Chikungunya

Background & objectives: Dengue virus activity has never been reported in the state of Goa. The present study was carried out to document a multilevel geographic distribution, prevalence and preliminary analysis of risk factors for the invasions of Aedes aegypti in Goa.

Methods: A geographic information system (GIS) based Ae. aegypti surveys were conducted in dry (April 2002) and wet (July 2002) seasons in the rural and urban settlements. The random walk method was used for household coverage. The non-residential area visits included ancillaries of roadways, railways, air-and seaports. Simultaneous adult mosquito collections and one-larva per container technique were adopted.

Results: The Ae. aegypti larval and adult prevalence was noted in all the four urban areas in both dry (Density index (DI)=3 to 6) and wet (DI=5 to 7) seasons and only one out of 3 villages showed Ae aegypti presence in wet season (DI=5 to 7). In the residential areas, hutments showed higher relative prevalence indices (Breteau index, BI=100; container index, CI=11.95; adult house index, AHI=13.33) followed by close set cement houses (BI=44.1; CI=12.0; AHI=11.24). Ae aegypti relative prevalence indices were also more for households with pets (BI=85.11; CI=12.5; AHI= 42.85); those with tap had higher risk (larval house index, LHI=32.03; relative risk, RR>2, n=256). Plastic drum was the most preferred breeding place (χ^sup 2^ = 19.81; P

Interpretation & conclusion: Established Ae aegypti prevalence in the urban settlements during dry and wet seasons and its scattered distribution in a rural settlement spell risk of dengue infection at macro-level. In the residential areas nature and types of the households, tap water supply and storage and communities' attitude and practices contribute to sustained meso-level risk of Ae aegypti prevalence dependant DEN. The non-residential areas offer transient meso-level risk as Ae aegypti prevalence was seasonally unstable and monsoon dependent. Risk at micro-level was due to the preferred larval habitats of Ae aegypti breeding viz., residential plastic-ware and tyres, and transport tyres in non-residential areas.

Key words Aedes aegypti * dengue * geographic information system * prevalence indices

Among Aedes aegypti (Diptera: Culicidae) borne viral diseases in India, dengue (Flaviridae: Fhvivirus) viruses (DEN) represent a case of continued re-emergence in the settlements across rural urban continuum; yellow fever (YF: Flavivirus) has a potential for emergence and the chikungunya (CHIK: Togaviridae: Alphavirus) has re-emerged about 18 yr after Barsi episode1-3. Risk factors for emergence of these diseases are continued urbanization, industrialization, and transport development, all of which contribute to the maintenance and spread of principal vector Ae. aegypti mosquito4. While the urban change and industrial development characterize the post independent India, rapid transport development that included road, rail, sea and airways changed ecology of the western coastal region considerably. DEN virus activity was reported in the West Coast at Kolencherry, Ernakulum district [National Institute of Virology (NIV), unpublished data 1983 & 1985], Thrissur (Trichur)5 in Kerala; Mangalore in Karnataka, Gujarat state, and Mumbai in Maharashtra6-8 but not yet in the state of Goa (NIV unpublished data and Govt. of Goa personal communication).

Until late 1960s the distribution of Ae. aegypti along the West Coast i.e., south of Mumbai was scant9. In 1971-1973 survey, restricted focal distribution in tires was noted in 16 of the 22 coastal towns of Maharashtra along the National Highway (NH)17 indicating recent spread10. Ae. aegypti distribution was recorded at Mapusa, Goa State during the vector studies on Japanese encephalitis11. However, no comparable prevalence data are available to enable risk assessments of Ae. aegypti spread adequately. Essential data on the knowledge, attitude and practices of the community to facilitate community participation in vector control and prevention of dengue are also not available for the country12.

The present study was undertaken to document the multilevel (viz., macro scale or settlement level, meso scale or ward premises level and micro scale or container habitat level) distribution and prevalence of Ae. aegypti in the state of Goa based on replicate surveys in dry and wet seasons. In addition, the role of people's perceptions, personal protection practices, housing types, house construction, water supply, differences of site and purpose of water storage on Ae. aegypti prevalence was also analyzed.

Material & Methods

Study area: Goa (14°53'54''-15°45'00'' N; 73°40' 33''-74°20'13'' E) had 1,343,998 population in 2001 showing 14.89 per cent decennial growth since 1991 which is lower than the current 21.34 per cent national growth. Its 3,702 km^sup 2^ area is bounded by Maharashtra state in the north, Karnataka in the east and south and the Arabian sea in the west. Terrain is hilly and spurs of the Western Ghats reach up to the seaboard. This region is linked with the west by a major seaport at Marmugao. By 1971 the National Highway (NH) 17 connected Goa with Karnataka and Maharashtra states; and by mid 1990s the Konkan railway has provided connectivity to the entire West Coast 13,14.

Site selection: Surveys were conducted in residential wards, households of merchandise areas, slums etc., representing the state's society and economy. The localities were selected so as to cover >1 house type or construction. Among the non-residential areas, a major seaport, an airport, automobile workshops, private jettys, railway associated areas (waiting rooms, restaurants etc.) and an industrial area were visited.

Study design and collection methods: A modular geographic information system (GlS) chart was modified to cover settlement formations, demographic, residential (housing types, house construction etc.), non-residential and environmental components and adopted for this study15-17. In residential areas, households with single kitchen were considered as premises whereas in non-residential areas workshop, depot, industry, airport, railway station, seaport etc., were considered as premises.

Mosquito surveys were conducted once in dry season (April-May 2002) and once during monsoon (July-August 2002) using aspirators for adults and one larva/pupa per container surveys simultaneously10. Households were selected by random walk method. Schedule-A (SA) was used to record the residential locality-wise data and Schedule-B (SB) for household data; and group data for the selected non-residential areas (SLS). Data recorded in SA included: (i) the housing types (e.g., farmsteads, homesteads, cement houses or huts etc.); (ii) arrangement of the houses (row, close-set, spaced or scattered houses); (iii) peoples perceptions of the presence of mosquito activity (day and /or night biting); and (iv) personal protection practices (repellents/ insecticides/bed-nets/ others). The SB was used to record (i) nature houses (residences/ shop houses/shops/mixed dwelling for men and live stock etc.); (U) the source of water (tap/ bore well/well/river/others) and; (Hi) types, purpose [potable-(p)=drinking and cooking nonpotable-(np)=washing and bathing / cooking / mixed purposes(mp)] and sites (indoor-ID/outdoor-OD) of water storage containers, which serve as larval habitats; adult Ae. aegypti positivity and the community efforts on vector control7,12. The collected specimens were identified following the key of Barruad18.

Container type-wise classification of larval habitats was as follows: iron drums (Id), plastic drums (Pd), plastic containers (Pc), metal potable (Mp), metal-nonpotable (Mnp), earthen-ware potable (Ep), earthenware-nonpotable (Enp), earthenware-partly buried (EPb), cement tanks (CT), tyres (T), miscellaneous (Misc.); number of containers with water (N); and the houses (H)7. Containers with water positive for mosquitoes are given as N+.

Statistical analyses: The relative prevalence indices of Ae. aegypti used were as follows (i) Breteau index (BI), denoted the number of containers with larvae per 100 houses visited; (ii) adult house index (AHI) and larval house index (LHI) denoted the per cent houses with the adults and larvae respectively; and (iii) the container index (CI) denoted the per cent containers with larvae. In addition, the concept of density index (DI) is used to facilitate mapping of Ae. aegypti prevalence10,19. Averages per household (x)or household-frequencies (F) were used to elucidate the variations in water storage practices at the settlement level.

2×2 χ^sup 2^ tests were performed to analyze differences in proportions, and assess the relative risk (RR) using STATCALC program of EPIINFO version (6.04)20. These tests compared the preferences of Ae. aegypti either for household attribute or a container characteristic as against the contrasting attributes/ characteristics following Southwood21.

Results

Prevalence at settlement level: Present study covered four towns and two villages in dry and wet seasons with 123 N+ for immature mosquitoes in 1803 N (Cl= 11.26%; n=447 residence visits). Four towns and one village were found positive for Ae. aegypti. In the dry season, larval prevalence was maximum in Marmugao and minimum in Panaji. The relative prevalence indices increased in the wet season with a maximum in Marmugao and a minimum in Madgao. Adult collections were maximal in Vasco and minimal in Madgao during dry season and in wet season it was maximal in Vasco and minimal in Marmugao (Table I).

Prevalence at household/premise level: 35 N+ for Ae. aegypti in 934 N were present in 234 H (BI-14.95; CI= 3.74; AHI= 1.70) during the dry season and 88 N+out of 869 N in 213 H(BI=41.31; CI=IO.13; AHI= 10.78) during wet season. Among the N+, Ae. aegypti was predominant during the both surveys (dry-71.43%; wet-57.14%) and this was followed by Ae. albopictus (29.36%) during the wet season only. Other mosquito species encountered were Culex quinquefasciatus (dry-20.4%; wet-9.74%), Cx. trilaeniorhynchus(dry-4.\%; wet-1.3%) and Anopheles spp (dry-4.1%; wet-0.65%); whereas Ae. vittatus (0.65%), Armigeres subalbatus (1.95%), Cx. (Lutzia) spp. (2.6%) and Toxorhynchites spp. (1.9%) only during the wet season. Some adults emerging from field caught pupae included Ar. aurolineatus and Cx. (Lutz i a) fus c anus.

Prevalence in non-residential areas: Exclusive non-residential premises chosen under this category included the areas of transport, its ancillaries, and some industrial areas. Smaller establishments which were inseparable physically (such as restaurants, shops, shop houses, etc.) were visited and analyzed along with the H. These areas contributed little to Ae. aegypti breeding during dry season. During wet season a total of 13 premises were visited and 13.75 man hours spent with 30.76 per cent CI (N = 247; BI=135.75;CI=7.69;AHI=28.54) (Table II).

Prevalence Aj/ housing types, house-construction, and by water supply: For majority of collections during dry season the SLS proforma was used. From grouped data thus gathered, scoring the H by the house type, by construction or household frequencies of water container types was difficult. During wet season data were scored using both the locality-wise (SA) and household schedules (SB). This facilitated summarizing data by settlements, localities and by household nature, household construction and with source of water.

Of 213 H surveyed exclusive residences were predominant (8 1.69%), followed by H with live pets (3.29%), shop houses (3.29%), and shops (11.74%) respectively. Ae. aegypli prevalence by housing types was not significantly different from those of exclusive human residences. However,/^, aegypli prevalence indices were more for households with pets(BI=85.11; CI= 12.5; AHI= 42.85) (Table III).

The prevalence of Ae. aegypli in flats and homesteads was lower than huts and close set cement constructions. While the overall larval containers were not higher in the huts (χ^sup 2^ =0.01, P>0.05); RR^sub households^ or RR of N examined did not differ from 1 when compared with the close set cement houses, which constituted maximum of households searched (Table IV).

Majority of H surveyed had tap water as exclusive source of water and LHI for Ae. aegypti breeding was 32.03 (82 / 256); nine had tap and well and seven had tap and bore-wells. No water supply was recorded in shops and in open areas where ownership of those premises could not be ascertained (Table V). The analysis showed higher RR of Ae. aegypti in the premises with tap as exclusive when compared with the houses with other sources (χ^sup 2^(for difference) P> 0.05 and RR>2).

Peoples 'perception and protection practices vis-a-vis Aedes spp. breeding: In a total of 81 (43.32%, n=187) households, people perceived day biting mosquitoes. This included exclusive perception of day biting mosquitoes in 2 households (1.07%), night + day biting in 79 (42.25%); among the rest 90 (48. ] 3%) had exclusive night biting, 4 (2.14%) felt no problem of mosquito bites and 12 were shops where the response was either uncertain or indifferent.

The proportionate distribution of households with people's perception of day biting mosquitoes differed non significantly from those with actual presence of Aedes (Stegomyia) spp (42.25%) breeding (χ^sup 2^= 0.02; P>0.05) but significantly from the households with Ae. aegypti (28.88%) breeding (χ^sup 2^= 8.45; P

Among the personal protection practices the repellents (mats and coil) (75.93%) were used in maximum households followed by none of the methods (9.60%), fan (8.02%), insecticide (7.49%), smoke (5.53%) respectively.

Prevalence of Ae. aegypti at container habitats: In general available outdoors containers were more than those available indoors during the wet season. These containers included rain filled and sundry were contributed by 230 plastic (44.92 %), 111 other artificial containers (21.68%), and 3 1 natural containers such as plant leaf axils, tree holes etc. (6.05%); They contributed to a total of 29 (CI=12.61%), 8(CI=7.21%) and 7 (CI=22.5%) mosquito breeding places respectively; these in turn included 29(CI=12.61%), 7(CI=6.31%), and 4(CI=12.9%) Ae.aegypti breeding places. Outdoor containers positive for Ae. aegypti were more than indoors positive containers during the wet season and the opposite was true in the dry season. Since none of the containers with potable water contributed to Ae.aegypti breeding, further analyses included only the containers with nonpotable water (Table VI).

In all, nine types of containers contributed to Ae. aegypti breeding. The household frequency for earthenware, cement tanks and natural containers were

Plastic drum ranked, as the most preferred breeding place among domestic containers and rubber tyres among sundry/rain filled container when compared with the other containers (Table VII). In the non residential premises three Aede.s spp.were encountered in different containers. In these areas the outdoor breeding were significantly higher than the indoor breeding (χ^sup 2^ = l 1.86; P

Discussion

This study records an increase in knowledge due to the presence of Ae. aegypti in six settlements, in addition to the sole earlier record at Mapuca". From the present ecological data multilevel epidemiological risk is being assessed using tools for spatial analysis22.

On the macro scale (i.e., rural-urban continuum of settlements) the detection of Ae. aegypli indicates a risk at nominal level. Its establishment in the rural and urban areas showing DI >3 is suggestive of the risk at interval level for DEN and YF viruses. These DI were further raised during wet season consistently19.

Tourism, transportation and trade links also augment the risk of dengue in the state. Panaji, the state's capital is linked by road with Mumbai (360 km), Pune (450 km) and Bangalore (450 km) all of which have known DEN virus activity8. Madgao, Vasco and Marmugao, the trade and commercial centers being the hubs of people's activity potentiate the formation of epicenters for DEN outbreaks. Invasion of Ae. aegypli is of significant nodal risk for dengue spread and rational risk is due to the inter-settlement linkages ".

Ae. aegypti is absent at Verna, where industrial development is underway by peri-sylvan clearance. While Ae. albopictus, a potential mosquito vector could invade these areas during the monsoon, presence of Ae. vittatus here could pose a risk of DEN transmission albeit at a low level24,23.

On the meso level both the residential and nonresidential biotopes contribute to Ae. aegypti breeding thus help formation of disease nidi (=niche) at community level23. Saliently, Ae. aegypti prevalence depends on the water storage practices in the residential areas of Panaji, Madgao, Marmugao and Vasco da Gama. Tap water supply among drinking water sources is known to be a risk factor in India26. Presently in Goa too the households with tap water supply predominate with increased risk of Ae. aegypti breeding.

The household types and/or nature of construction among the visited localities are heterogeneous. So, no single factor could account for the risk of Ae. aegypti prevalence in these wards. Exclusive human dwellings and close-set households with cement constructions had prolific Ae. aegypti breeding. The residences with livestock or pets were too few to make a generalized comment. Potential places of community interaction such as exclusive shops and shop-houses had poor water storage; so they supported Ae. aegypti breeding poorly. The restaurants or the cold drink shops or the vulcanizers, which occasionally interspersed the residentia, also contributed to Ae. aegypti breeding in the community. These factors could impose variations in Ae. aegypti source reduction strategy.

The bungalows (homesteads) with scattered distribution in our collections offered poorly to water storage dependent Ae. aegypti breeding, but the premises condition with rain filled sundry supported the breeding of Ae. albopictus, Armigeres spp., Ciilex quinquefasciatus etc. In addition, the phytotelmates or the plant containers too contribute to the species like Ae. albopictus as was also noted in the recently introduced countries23.

The households at Caranzale (Panaji) had closeset housing with linear formation, but have open yards or patios akin to bungalows. The junk water containers there in were usually tyres. Exposure of every house to open environment with green cover enhances risk of Ae. albopictus breeding. Tyres in shopping areas or the public transport systems with the varing tyre dump sizes (10Os or 1000s) require added control inputs.

Ae. aegypti breeding was found poorly in the multistoreyed flats/apartments, but predominantly in ID containers similar to that in Pune27. On the other hand, huts and cement/ brick constructions of middle or low economic areas contributed to copious Ae. aegypti breeding. Present assessments of risk based on water storage were similar to that in the Maharashtra and Gujarat coasts7,10.

Linking the community awareness of the day-biting mosquito profile, personal protection practices and the distribution of Ae. aegypli in the residential biotopes is attempted. Such analyses were considered important to prioritize community needs and help community-based interventions. We attempted documentation of responses to routine questions while seeking house entry. Observed frequencies of people's perception were similar to Ae. aegypli's relative prevalence. But the mosquito prevention practices help to reduce the nocturnal mosquito bites. These results are akin to the studies conducted in Trinidad and Tobago12. The residents in Goa used fans owing to warm humid climate, which was curiously felt to be preventive against day-biting mosquitoes as well. The potential use of such data in dengue prevention and/or disease surveillance was not documented hitherto. These clues could be utilized for optimal sampling for the vector populations or their susceptibility to the potential control agents.

At micro level 1 8 different potential habitat-sites and purpose were scored to assess their contribution to N+ of Aedes spp. in the residential areas. These categories included confounding variables N (OD/ID), and Mp, Mnp, Ep & Enp. Such recording facilitated exclusion of Np in further analyses. As the study progressed, four additional categories accounted for the profuse use of plastic ware viz., Pd ^sub (OD&ID)^ and Pc ^sub (OD&ID)^. Their inclusion improved the representation of the available habitats (N) in the coastal region significantly over the previous studies7,10,26. The characteristics of N+for ,4e. aegypti, N-distributions, household frequencies and their ranks adopted in this study are consistent with current trends12,28 and serve as precursors of the Ae. aegypli productivity studies and/or their contribution in the transovarial transmission of DEN viruses, an aspect gaining importance currently29.

The Pd, Pc and T are important N+ of the residential biotopes whereas only T at the nonresidential ones10. In the residential areas the RR>3 in N+^sub pd/[Pc/T^. Earlier in the Maharashtra coast10 N ^sub (Pd & pc)^ per cent and N+ ^sub (Pd & Pc)^ 1) in significant numbers during the wet season. This partly owed to the admixture of tyre shops and the neglected tyre in the ambit of homesteads. In addition to Ae. aegypli, Ae. albopictus breeding was common in the areas with vegetation cover. Otherwise Ae. viltatus colonized these habitats. This feature can be expected in the biogeographically similar areas.

The attempts to locate breeding of DEN vectors in the non-residential habitats during dry season were mostly futile except for N^sub np^ containers at tyre vulcanizers/restaurants. On the other hand in wet season N+owed more to the rain-filled containers, more frequently infested by the competing Aedes (Ae. albopictiifi, Ae. vittatus), Toxorhynchiies spp. Armigerex spp., Culex (Lulzia) spp., etc. The frequency of their habitats depends on species specific relative frequency in those areas28. The predacious larvae of Toxorhynchiies spp., Armigeres spp., Culex (Lutzia) spp., known to regulate populations and thus seem to exert pressure on Ae. aegypti at the Kadamba tyre dumps.

At the air-and seaports the larval frequency of Ae. aegypti breeding was low. Opportunities for Ae. albopictiis augmentation appear to be considerably high since areas were developed by the temporary vegetation clearance of the undulating laterite based landscape. This contrasted from that in the ports in Mumbai where N+for/4e. aegypti was considerable during both the dry and wet seasons10. On the other hand at Shewa port (JNPT), Raigadh district, N+was also found to be aggregated in the junk yard and stores (NIV unpublished data, 1995).

In all these non-residential and transport associated areas the rain filled containers showed limited focal build up of Ae. aegypti. Importance of tyres in the productivity of Ae. aegypti appears to be limited at the State Road Corporation (Kadamba) owing to its extra domestic locations and high frequency of the competing species present. Its prexence spells the risk of dengue as well as potential risk of vector spread. Hence, the transport areas demand massive control operations weather the diseases borne by Ae. aegypti are prevalent or not in that area.

In conclusion, the usage of a modular GIS format for the domiciliary vector distribution and prevalence studies has magnified the scope of collection, storage, retrival, analysis and interpretation of data on multiple scales. Databases generated by such structured sampling would hopefully be predictive in diverse geographic, demographic, and ecological settings for Ae. aegypti and diseases borne by it. In addition, increase in the tourism aided by international air travel could enhance vector spread30. Continued vigil on progressive environmental degradation31 and the increasing risk of DEN viral influx in Goa would be necessary in the wake of raised density indices of Ae. aegypti.

Acknowledgment

We thank Drs J.R. Bhatt and Ashok Bhatia, Ministry of Environment and Forests, Government of India for financial assistance for the Coastal Aedes Project; Dr R. Tamba (Epidemiologist), Dr M.B. Kaliwal (Entomologist), Directorate of Health services. Government of Goa State, Panaji, for useful discussions and local facilities; Dr P.D. Ajgaonkar, Chief Medical Officer, Goa Port-Trust, Marmugao and the Airport authority, Dabolim for permission to visit the sea and air ports; field staff of the project and the Division of Medical Entomology of NIV for assistance.

References

1. Pandit CG. India and the yellow fever problem. Indian J Med Res 1971; 59 : 1523-47.

2. Dhanda V. Recent trends in the spread of Aecles aegypti and dengue fever in rural areas of India and its significance. In: Ramachandran PK, Sukuinaran D, Rao SS, editors, Entomology for defense services. Proceedings of a symposium, 12-14 September 1990. Gwalior: Defense Research and Development Establishment; 1990 p. 52-7.

3. Mourya DT, Thakare JP, Gokhale MD, Powers AM, Hundekar SL, Jayakumar PC, et al. Isolation of Chikungunya virus from Aedes aegypti mosquitoes collected in the town of Yawat, Pune district, Maharashtra State, India. Ada Virol (Praha)2001; 45 : 305-9.

4. Gubler DJ, Clark GG. Community-based integrated control of Aedes aegypti: a brief overview of current programs. Am J Trop Med 1994; 50 (Suppl): 50-60.

5. Sreenivasan MA,. Rodrigues FM, Venkateshan CN, Panikar CJ. Isolation of dengue virus from Trichur District (Kerala state). Indian J Med Res 1979; 69 : 538-41.

6. Padbidri VS, Adhikari P, Thakare JP, Ilkal MA, Joshi GD, Pareira P, et al. The 1993 epidemic of dengue fever in Mangalore, Karnataka state, India. Southeast Asian J Trop Med Public Health 1996; 26 : 699-704.

7. Mahadev PVM, Kollali VV, Rawal ML, Pujara PK, Shaikh BH, Ilkal MA, et al. Dengue in Gujarat state, India during 1988 & 1989. Indian J Med Res 1993; 97 : 135-44.

8. Thakare JP, Walhekar B, Banerjee K. Hemorrhagic manifestations and encephalopathy in cases of dengue in India. Southeast Asian J Trop Med Public Health 1996; 27 : 471-5.

9. Rao '1'R. Distribution, density and seasonal prevalence of Aedes aegypti in the Indian subcontinent and south-east Asia. Bull World Health Organ 1967; 36 : 547-51.

10. Mahadev PVM, Dhanda V, Shetty PS. Aedes aegypti (L.) in Maharashtra state - distribution and larval habitats. Indian J Med Res 1978; 67 : 562-80.

11.Kulkarni SM, Naik PS. Breeding habitats of mosquitoes in Goa. Indian J Malarial 1989; 26 : 41-4.

12. Rosenbaum J, Nathan MB, Ragoonanansingh R, Rawlins S, Gayic C, Chadce DD, et al- Community participation in dengue prevention and control: a survey of knowledge, attitude and practice in Trinidad and Tobago. Am J Trop MedHyg 1995; 53 : 111-7.

13. Singh RL. India-? regional geography, Silver Jubilee publication. Varanasi, India: National Geographical Society of India; 1971 p. 1-992.

\4. Hand book of tourism - Goa. Panaji: Government of Goa; 1999 p. 1-8.

15. Bavia ME, Hale LF, Malone JB, Barraud DH, Shane SM. Geographic information systems and the environmental risk of Schistosomiasis in Bahia, Brazil. Am J Trop Med Hyg 1999; 60 : 566-72.

16.Kohn M. Occurrence of Aedes aegypti (L.) and Culex quinquefasciatus Say (Diptera, Culicidae) in houses of different construction in Phnom Penh, Kampuchea. Folia Parasitol (Praha) 1991; 38 : 75-8.

17. Beck LR, Lobitz BM, Wood BL. Remote sensing and human health: New sensors and new opportunities. Emerg Infect Dis 2000; 6 : 2.17-27.

18.Barraud PJ. Fauna of British India including Burma and Ceylon: Diptera : Family Culicidae, Tribes Megarhinini and Culicini, vol. 5. London: Taylor & Francis;1934 p. 1-452.

19. Bang YH, Down DN, Onwubiko AO. Prevalence of larvae of potential yellow fever vectors in domestic water containers in south-east Nigeria. Bull World Health Organ 1981; 59 : 107-14.

20. Dean J, Dean A, Coulombier D, Brendcl KA, Smith DC, Burton A, et al. EPI Info version 6.04: A word processing data base and statistics programme for epidemiology on micro computers. Public Domain Software for Epidemiology and Disease surveillance. Atlanta, Georgia USA: Centre for Diseases Control, Epidemiology Program Office, MSG 34, 2001.

21.Southwood TRE. Habitat the template for ecological strategies. J Anim Ecol 1977; 46 : 337-65.

22. Kitron U. Landscape ecology and epidemiology of vectorborne diseases: Tools for spatial analysis. J Med Entomol 1998; 35 : 435-45.

23. Lounibos LP. Invasions by insect vectors of human disease. Ann Rev Entomol 2002; 47 : 233-66.

24.Mavale MS, Ilkal MA, Dhanda V. Experimental studies on the susceptibility of Aedes vittatus to dengue viruses. Acta viVo/ (Praha) 1992; 36 : 412-6.

25.Mahadev PVM, Prasad SR, Ilkal MA, Mavale MS, Bedekar SS, Bancrjee K. Activity of dengue 2 virus and prevalence of Aedes aegypti in the Chirimiri colliery area, Madhya Pradesh, India. Southeast Asian J Trop Med Public Health 1997; 28 : 126-37.

26. Mahadev PVM, Gokhale MD. Water storage and Aedes aegypti (Diptera, Culicidae) breeding in multi-storeyed flats: A case study in Pune, Maharashtra, India. Entomon 1998; 23 : 173-83.

27. Nathan MB, Knudsen AB. Aedes aegypti infestation characteristics in several Caribbean countries and implications for integrated community -based control. J Am Mosq Control Assoc 1991; 7 : 400-4.

28. Joshi V, Mourya DT, Sharma RC. Persistence of dengue-3 virus through transovarial transmission passage in successive generations of Aedes aegypti mosquitoes. Am J Trop Med Hyg 2002; 67 : 158-61.

29. Eritja R, da Cunha Ramos H, Aranda C. Aircraft-mediated mosquito transport: new direct evidence. J Am Mosq Control Assoc 2000; 16 : 339.

30. Nag Choudhury BD. Introduction to environmental management. In: Environmental science series. New Delhi: Interprint; 1983 p. 1-156.

P.V.M. Mahadev, RV. Fulmali & A.C. Mishra

National Institute of Virology, Pune, India

Received September 1, 2003

Reprint requests: Dr l'.V.M. Mahadev, Deputy Director, National Institute of Virology (ICMR) Pashan Sus Road, Pune 411021, India

e-mail: mahadevpvm@yahoo.co.in

Copyright Indian Council of Medical Research Sep 2004
Provided by ProQuest Information and Learning Company. All rights Reserved