Indiscriminate use of agro-veterinary pesticides in plague endemic foci in Tanzania: potential risk for development of insecticide resistance in ea vectors

Background: Agro-veterinary pesticides are increasingly associated with the development of resistance in arthropod disease vectors. However, no has been conducted so far to assess such risk in ea vectors despite the indiscriminate use of pesticides in plague endemic foci in Tanzania. Methods: We identied risk factors associated with the use of agro-veterinary pesticides, which could enhance development of resistance in ea vectors in two plague endemic districts, using structured questionnaire and direct observations. Results: Excessive and injudicious use of agricultural and veterinary pesticides was common in both Lushoto and Mbulu district. Most farmers (80%, n=73) were applying agricultural pesticides over three times per cropping season, did not adhere to manufacturers recommended doses, and had limited or no knowledge on safety procedures and adverse effects associated with pesticides. Up to 49% were applying pesticides more than twice the recommended doses. About 91% of respondents in Lushoto and 93% respondents in Mbulu reported using agricultural pesticides. The three (3) out of fourteen (14) most commonly used agricultural pesticides in Lushoto were master kinga72WP (mancozeb 640g/kg+cymoxanil 80g/kg) (44%), suracron720EC (profenos 500g/l EC) (25.3%) and Sumo 5EC (lambda-cyhalothrin) (18.7%). The three (3) out of seventeen (17) most commonly used agricultural pesticides in Mbulu were Dursban50W (Chlorpyrifos) (29%), Duduban 450EC (Cypermethrin 10g/lt+chloropyrifos 35g/lt) (18%) and Dursban+farmerzeb (Chlorpyrifos 48%, Mancozeb 80%WP). Cybadip (Cypermethrin) ( ≥ 45%) and paranex (alphacypermethrin) ( ≥ 13%) were the most commonly used livestock pesticides. Conclusion: This study identied injudicious uses and/or other malpractices, which enhance Therefore, the ea vector populations in Lushoto and Mbulu districts are putatively under high risk of resistance development. Further studies are underway to conrm the insecticide resistance status, unravel distribution of the resistance, and involved resistance mechanisms.


Introduction
Plague is a life-threatening disease caused by a highly infectious bacterium, Yersinia pestis. The disease is transmitted to humans primarily by eas mainly from infected rodents [1]. More than eight ea species described so far can transmit the bacteria [2]. The most predominant and widely distributed ea vector species include Xenopsylla cheopis, X. brasiliensis, Dinopsyllus lypusus and Pulex irritans [3]. Xenopsylla cheopis and X. brasiliensis have played role in the past and recent plague outbreaks due to their overwhelming e ciency and broad host preference. 4 Several studies suggest that, the human ea, Pulex irritants, may be playing an important role in human-to-human transmission during plague outbreaks [5,6].
Commendable efforts have been made to develop plague vaccines; however, there is no any effective vaccine available to date. As such, measures targeted to ea vectors, mostly chemical insecticides, remain the most effective and widely used. This approach has signi cantly reduced plague in most if not all endemic countries; however, its long-term effectiveness is increasingly challenged by the development of resistance in the major ea vectors. Some endemic countries have reported resistance virtually against all WHO recommended classes of insecticide [7,8].
Based on the classical thinking, all cases of insecticide resistance in arthropod disease vectors including eas are exclusively attributed to selection pressure from the public health pesticides. Meaning that, all quiescent plague endemic areas without long-standing chemical control operations are safe from insecticide resistance. This assertion might put such areas in a great danger during the disease outbreaks, which are often unpredictable. There are increasing reports in mosquitoes associating insecticide resistance development to agro-chemicals [9][10][11][12]. To the best of our knowledge, no study has been conducted so far to assess such risk in ea vectors despite the indiscriminate use of agro-veterinary pesticides in Tanzania and elsewhere in Africa. Many small-scale farmers in different parts of the country are increasingly reported to misuse agro-chemicals particularly pesticides, thus predisposing the nontargeted organisms and ecosystems to negative impacts associated with such chemicals [13][14][15][16].
Therefore, this study aimed to establish malpractices and misuse of agricultural pesticides, which could be enhancing the exposure of ea vectors of plague to the pesticides.

Study area
This study was done in the selected villages of Lushoto and Mbulu district, Southern Tanzania (Figure1).
Lushoto district in Tanga region, is situated at the West Usambara Mountains, which forms part of the Eastern Arc Mountains (04 o 22'-05 o 08'S, 038 o 05'-038 o 38'E). The district lies at the altitude of 900 to 2,250 m above the sea level. The long rain season runs from March to May; and short rain season runs from November to December, with mean annual rainfall of 1070mm and temperature of 17 o C. The dry season runs from July to October. The district has a population of 492,441 people. Mbulu district in Manyara region, is bordered to the north by Arusha region and lake Eyasi; and to the west by Singida region (3 o 57'1"S, 35 o 18' 40"E). It covers a total surface area of approximately 3,800km 2 ; and lies between 1000-2400 m above the sea level. The long rain season runs from January to May; and short rain season runs from November to December, with mean annual rainfall of 994mm and temperature of 17.5 o C.The dry season runs from June to October. The district has a population of 320,279 people [17].

Study population
In Lushoto, the study was done in four villages, all with a history of plague outbreak, namely Viti, Lukozi, Ndabwa and Mavumo. Crop cultivation is the main economic activity in the area, and the main crops include maize, beans, potatoes, vegetables and fruits. In Mbulu, the study was conducted in two villages, all with a history of plague outbreak, namely Arri and Mongahay. Crop cultivation and livestock keeping are the main economic activities; and the main crops include maize, beans, potatoes, wheat and vegetables.

Study design and data collection
We employed a cross-sectional study design. The study villages were selected based on the criteria of having experienced plague outbreaks at least twice, and proximity to active and/or fallow agricultural elds. Two hundred (200) households, 100 per district, were used for this study. Head of the household (father or mother) was the main respondents. Otherwise, another adult permanent resident in the household participated in the study. The inclusion criteria for respondents were having lived in the area for ve years and above, ownership of active farmland and/or livestock and willingness to participate in the study. Salespersons from four agro-vet shops within the study villages were selected based on their involvement in such business within the last two years and above. The data were collected by means of a standard questionnaire with structured and semi-structured questions. The questionnaire was supplemented with direct observations to con rm certain responses where participants could provide evidence for example presenting containers of the pesticides they were using and demonstrating the preparation of spray solutions. Overall, the survey aimed at gathering information on indigenous knowledge (following manufacturers application instructions, safety procedures and health risks) and routine practices (commonly used chemicals, putting on protective attires, frequency of applying pesticides, use of agricultural pesticides on livestock and vice versa) in relation to the use of agricultural and veterinary chemicals.

Data analysis
The data were analyzed using Epidemiological Information, Epi Info TM version 7.2.3.1(CDC). Descriptive results were expressed as frequencies and percentages.

Commonly used pesticides against eas
In Lushoto and Mbulu district, the three commonly pesticides used against eas belong to two main classes: pyrethroids and carbamates (Table 5). Small portable hand sprayers were used to spray pesticides to infested animals or on the oor. The animals, mainly calves and goats, were only sprayed on body parts preferred by the eas. The powder formulation of carbamates was sprinkled inside living houses with hands. The targeted control of eas is so much restricted to certain parts of the house and was often done during the dry season when the ea population is high. The same applied for Mbulu only that the problem of eas seems to be relatively higher than in Lushoto district. Because of the cost of insecticides, most of the respondents reported to pour hot water inside their living houses as an alternative way of controlling eas. The study also revealed that, most people in the study villages in Mbulu district are still keeping livestock (cattle, goats etc) inside their living houses. Table 5 summarizes the commonly used chemicals for ea control in Lushoto and Mbulu district. The respondents from Lushoto district reported the highest application of pesticides per cropping season of three months, particularly on vegetables. Most of them (n=73; 80%) repeated pesticides application over thrice per cropping season. About 11% (n=10) and 6% (n=5) of the respondents applied pesticides once and twice per season respectively. The rest of the respondents (n=2; 3%) reported that they were not using pesticides at all. Similarly, respondents from Mbulu district reported high use of agricultural pesticides particularly on vegetables; and most of them were applying pesticides twice (n=39;42%) or over thrice (n=35;38%) per cropping season. About 8% (n=7) were applying pesticides only once per cropping season. The rest (n=11, 12%) were not using pesticides at all.

Discussion
This study aimed to establish risk factors associated with use of agro-veterinary chemicals, which could enhance the development of resistance in plague ea vectors. The thinking that resistance in arthropod vectors emanates exclusively from prolonged use of public health pesticides might predispose plague endemic countries to devastating outbreaks. Most of the studies conducted so far, mainly in mosquitoes, are increasingly emphasizing involvement of agricultural pesticides in the development of insecticide resistance [11,[18][19][20][21]. Similarly, veterinary pesticides are purported to contribute to development of resistance in public health pests. These assertions are both reasonable because the chemistries of public health and agro-veterinary pesticides are closely related. In fact, most if not all public health pesticides are reformulations of pesticides that were once used in agriculture [23].
Crop, animal and human arthropod pests interact either directly or indirectly via habitats and hosts. Due to that, they coincidentally contact pesticides and other control agents not meant for them. This is substantiated by Gratz [24] who reported that indoor residual spraying (IRS) with DDT against mosquitoes affected susceptibility of the ea vector, Xenopsylla cheopis such insecticide. As such, injudicious uses/practices of agro-veterinary pesticides, for whatever reason, which promote their exposure to ea vectors of plague, will constitute risk factors for resistance development. The present study identi ed several risk factors with huge potential for enhancing the development of resistance in ea vectors of plague.
Small-scale farmers and livestock keepers dominate Lushoto and Mbulu districts; and control pests almost exclusively using pesticides. Over 91% of respondents across the districts reported using pesticides to control agricultural pests. The use of pesticides against both agricultural and livestock pests is a common practice throughout Tanzania and elsewhere in the region [16,23]. Similarly, use of pesticides constitutes the most widely used and powerful means of controlling eas and other public health pests [1,25,26]. Pests are responsible for the largest burden in crop and livestock production [23,[27][28].
Excessive and injudicious use of agricultural and veterinary pesticides was common across the study districts. Most farmers reported applying agricultural pesticides up to four times per cropping season, particularly on horticultural crops. Even more worrying, these farmers did not adhere to manufacturers recommended doses and had limited or no knowledge on safety procedures and adverse effects associated with pesticides. Similar surveys in other parts of Tanzania and elsewhere in Africa indicated misuse and overuse of pesticides, non-adherence, poor storage and disposal of pesticide containers, use of banned and counterfeit pesticides and the use of a mix of pesticides in a single spray [15,16,29,30]. Respondents of the current study never reported the last two malpractices. It could be that the respondents feared of being penalized if they disclosed such extreme information. Our co-workers however witnessed the two forms of malpractices in the study areas and the people claimed that banned and/or mixed pesticides were relatively more e cient in terms of pest control outcomes. The misuse and/or malpractices observed in this and other studies could have been contributed by poor education background and inadequacy of the training programs on safe use of pesticides, particularly at the local levels [16,23,31]. Up to 83% of the respondents in the present study were primary school leavers. Farmers with this literacy level are mostly unable to understand manufacturer's labeling and instructions, particularly because they are written in English. Other studies argue further that most extension workers and salesmen in retail shops across Tanzania are unable to provide quality services because they are either incompetent or they did not undergo the relevant training [14].
Farmers reported using over fourteen agricultural and three veterinary pesticides, most of which were pyrethroids. Other than pyrethroids, there were carbamates, organophosphates, organochlorines, and avermectins. Most of the identi ed individual pesticides are also common elsewhere, particularly in horticulture and cotton industry [13,14,16]. Interestingly, development of resistance in public health pests, for instance mosquitoes has been reported from cotton and horticulture growing areas consequent to large quantities and frequent applications of pesticides [13]. These crops are relatively vulnerable to pests and diseases. As such, any other areas growing either of the aforementioned and/or other vulnerable crops are at risk of resistance development in public health pests, like eas, that are likely to contact contaminated soils and other surfaces/substrates. Horticultural production forms the largest agricultural industry in Lushoto and Mbulu district.
This study revealed infrequent application of pesticides recommended for ea control. At most, these pesticides were applied in cases of high ea infestations. The higher cost of these pesticides than that of agricultural pesticides could be a limiting factor. Farmers get tempted to use agricultural pesticides as alternatives against eas and other public health pests. This was outside the scope of this study, but our co-workers witnessed such scenarios in their previous visits to the current study sites. Switching pesticides to unintended uses is also not new elsewhere [32].
Because of the aforementioned injudicious uses and other malpractices, indoor and outdoor environments/surfaces are contaminated with chemicals. Immature and adult eas are constantly exposed to such surfaces, thus posing high risk of resistance development in Lushoto and Mbulu districts. Resistance to various chemicals in rodent eas may be selected at either larval or adult stage.
These eas are always found in loose soils of rodent burrows in peri-domestic and agricultural elds or animal/human dwellings. During rains, soils in rodent burrows and other habitats become contaminated via run-off of agricultural chemicals. Many studies have reported alarming contamination of soils, water, air and other type of environments with agricultural chemicals [33][34][35]. Furthermore, eas are exposed to chemicals via insecticide-treated livestock because these animals also serve as alternative blood-meal sources. This nature of exposure could be enhanced by high interactions between commensal rodents and domestic animals, particularly in areas where the tendency of keeping animals inside living houses during the night is still existent. The current study revealed that certain rural communities within Lushoto and Mbulu districts are still embracing such practice. As such, eas are constantly exposed to veterinary chemicals on top of exposure to agricultural chemicals.

Conclusion
This study revealed a suite of either injudicious uses and/or malpractices; excessive use of agricultural chemicals, maluse of agro-veterinary chemicals as well as poor adherence to the application and safety procedures, all of which potentiate contamination of environments/surfaces and exposure of the chemicals to eas thereof. As such, ea vector populations across Lushoto and Mbulu districts are putatively under high risk of insecticide resistance emanating from the indiscriminate use of agroveterinary chemicals. Further studies are underway to con rm the trend of insecticide resistance in the study area and elsewhere in view of informing the course of action in terms of awareness creation and development/implementation of resistance management strategies. Figure 1 A map of Lushoto and Mbulu districts showing the study sites.