Quality assessment of raw and pasteurized milk in Gondar city, Northwest Ethiopia: A laboratory-based cross-sectional study

Abstract

Milk is a complete and highly nutritious source of food for human beings. However, in many developing countries, including Ethiopia, the quality of milk products has become a major health concern for consumers, particularly for infants and children. Therefore, the aim of the present study was to assess the quality of raw and pasteurized milk marketed in Gondar city, Northwest Ethiopia. A laboratory-based cross-sectional study was conducted on 90 milk samples. The samples were chosen using a simple random sampling technique. For statistical analysis, ANOVA and the Pearson correlation coefficient were used. The specific gravity of pasteurized milk, farm milk, and milk vendors were found to be 1.021, 1.027, and 1.026, respectively. Farm milk, milk vendors, and pasteurized milk had fat contents of 3.38%, 3.22%, and 3.09%, respectively. The total bacterial count in pasteurized milk, farm milk, and milk vendors was found to be 7.08, 6.73, and 6.94 log₁₀ CFU/mL, respectively. In raw milk, hydrogen peroxide (7.7%), formalin (7.7%), and water (3.8%) were found, whereas in pasteurized milk, hydrogen peroxide (50%), formalin (50%), and water (19.8%) were found. Based on the findings of this study, the quality of both raw and pasteurized milk was found to be poor as per the milk quality standards. This may cause significant public health-related problems. Therefore, an appropriate intervention should be conducted to improve the quality of milk.

1. Introduction

Milk is a popular food for all age groups of humans, in particular for children, to support their biological functions and growth [1,2]. It contains important nutritional components such as fat, protein, carbohydrates, minerals, and vitamins, all of which are required for the body's growth and maintenance [3]. Milk, on the other hand, is an ideal medium for the growth of pathogenic microorganisms, the transmission of chemicals, and the transmission of other impurities [4]. Globally, consumers are increasingly concerned about the safety of their food [5,6]. Higher microbial loads in milk samples were found, which will result in public health hazards [7].

The quality of milk is being contaminated during milking, handling, processing, and storage, but it is considered sterile before milking [8]. Pasteurization (at 72 °C for 15 s or 63 °C for 30 min) is thus advised to kill spoiling microorganisms, resulting in improved microbial quality and shelf life. However, recontamination of milk after pasteurization might occur due to inappropriate handling and storage [9,10]. Many pathogens such as Staphylococcus aureus, Salmonella spp., Listeria monocytogenes, Bacillus cereus, and Campylobacter spp. may be present in contaminated milk and can be responsible for food-borne diseases [11].

The composition of milk may also be affected by various factors, such as genetic factors like the breed of the cow and nongenetic factors like stage of lactation, feeding regime, milking interval, and completeness of milking [12,13]. Consumers need safe dairy products that are produced and processed in a sound sanitary manner, free from pathogens and harmful toxic substances (such as antibiotic residues, aflatoxins, and dioxins). Moreover, milk needs to be free from suspended materials, bad flavors, and adulteration [13]. Consequently, quality milk production and delivery are important to fulfill consumers' demand [14]. Improvement of the physicochemical parameters of milk often plays a significant role in the health and safety of consumers [13].

Recently, adulteration of dairy products has become a common challenge in developing countries, including Ethiopia [6,15]. Adulteration alters the natural composition of milk and can introduce pathogenic bacteria and other harmful substances. Consumption of adulterated milk may have an adverse effect on human health because of the entrance of toxic substances or loss of nutritional value [16]. Milk adulteration with water and other substances has been widely reported in many developing countries, including Ethiopia [6,15,17]. Moreover, consumers could be forced to pay a higher price for a product that does not fulfill the quality standard [15].

In Ethiopia, about 95% of the marketed milk has been channeled through an informal marketing system. It is hard to control the quality of this informal marketing [12,18]. Quality should not be ignored at all stages, from farm to table [5]. The minimum total solid content of milk should be 12.8% [19]. Routine monitoring of milk quality is invaluable due to the importance of public health. However, there is limited information on the quality of milk in Ethiopia. Furthermore, there is no formal quality control system in place to ensure milk quality. Therefore, this study may have its own contribution and implications for regulatory bodies and decision-makers to take some corrective actions. Hence, the aim of the present study was to determine the microbiological safety, physicochemical quality, and adulteration conditions of raw and pasteurized milk marketed in Gondar, Northwest Ethiopia.

2. Methods and materials

2.1. Study area

The study was conducted in Gondar city, Amhara regional state, which is located 739 km northwest of Addis Ababa, the capital city of Ethiopia. Gondar is located at latitude 12°36′N and longitude 37°28′E, with an elevation of 2133 m above sea level. As it is endowed with resources necessary for production, it is an ideal place for market-oriented crop and livestock commodity development. At a regional level, the livestock population is estimated to be around 16,148,390 heads of cattle, according to the Central Statistics Agency (CSA) [20].

2.2. Study design, sample size, and sampling techniques

A laboratory-based cross-sectional study was conducted on selected dairy farms, milk vendors, and supermarkets in Gondar city from April to May 2020. The sample size was determined using an environmental sample size determination formula proposed by Manly [21] using Eq. (1).

$$n=\frac{8{\sigma }^2}{{\delta }^2}$$ (1)

where n = Number of samples, σ = Standard deviation, δ = Acceptable error.

σ was estimated at 0.15 in the pilot study of the present study, and δ was considered at 0.05. Therefore, the calculated sample size was 72.

Using proportional allocation, 33 raw milk samples from dairy farms and 39 raw milk samples from milk vendors' shops were collected. The samples were chosen using a simple random sampling technique. In addition to the 72 raw milk samples, 18 full fat pasteurized milk samples were collected from six brands of commercially available pasteurized milk, with three samples of each brand. Finally, a total of 90 milk samples were collected in Gondar city.

2.3. Sample collection and analysis

Before sample collection, the sampling glass bottles were sterilized in an autoclave for 15–20 min at 120 °C. Then 250 mL of raw milk samples in sterile glass bottles and 250 mL of plastic bottles of pasteurized milk were collected aseptically. Then, samples of raw and pasteurized milk were placed in the icebox (4 °C) for transportation. Milk samples were mixed thoroughly before analysis to ensure the homogeneity of the samples. In addition, analysis was performed within 24 h of sampling [22].

2.4. Experimental procedure

2.4.1. Preliminary quality analysis of milk

The Association of Official Agriculture Chemistry (AOAC) method was used to determine the alcohol test, clot on boiling test, and total acidity [23]. The alcohol test was determined using an equal volume of milk and 68% of ethanol in test tubes. If the milk has coagulated and fine particles are present after several inversions of the test tube, it indicates positive results. On the other hand, the clot in the boiling test was determined using the boiling of milk samples in the test tube; if it is clotted and coagulated, it indicates a positive result. For the acidity test, lactic acid was neutralized with a 0.1 N solution of sodium hydroxide. The amount of sodium hydroxide is measured, and the percentage of lactic acid is calculated from this amount using Equation (2).

$$Acidity(\%)=\frac{0.1NNaOH\left(mL\right)x0.009}{mLofmilksample}\times 100$$ (2)

2.4.2. Physicochemical quality analysis

The physicochemical quality of raw and pasteurized milk (specific gravity, freezing point, lactose, protein, fat, solid fat, and ash content of milk) was determined using a calibrated milk analyzer (Lactoscan), which is a product of Milkotronic Ltd. in Nova Zagora, Bulgaria, manufactured in 2016. The total solid content was calculated using the formula: % Total Solids = % Fat + % Solid Not Fat [2]. The other parameters, pH and temperature, were determined using a digital multiparameter probe (HI98196). It was performed according to the standard operating procedures [23].

2.4.3. Analysis of milk adulteration

Adulterants such as carbonate, starch, formalin, hydrogen peroxide, sodium chloride, sugar, and the addition of water were determined using standard procedures [23].

2.4.4. Microbial quality analysis

2.4.4.1. Media preparation

The media was prepared in accordance with the instructions provided by each manufacturer, as shown in Fig. 1. For serial dilution of the milk samples, peptone water that was autoclaved at 121 °C for 15–20 min and cooled to 30 °C was used [22].

Fig. 1.

Media and serial dilution preparation.

Total Bacteria Count (TBC): The Standard Plate Count Agar (SPCA) was used to determine the TBC.

In a sterile test tube containing 9 mL of sterile peptone water, 1 mL of each milk sample was added. It was serially diluted up to 10⁻⁷ after thoroughly mixing the milk suspension, and duplicate samples at the appropriate dilution (0.1 mL) were poured into a plate containing 15–20 mL of cooled SPCA solution and thoroughly mixed. Followed plates were allowed to solidify and incubated at 32 °C for 48 h [24]. The plates with colonies ranging from 30 to 300 colony-forming units (CFU) mL⁻¹ were selected for the determination of TBC [25].

Total Coliform Count (TCC): a sterile Violet Red Bile Agar (VRBA) was used to determine the TCC.

In a sterile test tube containing 9 mL of sterile peptone water, 1 mL of each milk sample was added. Then, a thoroughly mixed suspension of the milk was serially diluted up to 10⁻⁵, and duplicate samples (0.1 mL) were poured into a plate containing 15–20 mL of cooled VRBA. The plates were allowed to solidify and then incubated at 32 °C for 24 h after thoroughly mixing [24]. Plates with dark-red colonies were counted as coliforms. For the confirmatory tests, five to ten typical colonies from each plate were transferred into tubes containing 2% Brilliant Green Lactose Bile Broth (BGLBB) and incubated at 37 °C for 48 h. Production of gas and growth within the incubation period were considered sufficient evidence for the presence of coliforms [22]. Plates with colonies ranging from 15 to 150 CFU mL^-1 were selected to determine the TCC [25].

Escherichia coli (E. coli): Positive coliform confirmatory test tubes were sub-cultured with E. coli broth medium and incubated at 44.5 °C for 24 h. Tubes showing gas production were considered E. coli positives. The samples that were positive for E. coli contamination were confirmed by cultural and biochemical examinations. For the isolation and identification of E. coli, the enriched sample was cultured on the selective medium of Levine Eosin Methylene Blue (EMB) agar and incubated at 37 °C for 24 h [26]. Finally, blue-black colonies with a metallic green sheen were counted as E. coli. Biochemical tests (Indole and Methyl Red tests) were performed to confirm E. coli.

Yeast and Mold Count (YMC): Potato Dextrose Agar (PDA) was used to determine YMC. In a sterile test tube containing 9 mL of sterile peptone water, 1 mL of each milk sample was added. Then the thoroughly mixed suspension of the milk was serially diluted up to 10⁻⁵ and duplicate samples of 0.1 mL were spread on the pre-dried surfaces of media containing PDA acidified with 10% tartaric acid. The plates were then incubated at 25 °C for 5 days [27,28]. Yeast colonies with a blue-green or off-white tinge were counted as yeast, whereas colonies with a larger and more diffuse appearance were counted as molds [29]. Plates with 10–150 colonies were used for determining YMC [22].

The total number of CFU per milliliter of milk sample was calculated using the formula provided by the standard methods of the American Public Health Association (APHA) [22] using Equation (3).

$$N\left(\frac{CFU}{mL}\right)=\frac{\sum C}{\left[\left(1\times \mathrm{n}1\right)+\left(0.1\times \mathrm{n}2\right)\right]*\mathrm{d}}$$ (3)

where.

• N = number of colonies per milliliter of milk

• ΣC = sum of colonies counted on all positive plates.

• n1 = number of plates retained at the lower dilution.

• n2 = number of plates retained in the next higher dilution.

• d = the lowest dilution factor from which the first counts are obtained.

2.5. Quality assurance of data

To maintain the quality of the study, a pretest, proper sterilization of the equipment’s aseptic procedures, control media, and triplicate analysis were used. In addition, standard sampling techniques and analysis procedures were used.

2.6. Data management and analysis

The raw data was coded and entered into a Microsoft Excel 2016 spreadsheet. Microbial data was transformed to log₁₀ values before being subjected to statistical analysis to make the frequency distribution more symmetrical. After that, the data was exported to SPSS version 22 for statistical analysis. Mean values and standard deviations were calculated. ANOVA was used to test the statistical significance of differences among sample sources, and the Pearson correlation coefficient was used to assess the relationships among the tested parameters.

2.7. Ethical consideration

Ethical clearance was obtained from the Ethical Review Committee (ERC) of the Institute of Public Health, College of Medicine and Health Sciences, University of Gondar (Ref No./IPH/837/6/2012). In addition, a supportive and permission letter were obtained from the Gondar City Technical, Vocational, Educational, and Training (TVET) and Enterprise Office. After the purpose of the study had been identified, written informed consent was obtained from dairy farmers, milk vendors, and supermarket retailers. Participants were also informed that participation was on a voluntary basis and they could withdraw from the study if they were not comfortable. Moreover, the privacy and confidentiality of the information were assured by using anonymity during data collection.

3. Results

3.1. Quality analysis of milk

Alcohol and clot on the boiling test for pasteurized milk were found to be 83.3% and 67%, respectively. The titrable acidity for farm milk, milk vendors, and pasteurized milk was found to be 0.190%, 0.197%, and 0.22%, as depicted in Table 1 [19].

Table 1.

Quality analysis of raw and pasteurized milk in Gondar, Northwest Ethiopia.

Parameters	Raw milk (n = 72)		Pasteurized milk (n = 18)	Ethiopian standard (ES) [19]
	Farm milk (n = 33)	Milk vendors (n = 39)
Alcohol test (%)	27	38	83.3	Alcohol test should be negative and milk should not coagulate on clot on boiling
Clot on boiling test (%)	18	23	67
Titrable acidity (%)	0.190	0.197	0.22	Not Available

3.2. Physicochemical quality characteristics of milk

The specific gravity of pasteurized milk, farm milk, and milk vendors were found to be 1.021, 1.027, and 1.026, respectively. The total solid content of farm milk (11.56%), milk vendors (11.36%), and pasteurized milk (9.77%) was found. Whereas, the fat content of farm milk, milk vendors, and pasteurized milk was found to be 3.38%, 3.22%, and 3.09%, respectively, as depicted in Table 2 [19,30].

Table 2.

Mean ± SD of physicochemical quality of raw and pasteurized milk in Gondar, Northwest Ethiopia.

Parameters	Raw milk (n = 72)		Pasteurized milk (n = 18)	ES limits [19]	EU limits [30]
	Farm milk (n = 33)	Milk vendors (n = 39)
pH	6.55 ± 0.09∗∗∗	6.49 ± 0.09∗∗∗	6.24 ± 0.22∗∗∗	(6.6–6.8)	(6.6–6.8)
Temperature (°C)	28.2 ± 0.77∗∗∗	27.5 ± 0.71∗∗∗	22.1 ± 0.4∗	NA	NA
Specific Gravity	1.027 ± 0.002∗∗∗	1.026 ± 0.004∗∗∗	1.021 ± 0.002∗∗∗	(1.026–1.032)	(1.027–1.035)
Freezing point (°C)	−0.55 ± 0.04∗∗∗	−0.54 ± 0.03∗∗∗	−0.45 ± 0.03∗∗∗	(-0.525 to −0.55)	(<−0.517)
Protein (%)	3.07 ± 0.33∗∗∗	3.05 ± 0.17∗∗∗	2.46 ± 0.3∗	3.2	2.9
Lactose (%)	4.67 ± 0.64∗∗∗	4.63 ± 0.31∗∗∗	3.60 ± 0.2∗∗∗	NA	4.2
Fat (%)	3.38 ± 0.57∗∗∗	3.22 ± 0.85∗∗∗	3.09 ± 0.03∗∗∗	3.5	3.5
TS (%)	11.56 ± 0.09∗∗∗	11.36 ± 0.93∗∗∗	9.77 ± 0.7∗∗∗	12.8	12.5
SNF (%)	8.18 ± 0.48∗∗∗	8.14 ± 0.39∗∗∗	6.68 ± 0.5∗∗∗	NA	8.59
Ash (%)	0.62 ± 0.04∗∗∗	0.59 ± 0.06∗∗∗	0.50 ± 0.04∗∗∗	NA	NA

NA = Not Available, SNF = Solid Not-Fat, TS = Total Solids, ES = Ethiopian Standard, EU = European Union.

^∗Significance difference (p<0.05),

^∗∗∗Significance difference (p < 0.001).

3.3. Microbiological quality of milk

The TBC in pasteurized milk, farm milk, and milk vendors was found to be 7.08, 6.72, and 6.94 log₁₀ CFU/mL, respectively. E. coli in farm milk (3.20 log₁₀ CFU/mL), milk vendors (3.25 log₁₀ CFU/mL), and pasteurized milk (2.5 log₁₀ CFU/mL). The YMC in farm milk, milk vendors, and pasteurized milk was found to be 4.85, 4.97, and 3.64 log₁₀ CFU/mL, as depicted in Table 3.

Table 3.

Mean ± SD of microbiological quality of raw and pasteurized milk (log₁₀ CFU/mL) in Gondar, Northwest Ethiopia.

Parameters	Raw milk (n = 72)		Pasteurized milk (n = 18)
	Farm milk (n = 33)	Milk vendor (n = 39)
TBC	6.72 ± 0.13∗∗∗	6.94 ± 0.05∗∗∗	7.08 ± 0.1∗∗∗
TCC	5.37 ± 0.06∗∗	5.42 ± 0.06∗∗	3.59 ± 0.04∗∗*
E. coli	3.20 ± 0.01	3.25 ± 0.05∗∗∗	2.5 ± 0.05∗
YMC	4.85 ± 0.07∗∗∗	4.97 ± 0.03∗∗∗	3.64 ± 0.09∗∗∗

TBC = Total Bacteria Count, TCC = Total Coliform Count, YMC= Yeast and Mold Counts.

^∗Significance difference (p < 0.05).

^∗∗Significance difference (p < 0.01)

^∗∗∗Significance difference (p < 0.001),

3.4. Adulteration quality of milk

The positive tests for adulterants in raw milk samples were hydrogen peroxide (7.7%), formalin (7.7%), and added water (3.8%); in pasteurized milk samples, hydrogen peroxide (50%), formalin (50%), and added water (19.8%), as depicted in Fig. 2.

Fig. 2.

Adulteration conditions of raw and pasteurized milk in Gondar city, Northwest Ethiopia.

3.4.1. Correlation analysis of preliminary quality and bacteriological parameters in raw milk

The alcohol test was positively correlated with clot on the boiling test (r = 0.63), titrable acidity (r = 0.33), and TBC (r = 0.42) (p < 0.05). TBC was positively correlated with clot in the boiling test (r = 0.50), titrable acidity (r = 0.67), and TCC (r = 0.40) (p < 0.05). Whereas, TCC was positively correlated with titrable acidity (r = 0.45) and E. coli (r = 0.47) (p < 0.05), as depicted in Table 4.

Table 4.

Correlation analysis of preliminary quality and bacteriological parameters in raw milk samples in Gondar, Northwest Ethiopia.

	Alcohol test	Clot on boiling test	Titrable acidity	TBC	TCC	E. coli	YMC
Alcohol test	1
Clot on boiling test	0.63∗	1
Titrable acidity	0.33∗	0.34	1
TBC	0.42∗	0.50∗	0.67∗	1
TCC	−0.16	0.26	0.45∗	0.40∗	1
E. coli	−0.14	0.05	0.23	0.30	0.47∗	1
YMC	−0.14	0.28	−0.45	−0.44	0.28	−0.07	1

^∗Correlation is significant at the 0.05 level, TBC = Total Bacteria Count, TCC = Total Coliform Count, YMC= Yeast and Mold Counts.

3.4.2. Correlation analysis of physicochemical parameters in raw and pasteurized milk

The freezing point of raw milk was negatively correlated with SNF (r = -0.61) (p < 0.05), protein (r = -0.76), and lactose (r = -0.75) (p < 0.01). The freezing point of pasteurized milk was negatively correlated with specific gravity (r = -0.99) and fat (r = -0.98) (p < 0.01). The TS content was positively correlated with fat (r = 0.88) and SNF (r = 0.84) (p < 0.01). Whereas the TS content was positively correlated with SNF (r = 0.9) (p < 0.01), as depicted in Table 5.

Table 5.

Correlation coefficient of physicochemical parameters between raw and pasteurized milk samples in Gondar, Northwest Ethiopia.

Variables	F.P	S.G	Protein	Lactose	Fat	TS	SNF
Raw milk
F.P	1
S.G	−0.48	1
Protein	−0.76∗∗	0.45	1
Lactose	−0.75∗∗	0.43	0.9∗∗	1
Fat	−0.05	−0.19	0.29	0.27	1
TS	−0.36	0.26	0.52	0.52	0.88∗∗	1
SNF	−0.61∗	0.72∗	0.63∗	0.66∗	0.48	0.84∗∗	1
Pasteurized milk
F.P	1
S.G	−0.99∗∗	1
Lactose	−0.41	0.32	1
Protein	−0.59	0.48	0.3	1
Fat	−0.98∗∗	0.95∗∗	0.54	0.65	1
TS	−0.67	0.59	0.58	0.78	0.78	1
SNF	−0.28	0.18	0.46	0.67	0.43	0.9∗∗	1

^∗∗Correlation is significant at the 0.01 level (2-tailed), F.P= Freezing Point, S.G = Specific Gravity.

^∗Correlation is significant at the 0.05 level (2-tailed), TS = Total Solid, SNF= Solid Not-Fat.

4. Discussion

The current findings reveal that there are significant hygienic problems with all types of milk. These hygienic problems may cause public health risks due to the consumption of unsafe milk and milk products. This is due to the fact that the majority of milk suppliers are swindlers who dilute the milk with water and other substances. The quality of the milk was found to be poor as per the milk quality standards.

In this study, alcohol tests in raw milk samples were positively correlated with titrable acidity (r = 0.33) and TBC (r = 0.42) (p < 0.05). This positive correlation might be due to the fact that alcohol tests are an indicator of bacteriological contamination of milk [12,13]. In this study, alcohol and clots in the boiling test of pasteurized milk were found at 83.3% and 67%, respectively. It is inconsistent with the ES limit [19]. Therefore, the positive tests for alcohol and coagulation of milk might be due to long-time storage of milk, poor handling practices, and initial milk contamination [13].

This study shows that the titratable acidity of raw milk among farm milk was 0.190%, and milk vendors was 0.197%, which is consistent with the study conducted in Shashemena town, Southern Ethiopia [31]. In this study, the titrable acidity of pasteurized milk was 0.22%, which is slightly higher than that of farm milk and milk vendors. The findings were not within the normal range of 0.14–0.16 of lactic acid [32]. The variation might be due to a problem with the cleanliness of production or with the temperature at which the milk is kept. In addition, it might be due to bacterial growth and multiplication during storage and transportation of milk before sale, which leads to the production of more lactic acid [31]. Titrable acidity was positively correlated with TBC in raw and pasteurized milk. The determination of acidity in milk is an important factor in judging its quality. As a result, this positive correlation could indicate that milk quality has deteriorated due to bacterial substances [33].

Cow’s milk composition is nutritionally important for consumers. At a normal state, milk has unique physicochemical properties that are used as quality indicators [32]. The pH of raw and pasteurized milk samples in this study was significantly different (p < 0.001). The mean pH values of raw milk samples were found to be 6.55 for farm milk and 6.49 for milk vendors. Those findings were slightly higher than a previous finding (6.32) in Shashemene town, Southern Ethiopia [31], and they were slightly lower than (6.6) in central highland Ethiopia [34]. The mean pH value of pasteurized milk in this study was 6.24. This finding was higher when compared to the value (6.01) reported in Oromia, Ethiopia [35]. The findings revealed that the pH of both raw and pasteurized milk was not within the permissible limits (6.6–6.8) [19,30]. The possible reason for the inconsistency might be due to poor and prolonged storage of milk at room temperature, which facilitated microbial activity that may have led to the conversion of lactose to lactic acid and a lower pH [32,36].

In the present study, the specific gravity of raw and pasteurized milk was significantly different (p < 0.001). The mean specific gravity value of raw milk samples among farm milk (1.027) and milk vendors (1.026) was found. The finding of specific gravity in a raw milk sample was somewhat lower than the 1.030 reported in Shashemene, Southern Ethiopia [31]. The specific gravity value of pasteurized milk samples in this study was found to be 1.021. This finding was slightly lower than 1.023 reported in Addis Ababa, Ethiopia [37]. Even though the specific gravity of raw milk in this study was within the range of (1.026–1.032) [19] and (1.027–1.035) [30], the specific gravity of pasteurized milk in this study was lower than these standards. It might be due to pasteurized milk being significantly adulterated with water, as has been shown by the current study results.

The TS content of raw and pasteurized milk was significantly different (p < 0.001) in this study. The mean TS content of raw milk among farm milk (11.56%) and milk vendors' (11.36%) was found to be somewhat consistent with the (11.38%) reported in the central highlands of Ethiopia [34]. However, it was lower than (12.19%) reported in the West Shoa Zone, Ethiopia [14]. The TS content of pasteurized milk in this study was found to be 9.77%. This finding was almost consistent with the 9.78% reported in Oromia, Ethiopia [35]. The findings of TS in this study were not in line with (12.8%) [19] and (12.5%) [30]. The possible reason might be due to the addition of water, which reduces the percentage of soluble solids [38].

In the present study, the fat content of raw and pasteurized milk was significantly different (p < 0.001). The mean fat content of raw milk samples among farm milk (3.38%) and milk vendors (3.22%) was found, which was somewhat less than (3.56%) found around Addis Ababa, Ethiopia [39]. The mean fat content of pasteurized milk samples in this study was (3.09%). This finding was consistent with the (3.01%) reported in Oromia, Ethiopia [35]. The finding of fat content in this study was not in line with the recommended limit of 3.5% [19,30]. This might be due to milk being adulterated with water and other adulterants, which may lower the percentage of fat content in the milk. The fat content of raw milk was significantly correlated with TS (r = 0.88) (p < 0.01). The possible reason for this positive correlation might be due to the fact that fat constituents are solids. As a result, an increment in fat content leads to an increase in the TS in the milk [2]. However, it was significantly negatively correlated with the freezing point in pasteurized milk (r = −0.98) at (p < 0.01). This negative correlation might be due to the milk being diluted with water [13,17,35].

In this study, the TBC of raw and pasteurized milk was significantly different (p < 0.001). The TBC among farm milk and milk vendors was found to be 6.72 log₁₀ CFU/mL and 6.94 log₁₀ CFU/mL, respectively, in this study. The findings revealed that farm milk had lower milk contamination than milk vendors [40]. The TBC of pasteurized milk in this study was 7.08 log₁₀ CFU/mL in this study. The finding of TBC for pasteurized milk was significantly higher than for raw milk. This finding is supported by the pH and titratable acidity results found in this study. The present findings of TBC were not within the acceptable limits of <5 log₁₀ CFU/mL [19] and <5.6 log₁₀ CFU/mL [30]. The possible explanation for this variation might be the use of unclean milking utensils, the lack of knowledge about clean milk production, the poor hygienic quality of the milking area, and the initial high contamination of milk from milk handlers or the udder of the cow [32,41].

The TCC obtained from raw and pasteurized milk was significantly different (p < 0.01), in the present study. The TCC among farm milk (5.37 log₁₀ CFU/mL) and milk vendors (5.42 log₁₀ CFU/mL) was found to be within (<5.6 log₁₀ CFU/mL) [30]. However, it was higher when compared with the acceptable limit (<4.7 log₁₀ CFU/mL) [19]. Possible reasons for the presence of TCC in raw milk might be as a result of poor hygienic practices, unsanitary milking practices, the use of contaminated water, and improperly washed and maintained equipment [42,43]. The presence of TCC in a large number of dairy products is an indication that the products are potentially risky to consumers' health [14].

The TCC in pasteurized milk was 3.59 log10 CFU/mL in this study. Total coliform counts should not be over 1 log₁₀ CFU/mL, which is an acceptable limit for “Grade A″ pasteurized milk [19,30]. Coliforms in pasteurized milk may be present due to inefficient pasteurization and recontamination during transportation and storage [37]. The TCC was positively correlated with TBC (r = 0.4) and E. coli (r = 0.47) (p < 0.05) in raw milk and with TBC (r = 0.24) and E. coli (r = 0.5) (p < 0.05) in pasteurized milk. This might be due to the fact that coliform counts are part of the total bacteria and E. coli is a constituent of coliforms [22].

In the present study, the E. coli count obtained from milk vendors and pasteurized milk was significantly different (p < 0.001) and (p < 0.05), respectively. E. coli levels were found in farm milk (3.20 log₁₀ CFU/mL) and in milk vendors (3.25 log₁₀ CFU/mL) in this study. The results were greater than 2.58 log₁₀ CFU/mL a study conducted in Djibouti [44]. However, it was slightly lower than the 3.93 log₁₀ CFU/mL reported in Sudan [45]. Whereas the E. coli in pasteurized milk in this study was 2.5 log₁₀ CFU/mL. This finding was consistent with the finding of 2.5 log₁₀ CFU/mL study conducted in Dire Dawa [46]. According to World Health Organization (WHO) guidelines, E. coli should not be present in any sample for consumable purposes [47]. The presence of E. coli in milk typically indicates faecal contamination, which is a known cause of severe diarrhea and vomiting in infants and young children who consume contaminated milk and milk products [48].

The YMC obtained from raw and pasteurized milk was significantly different (p < 0.001) in this study. The YMC among farm milk (4.85 log₁₀ CFU/mL) and milk vendors (4.97 log₁₀ CFU/mL) were found. The findings were in agreement with the (4.9 log₁₀ CFU/mL) reported in the Haramaya district, Ethiopia [49]. However, it was higher than (3.9 log₁₀ CFU/mL) reported in the Hararghe Zone, Ethiopia [40]. Whereas the mean YMC of pasteurized milk in this study was (3.64 log₁₀ CFU/mL). The findings of YMC in this study were higher than the acceptable value (<1 log₁₀ CFU/mL) [50]. The presence of high amounts of YMC in the milk indicates that the milk has been contaminated with dust, air, soil, and other contaminants due to poor hygienic practices during milking production [51]. The presence of high YMC in milk may cause spoilage [52]. Moreover, some molds are major public health concerns due to their ability to produce toxic substances (mycotoxins), which may not be easily destroyed during food processing or cooking [50].

Adulteration alters the natural composition of milk and can have serious consequences for human health [6,53]. In the present study, the value of hydrogen peroxide and formalin in raw milk was found to be 7.7% for each. Whereas hydrogen peroxide and formalin in pasteurized milk were found at 50%. This might be due to the excess addition of hydrogen peroxide in pasteurized milk to activate the inherent lactose peroxidase enzyme system that is important for the improved quality of the milk even if cooling is not possible [54]. However, this high level of hydrogen peroxide in the milk may cause ulcers, perforated gut, and mouth, throat, and stomach burns. In severe cases, it may result in breathing problems, fainting, and even death [55,56]. Milk adulteration with preservative chemicals has been widely reported globally in Sudan, Pakistan, Brazil, India, and China [[57], [58], [59]]. The addition of chemical preservatives to milk to extend its shelf life is a common practice. Unfortunately, in the long run, these adulterants may cause carcinogenic and severe human health impacts [6].

In the present study, water adulteration in raw and pasteurized milk was found to be 3.8% and 19.8%, respectively. The finding of water adulteration in raw milk was slightly similar (3.64%) to a study conducted in Hossana, Ethiopia [17]. However, it was higher than the value (2.8%) reported in the central highlands of Ethiopia [34]. The presence of water adulteration in milk might be due to the need for suppliers to increase its quantity for financial gain. However, it may cause significant public health problems by reducing nutritional value and introducing microbes and toxicants. For example, if milk is adulterated with contaminated water, it may pose a serious health risk due to the possibility of waterborne disease [14].

Limitations of the study: This study was conducted as a cross-sectional study, cause-and-effect relationships could not be established. In addition, this study did not identify the determinant factors that can affect the quality of milk.

5. Conclusions

Based on the findings of this study, the quality of both raw and pasteurized milk was found to be poor as per the milk quality standards. There is a significant hygiene risk for all types of milk; there is fraud and dilution of milk. Due to this, it may cause significant public health-related problems. Therefore, to improve the quality of milk in Northwest Ethiopia, dairy farmers, milk vendors, and suppliers should obtain appropriate training on how to maintain the quality of milk from well-trained health care professionals. In addition, milk suppliers should apply good hygienic and sanitation practices during milk production, handling, processing, and storage. Moreover, routine assessment and inspection should also be conducted to safeguard the public health from the consumption of unsafe milk and milk products.

Declarations

Author contribution statement

Belay Desye: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Wrote the paper.

Bikes Destaw Bitew, Dagnachew Eyachew Amare, Tsegaye Adane, Alem Getaneh, Zenawi Hagos Gufue: Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Funding statement

This study was supported by University of Gondar, College of Medicine and Health Sciences, Ethiopia, [Ref NO./IPH/837/6/2012].

Data availability statement

Data included in article/supplementary material/referenced in article.

Declaration of interest’s statement

The authors declare no conflict of interest.