Citation:Morrison AC, Zielinski-Gutierrez E, Scott TW, Rosenberg R (2008) Defining Challenges and Proposing Solutions for Control of the Virus Vector Aedes aegypti. PLoS Med 5(3): e68. doi:10.1371/journal.pmed.0050068
Published: March 18, 2008
Copyright: © 2008 Morrison et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Support for the panel meeting and for this article was provided by the Bill & Melinda Gates Foundation.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: DSS, decision support systems;GIS, geographic information system;ITM, insecticide-treated material
The mosquito Aedes aegypti is the primary vector of three important viral diseases—dengue, yellow fever, and chikungunya—and is capable of transmitting a number of others. Ae. aegypti, which bites during daylight, is uniquely domestic among mosquito vectors: it mates, feeds, rests, and lays eggs in and around human habitation.
Control of this mosquito should lead to control of disease, and there are well-documented historical examples of both yellow fever and dengue being eliminated or significantly reduced through Ae. aegypti control . Construction of the Panama Canal was possible only after US Army Surgeon General William Gorgas stopped yellow fever transmission among workers by eliminating Ae. aegypti breeding sites. Fred Soper, of the Rockefeller Foundation, led the highly successful Ae. aegypti eradication program during the 1950s and 1960s that extinguished yellow fever and dengue transmission from most of Central and South America. More recently, Singapore and Cuba greatly reduced risk of dengue transmission by means of anti–Ae. aegypti legislation and actions. Use of the predatory crustacean Mesocyclops is preventing dengue transmission in parts of Vietnam .
Unfortunately, these successes are exceptions that were often too short-lived. Dengue has reoccupied Latin America and increases yearly; most of Southeast Asia remains highly endemic. Despite the availability of a vaccine, yellow fever outbreaks occur with dismaying frequency in Africa and Latin America. Chikungunya caused epidemics during 2005–2006 in the Indian Ocean region that were the largest yet recorded. In some regions, such as Sudan, all three viruses coexist. The reasons for failure are numerous. Spreading urbanization increases the habitat for expanding Ae. aegypti populations, rapid global migration increases the potential for vector and virus dissemination, poverty hobbles the efforts of individuals and communities to carry out effective protective measures, and even when resources for control exist, they are too often ineffectively applied .
- If done properly, Aedes aegypti suppression is a practical method to control urban dengue, yellow fever, and chikungunya viruses.
- The goal should be to reduce adult Ae. aegypti populations or their interactions with humans below that which can sustain an epidemic; it is unrealistic to expect to eradicate either the vector or the viruses.
- The reason Ae. aegypti control is not used more widely is that most endemic countries have poorly defined goals and are unwilling to commit resources except during epidemics.
- A new paradigm for control should include focused surveillance and strategies that kill adult mosquitoes, and development and testing of products that appeal to the consumer; this could make national programs more effective and cheaper, and therefore more attractive.
- Programmatic innovations and novel products can be most effectively and efficiently developed in areas where substantial baseline information is already available (e.g., Peru, Puerto Rico, Thailand, Singapore, Indonesia, and Vietnam). This should lead to large-scale intervention project(s) of sufficient duration to determine the longer-term intervention consequences, specifically to establish if transmission can be broken.
An international panel (see Acknowledgments) met at Fort Collins, Colorado, United States, in May 2006 to critically examine why Ae. aegypti control has so seldom been successful in eliminating disease and to recommend measures to increase the opportunities for success. Consideration of control methods was limited to technologies now available or that could be soon developed; therefore, methods depending on genetical manipulation, such as genomic transformations and sterile male releases, were not discussed. There was unanimous agreement that elimination or significant reduction of Ae. aegypti populations is an effective and proven method for disease prevention. After reviewing examples of successful and failed programs, two issues especially worthy of attention were identified: (1) program design and management—specifically, sustainability, goal-setting, and surveillance/assessment—and (2) the need for more effective mosquito control tools.
Program Design and Management
Common to all successes was the ability and commitment to sustain a mosquito control program. At its most basic level, Ae. aegypti–borne disease can be eliminated by the simple expedient of preventing vector access to water containers necessary for immature mosquito development. But the implementation even of such a straightforward program requires careful planning, blanket coverage, and conscientious execution; to have lasting effect it must be indefinitely sustained. Effective programs had two common elements consistent with a policy of determined leadership that stressed community responsibility: (1) a vertical component, usually governmental, that initiated, planned, and oversaw the program, and (2) a horizontal component, usually householders, who helped execute control measures and permitted access to their property.
Although community participation is crucial to success, there are no apparent examples of successful efforts initiated solely at the community level . Denying Ae. aegypti the opportunity to lay eggs or complete development in water in and around homes will work only when all members of the community participate—people who eliminate mosquito habitats are still vulnerable from neighbors who do not. In fact, participation is often high only during epidemics. No examples were found where education campaigns had lasting influence on behavior. Although “bottom-up” policy is attractive, it is unrealistic to expect it to work without strong “top-down” leadership and support. The most effective means of ensuring community participation would be arthropod control measures that are themselves attractive to the community, such as those that have broad impact by killing a wide variety of vectors and nuisance pests, not just Ae. aegypti.
Well-reasoned and specific public health goals will be the foundation for effective leadership. Sustainability will be enhanced by the availability of easy-to-use evidence-based decision-making tools that allow public health officials to effectively argue for policies that are scientifically based, make better use of limited resources, advocate for provision of the necessary funding, secure “buy-in” from higher levels of government, and obtain programmatic freedom for decision making.
How is success defined in an anti–Ae. aegypti campaign? What are the endpoints? Goals have often been undefined, ambiguous, or irrelevant. Although the ultimate objective must be to prevent disease, most current programs emphasize reduction of immature Ae. aegypti density, which is of little value because its relation to transmission risk is weak . Goals for preventing epidemics or maintaining consistently low vector populations require different programmatic strategies and necessitate different surveillance systems. One of the most important steps for improving efficacy of Ae. aegypti–borne disease control programs will be development of methods for setting quantitative goals that are spatially and temporally specific. This will require a shift in thinking from the current situation, in which little interpretation is needed of prescribed entomological measures that are uniformly applied, to a new approach that calls for assessing risk against locally derived goals that are based on location-specific dynamics in epidemiology, ecology, and availability of resources. In this new scenario, it is essential to understand that entomological thresholds are dynamic and that they only make sense in the context of local epidemiology .
Surveillance and assessment.
Surveillance is fundamental for setting goals and evaluating success. Unfortunately, except for vector eradication programs , current surveillance seems to play no significant role in strategically applied Ae. aegypti–borne disease prevention. There are epidemiological surveillance systems designed to detect introduction of novel viruses, but we are not aware of any systems that predict epidemic risk based on entomological information.
Measuring mosquito density is conceptually easier and less expensive than human diagnosis, but immature mosquito indices have a weak relation to risk from virus transmission. Methods that monitor production of late-instar larvae and pupae promise to be more informative [6,7] but require validation. Although the density of adult female Ae. aegypti, which transmit virus, is more closely associated with disease incidence, adults of this species are difficult to catch and rarely monitored. An inexpensive and effective Ae. aegypti–specific adult trap would be a significant surveillance breakthrough, and could also allow for virus testing.
It is not clear how useful any surveillance will be that does not measure the immunological status of the population at risk. Serologic assays of periodic blood samples from representative populations, including children, are impracticable in many communities. Development of noninvasive (e.g., using saliva, tears, or urine), serotype-specific, rapid, sensitive, and inexpensive methods for detection of antibodies would be a major advance for surveillance and assessment. Clinic networks that monitor and report clinical disease—syndromic surveillance—may forecast elevated risk, even though cases typically trail transmission, and subclinical or mild illness may evade detection.
There are no validated algorithms to predict, at a given location, to what extent a reduction in Ae. aegypti population will reduce transmission. An improved understanding of the relation of entomological factors to risk must be a priority. Information generated from surveying mosquitoes, virus, and sera needs to be synthesized into meaningful models of virus transmission risk. The complex natural history of arbovirus transmission contributes to the difficulty in setting goals and executing effective control. There is convincing evidence that mosquito-borne disease incidence is highly focal . Knowing the spatial distribution of cases in a given situation and the most productive sources of adult mosquitoes would allow planners to focus limited resources. Although surveillance data, both entomological and epidemiological, have often been collected, other than in Singapore  and research settings, there are few instances of control programs where disparate parameters have been usefully combined. Areas identified as needing urgent research attention are (1) development of entomological thresholds for different disease control goals and (2) effective, user-friendly virus transmission models that account for variation throughout time and space.
Four areas with potentially high impact for vector control tool development were identified based on products either currently on the market or whose development is nearly complete. They are (1) novel methods for control of immature Ae. aegypti, (2) novel delivery systems for adult control, (3) adult mosquito monitoring tools, and (4) quantitative assessment tools. The emphasis is on application within households or dwellings, recognizing Ae. aegypti's unique habits of feeding frequently on human blood and resting indoors .
Novel delivery systems for adult control.
It has been known since the early 1900s [10,11] that the most cost-effective means of preventing mosquito-borne disease is to target the adult vector, which transmits the pathogen. The prevailing paradigm for suppressing Ae. aegypti, however, targets immature mosquitoes, the vast majority of which will not survive long enough to transmit virus.
Aircraft-delivered and truck-mounted ultra-low volume spraying has limited efficacy  against Ae. aegypti because the vapor frequently does not penetrate into buildings where adult mosquitoes rest, although it is often used as a visible symbol of governmental action during emergencies. A better approach is to target adult vectors in places proximate to humans by delivering pesticides directly inside dwellings. This shifts vector population age structure to younger mosquitoes and reduces survival of infective or virus-incubating mosquitoes. Control in buildings can be accomplished with indoor residual or space spraying, but those approaches are often hampered by limited access into homes and resource limitations.
In another approach, which has been referred to as la casa segura, or “the safe home” (B. Beaty, personal communication), householders protect themselves in the home against a variety of vector-borne diseases and pest insects. This concept recognizes that householders in many developing countries already buy pesticides to control insects in the home. Residents in Thailand were estimated to have spent US$4–US$25/year/household on insecticides; this represents a greater amount than was spent per household on organized mosquito control . There are advantages in engaging market forces to promote products, rather than relying solely on public health appeals.
Insecticide-treated materials (ITMs), developed initially for malaria control, have not been sufficiently appraised against Ae. aegypti. In Mexico and Venezuela, lambdacyhalothrin- or deltamethrin-treated materials hung on windows and used as water jar covers significantly reduced Ae. aegypti densities in both intervention and controls clusters, which was attributed to community or “spill-over” effect . ITMs usually enjoy high acceptance. There has been steady improvement in ITM efficacy and economy—longer-lasting chemicals and ever-improving impregnation techniques—to meet malaria demand. Ae. aegypti–specific ITM delivery strategies have yet to receive the attention they merit, despite their obvious promise.
Ovitraps are faster and less expensive, use less pesticide, and are less likely to affect non-target species than interior residual spraying. A combination of lethal ovitraps and sticky ovitraps was used in North Queensland, Australia with encouraging results (S. Ritchie, personal communication; ). Lethal ovitraps are essentially a black bucket containing water with an attractant infusion (0.5 grams alfalfa pellet), a cloth strip treated with a residual pyrethroid insecticide, and a plastic mesh cover (, adapted from ). Use of oviposition repellents or source reduction might push gravid females away from hard-to-control natural sites toward lethal ovitraps or ITMs; i.e., a “lure and kill” strategy. Suitability and impact of these tools needs to be evaluated in endemic areas.
Space repellents are volatiles that expel mosquitoes from a large cubic area without necessarily killing them. In addition to killing mosquitoes, many insecticides repel at low doses . Metofluthrin (Sumitomo Chemical), a synthetic pyrethroid with spontaneous vapor action at room temperatures, has been highly lethal to mosquitoes in preliminary tests . Similar in concept to ITMs, it can be formulated as compressed paper or plastic strips to be hung from ceilings [20,21]. Passive space repellents, which could be used in combination with ITMs or lethal ovitraps, might provide low-cost, long-term mosquito repellency free of an external energy source and with minimal pesticide exposure for residents.
Novel methods for control of immature Ae. aegypti.
An ideal larvicide would be long-lasting, have low toxicity for both humans and other non-target organisms, and persist. Products should be evaluated to account for variation in container materials (plastic, metal, clay, glass, cement, wood) and in large trials under realistic field conditions.
Pyriproxyfen, an insect growth regulator, may be the most promising larvicidal product currently available. It is effective at inhibiting adult Ae. aegypti emergence at concentrations of less than or equal to one part per billion [22–27], can be applied in various formulations (e.g., sticks, granules), and is cost-competitive. It is already in veterinary and agricultural use. It remains effective up to five months, longer than Bacillus thuringiensis israelensis, methoprene, or temephos, and is less toxic. Adult mosquitoes exposed to pyriproxyfen have decreased fecundity. Importantly, contaminated adults can disseminate lethal doses from treated to untreated sites [25,27].
Adult monitoring tools.
Development of a cost-effective, field-appropriate method for estimating adult Ae. aegypti densities should be a priority. An adult trap would be less intrusive than current Ae. aegypti household surveys, require less labor, and allow for more complete coverage both spatially and temporally. Ideal characteristics of an adult trap would include low cost, ease of distribution, species exclusivity, a consistent sampling profile, and independence from electric power. An adult trap would benefit from an effective lure or attractant.
At present the best options are (1) backpack aspirators, (2) sticky ovitraps, and (3) the BG trap. Although current models of battery-powered backpack aspirators may be too expensive for most disease-endemic countries (>US$400), this is the most effective way to quickly collect large numbers of Ae. aegypti across different ages, sexes, and physiological statuses from large numbers of households [28–30]. In Thailand, backpack aspirators collect ~25%–30% of adult Ae. aegypti in a house (T. Scott and L. Harrington, unpublished data), and an individual collector can sample 25–30 households in a normal workday . The development of new, less expensive designs could extend their use into developing country settings. Sticky ovitraps  are an inexpensive method to collect adult Ae. aegypti. The sticky ovitrap consists of a plastic bucket, water, an infusion attractant, a fitted sticky surface, and a large-mesh covering. Collected mosquitoes can be identified and counted quickly; no electricity is needed, and traps can be left unattended for up to seven days. Their limitations are that they target egg laying rather than host-seeking females, and their effectiveness can be influenced by availability of natural oviposition sites. Large-scale validation studies are warranted. The BG trap is an attractant trap that requires electricity and costs US$100–US$300 per unit (http://www.biogents.com/en/index.html), which will limit its use in resource-strapped environments. It appears to be comparable in efficacy to backpack aspiration collections of Ae. aegypti adults .
Quantitative assessment tools.
A critical missing component of Ae. aegypti–borne disease control programs is quantitative assessment tools that can convert surveillance information into decisions. It is widely accepted that local conditions drive variation in Ae. aegypti–borne disease transmission patterns. Except for a few notable exceptions, most dengue prevention programs do not effectively use the entomological and clinical information that they routinely collect to make decisions [34,35]. Addressing this challenge requires: (1) improved, accessible transmission models, (2) evidence-based decision software, and (3) integrated use of geographic information systems (GIS).
There cannot be universal guidelines for preventing Ae. aegypti–borne disease because transmission dynamics vary across local fluctuations in immunity status, characteristics of human/mosquito contact, vector ecology, and climate patterns. Only quantitative models can adequately account for these kinds of key, site-specific variables. The ideal model will be user-friendly and facilitate easy importation of local surveillance data. Models for predicting success of different control strategies and transmission thresholds are under development ; increased functionality, ease of use, and rigorous validation are the principal goals for their improvement. Effective models will allow governments to make better use of limited resources by highlighting strategies and establishing priorities that are most likely to be effective.
Decision support systems (DSS) are overarching programs for synthesizing data into effective decision making. They could include models, GIS coordinates, relevant literature, and a decision tree in easily interpretable format for situation-specific advice . Existing DSS for other vector-borne diseases exemplify the power of this approach . For tsetse control web-based DSS, models, cost analyses, and general information on tsetse are available to assist in the planning and implementation of control operations (http://www.tsetse.org/).
GIS allows spatial display of entomological and epidemiological data and facilitates targeted interventions. Although developing GIS for a city may require an initial investment, benefits to program management and decision making are great. Base maps may exist for some municipalities. GIS software is becoming increasingly user-friendly and adaptable.
We hope this brief discussion will provoke a re-examination of how best to prevent dengue and other arboviruses using Ae. aegypti control. The universal reliance over the last 50 years on source reduction may appear logical, given the vector's domestic habitat, but obviously it is not working in most societies at risk. Dengue is more prevalent now than at any time in history. Malaria prevention, which is based on verifiable mathematical principles first derived nearly 100 years ago, preferentially targets adult mosquitoes, which transmit parasites. We recommend that far more attention should be given to methods directed toward adult rather than immature Ae. aegypti. Several classes of tools were identified that would make this easier. Immediate field testing and continued improvement of those ready for implementation is recommended, including those that may now be on the market. The use of various immature indices in surveillance was considered generally uninformative, and the refinement or development of more accurate indicators of risk and the means of measuring them was emphasized. The only metric that verifies that a program is working is a decrease in disease incidence. Finally, the crucial need for political commitment was repeatedly stressed as among the most important components of a control program, regardless of the methods used. Although participation of those affected is crucial, there has never been a successful program without enlightened, adequately funded, and well-organized leadership.
We thank the following meeting participants for helpful comments: Kathryn Aultman, Roberto Barrera, Barry Beaty, David Chadee, Gregory Devine, Ildefonso Fernandez Salas, Dana Focks, John Grieco, Samantha Hammond, Brian Kay, Vu Sinh Nam, Roger Nasci, Scott Ritchie, and Daniel Wang. Support for this meeting was provided by the Bill & Melinda Gates Foundation, the US Centers for Disease Control and Prevention, and the University of California at Davis.
- 1. Ooi EE,Goh KT,Gubler DJ (2006) Dengue prevention and 35 years of vector control in Singapore. Emerg Infect Dis 12: 887–893.
- 2. Kay B,Nam VS (2005) New strategy against Aedes aegypti in Vietnam. Lancet 365: 613–617.
- 3. Gubler DJ (2002) Epidemic dengue/dengue hemorrhagic fever as a public health, social and economic problem in the 21st century. Trends Microbiol 10: 100–103.
- 4. Heintze C,Garrido MV,Kroeger A (2007) What do community-based dengue control programmes achieve? A systematic review of published evaluations. Trans Royal Soc Trop Med Hyg 101: 317–325.
- 5. (2003) Aedes aegypti density and the risk of dengue virus transmission. In: Takken W,Scott TW, editors. Ecological aspects for application of genetically modified mosquitoes. Dordrecht (The Netherlands): FRONTIS. pp. 187–206. editors.
- 6. Tun-Lin W,Kay BH,Barnes A (1995) Understanding productivity, a key to Aedes aegypti surveillance. Am J Trop Med Hyg 53: 595–601.
- 7. Focks DA,Haile DG,Daniels E,Mount GA (1993) Dynamic life table model for Aedes aegypti (L.) (Diptera Culicidae). Analysis of the literature and model development. J Med Entomol 30: 1003–1017.
- 8. Lloyd-Smith JO,Schreiber SJ,Kopp PE,Getz WM (2005) Superspreading and the effect of individual variation on disease emergence. Nature 438: 355–359.
- 9. Scott TW,Morrison AC,Lorenz LH,Clark GG,Strickman D,et al. (2000) Longitudinal studies of Aedes aegypti (L.) (Diperta: Culicidae) in Thailand and Puerto Rico: Population dynamics. J Med Entomol 37: 77–88.
- 10. Ross R (1911) The prevention of malaria. London: John Murray.
- 11. Macdonald G (1957) The epidemiology and control of malaria. Oxford: Oxford University Press.
- 12. (1997) Surveillance and control of urban dengue vectors. In: Gubler DJ,Kuno G, editors. Dengue and dengue hemorrhagic fever. New York: CAB International. pp. 425–462. editors.
- 13. Mulla MS,Tharvara U,Tawatsin A,Kong-Ngamsuk A,Chompoorsi J (2001) Mosquito burden and impact on the poor: Measures and costs for personal protection in some communities in Thailand. J Am Mosq Control Assoc 17: 153–159.
- 14. Kroeger A,Lenhart A,Ochoa M,Villegas E,Levy M,et al. (2006) Effective control of dengue vectors with curtains and water container covers treated with insecticide in Mexico and Venezuela: Cluster randomised trials. BMJ 332: 1247–1252.
- 15. Hanna JN,Ritchie SA,Richards AR,Taylor CT,Pyke AT,et al. (2006) Multiple outbreaks of dengue serotype 2 in north Queensland, 2003/04. Aust N Z J Pub Health 30: 220–225.
- 16. Williams CR,Ritchie SA,Long SA,Dennison N,Russell RC (2007) Impact of a bifenthrin-treated lethal ovitrap on Aedes aegypti oviposition and mortality in north Queensland, Australia. J Med Entomol 44: 256–262.
- 17. Zeichner BC,Perich MJ (1999) Laboratory testing of a lethal ovitrap for Aedes aegypti. Med Vet Entomol 13: 234–238.
- 18. Grieco JP,Achee NL,Chareonviriyaphap T,Suwonkerd W,Chauhan K,et al. (2007) A new classification system for the actions of IRS chemicals traditionally used for malaria control. PLoS ONE 2: e716. doi:10.1371/journal.pone.0000716.
- 19. Ujihara K,Mori T,Iwasaki T,Sugano M,Shono Y,et al. (2004) Metofluthrin: A potent new synthetic pyrethroid with high vapor activity against mosquitoes. Biosci Biotechnol Biochem 68: 170–174.
- 20. Kawada H,Yen NT,Hoa NT,Sang TM,Dan NV,et al. (2005) Field evaluation of spatial repellency of metofluthrin impregnated plastic strips against mosquitoes in Hai Phong City, Vietnam. Am J Trop Med Hyg 73: 350–353.
- 21. Kawada H,Maekawa Y,Takagi M (2005) Field trial on the spatial repellency of metofluthrin-impregnated plastic strips for mosquitoes in shelters without walls (beruga) in Lombok, Indonesia. J Vect Ecology 30: 181–185.
- 22. Estrada JG,Mulla MS (1986) Evaluation of 2 new insect growth-regulators against mosquitoes in the laboratory. J Am Mosq Control Assoc 2: 57–60.
- 23. Hatakoshi M,Kawada H,Nishida S,Kisida H,Nakayama I (1987) Laboratory evaluation of 2-[1-methyl-2-(4-phenoxyphenoxy)-ethoxy]pyridineagainst larvae of mosquitoes and housefly. Jap J San Zool 38: 271–274.
- 24. Loh PY,Yap HH (1989) Laboratory studies on the efficacy and sublethal effects of an insect growth regulator, pyriproxyfen (S-31183) against Aedes aegypti (Linnaeus). Trop Biomedicine 6: 7–12.
- 25. Itoh T (1994) Utilisation of bloodfed females of Aedes aegypti as a vehicle for the transfer of the insect growth regulator, pyriproxyfen to larval habitats. Trop Med 36: 243–248.
- 26. Satoh T,Tsuda Y,Somboon P,Kawada H,Tagaki M (2003) Difference in the larval susceptibility to pyriproxyfen in nine colonies of six vector mosquito species. Med Entomol Zool 54: 155–160.
- 27. Sihuincha M,Zamora-Perea E,Orellana-Rios W,Stancil J,Lopez-Sifuentes V,et al. (2005) Potential use of pyriproxyfen for control of Aedes aegypti (Diptera:Culicidae) in Iquitos, Peru. J Med Entomol 42: 620–630.
- 28. Clark GG,Seda H,Gubler DJ (1994) Use of the CDC backpack aspirator for surveillance of Aedes aegypti in San Juan, Puerto Rico. J Am Mosq Control Assoc 10: 119–124.
- 29. Scott TW,Clark GG,Lorenz LH,Amerasinghe PH,Reiter P,et al. (1993) Detection of multiple blood feeding in Aedes aegypti (Diptera: Culicidae) during a single gonotrophic cycle using a histologic technique. J Med Entomol 30: 94–99.
- 30. Scott TW,Chow E,Strickman D,Kittayapong P,Wirtz RA,et al. (1993) Bloodfeeding patterns of Aedes aegypti in a rural Thai village. J Med Entomol 30: 922–927.
- 31. Morrison AC,Gray K,Getis A,Estete H,Sihuincha M,et al. (2004) Temporal and geographic patterns of Aedes aegypti (Diptera: Culicidae) production in Iquitos, Peru. J Med Entomol 41: 1123–1142.
- 32. Ritchie SA,Long S,Hart A,Webb CE,Russell RC (2003) An adulticidal sticky ovitrap for sampling container-breeding mosquitoes. J Am Mosq Control Assoc 19: 235–242.
- 33. Williams CR,Long SA,Webb CE,Bitzhenner M,Geier M,et al. (2007) Aedes aegypti population sampling using BG-Sentinel traps in north Queensland Australia: Statistical considerations for trap deployment and sampling strategy. J Med Entomol 44: 345–50.
- 34. Gubler DJ (1989) Aedes aegypti and Aedes aegypti-borne disease control in the 1990s: Top down or bottom up. Am J Trop Med Hyg 40: 571–578.
- 35. Gubler DG (2002) How effectively is epidemiological surveillance used for dengue programme planning and epidemic response. Dengue Bull 26: 96–106.
- 36. Focks DA,Daniels E,Haile DG,Keesling JE (1995) A simulation model of the epidemiology of urban dengue fever: Literature analysis, model development, preliminary validation, and samples of simulation results. Am J Trop Med Hyg 53: 489–506.
- 37. Hemingway J,Beaty BJ,Rowland M,Scott TW,Sharp BL (2006) The Innovative Vector Control Consortium: Improved control of mosquito-borne diseases in and around the home. Trends Parasitol 22: 308–312.
- 38. Vale GA,Torr SJ (2005) User-friendly models of the costs and efficacy of tsetse control: Application to sterilizing and insecticidal techniques. Med Vet Entomol 19: 293–305.