Insecticide-treated plastic tarpaulins for control of malaria vectors in refugee camps
Abstract
Abstract Spraying of canvas tents with residual pyrethroid insecticide is an established method of malaria vector control in tented refugee camps. In recent years, plastic sheeting (polythene tarpaulins) has replaced canvas as the utilitarian shelter material for displaced populations in complex emergencies. Advances in technology enable polythene sheeting to be impregnated with pyrethroid during manufacture. The efficacy of such material against mosquitoes when erected as shelters under typical refugee camp conditions is unknown. Tests were undertaken with free-flying mosquitoes on entomological study platforms in an Afghan refugee camp to compare the insecticidal efficacy of plastic tarpaulin sprayed with deltamethrin on its inner surface (target dose 30 mg/m2), tarpaulin impregnated with deltamethrin (initially ≥ 30 mg/m2) during manufacture, and a tent made from the factory impregnated tarpaulin material. Preliminary tests done in the laboratory with Anopheles stephensi Liston (Diptera: Culicidae) showed that 1-min exposure to factory-impregnated tarpaulins would give 100% mortality even after outdoor weathering in a temperate climate for 12 weeks. Outdoor platform tests with the erected materials (baited with human subjects) produced mosquito mortality rates between 86–100% for sprayed or factory-impregnated tarpaulins and tents (average ˜40 anophelines and ˜200 culicines/per platform/night), whereas control mortality (with untreated tarpaulin) was no more than 5%. Fewer than 20% of mosquitoes blood-fed on human subjects under either insecticide-treated or non-treated shelters. The tarpaulin shelter was a poor barrier to host-seeking mosquitoes and treatment with insecticide did not reduce the proportion blood-feeding. Even so, the deployment of insecticide-impregnated tarpaulins in refugee camps, if used by the majority of refugees, has the potential to control malaria by killing high proportions of mosquitoes and so reducing the average life expectancy of vectors (greatly reducing vectorial capacity), rather than by directly protecting refugees from mosquito bites. Mass coverage with deltamethrin-sprayed or impregnated tarpaulins or tents has strong potential for preventing malaria in displaced populations affected by conflict.
Introduction
Prevention of malaria is a major technical and operational problem in displaced populations affected by conflict. For control of malaria vectors by adulticidal treatment of their resting sites with residual insecticides (Najera, 1996), technical problems arise because conventional surfaces, such as walls and ceilings of houses, are not available for treatment in newly displaced or homeless populations, and because bednets are unsuitable for use in tented refugee camps (Bouma et al., 1996a; Rowland, 1999, 2001; Rowland & Nosten, 2001). Operational problems because of conflict, breakdown of health services, insecurity, and inaccessible populations may combine to make it impossible to organize anything better than an emergency humanitarian response. In the early acute phase the priority needs are provision of food and water, sanitation and distribution of blankets and shelter material (Anon, 1997). Agencies specializing in emergency response have neither the time nor the capacity to mount a considered preventive response against malaria. Hence, anything that can be done to prevent malaria is more likely to be taken up by these logistic or humanitarian agencies if it places no extra demands on their established response package. Because blankets and shelter materials (plastic tarpaulins) are always distributed as part of the emergency response, these materials may constitute the only surfaces suitable or readily available for insecticide treatment. Previous work has shown that top-sheets and blankets treated with residual pyrethoid can be a useful tool against malaria and cutaneous leishmaniasis (Rowland et al., 1999; Reyburn et al., 2000). Likewise, canvas tents sprayed with pyrethroid are a proven intervention in malaria epidemics (Hewitt et al., 1995; Bouma et al., 1996a). In recent years plastic tarpaulins have replaced canvas tents as the favoured shelter material for refugees; this is because polythene sheeting is cheaper to make, cheaper to air-freight, and easier to stockpile (Rowland & Nosten, 2001). If this material could be pre-impregnated with insecticide, be shown to kill malaria vectors, and give protection against malaria it would have major advantages, as it would require no additional resources or organization other than those already deployed at the outset of an emergency. Hence, the global malaria control initiative, Roll Back Malaria (Nabarro, 1999), has been working with industry to develop factory-impregnated plastic sheeting (Allan, 2001; Allan & Guillet, 2002; Frandsen, 2002). The present paper describes the first evaluation made under controlled conditions in a refugee camp.
Materials and Methods
Tarpaulins
The physical structure of the plastic tarpaulin is a core-weave matrix (90 µm thick) covered with two layers of laminate (each 45 µm), weighing 180 g/m2. The core weave is made of high-density polythene and the laminates of low-density polythene. The UNHCR (United Nations High Commissioner for Refugees) tarpaulin (made by Qin Gdao Gwhoa, Qingdao, China) is stained with blue dye. Vestergaard Frandsen (Kolding, Denmark) tarpaulin and tents were made of the same material, dyed white on the outside and black on the inside, and were impregnated with deltamethrin during manufacture. The core structure acts as a store for insecticide and the outer layers serve to physically and chemically protect the store and to regulate migration of insecticide to the surface. Owing to their physico-chemical properties, the laminates allow migration of insecticide, which builds up at the surface during storage. The laminates are impregnated with a low concentration of insecticide during manufacture. The concentration at the surface is a balance between the concentration in the core layer, migration and inactivation by ultraviolet light (UV). Through appropriate use of migration retarding chemicals and UV filters in the two laminates, a more constant effect at the surface is obtained. The concentration of deltamethrin during the mixing process was 45 mg/m2 in the surface laminates, and the total concentration was 2 g deltamethrin per kg of tarpaulin. Chemical analysis (by M. Galoux at Gembloux, Belgium) using acetone extraction, showed that 20–30% of deltamethrin was lost in the processing, because the operational temperature for tarpaulin production is similar to the evaporation temperature of the insecticide. This overall analysis does not reveal the final distribution at the surface.
Laboratory bioassays
Bioassays were carried out at the London School of Hygiene & Tropical Medicine (LSHTM). Samples of factory-impregnated sheeting were either stored indoors or weathered outdoors on the roof of the LSHTM building for 84 days between November and January. Tests were done in WHO (1981) resistance test kits lined with the impregnated sheeting; untreated polyethene was used as a control. Insecticide-susceptible females of Anopheles stephensi were exposed for 1 or 3 min (five replicates of 10 mosquitoes per replicate), then kept in the humidified holding chamber for 24 h with sugar solution before scoring mortality 24 h post-exposure.
Outdoor platform studies
The methodology of Hewitt et al. (1995) and Rowland et al. (1999) was used to simulate the type of outdoor contact that occurs naturally between host-seeking mosquitoes, shelters and sleepers. Giant trap nets (length 6 m × height 2 m × width 5 m) made of mosquito netting were erected above ant-proof platforms upon which were constructed λ-shaped shelters made from plastic sheeting, a ridge pole and two upright poles; the sheets were open at the ends and pegged to the floor along the edges. Within each shelter a man clothed in shalwar chemise and covered with a cotton sheet slept on a bedroll on the floor.
For the first half of the night wild, host-seeking mosquitoes, attracted to the platforms, were collected from the outside of the trap nets and released within. Numbers were supplemented with mosquitoes attracted to calves enclosed nearby within mosquito nets. The following morning mosquitoes were collected from the floor sheets and inner surface of the trap net, separated into dead or alive, and kept in humidified cups with sugar solution for a further 12 h before scoring delayed mortality. All mosquitoes were categorized as blood-fed or unfed, identified to genera and the anophelines to species.
The factory-impregnated plastic tarpaulin was tested against (a) a standard untreated UNHCR plastic tarpaulin as a control, (b) a UNHCR tarpaulin sprayed with deltamethrin on the inner surface at 30 mg/m2 (using a Hudson X-pert™ sprayer), and (c) a tent made from the factory-impregnated deltamethrin tarpaulin (manufactured by Vestergaard Frandsen A/S). Shelters of each treatment type were tested for one night on each of four platforms in rotation.
Statistical analysis
Statistical analyses were done using Stata 6 (www. stata.com). Proportional data were arcsine-transformed and subjected to analysis of variance to examine the effect of treatment on blood-feeding and mortality rates.
Results
Laboratory bioassays
On both weathered and unweathered sheeting, consistently 100% mortality of Anopheles stephensi females resulted from just 1-min exposure (and 24 h holding period), whereas the control mortality (on untreated sheeting) was less than 10%.
Outdoor platform bioassays
An average of 202 ± 15 (±standard error) culicines and 39 ± 7 anophelines were caught at each platform per night. The majority of anophelines were An. subpictus Grassi (18 ± 5) and An. stephensi (15 ± 2), plus small numbers of An. culicifacies Giles, An. fluviatilis James, An. splendidus Koizumi, An. pulcherrimus Theobald and An. annularis van der Wulp. With each species of anopheline present only in low numbers, the results were grouped by genera for presentation (Fig. 1). Tables 1 and 2 show the mortality and blood-feeding rates for culicines and the two most abundant anophelines. The majority of anophelines on platforms with insecticide-treated tarpaulins/tents died, whereas control mortality was never more than 6% (Culicines: F3,12 = 33, P < 0.001; An. stephensi: F3,12 = 24, P < 0.001; An. subpictus: F3,12 = 46, P < 0.001). There were no significant differences in mortality between the three insecticide treatments. Culicines showed slightly higher survival rates than anophelines. Blood-feeding rates were consistently low throughout the trial for anophelines and for culicines. There were no differences in blood-feeding rate between the insecticide and control treatments (Culicines: F3,12 = 0.47, P = 0.71; An. stephensi: F3,12 = 0.3, P = 0.82; An. subpictus: F3,12 = 0.61, P = 0.62). Figure 1 confirms that the majority of mosquitoes died unfed, presumably before making contact with the host.
Mean % bloodfed | |||
---|---|---|---|
Net treatment | Culicines | An. subpictus | An. stephensi |
Deltamethrin-sprayed UNHCR plastic tarpaulin | 8 (0–29) | 6 (0–44) | 7 (0–28) |
Vestergaard factory-impregnated plastic sheeting | 5 (0–16) | 20 (0–77) | 18 (0–78) |
Vestergaard factory-impregnated plastic tent | 5 (0–19) | 4 (0–28) | 5 (0–41) |
Untreated UNHCR plastic tarpaulin (control) | 3 (0–9) | 6 (0–46) | 11 (0–67) |
Mean % mortality | |||
---|---|---|---|
Net treatment | Culicines | An. subpictus | An. stephensi |
Deltamethrin sprayed UNHCR plastic tarpaulin | 79 (45–99) | 100 (100–100) | 97 (78–98) |
78 | 100 | 97 | |
Vestergaard pre-treated plastic sheeting | 98 (93–100) | 100 (100–100) | 99 (97–100) |
98 | 100 | 100 | |
Vestergaard pre-treated plastic tent | 66 (27–96) | 95 (59–95) | 86 (43–99) |
64 | 94 | 86 | |
Control | 5 (2–10) | 4 (0–33) | 5 (0–23) |
Discussion
Pyrethroid-impregnated tarpaulins show good potential for malaria prevention in displaced populations. The impressive insecticidal activity demonstrated in laboratory bioassays was corroborated in field tests in the Afghan refugee camp, where contact between treated material and mosquito was near to natural. There was little effect on blood-feeding. This contrasts with the demonstration of feeding inhibition (repellency) that occurred when pyrethroid-treated top-sheets were tested on the same platforms in earlier studies (Rowland et al., 1999; Graham et al., 2002). Crudely erected tarpaulins offer plenty of gaps through which host-seeking mosquitoes may pass en route to the host. Thus, the potential of treated tarpaulins as a means of malaria prevention will depend upon generating high mortality and greatly reducing vectorial capacity, mainly by reducing survival rates among the vector population (‘mass effect’) rather than giving direct personal protection from biting. The prospect for disease control would remain high because coverage in new refugee camps would approach 100% as a result of free distribution of tarpaulins on registering of refugees.
An earlier evaluation of permethrin-sprayed canvas tents in Pakistan showed decay of residue within a few months of spraying of inner surfaces (Bouma et al., 1996b). Better persistence was achieved with our factory-impregnated deltamethrin sheeting when weathered outdoors in London. The timing and location of this exposure (English winter) means that the sheeting would not have been subjected to particularly intense UV radiation. The UVA and B radiation, which accelerates the degradation of insecticides, is at higher levels closer to the equator and higher at comparable latitudes in the southern than in the northern hemisphere; cloud cover would also reduce UV levels. It is important that an examination of the resistance of the pre-treated sheeting to weathering also be carried out in a more severe, tropical climate. The weathering results were encouraging in showing that factory-impregnation is able to resist run-off of insecticide on exposure to frequent rain.
Although it is important that pre-treated sheeting is able to withstand weathering, this material will still be useful even if the period of residual activity lasts only a few months. The acute stage of any emergency – when human mortality rates are highest – is the period when conventional malaria control is often thwarted by logistic and security constraints (Rowland & Nosten, 2001). Plastic tarpaulins are distributed during that initial influx of refugees. A few months later, as the camp becomes better established, refugees usually erect their own homes using locally available materials such as mud and straw. Plastic tarpaulins may be retained as useful waterproofing for roofs or walls but may also be sold on. The insecticidal activity of the tarpaulin need only last as long as IDPs and refugees are using the tarpaulins as their main shelter. Once the camp moves into a chronic stage, conventional methods of malaria control (e.g. ITN, IRS) are more easily applied.
Diarrhoeal diseases are the most important cause of mortality in refugee camps (Toole & Waldman, 1997). The significant role of houseflies in the transmission of some diarrhoeal diseases (Cohen et al., 1991; Chavasse et al., 1999) indicates that the potential of pyrethroid-treated sheeting to reduce housefly numbers should be examined. Leishmaniasis is another vector-borne disease that can be controlled by residual spraying (Pandya, 1983; Vioukov, 1987; Reyburn et al., 2000). Insecticide-treated sheeting therefore has potential as a wider public health tool against various vector-borne diseases in refugee camps, alongside its promise as a weapon against malaria in the problematic acute phase.
Acknowledgements
HealthNet International's malaria control and research programme is supported by the European Commission. This project was supported by the Roll Back Malaria Secretariat Complex Emergencies Group with funding support from the US Department of State Bureau of Population, Refugees and Migration. M.R. and J.L. are supported by the U.K. Department for International Development and the Gates Foundation. None of these donors can accept responsibility for any information provided or views expressed. For chemical analysis we are grateful to M. Galoux, Station de Phytopharmacie, Ministry of Agriculture, Gembloux, Belgium. The mention of specific companies and/or products does not in any way imply that they are recommended or endorsed by the World Health Organization in preference over others that are not mentioned.
References
Accepted 22 June 2002