 15.8 38.4 4.9 3.4. Daily Milk Yield and Lactation Length The overall average number of lactating crossbred and local breed cows owned per household in the study area were 1.92 ± 0.12 and 1.86 ± 0.15, respectively (Table 4). The number of lactating crossbred cows varied significantly (P < 0.05) between urban and peri urban areas of the study district. The difference could be attributed to relatively better access to AI service and market opportunities of the urban farmers. The mean estimated daily milk yield/liter/cow obtained from crossbred cows (12.15 ± 0.26) was four times higher than that of the local cows (2.69 ± 0.04). The current report on milk yield of local cows was slightly higher than the earlier report by Asaminew et al. (2011) (average of 2 liter/day/cow) and Zewdu (2004) (1.8 liters/day/cow) in the first and second lactations in North Gonder Zone. The overall average lactation length of local and crossbred cows was 6.58 ± 0.22 and 9.19 ± 0.11 months, respectively. The average lactation length of the local cows and crossbred cows observed in this study is only slightly shorter than that reported by Debir (2016) in Sidama Zone, Southern Ethiopian which were 7.38 ± 10 and 9.79 ± 11 months for local and crossbred cows, respectively. Table 4. Daily milk yield and lactation length of crossbred and local cows in the study area. Variable Urban Peri-urban Overall mean SL N Mean± SE N Mean± SE N Mean± SE Number of lactating cows Local breed 6 1.16 ± 0.66 37 1.973 ± 0.17 43 1.86 ± 0.15 ns Crossbred 59 2.12 ± 0.85a 41 1.63 ± 0.12b 100 1.92 ± 0 .12 * Milk yield of cows (liters/day) Local breed 6 2.83 ± 0.16 37 2.66 ± 0.20 43 2.69 ± 0.04 ns Crossbred 59 13.14 ± 0.34a 41 11.52 ± 0.35b 100 12.15 ± 0.26 ** Lactation length (months) Local breed 6 6.14 ± 0.26a 37 6.66 ± 0.11b 43 6.58 ± 0.22 * Crossbred 59 9.28 ± 0.13a 41 9.04 ± 0.21b 100 9.19 ± 0.11 * Note: N = Number of respondents. Means with different superscripts in the same rows for the same parameter are significantly different at P < 0.05; ns = non-significant; SL = significance level. Alemnesh et al. East African Journal of Sciences Volume 14 (2) 151-162 156 According to the report of NSMoLF (2016), the total numbers of households using local cows for milk production were about 13,566 with about 2998 households using crossbred dairy cows. A total of 34,638,240 and 23,160,480 liters of milk were produced per annum from crossbred and local cows, respectively. High proportion of households in the district (80%) is engaged in selling raw milk. 3.5. Milk Hygienic Practices 3.5.1. Dairy cattle housing Almost all of the respondents (99.2%) had separate barn for dairy cattle. Most of the cattle houses had roofs made of corrugated iron sheet. Some of the households also used crop residue and grasses for thatching roofs (Table 5). Table 5. Daily milk yield and lactation length of crossbred and local cows in the study area. Variable Urban (N = 60) Peri-urban (N = 60) Overall mean (N = 120) N % N % N % House type Separate house 60 100.0 59 98.3 119 99.2 Inside family house 0 0.0 1 1.7 1 0.8 Floor material Concrete/cement 37 61.7 3 5.0 40 33.4 Mud/earthen 0 0.0 20 33.4 20 16.6 Stone 23 38.3 37 61.7 60 50.0 Type of roof Tin/corrugated iron 52 86.7 27 45.0 79 65.8 Thatched 8 13.3 23 55.0 41 34.2 Note: N = number of respondents. In the urban area, dairy producers built the floor with concrete and stone whereas 95% of the respondents in peri urban area used stone and earthen floor as bedding material. In contrast to this finding, Abebe et al. (2012), Berhanu (2012) and Kibru et al. (2015) reported that cattle share the same house with the family member during the night time in the southern Ethiopia. Asaminew (2007) also found that some of the households keep cattle in the same room with family members at Bahir Dar Zuria and Macha districts. In line with the findings of this study, Mustefa (2012) reported for Sululta and Welmera districts that about 94% of the dairy herd owners used earthen and stone floors. 3.5.2. Dairy animal health The major dairy cattle diseases reported in the study area were anthrax (Abba Sangaa), foot and mouth disease (FMD) (Manse), pasteurollosis (Gororsiisaa), blackleg (Abba Gorbaa), mastitis (Muchaa Dhiitessaa) and metabolic disorder due to imbalance feeding rations (kirkirsiisaa). Most of dairy herd owners (93%) encountered cow udder infection. Table 6. Udder health problems of dairy cattle and treatments practiced in the study area. Variable Urban (N = 60) Peri-urban (N = 60) Overall mean (N =120) N % N % N % Encounter Udder problem Yes 16 26.7 14 23.3 30 25.0 No 44 73.3 46 76.7 90 75.0 Milk animals with udder problem Yes 16 100.0 11 84.6 27 93.1 No 0 0.0 3 15.4 3 6.9 Milk from infected udder Dispose 11 68.7 3 27.3 14 51.8 Use as animal feed 5 31.3 8 72.3 13 48.1 Udder disease treatment Veterinary treatment 15 93.8 12 92.3 27 93.1 Veterinary and Traditional treatment 1 6.2 2 7.7 3 6.9 Note: N = Number of respondents. As a result, milk obtained from infected udders was either discarded (52.8%) or fed to animals (48%) (Table 6). In agreement with the current finding, Zewdie (2010) reported occurrence of anthrax, FMD, blackleg and mastitis as major diseases in the central highlands and central Rift Valley of Ethiopia. The same author also stated that these diseases usually occurred during the short rainy season (March to May) when animals are in poor body condition due to inadequate feed availability in the preceding dry period. About 93% of Alemnesh et al. ‘Producing, Processing, Marketing and Hygiene of Cow Milk 157 dairy herd owners reported udder infection but they had access to veterinary services for udder infection from the nearby government veterinary clinics. Milk produced from infected udders of milking cows were either discarded (52.8%) or fed to animals (48%). 3.5.3. Milking and milk hygienic practice In the study area, milking was practiced twice a day, in the morning and the evening. About 87.5% of the respondents used a wide-necked plastic vessel for milking whereas only 12.5% of the respondents used an aluminum milking can (Table 7). Similar studies by Teshome et al. (2014) and Teklemichael (2012) reported that 84.62% of the surveyed small-scale milk producer in Shashemane town and 75% of the surveyed farmers in Dire Dawa town, respectively used plastic utensils. This might be due to the fact that aluminum made vessels are very expensive, not affordable and are hardly available for most the farmers in the local markets. Table 7. Milk hygiene practices during milking in the study area. Variable Urban (N = 60) Peri-urban (N = 60) Overall mean (N = 120) N % N % N % Utensil used for milking Wide necked-aluminum can 10 16.6 5 8.3 15 12.5 Wide-necked plastic can 50 83.4 55 91.7 105 87.5 Cleaning cow’s shed before milking Yes 28 46.7 23 38.3 51 42.5 No 32 53.3 37 61.7 69 57.5 Wash hand before milking Yes 59 98.3 55 91.7 114 95.0 No 1 1.7 5 8.3 6 5.0 Wash udder before milking Yes 34 56.7 18 30.0 52 43.3 No 26 43.3 42 70.0 68 56.0 Use of towel while cleaning udder Individual towel 10 16.7 4 6.6 14 11.7 Collective towel 17 28.3 5 8.3 22 18.3 Not at all 33 55.0 51 85.0 84 70.0 Note: N = Number of respondents. Milking was usually done under poor hygienic conditions where milking rooms were contaminated with cow dung and urine. More than half of the sample households (57.5%) did not clean barn before milking. The influence of dirty cows on total bacteria counts depended on the extent of soiling of teat surface and cleaning procedures, followed immediately before milking (Teshome et al., 2015). About 43% of the respondents washed udder of milking cows before milking; however, 70% of these respondents did not use towel to dry up the udder after washing. Only 22% of the respondents used common towel to dry up udder after washing. Haile et al. (2012) and Amistu et al. (2015) also reported that 70−82.5% of smallholder farmers in Ethiopia did not practice drying up of udders using individual towel. The results of this study are consistent with the findings of Teshome et al. (2015) who reported that 71.79% of the household milk producers washed the teats and udder of the cows before milking, but without using detergents for cleaning udder and teats. 3.6. Milk Handling Practices at Collection Center Milk collection in the study area usually takes place in the morning time for both evening and morning milk. Milk was usually sold only in the morning times and hence milk producers store the evening milk in cold water to keep milk temperature lower until the next morning to reduce microbial multiplication. About 86.7% of milk was directly collected from dairy producer, while 13.3% of milk collectors buy milk from milk venders. All dairy farmers deliver milk to milk collection center by themselves (Table 8). The majority of milk collectors in the study area practiced milk quality test (Table 8). The common quality tests in the study areas were lactometer reading and alcohol test. However, 6.6% of milk vendors did not apply milk quality test. The major dairy processing plants (96.7) such as Lame Dairy PLC (Shola Milk Enterprise), MB PLC (Family Milk), Sebeta Agro- industry (Mama Dairy) and Elemtu Integrated Milk Industry were the formal customers that buy milk from those private milk collectors in the study area. Hotels and restaurants in Fitche town were also customers of the milk collectors. The equipment used to store and transport milk at collection centers was plastic container (80%) and only 20% of the collectors used stainless steel. Alemnesh et al. East African Journal of Sciences Volume 14 (2) 151-162 158 Table 8. Milk handling practices at collection centers of the study area. Variable Urban (N = 15) Peri-urban (N = 15) Overall (N = 30) N % N % N % Source of milk Farmers 12 80.0 14 93.3 26 86.7 Milk vender 3 20.0 1 6.7 4 13.3 Mode of delivery Farmer deliver the milk 15 100.0 15 100.0 30 100.0 Milk quality test methods upon delivery Organoleptic test 42 6.7 74 6.7 11 36.7 Lactometer and alcohol test 11 73.3 6 40.0 17 56.7 No test 0 0.0 2 13.3 2 6.6 Type of costumer Milk processing plant 14 93.3 15 100.0 29 96.7 Hotel and restaurant 1 6.7 0 0 1 3.3 Milk transportation utensils Stainless steel 5 33.3 1 6.7 6 20.0 Plastic water bottles 10 66.7 14 93.3 24 80.0 Milk cooling facility Yes 3 20.0 0 0 3 10.0 No 12 80.0 15 100.0 27 90.0 Note: N = number of respondents. 3.7. Milk Marketing Out of the total milk produced per day, the biggest share was supplied to the market. Producers also processed milk into butter and cottage cheese (Table 9). The total milk (TM) produced per day per household was significantly (P < 0.05) higher for urban (24.63 liters) than that of the peri urban (16.86 liters) households. All dairy producers who sell milk in the study area entered contractual agreements with milk collectors to deliver milk on daily bases and to collect milk money every fortnight. Table 9. Quantity of milk produced, processed, consumed and marketed in the study area. Variable Urban Peri urban Over all mean Mean ± SE Mean ± SE Mean ± SE TM produced/HH/day (liters) 24.63 ± 1.67a 16.86 ± 1.32b 20.75 ± 1.19 TM processed HH/day (liters) 2.33 ± 0.22 1.26 ± 0.19 1.80 ± 0.15 TM consumed HH/day (liters) 1.72 ± 0.10 0.97 ± 0.09 1.35 ± 0.77 TM sold/HH/day (liter) 20.57 ± 1.51a 14.77 ± 1.19b 17.67 ± 0.99 Time to arrive at market place (in min) 12.15 ± 1.07b 20.61 ± 1.68a 16.38 ± 1.6 Note: TM = Total milk and HH = house hold. Means with different superscripts in the same rows are significantly different (P <0.05). Private milk collectors and cooperatives/union buy milk from the producers on credit basis. Establishment of milk groups and milk-collection centers gave dairy farmers a broader choice of milk marketing instead of being dependent on local traders and neighborhood buyers. Thus, one entry point for intervention to improve the dairy sector could be the formation of new dairy cooperatives as well as strengthening the existing dairy cooperatives (Birhanu, 2013). 3.8. Constraint of Milk Production, Quality and Marketing The major constraints affecting dairy production in the study area were shortage of feed, lack of land, lack of productive dairy breeds, lack of clean water and presence of poor animal health services (Table 10). The current finding was in line with the result of Tsegaye et al. (2015) who reported feed shortage, animal health as well as water and labor scarcity problems being the major challenges which affect dairy cattle production and productivity in selected district of Sidama Zone, Southern Ethiopia. Moreover, limited awareness on hygienic handling, lack of appropriate materials used for milking and milk handling, shortage of capital and hygiene of the milker were found to be the other important constraint to the dairy sector. With regard to market related problems, majority of the respondents reported the absence of adequate milk markets during fasting seasons, high feed prices, and low milk and milk product prices. On the other hand, price regulatory mechanisms were not in place to make such important food item easily available with an affordable price to the large segment of the consumers. Alemnesh et al. ‘Producing, Processing, Marketing and Hygiene of Cow Milk 159 Table 10. Milk production, quality and marketing constraint in the study area. Production constraints 1st 2nd 3rd index Rank Poor quality and quantity feed 80 38 2 0.45 1st Lack of land 37 63 0 0.34 2nd Lack of productive dairy breeds 0 0 61 0.09 3rd Lack of clean water 0 20 16 0.08 4th Poor animal health 0 20 8 0.04 5th Milk and milk product quality related constraints 1st 2nd 3rd index Rank Limited awareness on hygienic milk handling 56 55 0 0.37 1st Lack of appropriate utensils for milking and milk handling 46 49 9 0.33 2nd Shortage of capital 16 34 8 0.17 3rd Poor hygiene of the milker 6 23 32 0.13 4th Marketing related constraints 1st 2nd 3rd index Rank Lack of adequate markets during fasting season 62 16 37 0.39 1st Increased feed prices 35 59 13 0.36 2nd Low price of milk and milk products 23 26 11 0.20 3rd Discarding of milk delivered to milk collector 0 8 19 0.05 4th Note: Index = the sum of (3 times first order + 2 times second order +1 times third order) for individual variables divided by the sum of (3 times first order + 2 times second order +1 times third order) for all variable. 3.9. The Microbial Quality of Raw Cow Milk in the Study area The mean value of aerobic mesophilic bacterial count (AMBC) of raw milk samples collected from producers (6.42 ± 0.07) was significantly (P <0.05) lower than that of the milk collectors (7.49 ± 0.10) (Table 11). However, significantly (P <0.05) lower bacterial counts of raw milk were observed in both sampling sources of urban areas of the district. The differences in the overall mean of bacterial counts observed in the study area might be attributed to the time elapsed after milking which is longer for collection centers. Similar values of AMBC 7.28 log10 cfu ml-1 was report by Haile et al. (2012) for milk samples collected from different farm sizes in Hawassa, Southern Ethiopia. Solomon et al. (2013) also reported 7.08 log10 cfu ml-1 for raw milk samples obtained from the selected large- scale dairy farms in Debre-Zeit town. Table 11. Bacterial counts of raw cow milk produced and marketed in study area. Parameters Milk Sampling sources Location Overall mean Urban Peri-urban AMBC Farm gate 6.22 ± 0.10c 6.62 ± 0.13b 6.42 ± 0.07 Collection center 6.99 ± 0.15b 7.99 ± 0.15a 7.49 ± 0.10 TCC Farm gate 3.87 ± 0.13d 5.10 ± 0.13c 4.49 ± 0.09 Collection center 6.96 ± 0.18 b 7.13 ± 0.18 a 7.05 ± 0.10 SFBC Farm gate 2.42 ± 0.74d 2.77 ± 0.10c 2.95 ± 0.05 Collection center 3.27 ± 0.10b 4.13 ± 0.10a 3.7 ± 0.07 Note: AMBC = Aerobic Mesophilic Bacteria Count; TCC= Total Coliform Count and SFBC = Spore Forming Bacteria Count. Means in all columns and rows bearing with different superscripts letters for the same parameter are significantly different from each other (P <0.05). However, the results obtained in this study are lower than the findings of Haile et al. (2012) and Teklemichael (2012) who reported a total bacterial count of 9−10 log10 cfu ml-1. According to Quality and Standards Authority of Ethiopian (Ethiopian Standards, 2009), good quality milk should not contain a total bacterial count of more than 5 log10 cfu ml-1 which indicated that the milk produced and marketed in the study area did not meet the quality standards set by the same Authority. The majority of the households in the study area reported the use of plastic containers for milking, transporting and storage. But these types of containers are not easy to clean with locally available cleaning methods and hence the milk residue may favor microbial multiplication that ultimately leads to having poor quality milk. The overall mean total coliform count (TCC) observed from farm gate (4.49 ± 0.09) is lower than the result of Teshome et al. (2014) who reported an average TCC of 4.99 ± 0.081log10 cfu ml-1 for milk marketed in Shashemene town. However, the current study showed higher TCC values than the finding of Abebe et al. (2012) who report 4.18 ± 0.01 log10 cfu ml-1 for raw milk samples in the Ezha districts of the Gurage Zone. In the current study, the TCC of raw milk sampled from collection centers (7.05 ± 0.10) were significantly (P < 0.001) higher than that of milk samples taken from the farm gate. According to ES (2008), the TCC Alemnesh et al. East African Journal of Sciences Volume 14 (2) 151-162 160 of good quality raw milk should not exceed 3 log10 cfu ml-1. The presence of high TCC in milk indicates unsanitary conditions of milk production, processing and storage. Moreover, presence of large number of TCC in dairy products is an indicator of potential hazard to consumer’s health due to possible presence of other enteric pathogens (Godefay and Molla, 2000). The Spore Forming Bacteria Count (SFBC) found from the samples of peri urban producers and collectors was significantly (P < 0.0001) higher than that of milk samples from urban producers and collector (Table 11). This result is lower than the report of Teshome et al. (2014) who indicated 4.703 ± 0.069 log10 cfu ml-1 in Shashemene town. 4. Conclusion The results of the current study have demonstrated that the major feed sources for cattle in the study area were a combination of grazing, grass hay, crop residues, concentrate and non-conventional feed resources like atella and bean hull. The overall average numbers of lactating local and crossbred cows per household in the study area were 1.86 ± 0.15 and 1.75 ± 0.16, respectively. About 99.2% of the farmers used separate house type barn for cows of which about 65.8% used corrugated iron sheet covered barns and the rest used grass thatched roof barns. Mean aerobic bacterial count (AMBC), coliform count (CC) and spore forming bacterial count (SFBC) for milk samples collected from milk collection center were significantly higher (P <0.05) than milk samples obtained from farm gates. Generally, the overall microbial count increased during milking at on-farm to collection centers, reflecting poor hygiene at milking, milk handling and transportation. This is mainly due to poor hygienic condition of the feeding system, milking environment, poor udder and teats cleaning practices, failure to use separate towel for each cow and the poor personal hygiene of the milkers. Therefore, awareness should be created on the importance of adequate udder preparation, hygienic milking environment, and use of appropriate milk equipment to produce and supply wholesome milk to the market. Moreover, milk collection centers should be equipped with cold chains, the necessary dairy equipment and quality water supply to minimize milk contamination. 5. Acknowledgements The authors would like to acknowledge Dilla University for financial support and Holleta Agricultural Research Center for technical support and laboratory facilities to conduct microbiological analysis. We also give special thanks to Girar Jarso Agriculture and Rural Development Office workers and the community for their support during data collection. 6. References Abebe, B., Zelalem, Y. and Ajebu, N. 2012. Hygienic and Microbial quality of raw whole cow’s milk produced in Ezha district of the Gurage Zone, Southern Ethiopia. Journal of Agricultural Biotechnology and Sustainable Development, 5(6): 91−98. Abrahamsen, R.K., Borg, G.I., Harstad, O.M., Haug A. and Wetlessen, A. 2007. Milk quality-future approach from a researcher’s point of view. Norwegian food research institute (Matforsk), Osloveien, Norway. Journal Animal and Feed Science 16(1): 209−226. Amistu, K., Degefa, T. and Melese, A. 2015. Assessment of raw milk microbial quality at different critical points of Oromia to milk retail centers in Addis Ababa. Food Science and Quality Management, 38: 1−9. APHA (American Public Health Association). 1992. Standard Method for the Examination of Dairy Products. 16th edition. APHA, Washington, DC, U.S.A. Pp. 213−223. Asaminew, T. 2007. Production, handling, traditional processing practices and quality of milk in Bahr Dar milk shed area, Ethiopia. MSc Thesis, Haramaya University, Ethiopia. Asaminew, T. and Eyassu, S. 2011. Microbial quality of raw cow’s milk collected from farmers and dairy cooperatives in Bahir Dar Zuria and Mecha district, Ethiopia. Agriculture and Biology Journal of North America, 2(1): 29–33. Azage, T., Berhanu, G., Dirk, H., Berhanu, B. and Yoseph, M. 2013. Smallholder dairy production and marketing systems in Ethiopia: IPMS experiences and opportunities for market- oriented development. Improving Productivity and Market Success of Ethiopian. Farmers Project Working Paper 31. ILRI, Nairobi. Berhanu, K. 2012. Market access and value chain analysis of dairy industry in Ethiopia: The case of Wolaita Zone. PhD Dissertation, Alemaya University, Ethiopia. Birhanu, M. 2013. The impact of market-oriented dairying on the socio-economic position of women farmers in Selale areas, Oromia National Regional State, Ethiopia. Pp. 157−171. In: Prah, M. (ed.). Insight into Gender Equality and Power Relations in Sub-Saharan Africa. African Books Collective, Uganda. Debir, L. 2016. Assessment of breeding practice and evaluation of estrus synchronization of dairy cattle in Sidama zone, Southern Ethiopia. MSc Thesis, Hawassa University, Ethiopia. Ethiopian Standards. 2008. Unprocessed whole/raw cow milk specification. ES: 1st ed., 3460:2008. Ethiopian Standards. 2009. Unprocessed whole/raw cow milk specification. 2nd ed., ES: 3460:2009. Alemnesh et al. ‘Producing, Processing, Marketing and Hygiene of Cow Milk 161 FDA (Food and Drug Administration). 2003. Bacteriological analytical manual online. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. Pp. 1–14. Godefay, B. and Molla, B. 2000. Bacteriological quality of raw milk from four dairy farms and milk collection center in and around Addis Ababa. Berl Munch Tierarztl Wochenschr, 113 (7-8): 276−8. Haile, W., Zelalem, Y. and Yosef, T. 2012. Hygienic practices and microbiological quality of raw milk produced under different farm size in Hawassa, Southern Ethiopia. Agricultural Research and Review. 1(4): 132−142. Kibru, B., Brihan, T. and Teka, F. 2015. Characterization of smallholder cattle milk production system in Aleta Chukko District, Southern Ethiopia. Journal of Advanced Dairy Research, 3: 132. doi:10.4172/2329-888X.1000132. Murphy, S.C. 1996. Sources and Causes of High Bacteria Count in Raw Milk: Abbreviated Review. Cornell University, Ithaca, N.Y. Pp1−4. Mustefa, A. 2012. Value chain and quality of milk in Sululta and Welmera Weredas, Oromia special zone surrounding Addis Ababa, Oromia, Ethiopia. MSc Thesis, Addis Ababa University, Ethiopia. O’Connor, C.B. 1994. Rural Dairy Technology. ILRI Training Manual No.1. 1995. International Livestock Research Institute, Addis Ababa, Ethiopia. Pp. 133. Pongruru, C.S. and Nagalla, V. 2016. Agribusiness review on milk and milk products in Ethiopia. International Journal of Economics and Business Management, 2(1): 143−152. Roberts, D. and Greenwood, M. 2003. Practical Food Microbiology. 3rd edition. Blackwell Publishing Ltd, Osney Mead, Oxford OX2 0EL, UK. SAS (Statistical Analysis System). 2009. Users Guide. Version 9.2 Institute, Inc. Carry, NC. Solomon, M., Mulisa, M., Yibeltal, M., Desalegn, G. and Simegnew, K. 2013. Bacteriological quality of bovine raw milk at selected dairy farms in Debre Zeit town, Ethiopia. Comprehensive Journal of Food Sciences and Technology Research, 1(1): 1−8. SPSS (Statistical Package for Social Sciences). 2011. Statistical Packages for Social Sciences. Version 20. SPSS Corporation. Teklemichael, T. 2012. Quality and safety of raw and pasteurized cow milk produced and marketed in Dire Dawa Town. MSc Thesis, Haramaya University, Ethiopia. Teshome, G., Fekadu, B. and Mitiku, E. 2014. Handling practices and microbial quality of raw row's milk produced and marketed in Shashemene Town, Southern Ethiopia. International Journal of Agriculture and Soil Sciences, 2: 153−162 Teshome, G., Fekadu, B. and Mitiku, E. 2015. Physical and chemical quality of raw cow's milk produced and marketed in Shashemene Town, Southern Ethiopia. Journal of Food and Agricultural Science, 5(2): 7−13. Tsegay, L., Agengew, A. and Ashenafi, S. 2015. Dairy cattle production at small holder level in Sidama Zone selected Districts, Southern Ethiopia. Food Science and Quality Management. 40(Online). Available at www.iiste.org. Zewdie, W. 2010. Livestock production systems in relation with feed availability in the highlands and central rift valley of Ethiopia. MSc Thesis, Haramaya University, Haramaya, Ethiopia. Zewdu, W. 2004. Indigenous cattle genetic resources, husbandry practices and breeding objectives in North western Ethiopia. MSc. Thesis, Haramaya University, Ethiopia. Alemnesh et al. East African Journal of Sciences Volume 14 (2) 151-162 162