AJTMH Transactions of the Royal Society of Tropical Medicine and Hygiene
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am. J. Trop. Med. Hyg., 77(5), 2007, pp. 897-902
Copyright © 2007 by The American Society of Tropical Medicine and Hygiene

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chansang, C.
Right arrow Articles by Kittayapong, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chansang, C.
Right arrow Articles by Kittayapong, P.
Related Collections
Right arrow Dengue
Right arrow Geographic Information Systems (GIS)
Right arrow Mosquitoes

Application of Mosquito Sampling Count and Geospatial Methods to Improve Dengue Vector Surveillance

Chitti Chansang AND Pattamaporn Kittayapong*
Center for Vectors and Vector-Borne Diseases and Department of Biology, Faculty of Science, Mahidol University, Bangkok, Thailand


ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Dengue hemorrhagic fever is a major public health problem in several countries around the world. Dengue vector surveillance is an important methodology to determine when and where to take the control action. We used a combination of the Global Positioning System (GPS)/Geographic Information System (GIS) technology and the immature sampling count method to improve dengue vector surveillance. Both complete count and sampling count methods were used simultaneously to collect immature dengue vectors in all houses and all containers in one village in eastern Thailand to determine the efficiency of the sampling count technique. A hand-held GPS unit was used to record the location of surveyed houses. Linear regression indicated a high correlation between total immature populations resulting from the complete count and estimates from sampling count of immature stages. The immature survey data and the GPS coordinates of house location were combined into GIS maps showing distribution of immature density and clustering of immature stages and positive containers in the study area. This approach could be used to improve the efficiency and accuracy of dengue vector surveillance for targeting vector control.

Dengue hemorrhagic fever (DHF) is a major public health problem in Thailand1 and many tropical regions of the world.2 Controlling the principal vector, Aedes aegypti (L.), is the only known method to reduce disease incidence. In Asia and the Americas, primary breeding sites of dengue vectors are man-made containers in and around houses. In Thailand, many types of containers such as water jars, ant traps, tires, bath basins, and metal drums serve as Ae. aegypti breeding sites.3 Surveying immature stages of Ae. aegypti is necessary for control plans. Visual larval survey4 is a standard method promoted for Ae. aegypti larval surveillance, but this approach does not provide an accurate estimate of the relationship between Ae. aegypti larval and adult densities,5 and indices from this survey and epidemic risk are hard to define.6 Other sampling approaches involve sampling immature Ae. aegypti using a net in metal drums7 and other containers such as tires.8 The premise condition index9 and pupae index obtained from Ae. aegypti surveys have been found to be the most accurate method for estimating dengue risk.10 The survey of Ae. aegypti immature stages from houses in villages has been reported.11 The complete count method (total number of larvae and pupae in a container) for estimating Ae. aegypti immature stages by sampling 10 houses in each month in 1 year has been studied.12 However, thus far, there has been no study to find the relationship between the complete count and sampling count to evaluate the accuracy of immature sampling.

Geographic Information Systems (GISs) were used for studying Ae. aegypti vectors by creating a map of the study areas13–15 and also for evaluation of a sampling count methodology for rapid assessment of Ae. aegypti infestation levels.16 These methods are likely to be used more and more in Ae. aegypti control programs in the future.17 The main objective of our study was to improve the surveillance and control of Ae. aegypti by combining the GIS technology and the immature sampling methodology to create a spatial density distribution maps and to identify the clusters of immature stages and the main breeding sources that should be the targets for vector control. Therefore, in this study, we surveyed immature stages of Ae. aegypti in all houses and all containers in a Thai village, and Global Positioning System (GPS) coordinates of houses were recorded at the same time. GIS was used to estimate the spatial distribution of Ae. aegypti immature stages and evaluate the relationship between the complete count and the sampling count methods.


MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Study areas and data collection. Village 10 (13°38'18''N, 101°17'32''E) in Hua Samrong Subdistrict, Plaeng Yao District, Chachoengsao Province, Thailand, was selected as the study area because there were high dengue infections from a serologic survey of anti-dengue immunoglobulin IgM and IgG in primary school children conducted by Kittayapong and others (unpublished data) in September 2001. Village 10 is located in a low-lying (~10 m above the sea level) flat area surrounded by rice fields. Housing in the village was not crowded and was interspersed with small orchards and rice fields. There were two basic types of dwellings made from local materials: wood and a combination of wood and cement. The dwellings were divided into four groups separated by rice fields, irrigation canals, and small roads. Water for household use came from subsurface wells, tap water, and rain water. In addition, residents stored water in various types of containers during the year. These containers included ones for human and livestock drinking water.

A survey of Ae. aegypti immature stages was conducted in Village 10. All containers in 151 houses were examined at the beginning of the hot and dry season in March 2002. The number of containers with or without Aedes immature stages was surveyed and recorded following standard WHO procedures.4 The complete count and the sampling count methods of Ae. aegypti immature stages were conducted following the methods of Strickman and Kittayapong.11,12 For the sampling count method, containers were sampled with a very fine round mosquito dip net with the diameter of 24 cm by dipping the net in the water, starting at the top of the container, and continuing to the bottom in a swirling motion. For the complete count method, the containers with the remaining immature stages from the sampling count method were emptied, and all larvae and pupae were collected with a strainer. The total numbers of larvae and pupae from the complete count method were calculated by adding the number of larvae and pupae from the sampling count method in the containers. For small containers, e.g., ant traps and small plastic containers, an immature survey was conducted by the complete count method. Only drinking water containers were carefully examined with a flashlight to count the number of larvae and pupae. The numbers of containers, the number of positive containers, and the number of larvae and pupae from both methods were recorded. The linear regression was used for finding the relationship between the two methods. The number of immature stages from the complete count method was set as the dependent variable, and the number of the immature stages from the sampling count method was set as the independent variable.

The geographic coordinates of each survey house were determined by GPS observations made with a Leica GS5+ (Leica Geosystems, Torrance, CA), which had an accuracy of ±3 m. The GPS unit linked to the pocket PC iPAQH 3850 (Compaq Information Technologies Group, Palo Alto, CA); ArcPad (ESRI, Redlands, CA) software was used to manipulate field data with maximum setting of the position dilution of precision (PDOP) = 2, and the field data were recorded in shape files that were transferred to a personal computer. Arc-View software (ESRI) was used to create a GIS map and to analyze the data obtained.

The density of immature stages in each house from both methods was mapped using GIS. The degree of association between the two methods was analyzed using Pearson correlation.18 Getis and others15 reported an application of the local spatial statistic Gi*(d) to find clustering of Ae. aegypti in Peru. This statistic was used to test whether a particular location i and its surrounding regions constitute a cluster of higher values than average values of a variable (x) of interest18 and is written as


Formula

where s is the sample SD of the x values (larvae, pupae, and positive containers), and wij(d) is equal to 1 if region j is within a distance of d from region i and 0 otherwise. The sum is calculated over all regions, including region i. Also, Wi* = {sum}jwij(d) and S1i* = jw2ij. The terms "location" and "region" refer to the ith house and surrounding areas (at distance < d) of the jth house. In this study, N = 151, and the local spatial statistic Gi*(d) was used under ArcView’s Arctoolbox to identify the houses that showed clustering of Ae. aegypti immature stages and positive containers. The local spatial statistic Gi*(d) was performed by setting conceptualization of spatial relationships = fixed distance band, distance method = Euclidean distance, and distance band or threshold distance = 0, 10, 20, and 30 m. The coordinate system and datum for GPS link with ArcPad, ArcView and the local spatial statistic Gi*(d) were Universal Transverse Mercator and Indian 1975, respectively. The units of measure were in meters. SPSS version 9.0 software package (SPSS, Chicago, IL) was used for statistical analysis.


RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Surveys of Ae. aegypti immatures were conducted in all houses in Village 10 (Figure 1Go). Ae. aegypti was the dominant species (99.88%), and Ae. albopictus represented 0.12% of the 2,374 pupae identified. The House Index (HI), Container Index (CI), and Breteau Index (BI) in Village 10 were 84.77, 31.21, and 327.15, respectively. The containers for the survey were classified into 12 types. For drinking water containers, mainly standard water jars, there were only 1,692 larvae and 172 pupae in 361 containers that were directly counted by flashlight. The prevalence of Aedes immature stages in each container type resulting from the complete count method is shown in Table 1Go. The main breeding sources were water jars of various types, especially standard water jars, and cement bath basins.


Figure 1
View larger version (38K):
[in this window]
[in a new window]

  		    FIGURE 1. Location of the study area in Village 10, Plaeng Yao District, Chachoengsao Province, Thailand.

 

View this table:
[in this window]
[in a new window]

 
TABLE 1
Prevalence of Aedes immature stages in each container types from the complete count method
 
Linear regression was used to find the relationship between the complete count and the sampling count methods. The total numbers of 494 positive containers were surveyed by the complete count, from which 269 containers were surveyed by the sampling count method. The results are shown in Table 2Go. There is a high correlation between these two methods. Linear regression is also used to find the relationship between the complete count and the sampling count methods in each important container type for larvae and pupae. The results are shown in Table 3Go.


View this table:
[in this window]
[in a new window]

 
TABLE 2
Regression models for estimating total larvae and pupae from larvae sampling and pupae sampling
 

View this table:
[in this window]
[in a new window]

 
TABLE 3
Regression models for estimating total larvae and pupae from larvae sampling and pupae sampling for various container types
 
The results from the complete count and the sampling count method including the locations of surveyed houses by GPS were entered into the GIS program to create a map to study the density distribution of Ae. aegypti immature stages between these two methods. The maps showing the number of larvae and pupae in each house are shown in Figures 2Go and 3Go, respectively. These maps show similar results between the complete count and sampling count methods. Pearson correlations between the two methods for larvae and pupae survey were 0.95 (P < 0.001) and 0.94 (P < 0.001), respectively. The map of the number of positive containers in each house is shown in Figure 4Go. The houses that were members of significant clustering of larvae are shown in Figure 2Go. From the complete count method, three houses were members of significant clustering of larvae within 10 m. From the sampling count method, five houses were members of significant clustering of larvae within 10 m and one within 20 m. The houses that were members of significant clustering of pupae are shown in Figure 3Go. From both methods, six houses are members of significant clustering of pupae, three houses within 10 m, and three houses within 20 m. The houses that were members of significant clustering of positive containers are shown in Figure 4Go. Three houses showed significant clustering of positive containers within 10 m.


Figure 2
View larger version (108K):
[in this window]
[in a new window]

  		    FIGURE 2. Maps of the number of larvae and the clustering of larvae based on the number of larvae in each house. A, From the complete count. B, From the sampling count method.

 

Figure 3
View larger version (104K):
[in this window]
[in a new window]

  		    FIGURE 3. Maps of the number of pupae and the clustering of pupae based on the number of pupae in each house. A, From the complete count. B, From the sampling count method.

 

Figure 4
View larger version (101K):
[in this window]
[in a new window]

  		    FIGURE 4. Map of number of positive containers and the clustering of positive containers based on the number of positive container in each house.

 

DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
From the complete count methods, the immature number of Ae. aegypti, especially pupae, in each container type indicated that the main breeding sources were standard water jars and cement bath basins, which accounted for 70.51% of total pupae in the study sites. Nearly the same results were obtained from the study of Strickman and Kittayapong.12 From this study, 72% of the breeding sources were standard and small water jars. Control of immature mosquitoes in these containers should have an impact on > 70% of pupae of Ae. aegypti in the village.

In Trinidad, Focks and Chadee10 showed that adult density could be estimated from pupae that were able to be counted in the containers. In Thailand, it was quite difficult to count all the number of pupae in all containers because there were many types of containers and a high number of immature stages. Kittayapong and Strickman3 reported that there were 10 types of breeding sources, whereas this study had 12 types of breeding sources. In a more practical approach, the sampling count method should be used. The surveyed data were analyzed using linear regression to find the relationship between the two methods. Some positive containers that were surveyed by the complete count method had low water levels. These containers could not be surveyed by the sampling count method using the mosquito dip net. Therefore, the positive container numbers for the sampling count method in Table 3Go were less than those for the complete count in Table 1Go. Results showed that the sampling count method was highly correlated with the complete count method in both the numbers of larvae and pupae. However, it is proposed to take into account the small container that can be more important in other areas and for which only the complete count method can be applied. Our study supported the idea that the sampling count method could be used to accurately estimate the density of immature stages of Ae. aegypti.

The GIS/GPS technology together with Ae. aegypti immature surveys at a village scale were used to create the GIS map and clustering of immatures that could be used for targeting control efforts to be more efficient and more cost effective. GIS and the local spatial statistic Gi*(d) were applied to study the density distribution of Ae. aegypti. Our result showed that this approach had a good potential for future use in the surveillance and control of Ae. aegypti. Maps of the density distribution of immature stages from the two survey methods were created for visual comparison. Results from both methods showed significant Pearson correlation, and the GIS maps created were quite similar. Both methods could characterize the densities distribution of Ae. aegypti mosquitoes in the village. Clustering of Ae. aegypti immature stages and positive containers in our study indicated that both surveillance methods yielded similar results, except that the sampling count method showed higher degree of clustering of larvae than the complete count method. Clustering of larvae, pupae, and positive containers was detected within the same house up to 20 m. One house showed clustering of larvae and pupae and one house showed clustering of larvae and positive containers. The groups of houses that had high Aedes densities could be characterized and be the target for prioritizing activities of the vector control plan. The Gi*(d) statistic is in part biased by the heterogeneous density of houses in the four areas. Our results indicate spatial clustering of Ae. aegypti immature stages and positive containers, which were different from the study in Peru where no clustering of larvae or pupae was found beyond individual households and only limited clustering of adult Ae. aegypti occurred.15 The GIS maps of the density distribution and Pearson correlation support the use of sampling count method for Ae. aegypti survey, which could give nearly the same information as that from the complete count.

The visual larval survey method was commonly used for Ae. aegypti survey. The number of containers with or without Ae. aegypti immature stages in houses was recorded, and the indices such as HI, CI, and BI were calculated to show Aedes indices in the study area. From this study, GIS/GPS could be used to analyze and show Ae. aegypti immature stages and positive containers in the houses. Our result showed that there were variations in Ae. aegypti density from house to house. Individual houses and a group of houses that had high Ae. aegypti density could be characterized. Therefore, individual households were an appropriate spatial unit for entomologic surveys.15 In addition, maps resulting from GIS and the local spatial statistic Gi*(d) could be used for developing an effective Aedes control plan by monitoring and controlling Ae. aegypti in the targeted groups of houses with clustering of immature stages and positive containers.

Received October 26, 2005. Accepted for publication April 7, 2007.

Acknowledgments: The authors thank Dr. John D. Edman for help in reviewing and editing the final manuscript; Dr. Laura C. Harrington for critical review and helpful comments; Somboon Srimarat, Tanong Aimmak, Sumas Jantamas, Suwanna Austthapornrungroj, Damrongrith Vinij, and Uruyakorn Chansang for field assistance; and residents of surveyed houses, local public health authorities, and local administrative authorities in Hua Sam Rong Subdistrict, Plaeng Yao District for cooperation.

Financial Support: This study was supported by the Thailand Research Fund (RGJ/PHD/0051/2544), the Mahidol University Research Grant (SCBI-47-T-217), and the UNICEF/UNDP/World Bank/WHO Special Programme for Tropical Diseases Research and Training (TDR/RCS/A00786).

* Address correspondence to Pattamaporn Kittayapong, Center for Vectors and Vector-Borne Diseases, Faculty of Science, Mahidol University, Rama 6 Road, Bangkok 10400, Thailand. E-mail: grpkt{at}mahidol.ac.th Back

Authors’ addresses: Pattamaporn Kittayapong and Chitti Chansang, Center for Vectors and Vector-Borne Diseases, Faculty of Science, Mahidol University, Rama 6 Road, Bangkok 10400, Thailand, Telephone: 662-201-5935, Fax: 662-201-5923, E-mail: grpkt{at}mahidol.ac.th.

Reprint requests: Pattamaporn Kittayapong, Center for Vectors and Vector-Borne Diseases, Faculty of Science, Mahidol University, Rama 6 Road, Bangkok 10400, Thailand. E-mail: grpkt{at}mahidol.ac.th.


REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Kantachuvessiri A, 2002. Dengue hemorrhagic fever in Thai society. Southeast Asian J Trop Med Pub Hlth 33: 56–62.[Medline]
  2. World Health Organization, 1995. Vector Control for Malaria and Other Mosquito-Borne Diseases. Geneva: World Health Organization.
  3. Kittayapong P, Strickman D, 1993. Distribution of container-inhabiting Aedes larvae (Diptera: Culicidae) at a dengue focus in Thailand. J Med Entomol 30: 601–606.[Web of Science][Medline]
  4. World Health Organization, 1972. An international system for the surveillance of vectors. Wkly Epidemiol Rec 47: 73–80.
  5. Tun-Lin W, Kay BH, Barnes A, Forsyth S, 1996. Critical examination of Aedes aegypti indices: correlations with abundance. Am J Trop Med Hyg 54: 543–547.[Abstract/Free Full Text]
  6. Reiter P, Gubler DJ, 1997. Surveillance and control of urban dengue vectors. Gubler DJ, Kuno G eds. Dengue and Dengue Hemorrhagic Fever, New York: CAB, 425–462.
  7. Tun-Lin W, Kay BH, Burkot TR, 1994. Quantitative sampling of immature Aedes aegypti in metal drums using sweep net and dipping methods. J Am Mosq Cont Assoc 10: 390–396.[Web of Science][Medline]
  8. Zhen TM, Kay BH, 1993. Comparison of sampling efficacy of sweeping and dipping for Aedes aegypti larvae in tires. J Am Mosq Cont Assoc 9: 316–320.[Web of Science][Medline]
  9. Tun-Lin W, Kay BH, Barnes A, 1995. The premise condition index, a tool for streamlining surveys of Aedes aegypti. Am J Trop Med Hyg 53: 591–594.[Abstract/Free Full Text]
  10. Focks DA, Chadee DD, 1997. Pupal survey: An epidemiologically significant surveillance method for Aedes aegypti, an example using data from Trinidad. Am J Trop Med Hyg 56: 159–167.[Abstract/Free Full Text]
  11. Strickman D, Kittayapong P, 2002. Dengue and its vectors in Thailand: Introduction to the study and seasonal distribution of Aedes larvae. Am J Trop Med Hyg 67: 247–259.[Abstract]
  12. Strickman D, Kittayapong P, 2003. Dengue and its vectors in Thailand: calculated transmission risk from total pupal counts of Aedes aegypti and association of wing-length measurements with aspects of the larval habitat. Am J Trop Med Hyg 68: 209–217.[Abstract/Free Full Text]
  13. Sithiprasasna R, Linthicum KJ, Lerdthusnee K, Brewer TG, 1997. Use of Geographic Information System to study the epidemiology of dengue haemorrhagic fever in Thailand. Dengue Bull 21: 68–72.
  14. Ali M, Wagatsuma Y, Emch M, Breiman RF, 2003. Use of a Geographic Information System for defining spatial risk for dengue transmission in Bangladesh: role for Aedes albopictus in an urban outbreak. Am J Trop Med Hyg 69: 634–640.[Abstract/Free Full Text]
  15. Getis A, Morrison AC, Grey K, Scott TW, 2003. Characteristics of the spatial pattern of the dengue vector, Aedes aegypti, in Iquitos, Peru. Am J Trop Med Hyg 69: 494–505.[Abstract/Free Full Text]
  16. Morrison AC, Astete H, Chapilliquen F, Ramirez-Prada G, Gloria D, Getis A, Gray K, Scott TW, 2004. Evaluation of sampling methodology for rapid assessment of Aedes aegypti infestation levels in Iquitos, Peru. J Med Entomol 41: 502–510.[Web of Science][Medline]
  17. Nelson MJ, 1994. The role of sampling in vector control. Am J Trop Med Hyg 50: 145–150.[Abstract/Free Full Text]
  18. Rogerson PA, 2001. Statistical Methods for Geography. London: SAGE Publications.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chansang, C.
Right arrow Articles by Kittayapong, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chansang, C.
Right arrow Articles by Kittayapong, P.
Related Collections
Right arrow Dengue
Right arrow Geographic Information Systems (GIS)
Right arrow Mosquitoes

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS