Rabies prevention and control practice and associated factors among dog owners in Aksum town and Laelay-Machew district, north Ethiopia: community based comparative cross-sectional study

Background Rabies is nearly 100% fatal zoonotic disease. One thousand seven hundred suspected rabies exposures reported in north Ethiopia, Tigray region in 2017, which has the highest rabies prevalence from Ethiopia. Almost half of them were from Central zone only. Of these 38% were in Aksum town and Laelay-Machew. Though Rabies exposure is prevalent in Tigray, there is scanty information on rabies prevention and control practices among dog owners. Thus, this deals with rabies prevention and control practice and associated factors among dog owners. Methods and materials Comparative community based cross-sectional study was conducted in Aksum town and Laelay-Machew district from March 01 to 20, 2018. A multi stage sampling was employed to recruit 558 households. Data were collected via structured and pretested questionnaire. Data were entered into Epi_info_7 and then exported to SPSS_20 for analysis. Both descriptive and inferential analysis was done with 95% confidence intervals at p value of 5% for the final model. Result The overall prevalence of poor rabies prevention and control practice was 56% [95%CI (50, 61.9)] in urban and 62% [95% CI (57.2, 67.7)] in rural dog owners. In urban; being government employee [AOR (95%CI) = 0.35 (0.13, 0.94)], private employee [AOR (95%CI) =0.39(0.16, 0.97)] and having poor attitude [(AOR (95%CI) =1.84 (1.04, 3.25)] were significantly associated with outcome variable. Whereas in rural dwellers; having no formal education [AOR (95%CI)=6.41(1.1,38.6)], poor attitude [AOR (95%CI)= 2.19 (1.18,4.05)], having one dog [AOR (95%CI)=3.31(1.34,8.15)], travel ≥30 minute to get vaccine [AOR (95%CI)= 4.26 (2.14,8.47]), no history of dog bite exposure [AOR (95%CI)= 4.16(1.49,11.6)] and neighbors as their source of information [AOR (95%CI) =3.64 (1.31,10.1)] have statistical significance with the outcome variable. Conclusion and Recommendation The prevalence of poor rabies prevention and control practice was higher among rural dog owners. Thus; interventions should be implemented both to urban and rural residents based on the identified findings so as to promote effective rabies prevention and control activities.


INTRODUCTION
Rabies is a zoonotic disease caused by virus in the Rhabidovirus family of the genus Lyssavirus.
It transmits to human through close contact with saliva (bite or scratch) of infected animals [1,2].
Carnivorous such as dogs, cats, foxes, jackals, bats, raccoons and skunks are rabies reservoirs depending on the continents. But, Dogs are responsible for more than 95% of all rabies transmission to humans in developing countries [3].
Rabies disease is widely distributed across continents of the world. Globally, It causes around 60,000 human deaths per year despite more than 15 million people receive post exposure prophylaxis (PEP). Above 95% of deaths occur in Asia and Africa. Africa accounts for 44% of deaths. More than 40% of deaths occur in children under 15 years old [4,5]. It was assumed that more than 2,700 human lives lose estimated annually in Ethiopia in 2015 [5]. By the year 2010-2013, About 4,734 dogs had vaccinated, 3,550 ownerless dogs killed by chemical applicants, 388 dogs sterilized and for 7,050 students awareness creation programs performed in Mekelle city and its surrounding [6]. Yet, rabies is still prevalent in Ethiopia particularly in Tigray region.
There were 2,928 cases exposed to suspected rabid dog and 31 rabies deaths in Ethiopia. Of these 1,439 (49.1%) were in Tigray region ranked first from the other regions as the national surveillance data reported by Ethiopian public health institute in 2016 [7]. As the regional health bureau annual report, about 1,723 persons exposed to rabies suspected dog were reported in Tigray region. The highest proportion of cases 827 (48%) were reported from central zone. Of which individuals exposed to suspected rabid dogs were highly reported in Aksum town 314 (38%) and but relatively lower 25 (0.03%) in Laelay Machew district in 2016/2017 [8].
Although rabies is 100% preventable but, nearly 100% fatal disease once its clinical symptoms appear. Globally, it leads to over 3.7 million disability-adjusted life years and an estimated of 8.6 billion dollars economic loss annually due to premature death (55%) and direct cost of PEP (20%) [5,9].
Rabies elimination is feasible through dog vaccination and prevention of dog bites. Dog vaccination is the preferred method of controlling and eliminating rabies worldwide. According to world health organization (WHO), dog vaccination coverage should at least 70% in rabies endemic zones to eradicate/block outbreak occurrence [2]. In most African countries, where dog vaccination is not free of charge, the coverage is as low as 9% (Tanzania) [10]. Anti-rabies vaccines are expensive and consequently out of reach for many peoples in developing countries [11]. However; recent successes of rabies control through mass vaccination of dogs have been reported in Philippines, Bangladesh and South Africa [12].
Dogs were the reason for 97% of rabies related human deaths in Ethiopia. Despite the availability of vaccines both to human and dogs in the country, only 3.9% dogs were found vaccinated as the study done from 2009-2012 in Addis Ababa [13]. The widespread use of traditional medications and religious approach to treat rabies cases as evil spirit are also the challenges for prevention and control of rabies. Rabies victim individuals especially from rural areas usually come to health institutions after failing traditional intervention and loss of life from family members [14].
Prevention of human rabies needs community effort involving both veterinary and public health officials. Baseline information on the coverage rates of practices (dog vaccination, dog restrain, [15] proper care of bite wounds, seeking PEP and management of rabid dogs and carcasses) is important to predict rabies control and elimination efforts of the urban and rural dwellers [9].
There is a scarcity of information on rabies prevention and control practice and associated factors among dog owners in Ethiopia, Knowledge on rabies, formal employment, higher education, male headed households, urban residence, history of dog bite and vaccine availability were identified as significant factors for rabies prevention and control practices in the studies conducted in Tanzania, Kenya, Hawasa, Addis Ababa and Bahirdar [13,[16][17][18]. Knowledge, attitude and practice (KAP) towards rabies prevention and control found to be associated with sex, age, level of education and occupation of study participants in Debretabor town and Debark district [19,20].
Descriptive studies conducted in Gonder Zuria, Gonder town and Dessie city assessed the overall KAP towards rabies. But, these didn't identify the contributing factors to rabies prevention and control practice [21][22][23]. A cross sectional studies conducted in Hawasa town, Jimma zone, Debretabor, Bahirdar and Debark are tried to assess the significant factors for KAP on rabies.
But these are not well informative in terms of the strength, precision and direction of the association between variables and did not compare among urban and rural households [18-20, 24, 25]. In most studies even though people were familiar with rabies, there is still a gap in practices towards rabies prevention and control measures. Therefore, this study was aimed to assess rabies prevention and control practice and associated factors among dog owners in urban (Aksum town) and rural (Laelay-Machew district). Community based comparative cross sectional study was employed to compare Rabies prevention and control practice status and associated factors.

Study population
All dog owner households in randomly selected ketenas of Aksum town and Laelay-Machew district during the study period.
Household heads or their spouses of ≥18 years age (in the absence of the household head) who lived at least for six months as permanent resident in the study area and having dog at least three months old prior to this study were included in the study. However, Dog owners' household head or their spouses aged ≥18 years old who have communication problem were excluded from the study.

Sample size and procedure
Sample size calculation was determined based on the risk factors of practice on rabies prevention in different literatures. A double proportion formula was used to calculate sample size by applying Epi info stat calc version 7. The following assumptions were considered in calculating sample size: 95% confidence interval, 80% power, 1:1 ratio of exposed to non-exposed group, OR of 2.7 and taking the proportion of P1=81.1% of urban households and P2=92.1% of rural households and multiplied by 1.5 of design effect and added 10% of non-response rate and then became 558 participants [17] (Table1).
Multistage sampling technique was employed for the selection of study unit. The study areas were selected based on burden of suspected rabies exposures. There were five kebeles in Aksum town and 16 kebeles in Laelay Machew district. At 3 rd stage, data was collected using systematic random technique from the selected ketenas/ Kushet every third for rural and second household for urban after randomly select the first household (figure 3).

Data collection tool and technique
A validated tool was adopted from previous studies [17,18,20,28]. Data were collected using a structured and pretested questionnaire by trained interviewer. Primary data on sociodemographic characteristics of households, six knowledge questions on rabies (description of the disease, mode of transmission, outcome, range of species affected and means of prevention and control), ten attitude questions and six practice questions towards rabies prevention and control strategies (dog vaccination, dog restrain, timely seeking PEP, first aid, action for suspected rabid dog and carcasses management) collected from the selected households. The questionnaire was developed first in English version and translated to local language (Tigrigna) for local appropriateness and easiness in approaching the study participants and it was also back translated to English by third person to check the consistency. Household heads or their spouses were interviewed on all over the questionnaire content. Data was collected by seven trained health extension workers and two supervisors (health officers).

Data quality control
Two day training was given to the data collectors on such issues as the techniques of data collection (face to face interview skills) and supervisors on how to check the completeness and consistencies of the questionnaires filled by the data collectors to ensure data quality. The questionnaire was pre-tested on 5% (28) of the sample size out of the sampled ketenas both in the urban and rural study areas then the questions that induce ambiguity based on the response obtained was revised and rephrased. Moreover; skip pattern and sequencing of questions was incorporated based on the pretest finding. The principal investigator visited and made close follow up to the data collectors to ensure whether appropriately collecting the data to prevent bias and give day to day feedback. At the end of each day filled questionnaire validity and completeness was checked before they return back from field. The principal investigator also verified the data completeness before data entry and analysis stages. Data was cross checked using double data entry.

Analysis
After collecting the questionnaire, data was cleaned and checked for its completeness.     of urban and 21 (7.5%) rural respondents were ever bitten by a suspected rabid dog. About 52 (18.6) of urban respondents and 25 (9%) rural respondents had family members ever bitten by dog. One hundred six (38%) rural respondents travel ≥30 minutes to get dog vaccine (Table 3).   (Table 4).  (Table 5).  (Table 6).

DISCUSSION
The finding of this study showed that the overall prevalence of poor rabies prevention and control practice among dog owners was 56% in Aksum town (urban) and 62.4% in Laelay Machew district (rural) with statistical difference in prevalence among urban and rural dwellers.
This is higher than the study done in and around Ambo town in 2017 (36.5%) [29]. This might be due presence of poor community attitude towards rabies prevention and control in the current study (53%) which was 49.5% in Ambo town. Moreover; the poor practice was as a result of lower dog vaccination coverage (55%) than WHO recommendation of annual rabies vaccination coverage for herd immunity at least 70% [2].
Among urban dog owners; statistically significant association was observed in level of practice with occupation and attitude of respondents. Being government and private employees were less likely to have poor practice on rabies prevention and control than farmers. This is consistent with the study done in Nigeria and Debretabor town [19,30]. This might be probably due to the government employee have higher educational background enable to access health information than farmers. Respondents having poor attitude were more likely to have poor practice than counterparts that is consistent with the study conducted in India [31]. This implies positive attitude drives to practice well.
Whereas among rural dog owners; households having only one dog were more likely to have poor practice than two and above. This might be to prevent cross infection of two and above dogs. In other way might be due to one vaccine vial contains five doses opened for five dogs come at the same time but most time it is difficult for those having only one dog. This implies less attention was given to one dog than two and above.
Respondents who travel ≥30 minute to get dog vaccination site were more likely to have poor practice than the counterparts. This might be due to not easily access to health education programs, vaccination schedules and timing of hospital presentation. Respondents who did not have any family exposure to dog bite were higher to have poor practice than having exposure history. This is consistent with the study done in Kakamega, Kenya and Hawasa [17,18,24].This implies having exposure history helps to have good practice towards rabies prevention and control. But everybody need to aware of prevention both before and after bite exposure.
Respondents with no formal education and attended primary schools were more likely to have poor practice than counterparts. This was consistent with the KAP studies conducted in Dedo district [32], Addis Ababa [33], Debretabor town [19], Hawasa town [24], and Debark district [20]. The possible explanation could be educated persons would have better information access and can easily understand the disease prevention and control measures. Schools have great role in acquisition of comprehensive knowledge on the disease.
Respondents that their sources of information got from personal contacts (relatives or neighbors) were 3.2 times more likely to have poor practice than heard from health professionals. This may be due to lack of appropriate and comprehensive information on rabies transferred from relatives/neighbors. Respondents having poor attitude were two times more likely to have poor practice than counterparts.
Generally; rural dog owners were two times more likely to have poor rabies prevention and control than urban ones. This is consistent with the study in Tanzania [16,34]. This might be due to behavioral change of urban dog owners on early hospital presentation of dog bite victims since they were living nearby to health facility and had better access to health information from different sources.

Strength and limitation of the study
As strength of the study; this is community based comparative cross-sectional study design of urban and rural communities with high (100%) response rate, however; there could be recall bias of respondents. To minimize the recall bias we included observation part to cross check the dog vaccination certificate and tying of dogs during time of data collection.

CONCLUSION
This study revealed that poor rabies prevention and control practice was higher among dog owners in Laelay Machew district than Aksum town. Occupation and attitude of dog owner households were identified as independent associated factors for rabies prevention and control practice in Aksum town. While Level of education, distance from dog vaccination site, number of dogs in the household, history of dog bite any of family member, source of information and attitude of respondents were identified in Laelay Machew district.
Then, it need to expand and strengthen the outreach dog vaccination campaigns especially to the households travel ≥30 minutes to get vaccination to increase vaccination coverage and all dog owners with dogs aged three months and above should be encouraged to take their dogs for rabies vaccination. And collaboration is basic with veterinary office and other sectors need to strengthen and expand outreach health education on rabies prevention activities.

Declarations Ethics approval and consent to participate
Ethical clearance was obtained from ethical review board of Mekelle University, College of Health Sciences. Support letter was obtained from Tigray Regional Health Bureau, Axum town and Laelay Machew district agriculture and health offices. Informed verbal and written consent was obtained from study participants after detailed explanation on the purpose, possible risks and benefits of the study before data collection. Participants also informed the right to withdraw from interview during data collection. Confidentiality of information was secured.

Consent for publication
Not Applicable.