Assessment of Milk Contamination, Associated Risk Factors, and Drug Sensitivity Patterns among Isolated Bacteria from Raw Milk of Borena Zone, Ethiopia

Background The contamination of raw milk depends on the number and type of organisms that can cause health risks, which can be judged by the presence of microorganisms and bacterial pathogens. This study evaluated bacterial contamination, the risk factor, and drug sensitivity patterns. Methods A cross-sectional study was carried out on conveniently selected 95 milk producers. Data were collected using the structured pretest questionnaire and the observation control list. Subsequently, 15 to 20 ml of milk samples were taken for laboratory analysis. The milk samples have been diluted and continuously inoculated on the number of standard plates and the blue Eosin methylene germs for the total number of bacteria and coliforms counted. Biochemical and drug sensitivity tests have been done. The version 21 Statistical Package for the Social Sciences was used for analysis. Analysis of the associated factors using binary logistical regression analysis and a P value less than 0.05 was considered to be statistically significant. Results The total bacteria (TBC) and coliform count (CC) average with the standard deviation were 7.57 ± 0.83 log10 and 6.54 ± 1.53 log10 CFU/mL, respectively. The prevalence of raw milk contaminated with TBC and TCC was 84 (88.4%) and 75 (78.9%), respectively. Lack of handwash practice before milking (AOR = 2.4 [95% CI: 0.35–16.4]) and using unclean milk containers (AOR = 7.47 [95% CI: 0.0023–28.64]) were found to be significantly associated with bacterial contamination of raw milk. The bacteria isolated were E. coli (30.7%), Staphylococcus aureus (16.7%), and Salmonella spp. (1.2%). Among isolated bacteria, 76.3% were extensive drug resistant, 13.2% were multidrug resistant, and 2.6% were resistant to all drugs tested in the current study. Conclusion Guaranteed appropriate hygiene exercise during time of milking and clean containers reduced milk contamination. Doctors should consider resistance to drugs during the treatment of patients with milk disease.


Introduction
Raw milk is a dairy product that promotes body tissue growth and maintenance by providing the necessary nutrients like proteins, energy, minerals, and vitamins [1]. Biologically important macromolecules such as proteins and casein in milk have been proven to be crucial for biochemical physiological functions which have an important effect on human health and metabolism. e immunoglobulin class in the milk is significant to protect newly born against different types of diseases [2].
Milk and other dairy products are useful for nourishing the pastoral community in Ethiopia. In lowland parts of the country where livestock rearing is the main occupation, the majority of social groups consume raw milk [3]. Raw milk is sold directly to consumers without low-temperature sterilization by producers and informal markets. Such a practice is a means of transmitting pathogens mediated by milk to humans, causing milk deterioration in the time of production, manipulation, transport, and transformation [4].
Milk contaminated with pathogenic bacteria is a major cause of foodborne disease, which is a serious health problem for millions of people in the world. Food-related diseases causing mortality or other complications due to contaminated milk increase every day and create a significant burden on the healthcare system [5]. Among food poisoning bacteria, salmonella causes the most widespread diseases in the world and is estimated at 1.3 billion gastroenteritis and 3 million deaths worldwide [6]. Similarly, food poisoning caused by Staphylococcus aureus (SFP) is also the most widespread cause of gastroenteritis in the world [7]. In addition, E. coli O157 : H7 is the other most important food mediator, causing human diarrhea, hemorrhagic colitis, and hemorrhagic therapy syndrome. Clear links between consuming raw milk and human illness were found in the cases of Campylobacter spp., Salmonella spp., Brucella melitensis, and Mycobacterium bovis [8].
Food-associated diseases are responsible for 33 to 90% of the deaths of children in Africa, and represent serious problems of the continent [9]. In developing countries, particularly Ethiopia, milk is an important source of foodborne diseases and other infectious diseases. is happens when the production of milk and various dairy products occurs under unsanitary conditions and bad production practices [10].
Overdose, misuse, and prolonged use of drugs for treatment of animals and humans lead to an alarming increase and distribution of antimicrobial resistant bacteria.
is is worsening the clinical scenario and is one of the greatest medical challenges of our time. It is also the reason for low cure rates, loss of human life, and animal life, and animal milk products [11].
Generally, raw milk and dairy product contamination control is not done on a daily basis in Ethiopia [12]. is situation increases the contamination of raw milk among pastoral communities of the country [13]. Likewise, the pastoral community of Borena zone, Gomole district, provides milk directly to consumers without taking appropriate measures. Consequently, this study was carried out to assess milk contamination, the related factors, and drug sensitivity testing in Borena zone, Gomole district, pastoral community, Southern Ethiopia.

Study Setting and Period.
is study was conducted in the Oromia regional state, Borena zone, Gomole district, Southern Ethiopia from March 1 to April 30, 2019. e Gomole District is one of the 10 districts in the Borena zone and 525 km South of Addis Ababa, located on the Addis Ababa-Moyale highway.
e Gomole district has 67,798 people in total and more than 90% of the population is in pastoral communities. is district is dry and semiaged, and the average annual daily temperature varies from 17°C to 30°C with a bimodal season. According to Animal Statistics in 2018, the region has 37,000 cows, 14,890 camels, and 108,222 goats. Raw milk and dairy products are the main export products to neighboring areas and cross the border to Kenya.

Study Design and Population Selection.
A cross-sectional study design was used on the conveniently selected study participants in the Gomole district of Borena zone. Raw milk not from selected kebele of Gomole district and their owners were excluded from the study. During the study period, milk from producers in selected Kebeles was included in the study, while the milk from vendors and milk imported from outside the selected kebeles was excluded from the study.

Sample Size and Sampling
Method. Using the double population proportion formula, the sample size was calculated by Epi Info Version 7.2.1.0 [14] with the following assumptions: power of the study � 80%, with a 95% confidence level, the ratio of unexposed : exposed is 1 : 1 including 15% nonrespondent rate. e final sample size obtained was 97. Out of 14 kebeles (locations) in the Gomole district, four kebeles were initially selected using a lottery method. e milk producers/owners in each kebele were proportionally allocated, and then participants were selected using a convenient sampling technique.

Data Collection Methods.
e pretested structured questionnaires and observation checklists were used to collect data through face-to-face interviews with the participants.
e questionnaire contains sociodemographic characteristics, such as age and sex, and other factors that can contaminate raw milk during observation to gather data on conditions around milk, containers, and the surrounding environment (S1 file).
en, aseptically collected 15-20 milliliter (mL) milk samples were directly collected from participant containers and placed into sterile screw cups. e collected samples were labeled, kept in a 4°C ice box, and transported to the microbiology laboratory for analysis.

Bacteria Enumeration and Isolation.
Total bacteria number and coliforms were counted by culturing on standard plate count agar and eosin methylene blue agar (EMBA), respectively. 1 mL of milk was transferred to a sterile test tube which contained 9 mL of sterile peptone water. e mixture was then serially diluted up to 10 −7 . One milliliter (mL) of diluted milk portion was taken and inoculated into sterile standard plate count agar media and eosin methylene blue agar (EMB) media (Oxoid, Basingstoke, UK) (S2 file). e inoculated plates were grown at 35°C overnight [15]. e plate growth colonies between 25-250 colony-forming units per milliliter (CFU/mL) of sample were taken to determine the total bacteria quality [16]. Total bacteria count became decided as the total number of CFU per milliliter of milk sample that was calculated by using this formula. CFU/mL � mean of enumerated number of colonies/Dilution factor × Volume plated. e counted TBC was determined according to the East African standard milk samples.
Escherichia coli bacteria were isolated from coliform contaminated raw milk by culturing on eosin methylene blue agar (EMBA) plates. en one or two typical greenish metallic sheen suspected colonies were examined with the Gram staining technique, and Gram-negative bacilli isolated were determined biochemically [18]. For Salmonella species isolation, one millileter of raw milk portion was added into 9 mL of lactose broth and incubated at 35°C overnight. en, one milliliter of diluted milk was inoculated in to 10 mL selenite F broth for preerichment and incubated at 35°C overnight. From enrichment media, a loop full of sample was inoculated on Xylose-lysine decarboxylase (XLD) and incubated at 35°C overnight (S2 file). After incubation, nonlactose fermenting Salmonella species suspected colonies were taken from XLD media and inoculated into nutrient agar plate and incubated at 35°C overnight. Biochemical tests were done on a pure single colony isolated from plate agar [13]. Regarding Staphylococcus aureus isolation, 0.1 mL aliquots from stored dilution of total bacteria contaminated samples were inoculated into mannitol salt agar (MSA) plates and incubated at 35°C overnight. When one or two yellowish suspected colonies grew on the media, pure colonies were transferred from MSA plate into nutrient broth (NB) tubes and incubated at 35°C overnight. en one loop full of suspension was inoculated onto nutrient agar and incubated at 35°C overnight.
en Gram staining examination, catalase, and coagulase test were performed on pure isolated colonies. e colonies formed clotting/clumping with coagulase test were considered as Staphylococcus aureus (S2 file) [19].

Drugs Sensitivity Test.
e bacteria isolates underwent drug sensitivity test which was carried out by the diffusion technology of the Kirby-Bauer disc modified by clinical and laboratory standards (CLSI). ree to five pure colonies of bacteria were collected and added into the tube that had 4 to 5 mL physiological saline and were gently mixed in the solution to adjust homogeneous turbidity with McFarland 0.5 standard.
e sterilized swab cotton in the prepared suspension was dipped and sprayed on the whole surface of Mueller-Hinton Agar (UK) in accordance with the directives of the Standard Clinical and Laboratory Research Institute [20]. en, fixed concentrations of drug discs were used for studies. e antimicrobials such as chloramphenicol, ampicillin, tetracycline, ciprofloxacin, and gentamycin were selected based on the possibility of using prescriptions for the treatment of animal infected with bacteria and the frequency of use in the country [21]. e discs were placed into the Muller-Hinton agar and incubated at 35°C overnight. e transparent inhibition zone was generated and dissemination of the bacteria after incubation, was determined and interpreted [20].

Quality Control.
e quality of the data obtained is guaranteed by following the standard procedure at each stage of the task. e questionnaire on this study was adopted in previous studies and was developed in the region. e questionnaire was pretested in advance with the completeness, and the adequacy of the regional context was tested on 5% of the participants at the District of Fichwa. Some questions have been set according to the results before the test. Daily data collectors, supervisors, and main researchers confirmed the questionnaire on site after data collection. e calibrated equipment was used to measure the reagents and other materials before using them in the process. e quality of the reagents, antibiotic discs, and disinfection solutions was guaranteed according to the manufacturer's management, and the expiration date has been confirmed. Before the distribution, the antibacterial disc was held at room temperature for 1 hour. e culture medium was applied and sterilized according to the manufacturer's directives, and then the sterile media were inspected by incubating 3 to 5% of the lot at 35°C overnight. Collection of American Culture of type (ATCC) reference species such as E. coli, S. aureus, and S. typhimurium were used for culture quality check.

Data Analysis.
e data were checked for completeness and consistency, and double data entry was made using Epidata version 3.5.1 software. en, the data were exported to SPSS version 21 for further analysis. e total bacterial and coliform count results were interpreted as direct and converted into Log10 CFU/mL. Descriptive statistics such as percentage and frequency were computed, and the mean with standard deviation or median with interquartile range was used to summarize the continuous variables accordingly. Different types of graphs or charts and tables were used to present the data. Bivariate and multivariate logistic regression analyses were used to identify factors associated with raw milk contamination. e variables in bivariate analysis with p values less than 0.25 (p < 0.25) were included in multivariate analysis. Multivariate logistic regression was performed to control for potential confounders. e model fitness was checked using the Hosmer-Lemeshow goodnessof-fit test. Finally, the strength of associations between outcome and determinant variables was expressed using adjusted odds ratios (AORs) with 95% confidence intervals, and the significance of associations was declared at a p value of less than 0.05.

Questionnaire and Observational Survey.
In total, 95 participants signed up for this study, and the response rate was 97.9%. e respondents' ages were between 19 and 67 years old, with an average and standard deviation of 38.1 ± 11.4. In terms of sex, most of the research participants 77 (81%) were female. Regarding education status, 84 (88.4%) could not write and read, only 4 (4.2%) went to primary school. On the other hand, 82 (86.3%) respondents were never trained on milk handling, while 74 (77.9%) among the participants in the research did not know the diseases transmitted via milk contamination (Table 1).
Among interviewed participants, the majority of milk producers 72 (75.8%) were using plastic containers, more near half of respondents 52 (54.7%) transported with motorcycles, and 46 (48.4%) transported milk for less than two hours ( Table 2).
Regarding hygienic practice observations, approximately 77 (81.1%) participants used containers with poor cleanliness, 45.3% had good personal hygiene, and 41.2% kept the milk under cool conditions (Table 3).

e Prevalence of Raw Milk Contamination and Bacterial
Isolates.

Risk Factors Associated with Milk Contamination.
Logistical regression analysis has shown that water supply sources for equipment cleaning, knowledge of milk collapse, handwashing practices before milking, and containers of milk were less than 0.25. However, multivariate logistical regression analysis, lack of hands wash before milking (AOR � 2.4 (95% CI: 0.35-16.4)), and unclean milk containers (AOR � 7.47 (95% CI: 0.002-28.64)) was significantly linked to bacterial contamination of raw milk (Table 6).

Antimicrobial Test of Bacteria
Isolates. Among isolated bacteria both E. coli and S. aureus were sensitive to ciprofloxacin. e isolated E. coli was highly resistant to ampicillin 19 (79.9%). From bacteria isolates, Salmonella spp. showed high resistance against the drugs tested in the current study (Table 7). e prevalence of multidrug resistance (MDR) and pandrug resistance of isolated bacteria were 13.2% and 2.6%, respectively (Table 8).
Multivariate analysis showed that milking without washing hands was nearly 2 times more likely to contaminate the milk and using unclean milk containers could contaminate raw milk nearly 7 times higher. is result was similar to studies carried out in Kampala [35], Hawassa [31], and Yabelo [22]. is may be due to the fact that the presence of bacteria on the hands and milk container goes through the milk and causes contamination.
Ampicillin was the drug highly resisted by (81.2%) E. coli. is finding is comparable with the report of previous studies [29,36,37]. Again, ampicillin was the drug mostly resisted by Staphylococcus aureus (64.3%) which means it is more comparable with the report of the previous study from Egypt [38] and Bangladesh [37] and less comparable with the   Journal of Tropical Medicine report of Adgrat [29]. Similarly, ampicillin was highly resisted by isolated Salmonella species. is finding was more similar to study done in Bangladesh [37] compared to the study conducted in Jigjiga [21] and less similar to the study reported from Tanzania [36]. e higher resistance of ampicillin by isolated bacteria in the present study may be due Statistical significance at P < 0.01 � * * * , P < 0.05 � * * and at P < 0.25 � * , COR � crude odds ratio; AOR � adjusted OR with CI � confidence interval.   Journal of Tropical Medicine to the communities' use of this drug as under-dose and extensively without prescription of antibiotics for the treatment of their diseases and also prolonged use for both prophylactic and therapeutic treatment. An overall prevalence of multidrug resistance was 13.2%. e result of the present study was less as compared to report from Jigjiga (55.2%) [21]. Multidrug resistant (≥3 drugs) of isolated S. aureus were 28.6%. is result was less than the report from Egypt at 34.8% [38]. However, S. aureus multidrug resistance was less prevalent among studied milk samples. It is one obstacle to promoting health and life saving conditions for human beings. In addition, the isolated Salmonella species have shown higher multidrug resistance, particularly among commonly available drugs like tetracycline and ampicillin.

e Study Limitations.
is study enrolled small sample size and also did not identify bacterial resistant gene to antimicrobials.
e correlation between phenotype characteristics and genetic MDR has not been evaluated.

Conclusion and Recommendations
e results of this study revealed that raw milk is contaminated by total bacteria count and total coliforms count. e isolated bacteria from contaminated raw milk were E. coli 23 (30.7%), Staphylococcus aureus 14 (16.7%), and Salmonella spp. 1 (1.2%). Factors significantly associated with milk contamination were lack of handwash before milking and unclean milk containers. All bacteria isolated showed ampicillin resistance . Resistance against commonly used drugs among the communities might be difficult for medical care. On the basis of the results of this study, the following recommendations are made: the pastoral community must clean the milk containers properly and wash their hands before starting milking. Government and nongovernmental organizations must strengthen cooperation within the sector to use or build safe sources for the community. Regional and national governments must establish a diagnostic center to identify and monitor antibacterial resistance and prevent the animal transmission of pathogens. Doctors should consider drug resistance during the treatment of patients with diseases mediated by milk.

Data Availability
Necessary data are available in the manuscript. Additional information can be received from corresponding author upon reasonable request..

Ethical Approval
e ethical approval was received from Haramaya University College of Health and Medical Sciences, Institutional Health Research Ethics Review Committee (IHRERC). e official letter was written from Haramaya University to the administration of Gomole district.

Consent
Informed letter written in the local language was provided to selected kebeles. Informed, voluntary, verbal consent was obtained from the study participant. en, 15 to 20 ml of milk samples were taken from the study participant's milk container. Study participation was voluntary.
e study participants were given the information on their right to continue/refuse their participation at any stage of the study if they did not want to participate. To ensure confidentiality of participant information, codes were used, and any identifier of participants was not written on the questionnaire.

Conflicts of Interest
e authors declare that there are no conflicts of interest.