Cattle trypanosomosis: the diversity of trypanosomes and implications for disease epidemiology and control.

Trypanosomosis is one of the most significant infectious threats to cattle in sub-Saharan Africa, and one form has also spread to Asia and South America. The disease is caused by a complex of trypanosome species, and the species and strain of parasite can have a profound influence upon the epidemiology of the host-parasite-vector relationships, the severity and course of infection, and, consequently, the implementation and development of control methods. This review will summarise our current knowledge of the relationship between trypanosome species/genotype and the phenotype of disease in cattle, and the implications that this has for ongoing efforts to develop diagnostics, drugs and vaccines for the control of cattle trypanosomosis.

Surveillance and sampling of disease vectors.

Improving the surveillance and sampling of vectors is associated with many issues, including: the relative merits of laboratory studies as against field studies of vector behaviour; the ability to track individual vectors; the cost-effectiveness of traps and confident interpretation of sampling data. In this paper, the authors offer examples of recent progress in these matters and suggestions for future progress, with an emphasis on the need for analytical approaches to be adopted more widely.

Mapping the benefit-cost ratios of interventions against bovine trypanosomosis in Eastern Africa.

This study builds upon earlier work mapping the potential benefits from bovine trypanosomosis control and analysing the costs of different approaches. Updated costs were derived for five intervention techniques: trypanocides, targets, insecticide-treated cattle, aerial spraying and the release of sterile males. Two strategies were considered: continuous control and elimination. For mapping the costs, cattle densities, environmental constraints, and the presence of savannah or riverine tsetse species were taken into account. These were combined with maps of potential benefits to produce maps of benefit-cost ratios. The results illustrate a diverse picture, and they clearly indicate that no single technique or strategy is universally profitable. For control using trypanocide prophylaxis, returns are modest, even without accounting for the risk of drug resistance but, in areas of low cattle densities, this is the only approach that yields a positive return. Where cattle densities are sufficient to support it, the use of insecticide-treated cattle stands out as the most consistently profitable technique, widely achieving benefit-cost ratios above 5. In parts of the high-potential areas such as the mixed farming, high-oxen-use zones of western Ethiopia, the fertile crescent north of Lake Victoria and the dairy production areas in western and central Kenya, all tsetse control strategies achieve benefit-cost ratios from 2 to over 15, and for elimination strategies, ratios from 5 to over 20. By contrast, in some areas, notably where cattle densities are below 20per km(2), the costs of interventions against tsetse match or even outweigh the benefits, especially for control scenarios using aerial spraying or the deployment of targets where both savannah and riverine flies are present. If the burden of human African trypanosomosis were factored in, the benefit-cost ratios of some of the low-return areas would be considerably increased. Comparatively, elimination strategies give rise to higher benefit-cost ratios than do those for continuous control. However, the costs calculated for elimination assume problem-free, large scale operations, and they rest on the outputs of entomological models that are difficult to validate in the field. Experience indicates that the conditions underlying successful and sustained elimination campaigns are seldom met. By choosing the most appropriate thresholds for benefit-cost ratios, decision-makers and planners can use the maps to define strategies, assist in prioritising areas for intervention, and help choose among intervention techniques and approaches. The methodology would have wider applicability in analysing other disease constraints with a strong spatial component.

Know your foe: lessons from the analysis of tsetse fly behaviour.

The emergence of new vector-borne diseases requires new methods of vector control. These diseases are often zoonoses associated with wilderness areas, and established methods of vector control used in domestic settings (e.g., indoor-residual spraying, insecticide-treated bednets) are therefore inappropriate. Similar difficulties are also emerging with the control of 'old' vector-borne diseases such as malaria. Understanding the host-finding behaviour of vectors assists the development and application of control methods and aids the understanding of epidemiology. Some general lessons are illustrated by reference to a century of research on the host-finding behaviour of tsetse flies which transmit trypanosomes causing human and animal trypanosomiases, including Rhodesian sleeping sickness, a zoonosis associated with wilderness areas of sub-Saharan Africa.

Estimating the costs of tsetse control options: an example for Uganda.

Decision-making and financial planning for tsetse control is complex, with a particularly wide range of choices to be made on location, timing, strategy and methods. This paper presents full cost estimates for eliminating or continuously controlling tsetse in a hypothetical area of 10,000km(2) located in south-eastern Uganda. Four tsetse control techniques were analysed: (i) artificial baits (insecticide-treated traps/targets), (ii) insecticide-treated cattle (ITC), (iii) aerial spraying using the sequential aerosol technique (SAT) and (iv) the addition of the sterile insect technique (SIT) to the insecticide-based methods (i-iii). For the creation of fly-free zones and using a 10% discount rate, the field costs per km(2) came to US$283 for traps (4 traps per km(2)), US$30 for ITC (5 treated cattle per km(2) using restricted application), US$380 for SAT and US$758 for adding SIT. The inclusion of entomological and other preliminary studies plus administrative overheads adds substantially to the overall cost, so that the total costs become US$482 for traps, US$220 for ITC, US$552 for SAT and US$993 - 1365 if SIT is added following suppression using another method. These basic costs would apply to trouble-free operations dealing with isolated tsetse populations. Estimates were also made for non-isolated populations, allowing for a barrier covering 10% of the intervention area, maintained for 3 years. Where traps were used as a barrier, the total cost of elimination increased by between 29% and 57% and for ITC barriers the increase was between 12% and 30%. In the case of continuous tsetse control operations, costs were estimated over a 20-year period and discounted at 10%. Total costs per km(2) came to US$368 for ITC, US$2114 for traps, all deployed continuously, and US$2442 for SAT applied at 3-year intervals. The lower costs compared favourably with the regular treatment of cattle with prophylactic trypanocides (US$3862 per km(2) assuming four doses per annum at 45 cattle per km(2)). Throughout the study, sensitivity analyses were conducted to explore the impact on cost estimates of different densities of ITC and traps, costs of baseline studies and discount rates. The present analysis highlights the cost differentials between the different intervention techniques, whilst attesting to the significant progress made over the years in reducing field costs. Results indicate that continuous control activities can be cost-effective in reducing tsetse populations, especially where the creation of fly-free zones is challenging and reinvasion pressure high.

Responses of tsetse flies, Glossina morsitans morsitans and Glossina pallidipes, to baits of various size.

Recent studies of Palpalis group tsetse [Glossina fuscipes fuscipes (Diptera: Glossinidae) in Kenya] suggest that small (0.25 × 0.25 m) insecticide-treated targets will be more cost-effective than the larger (≥1.0 × 1.0 m) designs currently used to control tsetse. Studies were undertaken in Zimbabwe to assess whether small targets are also more cost-effective for the Morsitans group tsetse, Glossina morsitans morsitans and Glossina pallidipes. Numbers of tsetse contacting targets of 0.25 × 0.25 m or 1.0 × 1.0 m, respectively, were estimated using arrangements of electrocuting grids which killed or stunned tsetse as they contacted the target. Catches of G. pallidipes and G. m. morsitans at small (0.25 × 0.25 m) targets were, respectively, ∼1% and ∼6% of catches at large (1.0 × 1.0 m) targets. Hence, the tsetse killed per unit area of target was greater for the larger than the smaller target, suggesting that small targets are not cost-effective for use against Morsitans group species. The results suggest that there is a fundamental difference in the host-orientated behaviour of Morsitans and Palpalis group tsetse and that the former are more responsive to host odours, whereas the latter seem highly responsive to visual stimuli.

Shoo fly, don't bother me! Efficacy of traditional methods of protecting cattle from tsetse.

Studies were made of the efficacy of using smoke and housing to protect cattle from tsetse (Diptera: Glossinidae) in Zimbabwe. The efficacy of smoke was assessed by its effect on catches in Epsilon traps baited with a blend of acetone, 1-octen-3-ol, 4-methylphenol and 3-n-propylphenol. The efficacies of different types of kraal (enclosure) were gauged according to the catches of electrocuting targets (E-targets), baited with natural ox odour, placed within various designs of kraal. Smoke from burning wood (Colophospermum mopane) or dried cow dung reduced the catch of traps by approximately 50-90%. Kraals with a continuous wooden or netting wall, 1.5 m high, reduced catches of E-targets by approximately 75%. Arrangements of electric nets were used to assess the numbers of tsetse attacking live cattle within kraals and/or near sources of smoke. The results confirmed findings with traps and E-targets: kraals reduced the numbers of tsetse that fed by approximately 80% and smoke reduced the numbers attracted by approximately 70%; the use of both reduced overall attack rates by approximately 90%. The inclusion of 4-methylguaiacol, a known repellent for tsetse and a natural component of wood smoke, halved the catches of traps and E-targets and the numbers of tsetse attacking cattle. The practical benefits and difficulties of using repellents and/or housing to manage trypanosomiases are discussed.

Prospects for the development of odour baits to control the tsetse flies Glossina tachinoides and G. palpalis s.l.

Field studies were done of the responses of Glossina palpalis palpalis in Côte d'Ivoire, and G. p. gambiensis and G. tachinoides in Burkina Faso, to odours from humans, cattle and pigs. Responses were measured either by baiting (1.) biconical traps or (2.) electrocuting black targets with natural host odours. The catch of G. tachinoides from traps was significantly enhanced ( approximately 5x) by odour from cattle but not humans. In contrast, catches from electric targets showed inconsistent results. For G. p. gambiensis both human and cattle odour increased (>2x) the trap catch significantly but not the catch from electric targets. For G. p. palpalis, odours from pigs and humans increased (approximately 5x) the numbers of tsetse attracted to the vicinity of the odour source but had little effect on landing or trap-entry. For G. tachinoides a blend of POCA (P = 3-n-propylphenol; O = 1-octen-3-ol; C = 4-methylphenol; A = acetone) alone or synthetic cattle odour (acetone, 1-octen-3-ol, 4-methylphenol and 3-n-propylphenol with carbon dioxide) consistently caught more tsetse than natural cattle odour. For G. p. gambiensis, POCA consistently increased catches from both traps and targets. For G. p. palpalis, doses of carbon dioxide similar to those produced by a host resulted in similar increases in attraction. Baiting traps with super-normal (approximately 500 mg/h) doses of acetone also consistently produced significant but slight (approximately 1.6x) increases in catches of male flies. The results suggest that odour-baited traps and insecticide-treated targets could assist the AU-Pan African Tsetse and Trypanosomiasis Eradication Campaign (PATTEC) in its current efforts to monitor and control Palpalis group tsetse in West Africa. For all three species, only approximately 50% of the flies attracted to the vicinity of the trap were actually caught by it, suggesting that better traps might be developed by an analysis of the visual responses and identification of any semiochemicals involved in short-range interaction.

Control techniques for Culicoides biting midges and their application in the U.K. and northwestern Palaearctic.

The recent emergence of bluetongue virus (Reoviridae: Orbivirus) (BTV) in northern Europe, for the first time in recorded history, has led to an urgent need for methods to control the disease caused by this virus and the midges that spread it. This paper reviews various methods of vector control that have been employed elsewhere and assesses their likely efficacy for controlling vectors of BTV in northern Europe. Methods of controlling Culicoides spp. (Diptera: Ceratopogonidae) have included: (a) application of insecticides and pathogens to habitats where larvae develop; (b) environmental interventions to remove larval breeding sites; (c) controlling adult midges by treating either resting sites, such as animal housing, or host animals with insecticides; (d) housing livestock in screened buildings, and (e) using repellents or host kairomones to lure and kill adult midges. The major vectors of BTV in northern Europe are species from the Culicoides obsoletus (Meigen) and Culicoides pulicaris (L.) groups, for which there are scant data on breeding habits, resting behaviour and host-oriented responses. Consequently, there is little information on which to base a rational strategy for controlling midges or for predicting the likely impact of interventions. However, data extrapolated from the results of vector control operations conducted elsewhere, combined with some assessment of how acceptable or not different methods may be within northern Europe, indicate that the treatment of livestock and animal housing with pyrethroids, the use of midge-proofed stabling for viraemic or high-value animals and the promotion of good farm practice to at least partially eliminate local breeding sites are the best options currently available. Research to assess and improve the efficacy of these methods is required and, in the longer term, efforts should be made to develop better bait systems for monitoring and, possibly, controlling midges. All these studies will need better methods of analysing the ecology and behaviour of midges in the field than are currently in use. The paucity of control options and basic knowledge serve to warn us that we must be better prepared for the possible emergence of other midge-borne diseases, particularly African horse sickness.

Towards a fuller understanding of mosquito behaviour: use of electrocuting grids to compare the odour-orientated responses of Anopheles arabiensis and An. quadriannulatus in the field.

The epidemiological role of and control options for any mosquito species depend on its degree of 'anthropophily'. However, the behavioural basis of this term is poorly understood. Accordingly, studies in Zimbabwe quantified the effects of natural odours from cattle and humans, and synthetic components of these odours, on the attraction, entry and landing responses of Anopheles arabiensis Giles (Diptera: Culicidae) and Anopheles quadriannulatus Theobald. The numbers of mosquitoes attracted to human or cattle odour were compared using electrocuting nets (E-nets), and entry responses were gauged by the catch from an odour-baited entry trap (OBET) relative to that from an odour-baited E-net. Landing responses were estimated by comparing the catches from E-nets and cloth targets covered with an electrocuting grid. For An. arabiensis, E-nets baited with odour from a single ox or a single man caught similar numbers, and increasing the dose of human odour from one to three men increased the catch four-fold. For An. quadriannulatus, catches from E-nets increased up to six-fold in the progression: man, three men, ox, and man + ox, with catch being correlated with bait mass. Entry responses of An. arabiensis were stronger with human odour (entry response 62%) than with ox odour (6%) or a mixture of cattle and human odours (15%). For An. quadriannulatus, the entry response was low (< 2%) with both cattle and human odour. Anopheles arabiensis did not exhibit a strong entry response to carbon dioxide (CO2) (0.2-2 L/min). The trends observed using OBETs and E-nets also applied to mosquitoes approaching and entering a hut. Catches from an electrocuting target baited with either CO2 or a blend of acetone, 1-octen-3-ol, 4-methylphenol and 3-n-propylphenol - components of natural ox odour - showed that virtually all mosquitoes arriving there alighted on it. The propensity of An. arabiensis to enter human habitation seemed to be mediated by odours other than CO2 alone. Characterizing 'anthropophily' by comparing the numbers of mosquitoes caught by traps baited with different host odours can lead to spurious conclusions; OBETs baited with human odour caught around two to four times more An. arabiensis than cattle-baited OBETs, whereas a human-baited E-net caught less ( approximately 0.7) An. arabiensis than a cattle-baited E-net. Similar caution is warranted for other species of mosquito vectors. A fuller understanding of how to exploit mosquito behaviour for control and surveys requires wider approaches and more use of appropriate tools.

Is there safety in numbers? The effect of cattle herding on biting risk from tsetse flies.

In sub-Saharan Africa, tsetse (Glossina spp.) transmit species of Trypanosoma which threaten 45-50 million cattle with trypanosomiasis. These livestock are subject to various herding practices which may affect biting rates on individual cattle and hence the probability of infection. In Zimbabwe, studies were made of the effect of herd size and composition on individual biting rates by capturing tsetse as they approached and departed from groups of one to 12 cattle. Flies were captured using a ring of electrocuting nets and bloodmeals were analysed using DNA markers to identify which individual cattle were bitten. Increasing the size of a herd from one to 12 adults increased the mean number of tsetse visiting the herd four-fold and the mean feeding probability from 54% to 71%; the increased probability with larger herds was probably a result of fewer flies per host, which, in turn, reduced the hosts' defensive behaviour. For adults and juveniles in groups of four to eight cattle, > 89% of bloodmeals were from the adults, even when these comprised just 13% of the herd. For groups comprising two oxen, four cows/heifers and two calves, a grouping that reflects the typical composition of communal herds in Zimbabwe, approximately 80% of bloodmeals were from the oxen. Simple models of entomological inoculation rates suggest that cattle herding practices may reduce individual trypanosomiasis risk by up to 90%. These results have several epidemiological and practical implications. First, the gregarious nature of hosts needs to be considered in estimating entomological inoculation rates. Secondly, heterogeneities in biting rates on different cattle may help to explain why disease prevalence is frequently lower in younger/smaller cattle. Thirdly, the cost and effectiveness of tsetse control using insecticide-treated cattle may be improved by treating older/larger hosts within a herd. In general, the patterns observed with tsetse appear to apply to other genera of cattle-feeding Diptera (Stomoxys, Anopheles, Tabanidae) and thus may be important for the development of strategies for controlling other diseases affecting livestock.

Less is more: restricted application of insecticide to cattle to improve the cost and efficacy of tsetse control.

Studies were carried out in Zimbabwe of the responses of tsetse to cattle treated with deltamethrin applied to the parts of the body where most tsetse were shown to land. Large proportions of Glossina pallidipes Austen (Diptera: Glossinidae) landed on the belly ( approximately 25%) and legs ( approximately 70%), particularly the front legs ( approximately 50%). Substantial proportions of Glossina morsitans morsitans Westwood landed on the legs ( approximately 50%) and belly (25%), with the remainder landing on the torso, particularly the flanks ( approximately 15%). Studies were made of the knockdown rate of wild, female G. pallidipes exposed to cattle treated with a 1% pour-on or 0.005% suspension concentrate of deltamethrin applied to the (a) whole body, (b) belly and legs, (c) legs, (d) front legs, (e) middle and lower front legs, or (f) lower front legs. The restricted treatments used 20%, 10%, 5%, 2% or 1% of the active ingredient applied in the whole-body treatments. There was a marked seasonal effect on the performance of all treatments. With the whole-body treatment, the persistence period (knockdown > 50%) ranged from approximately 10 days during the hot, wet season (mean daily temperature > 30 degrees C) to approximately 20 days during the cool, dry season (< 22 degrees C). Restricting the application of insecticide reduced the seasonal persistence periods to approximately 10-15 days if only the legs and belly were treated, approximately 5-15 days if only the legs were treated and < 5 days for the more restricted treatments. The restricted application did not affect the landing distribution of tsetse or the duration of landing bouts (mean = 30 s). The results suggest that more cost-effective control of tsetse could be achieved by applying insecticide to the belly and legs of cattle at 2-week intervals, rather than using the current practice of treating the whole body of each animal at monthly intervals. This would cut the cost of insecticide by 40%, improve efficacy by 27% and reduce the threats to non-target organisms and the enzootic stability of tick-borne diseases.

The use of aerial spraying to eliminate tsetse from the Okavango Delta of Botswana.

In Botswana, 16,000 km(2) of the Okavango Delta were aerial sprayed five times with deltamethrin, applied at 0.26-0.3g/ha, to control Glossina morsitans centralis Machado (Diptera: Glossinidae) over a period of approximately 8 weeks. The northern half of the Delta (7180 km(2)) was sprayed in June-September 2001 and the southern half (8720 km(2)) in May-August 2002. A barrier (mean width approximately 10 km) of 12,000 deltamethrin-treated targets was deployed at the interface of these two blocks to prevent tsetse from invading from the southern to the northern block. Prior to spraying, the mean catches of tsetse from man fly-rounds were 44.6 round/day in the northern block and 101 in the southern. Between September 2002 and November 2005, surveys ( approximately 820 daily fly-rounds and approximately 2050 trap-days) in the northern and southern blocks failed to detect tsetse. Simulations of tsetse populations suggest that while spraying operations can reduce tsetse populations to levels that are difficult to detect by standard survey techniques, such populations will recover to densities >100 tsetse/km(2) after 1000 days, at which density there is a very high probability (>0.999) that the survey methods will catch at least one fly. Since none was caught, it is argued that tsetse have been eliminated from the Delta. The particular success of this operation in comparison to the 18 aerial spraying operations conducted in the Delta prior to 2001 is attributed to the application of an adequate dose of insecticide, the use of a GPS-based navigation system to ensure even application of insecticide, and the large size and spatial arrangement of the spray blocks coupled with the use of a barrier of targets which prevented tsetse from re-invading the northern sprayed block before the southern one was treated.

The effects of host physiology on the attraction of tsetse (Diptera: Glossinidae) and Stomoxys (Diptera: Muscidae) to cattle.

In Zimbabwe, studies were made of the numbers of tsetse (Glossina spp.) and stable flies (Stomoxys spp.) attracted to cattle of different nutritional status, age and sex. Host odours were analysed to determine the physiological basis of these differences and improved methods are described for measuring rates of production of kairomones. Seasonal fluctuations in host weight, related to changes in pasture quality, had no significant effect on attraction of tsetse or Stomoxys. However, both attraction to different individuals and carbon dioxide production by these individuals were strongly correlated with weight, suggesting a possible link. Attraction to the odour from different types of cattle decreased in the order ox>cow>heifer>calf, and oxen were twice as attractive as calves of less than 12 months old. Lactation did not alter the relative attractiveness of cows. Calves less than six months old produced lower levels of carbon dioxide, acetone, octenol and phenols than oxen, but for older calves and cows, levels of production of known kairomones and repellents were similar to those of an ox. Carbon dioxide produced by cattle varied according to time of day and the animal's weight; cattle weighing 500 kg produced carbon dioxide at a mean rate of 2.0 l min(-1) in the morning and 2.8 l min(-1) in the afternoon compared to respective rates of 1.1 and 1.9 l min(-1) for cattle weighing 250 kg. Artificially adjusting the doses of carbon dioxide produced by individual cattle to make them equivalent did not remove significant differences in attractiveness for tsetse but did for Stomoxys. Increasing the dose of carbon dioxide from 1 to 4 l min(-1) in a synthetic blend of identified kairomones simulating those produced by a single ox, increased attractiveness to tsetse but not to the level of an ox. The results suggest that the main sources of differences in the attractiveness of individual cattle are likely to be variation in the production of carbon dioxide and, for tsetse, other unidentified kairomone(s). The biological and practical implications of these findings are discussed.

Blood-feeding behaviour of the malarial mosquito Anopheles arabiensis: implications for vector control.

Feeding behaviour of the malaria vector Anopheles arabiensis Patton (Diptera: Culicidae) was monitored for 12 months (March 2003-February 2004) in the Konso District of southern Ethiopia (5 degrees 15'N, 37 degrees 28'E). More than 45 000 An. arabiensis females were collected by host-baited sampling methods (light-traps, human landing catches, cattle-baited traps) and from resting sites (huts and pit shelters). In the village of Fuchucha, where the ratio of cattle : humans was 0.6 : 1, 51% of outdoor-resting mosquitoes and 66% of those collected indoors had fed on humans, human baits outdoors caught > 2.5 times more mosquitoes than those indoors and the mean catch of mosquitoes from pit shelters was about five times that from huts. Overall, the vast majority of feeding and resting occurred outdoors. In the cattle camps of Konso, where humans slept outdoors close to their cattle, approximately 46% of resting mosquitoes collected outdoors had fed on humans despite the high cattle : human ratio (17 : 1). In both places, relatively high proportions of bloodmeals were mixed cow + human: 22-25% at Fuchucha and 37% in the cattle camps. Anthropophily was also gauged experimentally by comparing the numbers of mosquitoes caught in odour-baited entry traps baited with either human or cattle odour. The human-baited trap caught about five times as many mosquitoes as the cattle-baited one. Notwithstanding the potential pitfalls of using standard sampling devices to analyse mosquito behaviour, the results suggest that the An. arabiensis population is inherently anthropophagic, but this is counterbalanced by exophagic and postprandial exophilic tendencies. Consequently, the population feeds sufficiently on humans to transmit malaria (sporozoite rates: 0.3% for Plasmodium falciparum and 0.5% for P. vivax, by detection of circumsporozoite antigen) but also takes a high proportion of meals from non-human hosts, with 59-91% of resting mosquitoes containing blood from cattle. Hence, classical zooprophylaxis is unlikely to have a significant impact on the malaria vectorial capacity of An. arabiensis in Konso, whereas treating cattle with insecticide might do.

User-friendly models of the costs and efficacy of tsetse control: application to sterilizing and insecticidal techniques.

An interactive programme, incorporating a deterministic model of tsetse (Diptera: Glossinidae) populations, was developed to predict the cost and effect of different control techniques applied singly or together. Its value was exemplified by using it to compare: (i) the sterile insect technique (SIT), involving weekly releases optimized at three sterile males for each wild male, and (ii) insecticide-treated cattle (ITC) at 3.5/km(2). The isolated pre-treatment population of adults was 2500 males and 5000 females/km(2); if the population was reduced by 90%, its growth potential was 8.4 times per year. However, the population expired naturally when it was reduced to 0.1 wild males/km(2), due to difficulties in finding mates, so that control measures then stopped. This took 187 days with ITC and 609 days with SIT. If ITC was used for 87 days to suppress the population by 99%, subsequent control by SIT alone took 406 days; the female population increased by 48% following the withdrawal of ITC and remained above the immediate post-suppression level for 155 days; the vectorial capacity initially increased seven times and remained above the immediate post-suppression level for 300 days. Combining SIT and ITC after suppression was a little faster than ITC alone, provided the population had not been suppressed by more than 99.7%. Even when SIT was applied under favourable conditions, the most optimistic cost estimate was 20-40 times greater than for ITC. Modelling non-isolated unsuppressed populations showed that tsetse invaded approximately 8 km into the ITC area compared to approximately 18 km for SIT. There was no material improvement by using a 3-km barrier of ITC to protect the SIT area. In general, tsetse control by increasing deaths is more appropriate than reducing births, and SIT is particularly inappropriate. User-friendly models can assist the understanding and planning of tsetse control. The model, freely available via http://www.tsetse.org, allows further exploration of control strategies with user-specified assumptions.

Could insecticide-treated cattle reduce Afrotropical malaria transmission? Effects of deltamethrin-treated Zebu on Anopheles arabiensis behaviour and survival in Ethiopia.

Anopheles arabiensis Patton (Diptera: Culicidae) is the most widespread vector of malaria in the Afrotropical Region. Because An. 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. Therefore, we conducted field experiments in southern Ethiopia to establish whether Zebu cattle (Bos indicus L.) treated with a pour-on pyrethroid formulation of 1% deltamethrin, widely used to control ticks and tsetse, would be effective against An. arabiensis or cause the female mosquitoes to feed more frequently on humans, due to behavioural avoidance of insecticide-treated cattle. Contact bioassays (3 min exposure) showed that the insecticide remained effective for about 1 month (kill rate > 50%) against mosquitoes feeding on the flanks of treated cattle. A novel behavioural assay demonstrated that An. arabiensis readily fed on insecticide-treated cattle and were not deflected to human hosts in the presence of treated cattle. DNA-fingerprinting of bloodmeals revealed that An. arabiensis naturally feeds most frequently on older animals, consistent with the established practice of applying insecticide only to older cattle, while allowing younger untreated animals to gain immunity against infections transmitted by ticks. These encouraging results were tempered by finding that > 90% of An. arabiensis, An. pharoensis and An. tenebrosus females feed on the legs of cattle, farthest from the site of pour-on application along the animal's back and where the treatment may be least residual due to weathering. Observations of mosquitoes feeding naturally on insecticide-treated cattle showed that the majority of wild female anophelines alighted on the host animal for less than 1 min to feed, with significantly shorter mean duration of feeding bouts on insecticide-treated animals, and the effective life of the insecticide was only 1 week. Thus the monthly application of deltamethrin to cattle, typically used to control tsetse and ticks, is unlikely to be effective against An. arabiensis populations or their vectorial capacity. Even so, it seems likely that far greater impact on anopheline mosquitoes could be achieved by applying insecticide selectively to the legs of cattle.

Insecticide-treated cattle against tsetse (Diptera: Glossinidae): what governs success?

The distributions of insecticide-treated cattle from sites in Tanzania and Zimbabwe were assessed from interviews with livestock owners, analysis of secondary livestock data and mapping technologies. The time-course of tsetse control operations at these sites were then simulated using a mathematical model that assumed diffusive movement and logistic growth in fly populations. A simulation of a tsetse control operation in Mudzi district, north-east Zimbabwe, was in accord with observations that the use of insecticide-treated cattle was unable to prevent substantial re-invasion of tsetse from Mozambique, consequent on the patchy distribution of cattle. The simulation was also consistent with the observed efficacy of a 10-km wide barrier of insecticide-treated targets deployed evenly at 4 km/(-2). Simulation of a control operation on Mkwaja Ranch in Tanzania was in accord with the observation that the use of insecticide-treated cattle reduced the tsetse population on the ranch by c. 90%. Insecticide-treated cattle were used to better effect in the Kagera Region of Tanzania. Simulation of this operation predicts that the deployment of 35,000 treated cattle in the area would result in > 99% control of the tsetse population, consistent with the observed decline, by 1-2 orders of magnitude, in cases of trypanosomiasis in the region. The greater success of the Kagera operation was due to the size and shape of the treated area and, particularly, to the restriction of re-invasion to 20% of the perimeter, compared with > 80% on Mkwaja. Simulation was used to assess how tsetse control could have been improved at Mkwaja. The results suggest that splitting herds into smaller, more numerous, units could have achieved some improvement but, in general, the disease problem would not have been solved by the use of insecticide-treated cattle alone. Only by deploying odour-baited targets in ungrazed areas, or in a 1-3-km barrier around the ranch, could substantially better control (99-99.9%) have been achieved. Sensitivity analyses of the Mkwaja simulation showed that the general conclusions were robust to assumptions regarding cattle distribution and the rates of fly movement and growth. Properly managed and appropriately applied insecticide-treated baits are powerful weapons for tsetse control but should not be used without regard to potential levels of re-invasion, consequent largely on considerations of the size and shape of the treatment area and the density and distribution of the baits.

A comparison of the feeding behaviour of tsetse and stable flies.

In Zimbabwe, observations were made of the behaviour of individual stable flies (Stomoxys spp.) (Diptera: Muscidae) and tsetse (Glossina spp.) (Diptera: Glossinidae) feeding on cattle during the wet (Stomoxys and tsetse) and dry (tsetse only) seasons. For Stomoxys landing on adult cattle, only 27% took a full meal (mean feeding time = 147 s). Most Stomoxys left the host before completing their meal, largely due to disturbance by the host's defensive behaviour (24%, mean time = 59 s) or other flies (44%, 71 s). The probability of a Stomoxys leaving the host progressively increased with time. Simultaneous observations of tsetse showed that, compared to Stomoxys, their feeding success was lower (15%), feeding was interrupted earlier (33 s) and the time taken to complete a meal was shorter (109 s). Further studies of tsetse across different seasons and hosts showed that feeding success varied according to host age (adult = 7%; calf = 3%) and was negatively correlated with the frequency of host defensive behaviour and the relative abundance of non-biting Diptera. Disturbances were more often caused by host behaviour (69%) than other flies (31%) and the probability of tsetse leaving decreased with time on the host. Overall, these results suggest that tsetse and Stomoxys have different feeding strategies. In particular, tsetse appear to be more responsive to host defensive behaviour, which reduces their feeding success relative to Stomoxys. These behavioural differences are consistent with the respective life-history characteristics of Stomoxys and tsetse.