Cross-sectional eld trial to assess the ecacy of restricted application of deltamethrin 1% ready to use formulation: To adapt and develop cost-effective tsetse control strategy to be applied in Ethiopia

Probability-based observational study design was used to collect the baseline data, data to isolate changes occurring overtime and data on last monitoring in which sampling involved stratied random selection of sampling units, since the surveying area is heterogeneous, at the outset, the area was isolated into subareas (strata) based on vegetation, climate, altitude, land use, distribution of human habitation, etc. into more or less homogeneous. Different trap traps design baited with odor attractants were deployed in a georeferenced locaitons. Then, animals were treated with deltamethrin 1%, using hypodermic syringe, at belly and legs body parts, as 0.06 ml of formulation per 1 kg of body weight which is less by 40% that needed for a whole body treatment regime The data processing was carried out based on quantitative data analyzing methods.


Background
An overview to the problem Tsetse y species that are of major economic importance in Ethiopia are con ned to semi-arid, sub-humid and humid lowlands of the southern and western regions between longitude 33 o and 38 o E and latitude 5 o and 12 o N which amounts to about 200,000 square kilometers. Across all tsetse infested areas of the country more than 14.8 million cattle are under direct exposure to the disease, though, at different degree of risks [1].
African animal trypanosomosis (AAT) is one of the`major constraints to arable agriculture and livestock production and productivity in Ethiopia. Tsetse-transmitted animal trypanosomosis is a complex disease that directly and indirectly impacts the livestock and crop agriculture, respectively. As a result of direct consequence of the tsetse transmitted trypanosomosis people are continually evicted. The direct impacts are mortality and morbidity, milk yields are low and carcass is poor, calving survival rate reduced, disruption of oestrus cycle and reduction of conception rate, reduction in libido, affected animals lack stamina and strength; even more signi cant may be the indirect impact on crop production, land use, ecosystem structure and function, and human welfare. As a conesequence of trypanosomosis, the potential development of livestock and crop production is seriously constrained. [2,3]. A signi cant effect of the disease is that it prevents the use of animal traction. Consequentially, farmers have to rely on manual tillage of the land which is associated with much lower crop yield. The absence of draft animals due to trypanosomosis leads to delayed planting, lower crop yields and higher production costs; above and all, it deters resettlement patterns. The lack of oxen, in turn, compelled many farmers to abandon their farm land as a result of the widely distributed ies. This may have implications on the agroecosystems by reducing the amount of cultivated land, manure availability and the quality of land preparation. Generally, the socio-economic and nutritional status of people in these tsetse infested areas, draft power facility and direct bene ts of animals are also seriously inhibited.
Currently, the most widely used method to contain trypanosomosis is through the distribution of imported trypanocidal drugs. This costly therapy is essentially a palliative and not a long-term solution to the trypanosomosis problem. Therefore, the heavy reliance of livestock owners on trypanocides, the alarming emergence of resistance together with the low adoption of other alternatives have been increasingly recognized as a major constraint to livestock production in the area. Previous studies conducted in the Ghbe valley, [4] proved that the three trypanocides (Isomethamedium chloride, Diminazine aceturate and Homidium) used to control tsetse-transmitted trypanosomosis in domestic animals have been in use for over six decades; hence, resistance of trypanosomes has emerged to these currently available trypanocidal drugs. As a result, the livelihood of millions of resource poor livestock owners is crippled where by causing ill health and death to their animals. Generally, shortage of drugs, high treatment cost and development of drug resistance by the parasite in one hand and reduced crop production due to the negative effect of the disease on draft animals on the other, have necessitated a costly vector control programme.
Application of pour-on insecticide to livestock is an important means of controlling tsetse. The use of insecticides is the major method currently employed for tsetse control in Ethiopia. There has been signi cant success where insecticide control measures are properly implemented. However, for resource poor livestock keepers, use of this method is constrained by the consistently increasing high cost of imported insecticides. Therefore, one major problem here is how to maintain community interest if the cost of insecticide constantly increases? The reliable solution to minimize this problem is treating cattle in more cost-effective and technically feasible manner; on only those regions of the animals' body where most tsetse feed, the legs and belly parts.
As previous studies on direct observation of the vectors proved, (Glossina morsitans morsitans, G. pallidipes, G. brevipalpis, G. austeni, G. tachinoides and G. palpalis palpalis) tsetse consistently land and feed on the legs of cattle, [5,6,7]. Following these results, the restricted application control technique was assessed at the eld to select and adapt, in Ethiopia situation, effective and cheaper (lesser by 40% amount of insecticide per individual animal) application method, [8]. This adaptable and problem oriented eld trial on deltamethrin 1% ready to use formulation, aimed at evaluating the technique e cacy and cost-effectiveness was carried out where three tsetse y species are found: namely Glossina morsitans submorsitans, G.pallidipes and G. fuscipes fuscipes surving as cyclical transimiters of Trypanosoma congolense, T. vivax and T. brucei, [9],causing signi cant depletions on agricultural production and socio-economic development at large.

Pre-intervention entomological and parasitological data
The base line entomological survey was carried out in different ecological situations, varying from densely infested woodland and gallery forests on the valley oor and it's escarpment from the acacia woodland into open savannah grassland nearly 12 kms of the Ghibe valley to areas where ies could hardly be detected. Accordingly, a total of 75 biconical, monoconical and pyramidal trap designs baited with odor attractants, placed in three different dispensers, were erected. Consequently, of total traps deployed, 1298 three species of tsetse ies namely G.pallidipes, G. m submorsitans and G. f. fuscipes were caught that accounted for an overall mean catch of 17.3±8.6 ies per trap per day and attributed to an overall apparent density of 8.7 ies per trap per day. Regarding their relative density, the most Glossina species caught was G. f. fuscipes with a mean catch of 10.7±6.3 (9.3-12.2) ies per trap per day followed by G. m. submorsitans with a mean catch of 4.9±2.9 (4.3-5.6) ies per trap per day and G.pallidipes 1.6±1.3 (1.3-1.9), ( Table 1). The one factor-ANOVA statistical test method depicted that the difference in mean catch, among the species, was found statistically signi cant; F 0.05 =95.4; p=0.000. Within each group of ay, more female ies were captured and the difference in sex proportion of the species was found statistically signi cant; F 0.05 =100.7; p=0.000.
On the other hand, to determine trypanosomes infection rate in tsetse, dissection on proboscis, salivary glands and mid-gut of the ies was done on a total of 239 ies of the aforementioned species. Of total dissected ies six were found infected with T.congolense and T.vivax that accounted for an overall y infection rate of 2.51%. Regarding their proportion, of total dissected ies, 174 (72.8%) were G. f. fuscipes while 19.7 %, n=47 and 7.5%, n=18 were G.m.submorsitans and G.pallidipes, respectively. An infection rate of T.congolense type, n=4, 66.7%, was prevalent in all species; in which the highest prevalence was observed in G.m.submorsitans while in the remaining two species were equally prevalent.On the other hand, n=2, 33.3% of the infection was arouse due to T.vivax and both were detected in G. f. fuscipes but none of the ies were found infected with T.brucei type alone and mixed infection in any combination.
Side by side to entomological survey, a total of 542 randomly selected animals were examined. Of these 63 were found infected with T.congolense, T.vivax, and mixed infection T.congolense occurring together with T.vivax/T.brucei. The overall trypanosome prevalence was as high as 11.62%±3.2 having an overall mean PCV value of 24.9%±3.2. The mean PCV-value recorded for aparasitaemic animals, 25.4%±2.9, exceeded that of parasitaemics, 21.6%±3.0, on average by 3.8% (3.0,4.6), while 39.1% (n=212) of examined animals recorded a mean PCV value below the threshold (25%). Regarding their relative proportion, (n=42, 66.7%) of the infection with a mean PCV-value of 21.2±2.9 was arouse due to T.congolense followed by T.vivax (n=17, 26.9%) with a mean PCV-value of 22.6±3.4 and mixed infection, (n=4, 6.3%) had mean PCV-value record of 21.5±2.9 while (n=479, 88.4%) of the animals with a mean PCV-value of 25.4±3.0 were found negative with BCT technique. The post hoc test which suggests where the difference lie, depicted that only the mean PCV-value of non-infected animals statistically differed from each group of trypanosome infected animals ( Table 1).

Discussion
The tsetse y survey well corresponds to the general classi cation of tsetse ies in terms of their habitat and requirements, particularly to the ecology of tsetse species found in the Ghibe valley, southwest of Ethiopia, [9]. The results obtained from this trial show that, tsetse overall apparent density signi cantly reduced from a mean catch of 17.3 ± 8.6 ies per trap per day of pre-control to 3.7 ± 2.6 ies per trap per day at last monitoring. This resulted in a 78.6% overall mean catch reduction. According to t-test statistics for equality of overall mean catch of ies pre-and post-intervention, the mean catch difference, 13.6%±1.2 SE, was found statistically signi cant; p=0.000.
The quantitative and qualitative variation in tsetse population, population dynamics, depend directly on biotic and abiotic factors in the environment, seasonal variation in the population size or more precisely in the apparent density is linked essentially to the life span of the imagos, on one hand, and the pupal mortality on the other, [10]. Having in mind the environmental factors contribution on tsetse density, the observed reduction in tsetse density was mainly associated with the pronounced effect of the insecticide based intervention in which it attributed on the adult tsetse shortened life span. This is because, the last monitoring survey was carried out after the short rainy season (April-May) in June 2019 when biotic and abiotic factors in the environment favored tsetse reproduction; during when the life span of the adults increases and the duration of the development of the pupa reaches its optimum, however, the presence of suitable habitat alone is not always in favor of the ies' reproduction. In this regard, Hargrove, J. W.
(2000), [11], stated that as vegetation is vital for providing shade and maintains a suitable microclimate for tsetse as well as habitat for their hosts, and provides ies enough moisture, replenishment of tsetse density rapidly gets better as soon as the rst rains have reduced the ambient and soil temperature. Thus, the rapid decline in tsetse density was caused by the pour-on based intervention. This can be evidenced by the drastic decline of tsetse (by 76.7 %) realized three months after nearly 18,311 animals were treated at rst time, in November 2017, (Graph 1). Incidentally, with this amount of insecticide used, only 13,104 animals could have been treated if the standard treatment regime was followed, and this signi es a 40% reduction in the amount of insecticide used. The Pearson Chi-Square test statistics too evidenced the presence of strong association between the intervention and tsetse density reduction; X 2 0.05 = 73.5; p=0.000. Graph 1. Impact of RAIC technique on tsetse density and its future linear trend to suppress tsetse Following the restricted application of insecticide at a density of 15 animals per km 2 area with nearly three months base, at last monitoring, the relative abundance of G.f.fuscipes was signi cantly declined from a pre-intervention mean catch by 91.3% followed by G.pallidipes, (76.1%) and G.m.submorsitans, (55.1%), (Table 2), and the relative reduction within the groups pre-and post-intervention was found statistically signi cant; p=0.000. In this context, two aspects are important for the largest reduction observed in G.f.fuscipes: rstly, this species largely have a preference for feeding on bovid bloodmeal as a result y-cattle contact was high; secondly, cattle in the present study area graze deep into the valley oor where there is suitable habitat for the speci ed species in which the y-cattle contact, once more was very high; thus, these facts permits us to conclude that G.f.fuscipes species in the present area, are the main vectors of trypanosomes; however, a blood meal analysis was not done. Similar conclusion can also be made regarding with G. pallidipes; since it was the second most reduced species. This conclusion is in consistence with the previous studies conducted by [12] who stated that G. pallidipes and G. f. fuscipes feed mainly on livestock and G. pallidipes is the main vector of animal trypanosomosis.
Table2.Relative FTD of tsetse ies pre-and post-intervention time and their relative reduction Despite a high prevalence of multiple-drug resistant trypanosomes widespread in the target area, the overall mean trypanosome prevalence among the herds dropped from 11.6%±3.2 in November 2017 to the current level of 3.9%±2.3. This resulted in an overall of reduction by 66.4% in which the reduction was found statistically signi cant; p=0.000. Unlike to pre-intervention period, the dominance of T.congolense being a major cause of infection (63.5%) was reverted, at last monitoring by T.vivax type which accounted for 70.6% of the overall prevalence while the former accounted for a proportion of 23.5%. Signi cantly higher proportion of infection due to T. congolense, during pre-intervention period, signi ed cyclical transmission domination than mechanical. The widespread occurrence of T.vivax type infection at last monitoring, on the other, has got its own advantage for the improved general health levels and productivity of livestock, as it is considered being less virulent for cattle than T. congolense, which appears to cause more sever effect than T.vivax, which could be manifested by the mean PCV-value. Additionally, the possible situations of the in uence of trypanocidal drugs on trypanosome infection type never be neglected, [14]. They referred as the higher in T.congolense infection rate in tsetse-infested area could be resulted due to ineffective drug treatments against it to the same degree as T.vivax. This is in compliance with the present area pre-intervention time, where the curative and prophylactic capacities of the three trypanocides have been impaired as a result of widespread occurrence of drug resistance by the parasites. Nonetheless, the presence of high proportion of T. congolense infection is mainly associated with its number of antigenic variability.
Above all, the rare detection of T. brucei infection in sampled animals and its absence in dissected tsetse, may be attributed to the longer development cycle and complex pathway that it takes in the y which hinder its survival and maturation especially when the life span of the y is shorten,. Besides, in contrast to T.congolense and T.vivax, infection with trypanzoon species is characterized by generally low parasitaemia and a marked invasion of tissue, the parasite has got a lesser probability to be picked up by the y while ingesting the blood meal, consequentially, infection rate in tsetse is very low [13], and this has been manifested by the lowest infection rate in cattle.
The epidemiology of African animal trypanosomosis (AAT) is almost entirely dependent on tsetse ies. As a result, the trypanosomal infection type in tsetse is of prime importance in determining the type of infection the animals to be contracted with. Fly species differ in their capacity to transmit trypanosomes. Morsitans group ies, except for G.austeni, are good vectors of all trypanosome species. To the contrary, Palpalis group species appear to be poor vectors of most trypanosome species except certain stocks of West African T.vivax, although G.f.fuscipes can be important vector of human infective trypanosomes, [15]. This vectoral capacity difference is associated mainly with the type of lectins found in the gut of different species of ies. It has been established that lectin plays a role in determining refractoriness to infection by killing procyclic trypanosomes, whilst symbiotic bacteria are involved in determining susceptibility to infection by a process that results in inhibition of these midgut lectins and thus reduces their killing effect (blocks the lectin-mediated trypanocidal activity), [16]. Glucosil lectin is found in morsitans and fusca groups whose action can be inhibited by D-(+) glucosamin which is secreted by maternally inherited Rickettsia-like organisms (RLO) while palpalis group species possess the galactosil lectin whose action is not inhibited by D-(+) glucosamin. That's why, palpalis group tsetse species are less refractory to trypanosomes infection as a result of the effect of the non-cellular factor, galactosyl lectin. It acts as agglutinin, which interferes on trypanosome survival and maturation. Based on these general in uences on Glossina infection rates, neither all species nor all individuals within a species are equally e cient, [17].
Provided that G.f.fuscipes outnumbered the sum density of G.m. submorsitans and G.pallidipes before the onset of the intervention, and the former is considered good vectors of T.vivax, the vivax type infection proportion in cattle was lower than congolense type. As shown in the present study, however, both groups of tsetse (riverine and savannah) harbored T.congolense, the highest, (75%), infection rate of T.congolense in the ies was observed in G.pallidipes and G.m.submorsitans which indicates the important role of these species, playing signi cant role in transmitting T.congolense while they were prevalent with relatively lower number; this is in agreement with many studies which has shown that palpalis group ies are poor vectors of trypanosomes especially, T.congolense. Additionally, the possible situations of the in uence of trypanocidal drugs on trypanosome infection type never be neglected, [14]. They referred as the higher in T.congolense infection rate in tsetse-infested area could be resulted due to ineffective drug treatments against it to the same degree as T.vivax. This is in compliance with the present area pre-intervention time, where the curative and prophylactic capacities of the three trypanocides have been impaired as a result of widespread occurrence of drug resistance by the parasites. Nonetheless, the presence of high proportion of T. congolense infection is mainly associated with its number of antigenic variability.
Nonetheless, post intervention period even if these savannah species were more abundant than G.f.fuscipes, congolense type infection in cattle, by far lessen from vivax type. In this context, some possible explanations can be given based on the epidemiology of trypanosomosis: rstly, higher proportions of infection were transmitted mechanically as compared to cyclical transmission, which in turn signi es the importance of nuisance ies in the area, [18]; this is because T.vivax is the species most likely to be transmitted mechanically, other species of trypanosomes pathogenic to cattle can also be transmitted and Glossina spp. may also act as mechanical as well as cyclical vectors [19]; secondly, the proportion of T.vivax infections in tsetse increases under adverse environmental conditions (in our case the progressed tsetse control) when the ees' shortened life span prevented the completion of cyclical development of the Nannomonas (T.congolense) and Trypanzoon (T.brucei) trypanosome species [20]; consequentially, transmission of these parasites were hindered; thirdly, high T.vivax infection rates in ies arouse for the fact that T.vivax develops in the proboscis, it is far from the action of anti-trypanosomal factor of the ies' gut, in which beside to shorter development cycle it follows, absence of this trypanosomal action on T.vivax enabled it to be found in higher proportion in tsetse consequentially, in naturally infected cattle. Finally, higher T.vivax type infection in cattle usually encounters when the overall prevalence of trypanosomosis in cattle is low. Off course, the species of animal on which tsetse fed, exerts the greatest in uence on the infection rate in ies. Whiteside, E.F (1958), [21] and Jordan, A. M.
(1986), [22] identi ed at least 18 major variables in uencing the potential of trypanosomes to develop in tsetse ies (the epidemiology of African animal trypanosomosis), which relate to interactions of tsetse (as endogenous factors associated with tsetse), ecological factors and parasite and host factors, .
One of the major effects of infection with pathogenic trypanosomes is anemia. Measurement of anemia gives a reliable indication of disease status and is correlated with parasitaemia-the higher the parasitaemia, the lower the PCVs. As a result of decline of trypanosome prevalence in the herds, the overall mean PCV-value of animals improved on average by 2.0%. Thus, PCV is a good indicator of the health status of the herd in an endemic area, which negatively correlates to disease incidence. Nonetheless, 44.3% of the animals' mean PCV values during pre-intervention period and 28.9% of the animals during last monitoring were below the threshold of 25% normal PCV-value.
In this regard, partially or semi-trypanotolerant cattle living in and in close proximity to the target areas were observed. Some animals achieved better control of parasitaemia while they were infected by maintaining PCV in its normal range and this could be an indicative that deferent serodemes of trypanosomes are circulating in the area, against to which individual animals developed resistance, [23].
On the other hand, some animals PCV-value was lower than the threshold level, while they were not infected with trypanosomes, which suggests that anemia is a multi-factorial problem in the area in which management, exercise and heat stress could be attributable [24].However, management factors share the major contributions to anemia, contribution of concurrent infections such as internal, external parasites and other heamoparasites is not negligible. The degree of anemia is directly correlated with the loss of productivity performance and could be a major contributor to death of trypanosome-infected cattle.
Therefore, in areas where multiple drag resistance is widespread, the use of 'pour-on' insecticides, may help as a more sustainable method of tsetse control, consequentially, trypanosomosis, even without integrating the technique with other tsetse control methods. Nonetheless, the effectiveness of this technique works best where livestock are the main host of tsetse. The results of this eld trial are in consistent with the ndings of Leak, S.J.A., 1995, in which  [25].
The standard (application of insecticide on the back of cattle) tsetse control technique so far employed, has got a potential to reduce both tsetse population and the disease prevalence within a short time.
However, treating cattle with insecticide is an increasingly important means of controlling tsetse ies as livestock keepers in particular and the national economy at large, it has been hindered by high cost of the insecticide. Use of this control method with lesser amount in more cost-effective manner could be a reliable solution to the problem. In this regard, restricted application of insecticides to cattle has been recognized as a cheap, safe and environment friend farmer-based method to control tsetse and trypanosomosis, [26]. The present ndings have assured that the restriction of pyrethroid application to only the belly and legs parts of the body by far reduces not only the amount of insecticide needed per application, according to the present trial by 40% that needed for whole body application. They have also proved its great potential to suppress tsetse population, consequentially, trypanosomosis incidence to prescribed low levels. The technique could be more useful in areas where the creation of y-free zones is challenging and reinvasion pressure high. The technique provides an extra bene t, compared with targets as nuisance ies and ticks may also be controlled. Since the whole body treatment regime can contaminate the dung su ciently to affect dung fauna, thereby threatening the important role that such fauna play in dung dispersal, and hence soil fertility and the productivity of pastures and crops [27]. The occurrence of trypanosomal infections in areas apparently free of tsetse promoted that infection can be maintained in nature by the mechanical transmission of trypanosomes by other heamatophagous ies [28,29]. As other study has shown, since other biting ies concentrate on the lower legs and belly [30] the technique is applicable even for biting ies (e.g. Stomoxys, Tabanus spp. etc.) especially during the late rainy season when these ies are more abundant. Besides, as is proved by [31], Anopheles arabiensis Patton (Diptera: Culicidae) is the most widespread vector of malaria in the Afrotropical Region. Because Anopheles arabiensis feeds readily on cattle as well as humans, the insecticide-treatment of cattle as employed to control tsetse (Diptera: Glossinidae) and ticks (Acari: Ixodidae) might simultaneously affect the malaria vectorial capacity of this mosquito.
Ultimately, application of insecticide in standard or restricted scenario does not have a signi cant difference in tsetse suppresion except the later system lessens the amount of insecticide used and the impact on non-target organisms. As is shown in the above graph, the technique linear trend to suppress the vector can lead to dramatic decline of tsetse populaiton to close to zero, if repeated treatement of animals is achieved. Tsetse control (eradication if possible) is the most reliable and effective alternative strategy available to date, towards adequately reducing and nally removing tsetse transmitted trypanosomosis risks and losses.

Conclusions
The result of this eld trial demonstrated that RAIC control technique not only reduced the population size of tsetse ies but also the prevalence of Trypanosoma spp in cattle with a relatively cost-effective manner. First and foremost, the control of tsetse would enable the livestock economy to grow, thereby reducing mortality and improving the general health of animals. The present ndings con rmed the costeffectiveness of using ITC, especially when applied in restricted scenario, to control both tsetse ies and the attendant disease they transmit. To sum up, both direct and indirect bene ts derived from the technique have large impacts on the overall balance between livestock and agricultural production and hence contribute to the goals of poverty reduction thereby enhancing incomes of the rural community.
Since, the technique has not yet exhausted its potential to suppress tsetse with relatively lower amount, further study need to be carried out to maximize the economic bene t derived from it either alone or integrating it with other tsetse control techniques.

Materials And Methods
Background to the study area This initial trial of RAIC in Ethiopia was conducted in area of >850 km 2 located along side of the Ghibe valley in Boter Tolay district located at grid references of 8°19' to 8°45'35"N latitude and 37°45' to 38°71'42"E longitude where a high drug-resistant trypanosomes are widespread. The Ghibe River is big perennial and is tributary of the Omo River which ows into south wards along the rift valley into Lake Turkana on the Ethiopian/Kenyan border. The area has got sub humid lowland climate that occurs between 500 at the valley oor and 1500 meters above sea level at its escarpment. The area receives reliable rainfall with an average of 900 mm precipitation per annum and the mean monthly maxi-mum temperature ranges from 29.8 to 44.0 •C [32]. The main rainy season lasts from mid-June to late September and the shorter one from April to May.
The predominant farming system of the area is characterized by mixed livestock and crop production, with livestock playing a vital role in agricultural activities, nutritional values and income generation. An East African Zebu type breed of cattle, which are also known as Abyssinian short-horned zebu, found in the study areas. This type of cattle breed is characterized by a pronounced susceptibility to trypanosomosis; as high prevalence rate, high drug resistance and high number of treatments [33]. Cattle, in the target area, are kept under traditional extensive husbandry system with communal herding, [34].
Along the Ghibe drainage lines and other its principal tributaries valley anks, high gallery forests are found. The most common woody vegetation in the valley oors and their escarpments identi ed as Accacia abyssinica, Accacia seyal, Croton macrostachys and Combretum species. Wild animals that have been seen within the total area of the districts that serve as potential tsetse host include, roan antelope, warthog, bush pig, Colobus monkey, rabbit hyena, hippopotamus bush back, baboon (Personal communication with Regional Resource Bureau).

Statistical test analyses
To this end, data on individual animals parasitological and hematological and data on entomological The prevalence assumption was made based on previous survey that had been carried out by NTTICC, 20%, [35] Subsequently, to accurately estimate the mean number of cattle infected with trypanosomosis, the sample size determination was done based on the standard deviation of previous trypanosomosis prevalence, at 99% con dence level and 5% acceptable error. Sample size was determined using formula of [36]. Accordingly, 429 animals were su cient to estimate the prevalence, but 542 animals were sampled during a reference data collection and 432 animals at last monitoring. And this allowed us being con dent that amongst the whole population there is 99% chance that the mean is within the acceptable error limit of 5%.
The data processing was carried out based on quantitative data analyzing methods. The collected data were subjected to univariate one-way ANOVA (Analysis of Variance) to explore the relative proportion of tsetse species caught, the identi ed trypanosomes relative prevalence and their respective mean PCVvalue. And to determine where the difference lays Bonferonni procedure was used. On the other hand, independent t-test, Matched pair t-test, Student t-test were used to analyze the difference in overall tsetse density and trypanosome prevalence pre-and post-control data, and Chi-square test statistics was employed whether the results obtained were arose due to the intervention or by chance for both entomological and parasitological data. Descriptive statistics, such as table, frequency distribution, means, percentiles, standard error, standard deviation and graphs were employed.

Study design
Entomological Probability-based cross-sectional study design was used to collect the baseline data, data to isolate changes occurring overtime and data on last monitoring in which sampling involved strati ed random selection of sampling units [37]. Regarding entomological surveys, since the surveying area is heterogeneous, in order to avoid the possible biasness while determining the apparent density of ies that could be arisen due to their seasonal changes in distribution, surveys were carried out by the method of strati ed sampling that repeated for each surveying time after designing a sampling strategy. At the outset, the target area was isolated into subareas (strata) based on vegetation, climate, altitude, land use, distribution of human habitation, etc. into more or less homogeneous [38]. Having done this, biconical, monoconical and pyramidal trap designs baited with odor attractants such as octenol (1-oct-3-nel), acetone (C 8 H 16 O); and 3 weeks old treated cattle urine, [39], placed in three different bait dispensers, were deployed along transects from the acacia woodland into open savannah grassland nearly 12 kms of the Ghibe valley and its escarpment in both the late rainy and the dry seasons at the same trapping sites. Traps were positioned for two consecutive days with mean interval between traps of nearly 300 mts, in most likely areas for nding tsetse, based on the presence of forests, bushes and the location of rivers and streams after clearing up to 2-3 meters radius of the trap site. The traps were deployed in a georeferenced locaitons to map and easily display the distribution of tsetse ies on the GIS map. After the identity of tsetse and their sex determined, the arithmetic mean catches per trap per day was calculated for the full sampling periods according to [40,41,42].
Fly dissection and observation of the parasites This was achieved on freshly killed ies after immobilizing them by squeezing the thorax gently. After making two small incisions on either side of the tergites 1-3, with the y on its ventral surface, the gut was allowed be seen, and by placing a cover slip over it and gently pressed down gut contents in a drop of saline solution, the gut contents were examined under a compound microscope using a 10× eyepiece and 25× objective. Proboscides from ies were dissected into a separate drop of saline solution and teased apart, gently rubbing a ne needle down the length of the proboscis to dislodge the trypanosomes. The salivary glands were also examined according to guidelines of dissection for trypanosome infection rate in tsetse, [42]. Trypanosome infections in the hypopharynx and labrum (proboscis) only, are classed as Trypanosoma vivax-type; in the proboscis and midgut only as T. congolense-type and in the proboscis, midgut and salivary glands as T. brucei-type,

Parasitological
For parasitological surveys, strati ed random sampling method was followed. The target populations were separated at village level, in a non-overlapping strata or sub-populations that are thought to be homogenous such that there is less variation among the sample units in the same stratum and among sampling units in different strata. Afterward, approximation of trypanosome prevalence and identi cation of the parasite were achieved by collecting blood from sample animals puncturing, aseptically, marginal ear veins with lancet and the blood sucked up by heparinized micro-hematocrit capillary tubes allowing the tubes were lled up to three fourth of their length and sealed at one end with crista-seal. The capillary tubes were centrifuged at 12,000 rpm for 5 min to concentrate the trypanosomes in the buffy coat layer. After the Packed Cell Volume (PCV) was measured, the buffy-coat plasma junction including one mm of the red blood cells was cut using diamond pencil to pour on the BC on microscope slide and covered with cover slip of 24 mm×24 mm and examined for the presence of trypanosomes, [43]. Thin blood smears were prepared from positive and anemic animals (PCV of 25% or less). The thin blood smears also checked for other hem parasites such as Anaplasma spp. Babesia spp and Thileria spp. The obtained results recorded for further prevalence determination using the formula for prevalence [44].

Treatment of animals with insecticide
After a cross sectional baseline data collection to determine the ies' spatial and temporal distribution and risk of trypanosomosis among the herds, animals were treated with deltamethrin 1%, using hypodermic syringe, at belly and legs body parts, as 0.06 ml of formulation per 1 kg of body weight which is less by 40% that needed for a whole body (standard) treatment regime applicable as 0.1 ml of formulation per 1 kg of body weight. To this end, though, other studies recommended that treating 4 large cattle per km 2 is considered su cient to control or eliminate tsetse [27], the current study followed a density of 15 treated animals/km 2 . This is because, communal herds in the area typically comprise 25% young animals (> one year old) and 60-70% females, and are not evenly distributed spatially or temporally; to achieve su cient control, the identi ed regime was followed over a relatively large area (>850 km 2 ), where the participation of many livestock keepers were involved. Meanwhile, many herds from the district were involved in the study wherein, >55,000 cattle from 17 villages were treated on three months base, where they were watered at different watering points located in the Ghibe valley and its principal tributaries.
A treatment regimen, in the course of the entire survey times, was followed in which animal found to be parasitaemic using the buffy coat phase contrast technique or animal in which trypanosomes were not detected but with a packed red cell volume (PCV) less than 20% or any animal suspected of the disease, based on clinical signs, was treated with diminazine aceturate at a dose rate of 7 mg/kg BW or alternatively with isomethamedium chloride at a dose rate of 1.0 mg/kg body weight without charge. But no blanket treatment of animals was achieved and no other tsetse control techniques were employed and single blindness protocol was followed.

Limitations
Two limitations of this study can be speci ed; rstly, however, too many other biting ies were collected and identi ed (Tabanides, stomoxis, heamatopota, etc.) they were not recorded in which the important role of these ies in transmitting trypanosome spp. was neglected. Secondly, however, identi cation of trypanosome infection rates in tsetse were determined by dissection and observation of the parasites in organs of the y, an imprecise identi cation of the trypanosome species subgenera can be made.
Therefore, molecular biological diagnostic procedures (DNA probes and polymerase chain reaction, PCR) should have been used. However, since it requires sophisticated laboratories, specialist staff and reagents and is more expensive to perform, we were unable to do so. Besides, ovarian ageing and wing fray analysis for age structure determination should have been done, especially at last monitoring.

Abbreviations
African animal trypanosomosis (AAT); packed red cell volume (PCV); CI; Con dence Interval; ANOVA; Declarations DA, conception of the trial, study design preparation, data analysis, data results interpretation, writhing the manuscript and submission; AB, data arrangement, reviewed the study design, eld monitoring data collection, supervising the activity; BM, reviewed the study design, baseline data collection, following the activity, organizing stakeholders; TA, monitoring data collection, following the activity, organizing stakeholders, reviewed the study design; Authors

Availability of data and materials
All data generated or analyzed during this study are included in this published article and its supplementary information les and are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
The study design and the procedures followed in this eld trial were reviewed and approved by the Bedele