Seroprevalence of brucellosis in dairy animals and their owners in selected sites, central highlands of Ethiopia

ABSTRACT

A cross-sectional study was conducted from December 2019 to May 2020 to determine the seroprevalence and identify potential risk factors of brucellosis in dairy cows and owners. Purposive sampling was performed in the respective farms and kebeles to screen recent cases of abortion. The overall seroprevalence of bovine and human brucellosis was 0.61% (95%CI:0.016–2.09) and 1.21% (95%CI:0.032–4.27) respectively using combined RBPT and CFT tests. Late stage of abortion (OR = 14.74, p = 0.0002), retained fetal membrane (OR = 32.74, p = 0.006), market-based stock replacement (OR = 16.55, p = 0.002), and presence of parturition pen (OR = 11.511, p = 0.027) was found to be significantly associated with seropositivity for Brucella infection in dairy cattle. Human housing (OR = 1.8, p = 0.002), contact with aborted fetus (OR = 21.19, p = 0.017), and drinking raw milk from aborted (OR = 5.72, 0.019) and retained fetal membrane (OR = 4.22, p = 0.029) cows all had a significant influence on human brucellosis seropositivity. A structured interview question was administered to 284 respondents. Accordingly, most respondents had no knowledge of bovine brucellosis and their zoonotic effects (93.3%) in contrast most of them drink raw milk. Implementation of a test and slaughter strategy with compensation is recommended with this low prevalence. In the case of human brucellosis, implementing one health approach framework should be practiced.

KEYWORDS:

• Abortion
• bovine brucellosis
• human brucellosis
• risk factors
• seroprevalence

Introduction

One of the world's most neglected zoonotic diseases, brucellosis is a contagious and economically significant bacterial disease of animals. Although it is still endemic in Africa, it has nearly been eradicated from the majority of developed nations (WHO 2001; Donev et al. 2010). Bovine brucellosis is an infectious and contagious disease and is predominantly a disease of sexually mature animals which usually caused by B. abortus; occasionally by B. melitensis and B. suis. It affects approximately 5% of the livestock population worldwide. The disease poses a barrier to trade of animals and animal products, represents a public health hazard, and is an impediment to free animal movement. Economic loss due to loss of calves, reduced milk production, culling and economic losses from international trade bans in tropics and subtropics (OIE 2009).

When it comes to cattle, the route of transmission typically involves contact between animals after an abortion and a retained placenta. Ingestion of contaminated feed is the most common way to contract the organisms. Other options include inhalation and conjunctival innoculation. It's also possible to spread an infection when feeding newborn calves pooled colostrum. Most of the time, sexual transmission has little impact on the epidemiology of bovine brucellosis. Semen must only be taken from animals known to be disease-free because artificial insemination can spread the illness (OIE 2009).

After rabies, brucellosis is the second most significant zoonotic disease in the world. Zoonotic pathogenic species are B. melitensis, B. abortus, and B. suis (Pappas et al. 2006). Human brucellosis is also referred to as ‘undulant fever,’ ‘Mediterranean fever,’ or ‘Malta fever.’ Infected tissues, raw milk, blood, urine, vaginal discharges, aborted animal fetuses, and placentae are the main sources of brucellosis infection in humans. It can also be spread by the inhalation of aerosols, which can occur in labs, abattoirs, and animal pens and stables. A few cases have happened as a result of unintentional self-inoculation with strain 19 Brucella vaccinations (OIE 2009). It has an impact on people of all ages and genders. A severe flu-like illness, a high, recurring fever, arthritis, and internal organs (heart failure) are the symptoms of human clinical disease (Berhe and Belihu 2007).

How and when bovine brucellosis was introduced and established in Ethiopia is not fully understood. But during the past 20 years, numerous serological investigations have shown that it is endemic and widespread (Berhe and Belihu 2007; Ibrahim et al. 2010). Both highland and lowland areas are affected by the disease in cattle (Ahmed et al. 2008; Kebede et al. 2008; Edao et al. 2018). Even though a large number of studies on bovine brucellosis have been reported in different parts of the country, studies conducted on the role of bovine brucellosis in relation to public health significance in occupationally exposed individuals are limited in intensive and extensive production systems in the study areas (Tolosa et al. 2010; Megersa et al. 2011a).

Though there is limited information on the seroprevalence of bovine brucellosis in some farms in Holeta Town, there is no previous seroprevalence report of brucellosis in small holder dairy cattle and in exposed farm owners and farm employees found in smallholder and government-owned farms in the study areas, which are located in the milk-shed areas for the capital city of Ethiopia, Addis Ababa, and its surroundings. In addition, assessment of the status of the disease and understanding of the disease awareness among the community has paramount importance in order to identify the risk factors for infection and zoonotic transmission and design appropriate measures to reduce the public health significance of brucellosis. Therefore, this study was carried out to determine the current seroprevalence status of brucellosis in bovines with a recent history of abortion in the study area; to determine the seroprevalence of brucellosis in animal owners and government farm employees; and to assess the associated risk factors of bovine and human brucellosis in the study area.

Material and methods

Study design

A cross-sectional study was conducted from November 2019 to May 2020 to study brucellosis in dairy cows with a recent history of abortion and workers of the Holeta and Adda Berga Agricultural Research Center dairy farms and dairy cow owners in Holeta town and Wolmera District who had direct contact with dairy animals.

Description of study areas

The study was carried out in the central highlands of Oromia, Ethiopia, at Holeta Town, Wolmera District, and Adea Berga EIAR dairy farm, which are well-known for their well-developed dairy production and serve as the main milk sheds for Addis Ababa. The dairy farms at the Ethiopian Institute of Agricultural Research are located in Holeta Town.

Holeta is located in the Oromia special zone, which surrounds Addis Ababa, the country's capital. The town, which is a part of Ethiopia's central highlands, is situated 29 kilometers west of Addis Ababa at 9° 30’ N and 38° 30’ E. Its altitude ranges from 2300 to 3800 meters above sea level. The average annual minimum and maximum temperatures were 6 and 22 °C, respectively. The town has a population of 23,296 people (men: 11,512; women: 11,784) as of the 2007 population and housing census (CSA. POPULATION and HOUSING CENSUS OF ADMINISTRATIVE 2007). Mixed crop-livestock farming around the towns is one of the main livestock production systems in the region (CSA 2016). The estimated total number of cattle in the study area is 175 741, of which 172 769 (98.3%) are native breeds, and 2972 (1.7%) are crossbred (WoWAHA 2015).

Adda Berga is a woreda in Ethiopia's Oromia Region, situated at 9° 15′ 0′′ N and 38° 25′ 0′′ E. The dairy farm substation of the Ethiopian Institute of Agricultural Research is located there. In 1986, 400 pregnant pure Jersey heifers and two Danish sires (foundation stock) were introduced to the Adea Berga wetland to start the Adea Berga dairy farm, which was a government state farm and produced commercial milk (Siyoum et al. 2016). In addition to serving as a bull dam station for the national artificial insemination center, the farm has been producing and raising pure Jersey breeds from the foundation stock to produce milk for dairy development firms (NAIC). The farm was then given to the Holeta Agricultural Research Center in 2007 in order to conduct genetic improvement research.

Study population

Dairy cattle with recent abortions were the study's target group. The occurrence of abortion cases in one month, referred to as ‘recent abortion,’ was assessed at the respective site during the entire period of this study. All the pure Holstein Friesian and Jersey breeds, native breeds, and Boran Holstein Friesian and Boran Jersey cross breeds, which have no vaccination history, were included in the study's dairy cow population. At the time of sampling, information on study animal-related features such as species, age, breastfeeding, reproductive status, parity number, abortion period, and history of abortion was gathered and recorded. Based on Ibrahim (Ibrahim et al. 2010), dairy cows were divided into three age groups: young, adult, and old, or <4years, 4–8 years, and >8 years (Svendsen 1997) respectively.

Employees of the Holeta and Adda Berga Agricultural Research Center dairy farms, as well as dairy cow owners in Holeta Town and Wolmera District, who have direct contact with dairy cows by the age of 18 and were willing to participate in the study and sign the informed consent form were included.

Sampling technique and sample size determination

Medium-, large-, and small-scale farms were selected purposively (Shanko and Shanko 2017). As a result, the Ethiopian Institute of Agricultural Research (HARC) dairy farm, all eight kebeles, and eleven medium-scale farms from holeta Town were all included. On the other hand, fourteen of the twenty-three kebeles in the Wolmera District were purposively chosen based on their accessibility and the number of dairy cows present. The study also includes a large-scale farm operated by the Ethiopian Institute of Agricultural Research in the Adda Bberga District on purpose.

The sample size was estimated based on the previous study results by (Shanko and Shanko 2017) in Holeta Town in dairy farms, which were 0.92% individual seroprevalence. Using the formula provided by (Thrusfield 2007), the sample size for the study was computed with a defined precision of 5% and a 95% level of confidence interval. $n={1.96}^2\mathrm{XPexX}\left(1-\mathrm{Pex}\right)d^2$where, n = required sample size, Pex = expected prevalence, and d = desired absolute precision.

Hence, based on the above formula and taking into account 0.92% individual prevalence, the minimum sample size is: $\begin{array}{rl}& n={1.96}^{2}X0.0092X\left(1-0.0092\right)(0{.05)}^2\mathrm{n}=14\\ & \end{array}$However, to increase precision and reduce standard error, all recently aborted cows in the study area during the study duration were included. Therefore, a total of 352 recently aborted cows were sampled during the study duration. On the other hand, 149 voluntary animal owners from Holeta Town and Wolmera District and 17 farm employees from Holeta and Adda Berga EIAR dairy farms were also included in the study.

Sample collection

Blood sample collection

Blood samples of 7–10 mL were collected from each cattle after disinfecting the site of the jugular vein, and 5 mL of blood samples were collected from the cephalic vein of voluntary animal owners and farm workers by a registered nurse in sterile plain vacutainer tubes. The blood samples were kept in a slanting position overnight at room temperature to separate the serum according to the (3) manual. Then each serum was gently decanted into sterile screw-capped Eppendorf tubes (1.8 ml), labeled and stored at −20°C until tested. All serum samples collected from animals were tested using RBPT and CFT (produced by Lillidale diagnostic manufacturer United Kingdom) in the serology laboratory of NVI (National Veterinary Institute), Bishoftu. Human serum samples were screed by RBPT in the Holeta Town human health clinic and confirmed for CFT.

Laboratory diagnosis

Rose bengal plate test (RBPT)

All serum samples collected from bovine and humans were screened using RBPT according to the procedures described by (Alton et al. 1988; Stear 2005) and the manufacturer’s instructions.

Complement fixation test (CFT)

The CFT test was used to confirm the results of the RBPT test on serums that were positive for the standard Brucella antigen. Titration was used to assess the reagent preparation, which was done in accordance with the World Organization for Animal Health's recommended protocols (OIE 2009).

Questionnaire survey

A structured interview questionnaire was prepared and applied to all concerned workers of the farm and animal owners who have direct contact with animals and animal products in the study area. Age, sex, educational background, awareness of brucellosis, raw milk consumption, handling of aborted fetuses, history of abortion, chronic headache, knee pain, and testicular swelling were among the personal demographic data collected in the format created.

Animal contact with other herds, abortions, stillbirths, retention of fetal membranes, and separate parturition pens were all recorded. Artificial insemination (AI), bulls, or both were used to differentiate breeding techniques. Placenta, abortion waste, and dead fetuses disposal methods included burying, feeding to dogs, and throwing into open waste bins.

Data management and analysis

Data collected from field and serological tests was coded and stored in a Microsoft Office Excel spread sheet and transferred to R. software version 4.0 for statistical analysis. The seroprevalence for animal level was calculated on the basis of RBPT and CFT positivity by dividing the number of Brucella reactors by the total number of tested animals. The Chi-square (χ2) was used to analyze descriptive questioner results and Firth's Bias-Reduced Logistic Regression analysis was employed to identify the association of seropositivity with the potential risk factors.

Result

Seroprevalence of brucellosis in dairy cattle

The present study revealed that out of 352 sampled cows (222 cross and 130 local breed), 4 (1.2%) (95%CI: 0.0047–0.0297) tested positive by RBPT. The RBPT positive were samples were further screened using CFT for confirmation. Accordingly, only 2 of them were confirmed by CFT. The overall seroprevalence of bovine brucellosis was thus 0.6% (95%CI: 0.0016–0.0209) in the study area.

Farm characteristics

Two large scale farms have semi-intensive management systems, eight medium-scale farms have intensive and four of them have semi-intensive management systems, while most small-holder (164) farmers have extensive management systems. It was also found that 57(34.8%) of sampled cows from large scales were using the AI breeding system, while 107(97.3%) of small-holder farmers were dependent on natural mating and also 76(97.4%) of small-holder farmers were using both AI and natural mating (Table 1).

Table 1. Farm characteristics of different scale farms.

Variables	Proportion of Respondents
	Large scale farm (%)	Medium scale farm (%)	Small holder farmer (%)
Type of farming
Extensive	0	0	164(100)
Intensive	0	8(25)	24(75)
Semi intensive	2(2.8)	3(4.2)	66(93)
Source of replacement stock
Own farm raised	55(19.4)	29(10.2)	200(70)
Market purchased	3(6.1)	10(20.4)	36(73.5)
Government gift	0	1(5.3)	18(94.7)
Service type
AI	56(34.8)	33(22)	48(43.3)
Natural mating	1(0.9)	2(1.8)	107(97.3)
Both	0	2(1.6)	76(97.4)
Having parturition pen
No	0	8(3.1)	251(96.9)
Yes	2(28.6)	2(28.6)	3(42.9)
Cleaning parturition area after birth
Yes	2(0.75)	2(0.75)	4(1.49)
No	0	9(3.37)	250(93.63)
Hygiene of barn
Good	30(29.7)	12(11.9)	59(59.4)
Medium	28(20)	23(16.4)	89(63.6)
Poor	0	5(4.3)	106(95.5)

Association of risk factors with brucellosis seropositivity

Univariable Firth's Bias-Reduced Logistic Regression analysis was computed to evaluate the association between brucellosis seropositivity and different risk factors. The seroprevalence of bovine brucellosis in late stage of abortion (OR = 14.75, p = 0.000), retained fetal membrane (OR = 32.73, p = 0.006), market source of stock replacement (OR = 16.55, p = 0.002), natural breeding method (OR = 7.5, p = 0.05) and parturition pen (OR = 11.54, p = 0.02) were statistically significant (Table 2).

Table 2. Univariable Firth's Bias-Reduced Logistic Regression analysis of risk factors.

Variables	No. of cows sampled	No. positive		OR(95%CI)	P-Value
		RBPT (%)	CFT (%)
Origin/Site
Adea Berga farm	30	0	0	1.0
Holeta Town	163	4(1.2)	2(0.6)	1.944(0.0744–13.1495)	0.97
Wolmera	159	0	0	1.912(1.0271–35.596)	0.4334
Herd size
Large scale	58	1(0.3)	0	1.0
Medium scale	40	1(0.3)	0	1.44(0.7754–2.6907)	0.8552
Small holders	254	2(0.6)	2(0.6)	1.158(9.2643–60.6058)	0.923
Breed
Cross	222	3(0.9)	1(0.3)	1.0
Local	130	1(0.3)	1(0.3)	1.71 (0.137–21.224)	0.645
Age
Young	78	1(0.3)	2(0.6)	1.0
Adult	233	2(0.6)	0	3.356 (0.269–42.767)	0.312
Old	41	1(0.3)	0	9.245 (0.061–18.527)	0.274
Parity status
Primipareous	57	1(0.3)	1(0.3)	1.0
Pluripareous	295	3(0.9)	1(0.3)	5.212 (0.417–65.049)	0.176
Stage of abortion
Late Sage	13	3(0.9)	2(0.6)	14.76(1.1211–2.042)	0.0002***
Early stage	339	1(0.3)	0	1.0
Source of replacement stock
Government	1	0	0	1.0
Own source	198	3(0.9)	1(0.3)	0.431(2.326–7.974)	1.000
Market purchase	16	1(0.3)	1(0.3)	16.548(8.436–24.612)	0.0022**
Both	52	0	0	4.981(2.607–7.324)	0.143
Hygiene of barn
Poor	111	2(0.6)	2(0.6)	4.635 (3.713–64.2)	0.256
Medium	140	2(0.6)	0	0.722 (3.902–133.75)	0.871
Good	101	0	0	1.0
Type of farming
Intensive	32	0	0	1.0
Extensive	164	1(0.3)	1(0.3)	1.564(8.272–22.933)	0.776
Semi-intensive	156	3(0.9)	1(0.3)	3.638(19.105–53.43)	0.394
RFM
Yes	48	4(1.2)	2(0.6)	32.74(2.611–4.544)	0.0063**
No	304	0	0	1.0
Mating practice
Natural mating	80	0	0	1.0
AI	137	4(1.2)	2(0.6)	7.581(0.609–1.049)	0.05*
Both	135	0	0	2.095(1.131–3.882)	0.715
Presence of parturition pen
Yes	7	2(0.6)	0	1.0
No	260	2(0.6)	2(0.6)	11.533(0.0604–0.22)	0.027*
Separation of cow during parturition
yes	335	4(1.2)	2(0.6)	1.0
No	17	0	0	0.262(0.02–26.72)	0.461
Cleaning of calving area after parturition
No	280	3(0.9)	2(0.6)	0.768(0.554–9.57)	0.861
Yes	72	1(0.3)	0	1.0

* = statistically significant, OR = Odds ratio, CI = Confidence interval.

The stepwise multivariable Firth's Bias-Reduced Logistic Regression analysis results showed important risk factors for bovine brucellosis seropositivity. Therefore, stage of abortion, retained fetal membrane, source of animal, and presence of parturition pen were included in the final model. Cows with a late stage of abortion, a retained fetal membrane, and market purchase herd replacement were 1.283, 1.046, and 1.0638 times more likely to be seropositive to Brucella infection, respectively, than those with an early stage of abortion, no retained fetal membrane, and herd replacement from their own or a government source (Table 3).

Table 3. Stepwise Multivariable Firth's Bias-Reduced Logistic Regression analysis of risk factors.

Variables	No. of cow tested	CFT (%)	OR(95%CI)	P-Value
Stage of abortion
Early stage	339	0	1
Late Sage	13	2(0.6)	1.283(1.215–1.3557)	0.000***
RFM
No	304	0	1
Yes	48	2(0.6)	1.046(1.0187–1.0754)	0.0014**
Source of stock replacement
Government	1	0	1
Own source	198	1(0.3)	6.549(30.066–76.357)	0.075
Market Purchase	16	1(0.3)	1.0638(1.026–1.1029)	0.0008***
Both	52	0		0.399
Presence of parturition pen
Yes	7	0	1	0.281
No	260	2(0.6)	0.025-0.032

OR = Odds ratio, CI = Confidence interval.

Seroprevalence of brucellosis in human

The present result revealed that out of 284 dairy owners, only 166 volunteered to be sampled. From serum samples tested (109 male serum and 57 female serum), a total of 7 (4.2%) (95%CI: 0.0205–0.0843) were tested positive by RBPT. Two sera were confirmed positive, giving an overall seroprevalence of human brucellosis of 1.2% (95%CI: 0.0033–0.043) in the study area.

Demographic characteristics of the respondents

A total of 284 participants were interviewed (213 male and 71 female). From those respondents, 2 (0.7%) were large-scale farm managers, 11 (3.9%) medium-scale farm owners, 254 (89.4%) small-holder farmers, and 17(5.9%) government farm employees. About 3.9% of respondents were university graduates, while 20.4% could not read and write, and 13.7% had no formal education. The majority of the respondents (49.3%) were in the 35–45 age groups. Also, 242 (85.2%) of the participants were married, and 42 (14.8%) were single (Table 4).

Table 4. Demographic characteristics of respondents.

Demographic variables	Categories	No. Respondents	Percentage (%)
Gender	Male	213	75
	Female	71	25
Age categories	<30 years	57	20.1
	31–45 years	140	49.3
	46–60 years	71	25
	>60 years	16	5.6
Marital status	Married	242	85.2
	Single	42	14.8
Educational status	Illiterate	58	20.4
	Informal	39	13.7
	Non academic	11	3.9
	Primary school	85	29.9
	Secondary school	47	16.5
	College	33	11.7
	University	11	3.9
Animal ownership status	Farm employees	17	5.9
	Small holder animal owners	254	89.4
	Medium scale farm owners	11	3.9
	Large scale farm owners	2	0.7

Association of risk factors with human brucellosis seropositivity

Human brucellosis was associated with being housed with dairy animals (OR = 1.8, p = 0.002), having contact with an aborted fetus (OR = 21.19, p = 0.017), drinking raw milk (OR = 24.99, p = 0.012), drinking raw milk from aborted animals (OR = 5.72, p = 0.019), and drinking raw milk from cows with RFM (OR = 4.217, p = 0.029) (Table 5).

Table 5. Univariable Firth's Bias-Reduced Logistic Regression analysis of risk factors associated with human brucellosis seropositivity.

Variables	Categories	No. sampled	CFT (%)	OR(95%CI)	P-Value
Origen	HARC Berga farm	4	0	1.0
	HARC Holeta farm	13	0	3.3(−1.677–6.601)	0.603
	Holeta Town	55	2(1.2)	4.2(0.276–6.145)	0.623
	Wolmera District	94	0	4.7(−0.243–9.281)	0.191
Sex	Male	109	2(1.2)	2.6(0.212–3.713)	0.486
	Female	57	0	1.0
Age	<30 years	41	0	1
	31–45 years	91	2(1.2)	1.9(0.148–2.661)	0.658
	46–60 years	30	0	0.6(−0.003–0.123)	8.38
	>60 years	4	0	6.9(−0.035–1.377)	0.373
Educational status	College	18	0	1.0
	Illiterate	34	1(0.6)	1.6(0.0839–2.468)	0.752
	Non-academic	4	0	4.1(−0.02–8.131)	0.507
	Informal	6		2.8(−0.014–5.529)	0.616
	Primary school	69	1(0.6)	0.8(0.041–1.202)	0.9004
	Secondary school	28	0	0.6(−0.003–1.223)	0.831
	University	7	0	2.4(−0.012–4.766)	0.663
Human housing with dairy animals	Common housing	14	2(1.2)	1.8(0.013–2.426)	0.0022**
	Separate housing	135	0	1.0
Contact with aborted fetus	Yes	89	2(1.2)	21(1.167–2.951)	0.017*
	No	77	0	1.0
Condition of contact with aborted material	Bare hands	89	2(1.2)	1.0
	Wear gloves	20	0	0.8(−0.602–1.090)	0.912
	Using plastic	55	0	0.3(−0.022–3.854)	0.394
	With other material	2	0	6.9(−0.045–1.258)	0.331
Washing hands after handling aborted materials	Yes	147	2(1.2)	1.0	0.609
	No	19	0	0.6(−0.051–9.375)	0. 81
Consumption of raw milk	Yes	137	2(1.2)	24(1.196–3.485)	0.012*
	No	29	0
Consumption of milk from aborted cow	Boiled	47	0	1.0
	Raw	43	2(1.2)	5.7(0.448–7.981)	0.019*
	No consumption	76	0	0.6(−0.0333–1.15)	0.813
Consumption of milk from RFM cow	Boiled	75	0	1.0
	Raw	91	2(1.2)	4.2(0.33-5.852)	0.029*
History of miscarriage	Yes	9	0	3.2(−0.023–4.422)	0.51
	No	157	2(1.2)	1.0
History of sterility	Yes	1	0	21(−01359–5.457)	0.167
	No	165	2(1.2)	1.0
History of chronic headache	Yes	66	2(1.2)	7.7(0.621–10.813)	0.117
	No	100	0	1.0
History of knee pain	Yes	68	2(1.2)	7.5(0.605–10.541)	0.123
	No	98	0	1.0
History of testicular swelling	Yes	5	2(1.2)	230(1.0533–3.43)	0.000***
	No	52	0	1.0

The most common symptoms of human brucellosis encountered by respondents were abortion 9(5.42%), sterility 1(0.6%), chronic headache 66(39.8%), knee pain 68(40.9%), and testicular swelling 5(3.01%). A statistically significant association was observed in respondents with a history of testicular swelling (Table 5).

Stepwise multivariable Firth's Bias-Reduced Logistic Regression analysis showed that human housing, contact with an aborted fetus, and drinking raw milk were included in the final model. Thus, the analysis revealed that humans who shared housing with dairy animals, had bare-hand contact with aborted fetus, and drank raw milk were 1.947, 1.022, and 1.019 times more likely to be seropositive for Brucella infection, respectively, than humans who shared separate housing, used protective materials when in contact with aborted fetus, and boiled before drinking raw milk (Table 6).

Table 6. Stepwise Multivariable Firth's Bias-Reduced Logistic Regression analysis of risk factors associated with human brucellosis seropositivity.

Variables	No. Sampled	CFT (%)	OR (95% CI)	P-Value
Human housing
With dairy animals	14	0	1
Separate housing	135	2(1.2)	1.947(0.912–0.983)	0.0049**
Contact with aborted fetus
No	77	0	1
Yes	89	2(1.2)	1.022(0.989–1.057)	0.012*
Drinking row milk
No	29	0
Yes	137	2(1.2)	1.019(0.9849–1.055)	0.046*

OR =  Odds ratio, CI = Confidence interval.

Knowledge, attitude and practices towards brucellosis

In current study the majority of respondents, 265(93.3%) had no information about brucellosis while 19(6.7%) of respondents had information and knowledge about brucellosis in human and animal (Table 7). The analysis result showed that most of the dairy animal owners, 173(60.9%) consumed raw milk and 111(39.1%) had not reported to consume raw milk. Respondents also reported that milk from aborted cows were not consumed by 128(45.1%), consumed without boiling by 80(28.2%) of respondents and 76(26.8%) of respondents were boiled before consuming (Table 8). Majority of the respondents, 250(88%) mentioned that they had poor knowledge on zoonotic diseases and its transmission mechanism via dairy animals and animal products. Only 27(9.5%) of respondents had knowledge about zoonotic diseases transmitted through raw milk consumption. On the other hands 59.2% of respondents contact with aborted material with bare hands (Table 9).

Table 7. Knowledge of respondents regarding brucellosis.

Variables	Categories	No. Respondents	Frequency (%)	Knowledge level
Knowledge on brucellosis	No	265	93.3	Poor
	Yes	19	6.7
Knowledge on zoonotic diseases transmitted by handling of infected animal and animal products	No	250	88	Poor
	Yes	34	12
Knowledge on zoonotic diseases due to raw milk consumption	No	257	90	Poor
	Yes	27	9.5

Table 8. Attitudes of respondents towards brucellosis.

Variables	Categories	No. Respondents	Frequency (%)	Attitude
Consumption of raw milk	No	111	39.1	Negative
	Yes	173	60.9	Positive
Milk from aborted cow	Boil before consumption	76	26.8	Positive
	Consume without boiling	80	28.2	Negative
	Not drink	128	45.1	Negative
Milk from cow with RFM	Boil before consumption	118	41.5	Positive
	Consume without boiling	166	58.5	Negative

Table 9. Practices of respondents regarding brucellosis.

Variables	Categories	No. Respondents	Frequency (%)	Practice
Contact with aborted fetus	No	116	40.8	Good
	Yes	168	59.2	Bad
Condition of contact with aborted material	Bare hands	168	59.2	Bad
	Wear gloves	25	8.8	Good
	Wear plastic	80	28.2	Good
	With other material	11	3.9	Good
Hand washing after contact with aborted fetus	No	259	91.2	Bad
	Yes	25	8.8	Good

Discussion

The current study revealed that the overall seroprevalence of brucellosis in animals with a history of recent abortion was 0.6%. The seroprevalence in our investigation was marginally greater than that reported by (Edao et al. (2018)), who estimated a 0.06% overall prevalence in the Addis Ababa region. The different study animal management systems may be the cause of this discrepancy in seroprevalence. In this research, small-holder farmers who kept their animals under a strict management regime provided the majority of the reactive animals. One potential route for spreading the disease to herds that are not already infected is the dependence of the majority of farmers on outside sources for stock replacement (Radostits et al. 2000).

The results of the study were in close agreement with the findings of (Tesfaye) (0.69%); (Tolosa et al. 2007) 3 (0.77%); (Gumi et al. 2013) (0.9%); (Adugna et al. 2013) (1.0%) from Ethiopia but lower than the findings of (Berhe and Belihu 2007) (3.19%); (Ibrahim et al. 2010) (3.1%); (Kebede et al. 2008) (11.0%); (Adugna et al. 2013) (2.9%); (Degefu et al. 2011) (1.38%); (Haileselassie et al. 2011) (6.1%); (Megersa et al. 2011b) (29) (3.5%); (Asmare et al. 2013a) (1.9%); (Tibesso et al. 2014) (4.3%); and (Minda et al. 2016) (1.4%) in different parts of the country. Other authors have reported relatively higher seroprevalence in other African countries: Some of the reports: 8.5% (Omer et al. 2000) from Eritrea; 24.5% (Angara et al. 2004) from Sudan; 24.0% (Matope et al. 2011) from Zimbabwe; and 5.5% (Mai et al. 2012) from Nigeria were some of the reports. The difference observed in prevalence could be due to variation in production systems and animal management (Coetzer et al. 1994; Solanki et al. 2022).

The seroprevalence of brucellosis seropositivity and breeding practices are significantly associated. The majority of farms used artificial insemination (38.9%) more frequently than bulls (22.7%) for breeding purposes in the current study areas. In contrast to the bull mating method, where there was no record of any seropositives, the AI service had a greater seroprevalence rate. It was determined that the sources of replacement stock had a substantial impact on the incidence of bovine brucellosis in the research areas (p = 0.021). Comparatively speaking, cattle brought in from other regions were more prone to brucellosis than cattle raised to replace the stock.

The study's findings indicated that there was no discernible difference in Brucella infection between breeds. The current finding is consistent with earlier reports from (Kebede et al. 2008; Asmare et al. 2013b; Geresu et al. 2016a) who claimed that the seropositivity of Brucella infection was not breed-dependent. There may not have been a substantial difference in Brucella infection between the breeds due to the differences in the number of animals sampled in each breed group. On the other hand, farmers in the research area were raising more crossbred cows, and they received greater care than native breeds.

The present study found no statistically significant difference in Brucella seropositivity between age groups. In contrast to the young and elderly age groups of dairy cattle in the study sites, where no Brucella seropositivity was seen, all positive cows (0.6%) were discovered within the adult age group. Similar findings were published by (Jergefa et al. 2009; Geresu et al. 2016b).

The higher seroprevalence of brucellosis among adult cows may be related to their advanced age as the organism may remain latent or chronic for an unspecified period before manifesting as a clinical disease. The other justification is also possible as age is one of the intrinsic factors which influences the susceptibility to Brucella infection. Brucellosis appears to be more associated with sexual maturity (Radostits et al. 2000). It is essentially a disease of sexually mature animals, and susceptibility increases with sexual maturity and pregnancy due to the influence of sex hormones and placenta erythritol on the pathogenesis of brucellosis (Solanki et al. 2022). Younger animals, on the other hand, are more resistant to infection and frequently clear infections, though latent infections can occur (Shanko and Shanko 2017).

Farm hygiene was observed based on manure disposal, drainage, and physical appearance of the animals and ventilation status of the farms. Accordingly, most of the farms’ hygiene was fair, and some were good, while others were poor. Even though this risk factor had no statistical significant association with the occurrence of brucellosis, the high seroprevalence was recorded in poor hygienic barns (0.6%). Though good hygiene on dairy farms (Mugizi (2009) is thought to be a protective factor against brucellosis, unsanitary practices have been identified as factors that will facilitate the spread of Brucella infections (Adesokan et al. 2013).

The overall seroprevalence of human brucellosis in the study area was 1.2%. It is not surprising to get people infected with brucellosis in an area where there are infected animals since the prevalence of brucellosis in humans is largely influenced by the prevalence of disease among domestic animals around us (Omer et al. 2000). The findings of this study are relevant for the country's development of a national brucellosis control program by the medical and veterinary sectors. Our study provides evidence that brucellosis is one of the public health problems among the rural and urban populations of Ethiopia.

There were many studies done in Ethiopia to figure out the seroprevalence of human brucellosis. Similar studies were reported by (Haileselassie et al. 2011) (1.2%) in Western Tigray. However, the relatively higher brucellosis seroprevalence reported by (Ahmed et al. 2008) (16.5%), (Haileselassie et al. 2011) (2.2%), (Hailemelekot et al. 2007) (3.8%), (Tibesso et al. 2014) (2.2%), and (Kassahun et al. 2006) (4.8%) compared to the current study may be due to the large sample size and/or the two studies’ use of different confirmatory tests, the CFT versus the 2-mercaptoethanol test (MET) (Coetzer et al. 1994).

Human brucellosis seroprevalence was found to be strongly associated with human housing, contact with aborted fetus, drinking row milk, drinking raw milk from aborted animals, and drinking raw milk from cows with RFM in the current study.

Statistically significant associations were observed between human Brucella seroprevalence and human housing, as well as contact with an aborted fetus, as reported by (Ahmed et al. 2008). This may be due to more cases of human brucellosis occurring in rural areas where most of the people are farmers or in close contact with animals. The other possible explanation that could be given from this finding is that both farmers, animal health personnel, and farm government employees could be infected while helping infected cows during parturition, either through abrasions or the conjunctiva, acquire infections by handling tissues containing Brucella organisms, and also contract brucellosis either by handling infected animals or by living with infected animals in similar houses.

In the present study, a strong association was observed between seropostivity and handling of parturient materials, drinking row milk, drinking raw milk from aborted animals, and drinking raw milk from cows with RFM, which is in agreement with other studies by (TADELE AD. FACULTY OF VETERINARY MEDICINE 2004; Jergefa et al. 2009). The possible reason for this may be that in our study area, most participants drank raw milk, including raw milk from aborted and retained fetal membrane cows, as it was assessed with interview questions. The primary methods of transmission of Brucella are through raw milk and contact with aborted materials.

The clinical manifestations pertaining to human brucellosis found in this study were similar to previous findings reported by (Mussie 2005; Ahmed et al. 2008; Jergefa et al. 2009). Apart from general clinical manifestations, participants with a history of testicular swelling had a statistically significant association with Brucella seroprevalence. This might be due to the fact that the primary manifestation of brucellosis in man is chronic head ache at the early stage and swelling of the testicles due to Orchitis.

A total of 267 cattle owners and 17 government farm workers were interviewed to assess their awareness levels regarding animal management, brucellosis and occupational risks using structured interview questions. Knowledge of diseases is a crucial step in the development of prevention and control measures (Shanko and Shanko 2017). Despite huge efforts by the government and non-government institutions to improve animal production in the areas, general knowledge of brucellosis among the farmers was still poor. The educational status attained by the majority of the respondents was low, which falls between reading and writing and lower grades. Furthermore, barn hygiene, proper disposal of aborted materials, and the use of a separate parturition pen were not well cooperated, particularly among small-holder farmers. This could have led to high risks of transmitting the disease within and between the herds and humans. This is in agreement with previous studies (Genene et al. 2009; Megersa et al. 2011a). Likewise, mixing of different animal species has its own economic importance by increasing the chances of transmission of brucellosis to cattle.

The knowledge level of brucellosis and other zoonotic diseases among small-holder farmers, medium-scale farm owners, and government farm employees was found to be relatively low. In addition, most of them did not wear protective gloves or other material while handling aborted animals and aborted material. A similar finding was also reported by (Genene et al. 2009). This might be attributed to their educational status, since most of the farmers could not read and write or have a primary education background. Generally, a lack of sufficient knowledge of brucellosis and other zoonotic diseases, unprotected working conditions, regular exposure from aerosol and contact through cuts and abrasion to infected materials such as aborted materials, carcasses, viscera, organs, blood and urine are considered as fertile grounds for exposure and transmission of the diseases to humans. In this regard, very little has been done by way of awareness creation about brucellosis.

Conclusion and recommendations

In Holeta towns, Wolmera District, and HARC Adea Berga dairy farm, West Shoa, Oromia Region Ethiopia, the overall seroprevalence of bovine brucellosis with recent abortion history and human brucellosis from animal owners and farm workers were 0.6% and 1.2%, respectively. The finding of positive serological reactors did not only suggest the presence of the disease in the cattle population in the areas, but also indicated the presence of foci of infection that could serve as sources of infection for the spread of the disease into unaffected animals and humans. In this finding, stage of abortion, retained fetal membrane, source of stock replacement, and breeding methods were statistically significant risk factors associated with dairy animal brucellosis seropositivity. Human housing, contact with aborted fetus and RFM, drinking raw milk from aborted animals, and drinking raw milk from cows with RFM, on the other hand, were all statistically significant risk factors for human brucellosis seropositivity. This study also provided important information on the knowledge, attitude, and practice of livestock owners and occupational workers about brucellosis that result in significant zoonotic importance of using raw milk for human consumption. This emphasizes the impact of brucellosis on animals, public health, and the need to control and prevent brucellosis in the study areas.

Based on the above conclusions, the following recommendations are forwarded to curb the further spread of the disease in both cattle and human populations:

• ✓ Isolation of aborted animals and proper disposal of aborted fetuses and fetal membranes, preferably by incineration.

• ✓ Replacement stock should be purchased from a herd known to be free of brucellosis.

• ✓ Strict movement control of animals from one area to another is necessary in order to prevent the spread and transmission of the disease from infected cattle to the non-infected ones.

• ✓ The implementation of test and slaughter policies with compensation payment to the farmers as the prevalence of the disease is low in the study area.

• ✓ Adoption of replacement stock vaccination in order to eradicate diseases and reduce their impact on the public and economic sectors.

• ✓ Awareness creation among farmers, butchery men, abattoir workers, and animal health workers about the nature and effect of the disease through formal and informal educational channels is required.

Consent for publication

I am fully agreed that this paper can be published on our journal.

Competing interests

No conflict of interest between authors.

Availability of data and materials

All required raw data is readily available.

Additional information

Funding

This work is funded by Ethiopian Institute of Agricultural Research Holeta Agricultural Research Center Animal Health National Program and Addis Ababa University College of Veterinary medicine and Agriculture. HARC provide logistics and AAU provides Rose Bengal and CFT.