Livestock Research for Rural Development 27 (3) 2015 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Status and factors affecting milk quality along the milk value chain: a case of Kilosa district, Tanzania

E D Karimuribo, P L Gallet, N H Ng'umbi, M K Matiko1, L B Massawe, D G Mpanduji1 and E K Batamuzi1

Department of Veterinary Medicine and Public Health, Sokoine University of Agriculture, Box 3021, Morogoro, Tanzania
1 Department of Veterinary Surgery and Theriogenology, Sokoine University of Agriculture, Box 3020, Morogoro, Tanzania


A study was carried out to establish the status of milk quality along the milk value chain (MVC) in Kilosa district. The study further assessed factors contributing to the reduced quality of milk along the MVC. This involved determination of microbial contamination (MC), adulteration, presence of antibiotic residues as well as the prevalence of subclinical mastitis. A total of 201 milk samples were collected from pastoral and agro-pastoral households, milk vending and selling points as well as from packed milk between October 2012 and May 2013. The MC was assessed by Total Bacterial counts and contamination with Escherichia coli, antibiotic residues using DelvoTest and adulteration by specific gravity.

Overall, 13.4% of milk samples tested had significant bacterial contamination with street vended milk having the highest contamination rate (41.9%). The range of bacterial contamination was between 38 and 1.92 x 106 colony forming units per ml (cfu/ml). E.coli contamination, detected only at farm (2.0%) and milk vending points (2.3%), was minimal. Presence of antibiotic residues in milk was at 35.3%, 30.8%, 20.5%, and 12.5% of milk samples from households, milk vending/milk serving points and packed milk, respectively. Thirty two percent of raw milk from street vending points, 18% from milk serving points and 5% from pastoralist households had low specific gravity. The prevalence of subclinical mastitis based on California mastitis test at cow (n=108) and quarter (n=430) levels was 33.3% and 17.2%, respectively.

Findings from this study suggest reduced milk quality at various levels is attributed to diseases, antibiotic residues, adulteration and contamination. This calls for concerted efforts to address these challenges in order to enhance contribution of milk to improved livelihoods of various actors in the study area.

Keywords: antibiotic residues, microbial contaminants, milk quality, pastoral system, Tanzania


Livestock keeping is one of dependable sources of livelihoods for inhabitants of Kilosa district. A number of different actors along the milk value chain are involved in selling milk as their source of income. These includes those selling raw milk such as livestock keepers (pastoralists and agro-pastoralists), milk vendors and collectors as well as those selling ready to consume milk at different selling points, restaurants and shops. As milk passes along different handlers from producer to the consumer, it is prone to contamination resulting to poor quality of milk associated with poor hygienic and handling practices as well as adulteration which poses various health risks. In Tanzania, scanty information is presently available on the quality of milk consumed in the country (Kivaria et al 2006).

High prevalence of infectious diseases may also influence quality of milk produced by sick lactating animals. This may be through direct effect of diseases such as subclinical mastitis. Milk quality may also be lowered by the presence of antimicrobial drug residues after their applications in lactating animals. Mastitis has been reported in Tanzania to be one of the diseases that affect cattle hence leading to production of poor quality milk. Previous studies reported cow-based prevalence of subclinical mastitis to be 26.3% (Mdegela et al 2005). Kilosa, like some other parts of Tanzania, is faced with high prevalence of infectious animal diseases. This situation influences use of antimicrobial drugs to prevent, control or treat livestock diseases. Failure of any actor along the milk value chain to play his or her role could contribute to the poor quality milk reaching consumers hence contributing to various health risks associated with consumption of poor quality milk. Such risks may include milk-borne diseases, antimicrobial drug resistance and associated socio-economic impacts.

The present study was designed to assess status and identify factors affecting quality of milk along the milk value chain in Kilosa District. It was anticipated that results of the study would be used to inform policy makers so as to improve quality of milk and consequently safeguarding the public health.

Materials and methods

Study area

Kilosa District is one of seven districts of Morogoro region in Tanzania. It is geographically situated between latitudes 5°55’ and 7°53’ South of Equator and longitude 36º31’ and 37º30’ East of Greenwich. Kilosa district receives bi-modal rainfall with short rains experienced from November to January and long rains from March to May. The amount of rainfall ranges from 1000 to 1400 mm in the southern flood plains whilst in the north, it ranges from 800 to 1100 mm. The average annual temperature of Kilosa is 25°C and it ranges from 19°C in July to 30ºC in March. According to the National census (2012), the population of the then Kilosa district is 631,186 (constituting 438,175 in the current Kilosa district and 193,011 in Gairo district after split in March 2012). Before March 2012, Kilosa district comprised of both Kilosa and Gairo districts that were formed after the split of the then district. In this paper, Kilosa district has therefore been used to represent the district before it was split in March 2012.

Data collection

The study was carried out in Kilosa district between October 2012 and May 2013. A cross-sectional study design was used to collect data from respondents along the milk value chain as well as milk samples that were later on analysed in the laboratory. A total of 201 milk samples were collected from study sites as shown in Figure 1. From livestock keepers and street vendors, approximately half a litre of milk samples were bought. From milk serving points, a glass of boiled fresh milk which did not contain sugar was bought. From supermarkets different sachets of fresh milk with different batch number were bought. The milk samples collected in this study were aliquoted into two portions, one portion was used to determine specific gravity in the field and the second portion was placed in a sterile container and labeled for future laboratory analysis (i.e. microbiological and antibiotic residues assays).

Figure 1. Distribution of 201 Milk samples collected along the milk value chain in Kilosa
district, Tanzania.

The assessment of subclinical mastitis status of lactating cows was done using California Mastitis Test (CMT). The CMT was performed as a cow-side test. A total of 11 households were randomly selected from Msowero, Kitete and Magole wards where 430 milk samples from 108 lactating cows were screened. Equal volumes of milk and CMT solution (Kruuse®) was added to the four cups of the CMT paddle and mixed by rotating the paddle. The results were read and graded according to the manufacturer’s recommendation based on the amount of gel formed (Radostitis et al 2006). A quarter was defined as CMT positive if it had a score of 1+ or above. A cow was defined as CMT positive if it was having at least one quarter with a CMT score of 1+ or above.

Laboratory analysis
Microbiological analysis of milk quality

Laboratory analysis was done immediately upon arrival from the field. This involved determination of total bacterial counts (TBC) and establishment of contamination of milk with Escherichia coli. For determination of TBC, tenfold serial dilution was carried out followed by transfer of 1 ml of the mixture onto the nutrient agar in sterile petri dishes. The petri dishes were then shaken for thorough mixing then incubated at 37°C for 24 h and thereafter plate counts were recorded using standard procedures (Kivaria et al 2006).

Isolation of Escherichia coli was used as an indicator for faecal contamination of milk. Mac Conkey’s agar (HiMedia Laboratories Pvt. Ltd, India) was used. Collected milk samples were directly cultured on sterile Mac Conkey’s agar petri dishes and then incubated at 44°C for 24 h. Individual colonies were then subcultured on sterile Mac Conkey’s agar petri dishes and incubated at 37°C for 24 h to obtain pure cultures. Microscopic examination was carried out according to Donkor et al (2007) to identify E.coli colonies. Confirmation of suspected E. coli isolates was done using biochemical tests which included Oxidase and Indole tests. In biochemical tests, the isolates were considered to be E.coli if they were indole positive and oxidase negative.

Assessment for of the presence of antibiotic residues

Qualitative assessment for the presence of antimicrobial residues in milk was carried using the Delvo SP® test kit (Delft, Netherlands). Briefly, the procedure involved pipetting 0.1 ml of each milk sample that was placed in an ampoule with nutrient tablet. A negative control was included in the test system by using 0.1 ml of reconstituted UHT milk considered free from antibiotic residues. The ampoules were incubated at 64 ±2°C for 3 hrs using a water bath according to manufacturer’s instruction. Positive results for samples with antibiotic residues showed a bluish purple colour as antimicrobial residues had inhibited growth of the test organism, Bacillus stearothermophilus var.colidolacti (Perme et al 2010).

Data analysis

Data was entered in Excel spread sheet and analysis was done using Epi InfoTM version 7, (CDC 2011). Descriptive statistics for variables was computed. These included determination of frequency (for categorical variables) and; mean and standard deviation (for continuous variables). Graphical presentation of data was carried out using Microsoft Office Excel 2007. The Chi square test was used to assess the significance of observed differences in proportions, and results with P-value < 0.05 were considered statistically significant.


Milk quality based on microbial contamination

The trend of bacteria load in milk samples collected along the value chain is summarized in Figure 2. It was found that the TBC levels increased sharply at street vending level followed by gradual decline at milk collection centres. The results shows that 9.8% (n=10) of raw milk from livestock keepers had higher count above recommended COMESA standard. It was further observed that 23% and 42% of raw milk sampled from vendors at milk collection centres and those from street vendors, respectively had higher total bacteria count above recommended standard. Furthermore, 5% of ready to drink milk had higher total bacteria count above recommended standard.

Figure 2. Trend of bacteria load along the milk value chain in Kilosa District Tanzania,

The status of milk quality based on bacterial contamination with Escherichia coli showed that 2% and 2.3% of raw milk from livestock keepers and vendors, respectively were contaminated. No E.coli was isolated from milk samples collected from restaurants and supermarkets.

Status of milk adulteration

The status and trend of milk adulteration defined by specific gravity is summarized in Table 1. It was found that milk adulteration was higher at street vending and ready to drink nodes. The level of adulteration was low for milk samples from livestock keepers and vendors at milk collection centres. The milk specific gravity for processed milk was within recommended East African standards (EAS). Milk specific gravity observed in all stages from producer to consumer was ranging between 1.022 g/ml and 1.042 g/ml. The proportion of adulterated milk from street vendors was significantly higher than the proportion of milk samples adulterated by farmers (5%), (P < 0.05). However there was no statistical significant difference in proportion of milk adulteration between milk from street vendors and milk from vendors at milk collection centres (P > 0.05).

Table 1. Status and trend for milk adulteration along the milk value chain in Kilosa district, Tanzania
Node Samples Mean SG±SD Range Adulteration, No. (%)
Below EAS Above EAS
Livestock keeper 102 1.034±0.002 1.03-1.04 5 (4.91) 0 (0.00)
Street milk vendor 31 1.03±0.003 1.02-1.04 10 (32.3) 0 (0.00)
Milk vendor at MCC 13 1.03±0.00 1.03-1.03 1 (7.69) 0 (0.00)
Ready to drink milk 39 1.03±0.00 1.02-1.04 7 (17.9) 2 (5.13)
Processed milk 16 1.03±0.01 1.03-1.03 0 (0.00) 0 (0.00)
Presence of antibiotic residues in milk

The presence of antibiotic residues in milk at various nodes along the milk value chain is shown in Figure 3. It was observed that there was a decreasing trend of contamination as you move from production (livestock) to consumption (processed milk). However there was no statistical significant difference in proportion of antibiotic residues among actors in value chain node (P > 0.05).

Figure 3. Percentages of samples with antibiotic residues along the milk value chain
in milk samples screened with Delvo test in Kilosa district, Tanzania
Prevalence of subclinical mastitis

The prevalence of subclinical mastitis as defined by CMT was 33.3% as cow (n=108) level and 17.2% at quarter (n=430) level.


The findings from this study suggest that milk from Kilosa district was of better quality compared to other places such as in Sudan where percentage of milk adulteration was much higher (35%) as reported by Adam (2009). The proportion of adulterated milk recorded in milk samples from street vendors was significantly higher than the proportion of milk samples adulterated by farmers. Such an observation may be explained by the vendors being more business oriented and consequently want to maximize profit through increasing the volume of milk sold. However there was no statistical significant difference in proportion of milk adulteration between milk from street vendors and milk from vendors at milk collection centres. For normal whole cow milk specific gravity ranges from 1.028 g/ml – 1.036 g/ml based on the COMESA/EAS (2006). Findings from this study suggest that the majority of livestock keepers in Kilosa District do not adulterate milk as opposed to the milk vendors who, some of them, had adulterated milk. Having specific gravity below recommended level implies that there was adulteration of milk with water which contributes to production of poor quality milk. Findings from a similar study suggest that milk sellers add water to milk because is cheaper rather than starch which may be expensive or difficult to be homogenized and obviously can be detected and discovered by the consumer (Adam 2009). It is not uncommon to find that street vendors add water in milk while sellers of ready to drink milk add both water and solids. Addition of water increases the volume of milk while addition of solids increases density of milk. Most milk sellers adulterate milk so to increase their profit. Milk is a very perishable product and its shelf life is few hours (Chanda et al 2012). Adulteration of milk with water, which is very common, does not only cause dilution of milk but also increases the risk of introducing germs into the milk, further decreasing its quality (Hossain et al 2011). Adulteration can also lead to reduced nutritive value of milk due to dilution effect which may also affect further processing of milk to produce other dairy products like cheese and yoghurt. In addition, milk adulteration may lead to introduction of hazardous pathogens that are of public health importance.

Based on the East African Standards (2006) and TZS 185: (1983), the average total bacteria count was within recommended range. It was found that the total bacteria count (TBC) levels increased sharply at street vending level followed by gradual decline at milk collection centres. Such an increase of TBC may be explained by higher rates of milk adulteration with water and use of water from unsafe sources by milk vendors. However lack of cooling technology, the use of plastic containers which are difficult to wash and long time used for transportation of milk from farms to selling points could also contribute to the increased TBC as all these factors favour bacterial multiplication in milk. Finding from this study are in agreement with other studies carried out by Kivaria et al (2006) and Adesina et al (2011). Findings by Hossain et al (2011) and Dehinenet et al (2013) showed higher counts compared to those found in the present study. The most frequent causes of high bacterial load are poor cleaning of the milking system, milking dirty udders, maintaining an unclean milking and housing environment, and failure to rapidly cool milk to less than 4°C after milking (Hossain et al 2011). High microbial counts and the occurrence of pathogens are likely to affect the keeping quality and safety of raw milk as well as products derived from it (Mubarack et al 2010). The bacterial load of milk at milk selling points and processed milk was low as most milk sample in those categories were within recommended standards. Such an observation may be attributed to treatment of milk such as boiling or pasteurization which reduces bacteria load in milk.

Contamination of milk with Escherichia coli as documented in this study was low. These findings are similar to the study carried out by Adesina et al (2011) in Nigeria who reported low prevalence of Escherichia coli (6.7%). The presence of Escherichia coli was used as an indication of faecal contamination which indicates possible presence of enteropathogenic bacteria in milk according to Abeer et al (2012). Contamination of milk with E.coli may be attributed to unhygienic milking environment or other sources of faecal contamination including unclean udder of the cow during milking. Faeces may as well be of human origin if people neither use toilet nor wash hands. The presence of Escherichia coli in milk may pose health risks through infection to people especially if not well pasteurized.

Findings from this study showed that the presence of antimicrobial residues in milk at all stages along the milk value chain. It was found that the proportion of milk samples with antimicrobial residues declined gradually from livestock keepers to the milk processing node. Reasons that could explain such a trend may be the dilution of milk from different animals that were not treated with antimicrobial drugs. It is likely, for example, that milk vendors collect milk from various livestock keepers in which case if milk containing antimicrobial residues is mixed with clean milk, the overall result would be dilution effect in terms of the presence of antimicrobial residues. Similarly, the dilution effect would prevail at milk collection centres where milk from vendors is mixed together. The percentages of milk samples with antimicrobial residues in this finding were higher than findings in other studies by Aboge et al (2000) and Kivaria et al (2006). Such variation may be attributed to the type of animal production system as in the studies by Aboge et al (2000) and Kivaria et al (2006), milk samples were smallholder farmers who apply antimicrobials at relatively less frequency compared to the pastoralists. The presence of antimicrobial residues in milk samples along the value chain implies that most people are at high risk of consuming antimicrobial residues in milk without knowing that its consumption has negative impacts to the body. According to Doyle (2006), all antimicrobials have the potential to cause allergic reactions; penicillins are most commonly implicated, affecting up to 10% of people receiving these drugs therapeutically. Potential for development of antimicrobial-resistant organism in humans is one of the impacts of the presence of antimicrobial residues in food (Jones 2009). Inhibition of starter cultures used to produce cultured milk products such as cheese and yoghurt is another problem of antimicrobial residues faced by milk processors (Khaskheli et al 2008, Jones 2009).

This study aimed to assess the udder health of lactating cows in pastoral and agropastoral households of Kilosa district. This study indicated that subclinical mastitis is prevalent in pastoral and agropastoral households of Kilosa district. The prevalence of subclinical mastitis recorded in this study is lower than that which was reported by Ali et al (2011) who reported the prevalence of subclinical mastitis at 44%. This difference may be explained by type of animals kept by livestock keepers in the study area.

Conclusion and recommendations

This study attempted to assess and document factors affecting milk quality along the milk value chain. Major conclusions and recommendations are:

All these findings contribute to reduced quality of milk calling for institution of appropriate corrective measures to reverse the situation.


This study was supported by the Enhancing Pro-Poor Innovation in Natural Resources and Agricultural Value Chains (EPINAV) programme at SUA. We acknowledge cooperation received from Kilosa District Council authorities, with particular thanks to the District Veterinary Officer, Dr. Y. Mgeni and District Community Development Officer, Mrs. R.S. Ngowi and Mr. A. Buhore for their key roles in facilitating field work in the study area. We also thank livestock keepers who participated in the study.


Abeer A A, Gouda A S A, Dardir H A and Ibrahim AK 2012 Prevalence of some milk borne bacterial pathogens threatening camel milk consumers in Egypt. Global Veterinaria 8(1): 76 – 82.

Aboge G O, Kangethe E K, Arimi S M, Kanja L W, Omore A O and McDermott J J 2000 Antimicrobial agents detected in marketed milk in Kenya. Paper Presented at the 3rd All Africa Conference on Animal Agriculture, 6 – 9 November, Alexandria, Egypt. pp. 1 – 3.

Adam A A H 2009 Milk adulteration by adding water and starch at Khartoum state. Pakistan Journal of Nutrition 8(4): 439 – 440.

Adesina A, Oshadi A A, Awoniyi T A M and Ajayi O O 2011 Microbiological assessment of cow milk under traditional management practices. Pakistan Journal of Nutrition 10(7): 690 – 693.

Ali M A, Ahmad M D, Muhammad K and Anjum A A 2011 Prevalence of subclinical mastitis in dairy buffaloes of Punjab, Pakistan. The Journal of Animal & Plant Sciences 21(3): 477-480.

CDC, Centers for Disease Control and Prevention 2011 Epi Info Statistical software for public health professionals.

Chanda T, Debnath G K, Hossain M E, Islam M A and Begum M K 2012 Adulteration of raw milk in the rural areas of Barisal district of Bangladesh. Bangaladesh Journal of Animal Science 41(2): 112 – 115.

COMESA/EAC 2006 Raw Cow Milk Specification.East African standard. Retrieved on December 4, 2014.

Dehinenet G, Mekonnen H, Ashenaf M and Emmanuelle G 2013 Determinants of raw milk quality under a smallholder production system in selected areas of Amhara and Oromia National Regional States, Ethiopia. Agriculture and Biology Journal of North America 4(1): 84 – 90.

Donkor S E, Aning G K and Quaye J 2007 Bacterial Contaminations of informally marketed raw milk. Ghana Medical Journal 41(2): 58 – 61.

Doyle M E 2006 Veterinary drug residues in processed meats. Potential health risk. A Review of the Scientific Literature. Retrieved on December 4, 2014 from FRIBrief_VetD rgRes.pdf

Hossain T J, Alam M K and Sikdar D 2011 Chemical and microbiological quality assessment of raw and processed liquid market milks of Bangladesh. Continental Journal Food Science and Technology 5(2): 6 – 17.

Jones G M 2009 On farm tests for drug residues in milk. Virginia cooperative extension, virginia polytechnic Institute and State University, and Virginia State University. Retrieved on December 5, 2014 from http://

Khaskheli M, Malik R S, Arain M A, Soomro A H and Arain H H 2008 Detection of ß - lactam antibiotic residues in market milk. Pakistan Journal of Nutrition 7(5): 682 – 685.

Mubarack H M, Doss A, Dhanabalan R and Balachander S 2010 Microbial quality of raw milk samples collected from different villages of Coimbatore District, Tamilnadu, South India. Indian Journal of Science and Technology 3(1): 0974 – 6846.

Kivaria MF, Noordhuizen M T P J and Kapaga M A 2006 Evaluation of the hygienic quality and associated public health hazards of raw milk marketed by smallholder dairy producers in the Dar es Salaam region, Tanzania. Tropical Animal Health Production 38(3): 185 – 194.

Mdegela R H, Kusiluka L J M, Kapaga A M, Karimuribo E D, Turuka F M, Bundala A, Kivaria F, Kabula B, Manjurano A, Loken T and Kambarage D M 2004 Prevalence and determinants of mastitis and milk borne zoonoses in smallholder dairy farming sector in Kibaha and Morogoro districts in eastern Tanzania.Journal of Veterinary Medicine B 51: 123 – 128.

Mdegela R H, Karimuribo E, Kusiluka L J M, Kabula B, Manjurano A, Kapaga A M and Kambarage D M 2005 Mastitis in smallholder dairy and pastoral cattle herds in the urban and peri-urban areas of the Dodoma municipality in Central Tanzania. Livestock Research for Rural Development. Volume 17, Article #123. Retrieved December 18, 2014, from

Perme T, Bizjak M, Gacnik K S and Kirbis A 2010 Validation of twinsensorbt, screening test for the Detection of ß-lactams and tetracyclines in milk, and comparison to DELVOTEST® SP NT. Slovenia Veterinary Research 47(3): 97 – 106.

Radostitis O M, Gay C C, Hinchcliff K W and Constable P D 2006 Veterinary Medicine, A textbook of the Diseases of cattle, Sheep, Pigs, Goats and Horses, Tenth Edition, Elsevier Saunders, New York, USA.

Received 20 December 2014; Accepted 24 January 2015; Published 3 March 2015

Go to top