Dengue viruses are responsible for over 100 million infections a year worldwide and are a public health concern in Bangladesh. Although risk of transmission is high, data on vector population characteristics are scanty in Bangladesh; therefore, a comprehensive prediction of the patterns of local virus transmission is not possible. Recognizing these gaps, multi-year entomological surveys were carried out in Dhaka, where the disease is most frequently reported. The specific objectives of the present study are threefold: i) to determine the risk factors for the presence of Aedes mosquitoes; ii) to identify the types of most productive and key containers; and iii) to estimate the effects of climatic factors on Aedes abundance in the city of Dhaka, Bangladesh. Entomological surveys were conducted in 12 out of 90 wards in Dhaka. These wards were selected using a probability proportional sampling procedure during the monsoon seasons in 2011, 2012 and 2013 and in the dry season in 2012. All containers inside and around sampled households were inspected for mosquito larvae and pupae, and containers were classified according to their relative size, use pattern, and materials of construction. During the study period (2011–2013), 12,680 larvae and pupae were collected. About 82% of the identified immature mosquitoes were Aedes aegypti, while the remainder were Ae. albopictus and other mosquito species. The largest number of immature mosquitoes was collected from tires and refrigerator trays during 2011 and 2012 monsoon seasons. Conversely, plastic drums were the most productive during the 2012 dry and 2013 monsoon season. Vehicle parts and discarded construction materials were the most efficient producers of Aedes mosquitoes in all surveys. The presence of Aedes mosquitoes was significantly (p < 0.05) higher in low socio-economic zones of Dhaka. Container location, presence of vegetation, and availability of shade for containers were also significantly associated with finding immature Aedes mosquitoes, based on multivariable analysis after confounder adjustment. Rainfall, temperature, and relative humidity also significantly affected the mean abundance of mosquitoes. Proper use, disposal, and recycling of the containers that effectively produce large numbers of Aedes vector mosquitoes may decrease the risk of arboviral transmission.
Citation: Paul KK, Dhar-Chowdhury P, Haque CE, Al-Amin HM, Goswami DR, Kafi MAH, et al. (2018) Risk factors for the presence of dengue vector mosquitoes, and determinants of their prevalence and larval site selection in Dhaka, Bangladesh. PLoS ONE 13(6): e0199457. https://doi.org/10.1371/journal.pone.0199457
Editor: Olle Terenius, Swedish University of Agricultural Sciences, SWEDEN
Received: January 10, 2018; Accepted: June 7, 2018; Published: June 21, 2018
Copyright: © 2018 Paul 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.
Data Availability: As per the agreement between University of Manitoba (UM), International Development Research Centre (IDRC) and International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b), summary of data can be publicly displayed or can be made publicly accessible. To protect intellectual property rights of primary data and to respect the legal restrictions of the agreement, International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) cannot make primary data publicly available. However, upon request, Institutional Data Access Committee of International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) can provide access to primary data to any individual, upon reviewing the nature and potential use of the data. Requests for data can be forwarded to: Ms. Armana Ahmed, Head, Research Administration, icddr,b, Dhaka, Bangladesh, Email: email@example.com, Phone: +88 02 9827001-10 (ext. 3200).
Funding: This work was funded by International Development Research Centre (IDRC), Ottawa, Canada (https://www.idrc.ca/) (Grant no. 106040-001) to CEH. In addition, the Government of Bangladesh, Canada, Sweden and the UK provided core/unrestricted funding support for this work. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: BI, Breteau index; CI, container index; HH, household; HI, house index; PC, positive container; PI, pupal index; SESZ, socio-economic status zones; WC, wet container
Dengue fever is a self-limiting illness and the four dengue virus serotypes are the most frequent global arboviral infection of humans. Over 3.9 billion people from 128 countries live under the threat of dengue and 390 million infections occur every year, with 96 million people experiencing clinical manifestations of dengue [1,2]. In Bangladesh, the first dengue cases were detected in 1964 and the disease was referred to as ‘Dacca fever’ . From 1964 to 1999, sporadic cases of dengue were reported but it was not considered as a major public health concern. However, in 1999 a major outbreak took place the capital city, Dhaka, followed in 2000 by a countrywide epidemic involving 5,551 cases and 93 deaths [4–7]. In addition to dengue, chikungunya fever outbreaks, another Aedes-borne viral disease, was reported in two different geographical regions of Bangladesh in 2008 and 2011 [8,9]. In 2016, with the help of World Health Organization, Zika virus infection was detected for the first time in Bangladesh .
Further outbreaks of dengue were reported from 2000 to 2009 and more than 90% of all of the dengue cases were reported from Dhaka, and hence, this capital city was identified as the most endemic urban area for dengue in the country . Further to this, a serosurvey conducted during 2012 in 12 administrative wards of Dhaka observed dengue antibodies among more than 80% of the study participants .
Aedes (Stegomyia) aegypti Linnaeus and Aedes (Stegomyia) albopictus Skuse are the known vectors of dengue virus and other arboviruses including chikungunya and zika virus . These species are highly efficient vectors of arboviruses and live in close proximity to humans . Both of these species were first recorded in Bangladesh during 1952  and studies conducted in 1980 showed high abundance of both of these species in Dhaka . In 1997, a Breteau Index (BI) of 30.8 was recorded which was well above the risk levels for dengue transmission . During the 2000 dengue outbreak a comprehensive entomological survey was conducted in all wards of Dhaka, and an equally high BI of 24.6 was reported . Aedes aegypti and Ae. albopictus mosquitoes lay their eggs and larvae subsequently develop in domestic water storing containers, rainwater-holding objects including a wide variety of discarded materials as well as natural water retaining structures like tree holes and plant axils that are often abundant in peri-domestic environments. Thus there is a wide range of potential larval development sites for these mosquito species that vary in size, shape and location in the environment. Though many different types of containers can serve as development sites for vector mosquitoes, in some cases selected containers produce large numbers of larvae/pupae or some containers are sufficiently abundant to be efficient sources of larvae/pupae. Thus, these “key containers” are either highly productive or highly efficient, and both drive the local abundance of vector mosquitoes and are the targets for source reduction so important to vector suppression and dengue prevention. Vector abundance varies seasonally because of local changes in temperature, humidity and rainfall that affect the availability of larval development sites . In addition, the larval development sites for Aedes mosquitoes also vary spatially . Mosquitoes may concentrate in some parts of urban centres and display local scale variations within an urban centre . As a result, a clear and comprehensive understanding of the spatial distribution of key containers and how these can change on a seasonal basis is essential for designing effective vector control programs. Notably, data on Aedes abundance, and how it changes over the seasons in Dhaka are scanty.
International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) in collaboration with the University of Manitoba, Canada, the Public Health Agency of Canada, North South University, Bangladesh, and Directorate General of Health Services under Ministry of Health and Family Welfare of the Government of Bangladesh, pursued the present study to understand the dynamics of dengue vectors in Dhaka by conducting entomological surveys for Aedes mosquitoes. The results of the first survey, conducted during the monsoon season in 2011, were published elsewhere . In this study, we have combined the outcomes of four entomological surveys conducted in 2011, 2012 and 2013 in order to examine seasonal patterns in the spatial and temporal abundance of the immature stages (larvae and pupae) of Aedes mosquitoes, to identify the most productive and efficient container types for these species and to determine some of the factors affecting the abundance of Aedes larvae in Dhaka. The specific objectives of the present study are threefold: i) to determine the risk factors for the presence of Aedes mosquitoes; ii) to identify the types of most productive and key containers; and iii) to estimate the effects of climatic factors on Aedes abundance in the city of Dhaka, Bangladesh.
Three household based entomological surveys were conducted in 12 of 90 wards in Dhaka  at times when dengue incidence was anticipated to be high  and included the monsoon periods of 2011 (July-August), 2012 (July) and 2013 (August-September). In addition, a pre-monsoon (dry season) entomological survey was carried out in 2012 (March) when dengue incidence was assumed to be low.
Dhaka, the capital of Bangladesh, is one of the most densely populated cities in the world with more than 12 million people living in an area of about 360 km2, and it is divided into 90 local administrative units—called ‘wards’. The city usually experiences a hot, wet, and humid tropical climate, with monthly mean temperature varying between 20°C (68°F) in January and 32°C (90°F) in May. Using the Delphi method , the administrative wards were categorized into three socioeconomic status zones (SESZ), which produced 36 wards designated as low SESZ (LSESZ), 40 wards as medium SESZ (MSESZ), and 14 wards as high SESZ (HSESZ) (Fig 1). The indicators used for SESZ categorization included: a) municipal property tax rates, b) property market value, c) rate of property rent, d) proximity of types of markets and shopping areas, e) types of building structure, f) proximity to public services, g) state of infrastructure, and h) state of transport in each ward. With respect to cost, human resource requirements, and the time needed, as well as to minimize the margin of error in making population inferences, we determined that 12 wards were adequate for representative entomological surveys for the entire city. By employing a probability proportional to the number of wards in SESZs sampling method, a total of 12 administrative wards (2 wards from HSESZ, 5 wards from MSESZ, and 5 wards from LSESZ) were selected. Because the primary sampling unit was a household (HH), a random sample of 100 HHs were targeted from each ward, resulting in 1200 HHs as sampling units for each survey. For HH surveys, a spatial randomization procedure was followed by drawing grid cells on a map of each ward and a total of 100 HHs were selected using a random number table as described by Dhar-Chowdhury et al . The survey was repeated, where possible, in the same HHs for each of the subsequent three surveys.
All selected sites were physically marked on the map and their addresses were recorded on both field data sheets and GPS units. All containers inside the HHs as well as within a 50-meter radius of the HHs were inspected for immature mosquitoes. Any containers that held water for more than 3 days were considered “wet containers” . The amount of water in these containers at the time of the survey was recorded, and respondents were asked about the source and use of water within each container type. Water within any outdoor containers, without any nearby water source, during the monsoon season was considered “rainwater”. The number and type of “dry containers” (ones holding water for less than 3 days) were also recorded in order to calculate the container index (CI). All wet containers were examined for Aedes larvae and/or pupae. Only single storied or ground floor apartments were inspected because of the lack of access to the upper floors and roofs of the buildings and any containers on rooftops such as rooftop water tanks, flower tubs and other discarded materials were not inspected. Collection of larvae and pupae and the identification procedures were described elsewhere . Species determinations were performed only on a subset of the immature mosquitoes collected during each survey but the sub-sampling was done randomly and the subset of data is representative of the species composition across all households and years when the surveys were conducted.
Classification of containers used as larval development sites
All wet containers were categorized into seven classes depending on their use in daily life, manufacture materials, and their relative size or volume. Class 1 containers were comprised of small (i.e., 1 ml to 1 l of water) plastic containers that are used for domestic purposes (e.g., water storage). Class 2 containers were discarded but recyclable vehicle parts and construction materials. Class 3 containers consisted of medium (i.e., >1 and < 5 l) to large (>5 l) water reservoirs that are used regularly for water storage. Class 4 containers included small to medium sized non-plastic containers, most of which were used for temporary water storage. Class 5 containers were discarded HH materials. Class 6 containers consisted of ornamental containers typically used to capture water run-off from plants. Class 7 containers included natural plant based materials such as coconut shells, plant axils and tree holes. Immature Aedes mosquitoes were detected in these classes of containers during the course of this study.
Collection of meteorological data
Meteorological data including rainfall, temperature and relative humidity for monsoon (June-September) and pre-monsoon (February-March) from 2011 to 2013 were collected from Bangladesh Meteorological Department, Dhaka, Bangladesh. Such temporal rainfall, temperature and relative humidity daily data coverage ensured corroboration of meteorological data for previous 30 days of the entomological survey periods (2011: July-August; 2012: March; July; 2013: August-September) enabling to assess lag time effects of climatic factors on vector abundance.
Stegomyia indices including house index (HI), the percentage of houses infested with larvae and/or pupae; container index (CI), the percentage of positive containers; BI, the number of positive containers per 100 HHs and pupal index (PI), number of pupae per 100 houses inspected  and their 95% confidence intervals (CI) were calculated using the exact binomial test. Descriptive analysis was conducted for the distribution of wet containers and Aedes immatures for the four surveys. Wet containers with any number of larvae or pupae were considered “positive containers” (PC). Houses with PCs were considered “positive premises”. Following Tun-Lin et al.’s study , we considered “Key premises” (KP) as those contained ≥3 PCs either indoors or outdoors. The contribution of different types of containers on the production of Aedes mosquitoes was analyzed by calculating the percentages of positive containers, mean immatures with 95% confidence interval for each category of container, container productivity (number of immatures/all immatures X 100), and container efficiency (productivity/prevalence of container) . A container efficiency of 1.0 was considered when all containers were assumed equally efficient .
Prevalence of container was calculated by dividing the number of wet containers with all containers. We performed the univariate logistic regression method for each Aedes species to calculate odds ratio to identify associations between individual factors and the presence of larvae and/or pupae. As explanatory factors, we considered six variables including socio-economic status of the HHs, type or source of water, location of container (in-or out-doors), presence of vegetation near containers, and the amount of shade for the containers. Following these procedures, we constructed a conceptual framework to identify factors of interest for the final multivariable logistic regression model. A generalized linear model was used to investigate the association between climatic variables and the abundance of Aedes immatures. In the analysis, the abundance of mosquitoes per container was considered as the outcome variable and mean temperature, total rainfall, and percent relative humidity of the preceding month were used as predictor variables to make possible adjustments. All data collected were analyzed using the statistical software package STATA 13.1 (StataCorp LP, TX, USA).
Ethics approval and consent to participate
This study was approved by the Bangladesh Medical Research Council, icddr,b research and ethical review committees, and the Joint Faculty Research Ethics Board at the University of Manitoba, Canada. The purpose and objectives of the study, benefits and risks, and their voluntary participation option were explained to the head of each HH. As culturally appropriate, such informed-consent procurement sessions were witnessed by the neighbours and close kin to validate and witness verbal consent. After obtaining informed oral consent, the HH premises were inspected for the presence of containers and discarded materials that can store water and immature stages of mosquitoes.
The annual response rate was 74%, 60% and 53% in 2011, 2012 and 2013 respectively, almost equally distributed among SESZs, and thus majority of the respondents volunteered to participate in the study each year. A total of 884 HHs participated in the first survey (2011). Depending on the availability of survey time, alternative HHs were selected if the residents of the randomly selected HH refused to participate or were not at home during survey. In dry months, dengue incidence usually becomes very low in Bangladesh . This influenced the second survey (2012 dry season) as many previous participants refused to continue to participate. We observed a large number of people who participated in the previous survey were not available in subsequent surveys as moving residential location of HHs within Dhaka in a short period of time is frequent, attributed to the fact that most of the city dwellers live in rented apartments. As a result, members of only 546 HHs consented in the second survey (2012 dry season). The third (2012 monsoon season; n = 899) and fourth (2013 monsoon season; n = 639) surveys were conducted during the monsoon months and the number of respondents were higher, primarily due to restricted residential shifts during monsoon seasons (Table 1).
Numbers of water-holding containers inspected
Our entomological survey results revealed that that the proportion of positive containers has been decreasing gradually over the successive rounds of sampling, from a high of 39% in 2011 to only 24% in 2013 (Table 1). Such a trend was reflected in the consistently declining gradient from 2011 to 2013, with an expected low proportion during the dry season (11% during 2012 pre-monsoon).
Species determination of collected Aedes immatures
A total of 12,680 Aedes larvae and pupae as well as 268 larvae of other mosquitoes were collected during the four surveys. Highest (n = 5,554) number of immatures were collected during the monsoon season of 2012 and lowest number (n = 514) were collected during dry season of 2012 (Table 1). Species determination of 5,078 (39%) immature mosquitoes was carried out, of which 82% (4,152) immatures were identified as Ae. aegypti, 13% (658) were Ae. albopictus, and 5% (268) were of other mosquito species including Culex quinquefasciatus and Armigeres subalbatus. The overall proportion of Ae. aegypti ranged from 72% to 85%, and was highest during the 2011 monsoon and lowest during the 2012 monsoon. In contrast, the overall proportion of Ae. albopictus ranged from 8% to 23%, and was highest during the 2012 monsoon and lowest during the 2013 monsoon season.
Stegomyia indices according to seasons and SESZ
The HI, CI and BI were found to be lower during 2013 monsoon compared to 2011 monsoon season (Table 1). However, PI was higher in 2012 than in 2011. Noticeably, the Stegomyia indices were highest in wards located in LSESZ and were lowest in wards located in MSESZ (Table 2). The PI did not show any successive pattern; it was highest in HSESZ relative to two other zones.
Positive and key premises for vector production over 2011–2013
It is important here to analyse and reiterate positive and key premises to provide the context of container analysis. During the three consecutive monsoon seasons of 2011, 2012 and 2013, 25% (220 of 884), 24% (220 of 899), and 17% (106 of 639) of the inspected houses were positive for Aedes larvae and/or pupae, respectively. During 2012 dry season, only 7% (39 of 546) of the inspected houses were observed to be positive for Aedes (S1 Table). Percentage of KP decreased from 37% in 2011 monsoon to 10% in 2012 monsoon; however, it increased to 22% during the 2013 monsoon season (S1 Table).
Main types of positive containers and most productive containers by season
Types of Positive Containers (PCs) varied from 20 to 32 during all four surveys. In each sampling period, only 10 different types of containers produced about 80% of the Aedes immatures. Tires, usually discarded (in open place), produced highest number of immatures during 2011 monsoon (11.3%) and 2012 monsoon seasons (12.5%) while plastic drums, which are used to store water indoors, yielded most Aedes during 2012 dry season (28.2%) and 2013 monsoon seasons (17.6%). Overall, tires, flower tub and tray, refrigerator tray, and plastic drum (sealable) produced higher proportion of Aedes immatures compared to other containers (Table 3).
Determining the key containers
Considering immature production in all four survey periods, the class 1 containers were most abundant in the inspected HHs and their surroundings. These were the most productive (29.4) but least efficient (0.6) containers among all the container classes. Although class 2 containers were neither most abundant nor most productive, they were the most efficient containers (3.3) to produce immature Aedes. Class 1 and class 2 containers can therefore be distinguished as the key containers for Aedes immature production in Dhaka. The remaining classes of containers were moderately abundant and efficient (Table 4).
Risk factors for the presence of Ae. aegypti
From bivariate analysis, it was observed that the SESZ, type of water, location of container, presence of vegetation and shade was associated with presence of Ae. aegypti immatures. In the multivariable analysis after adjusting for possible confounders, the presence of Ae. aegypti was found to be significantly high in the low SESZ area (OR 1.69, CI: 1.17–2.44) and in containers holding rain water (OR 1.92, 95% CI: 1.35–2.72). Containers located outdoors (OR 1.92, 95% CI 1.48–2.50) and wet containers which were proximal to vegetation (OR 1.50, 95% CI: 1.10–2.04) were more likely to have Ae. aegypti. More larvae and pupae were detected in containers without shade compared to the containers located in areas with full shade (OR 2.30, 95% CI 1.55–3.43) (Table 5).
Risk factors for the presence of Ae. albopictus are similar to Ae aegypti
From the multivariable analysis after adjustment for possible confounders, the presence of Ae. albopictus was observed to be significantly high in containers with rainwater (OR 2.98, 95% CI 1.09–8.13). Containers located outdoors were also more likely to harbour Ae. albopictus than containers located indoors (OR 5.05, 95% CI 2.33–10.96). Detection of Ae. albopictus was significantly more likely when the container was in close proximity to vegetation (for container under vegetation—OR 3.94, 95% CI 1.70–9.16; for container located nearby vegetation—OR 3.12, 95% CI 1.45–6.68). The containers with no shade were also significantly more likely to harbour immature Ae. albopictus (OR 3.96, 95% CI 1.33–11.82) (Table 6).
Effect of climatic factors on the abundance of Aedes mosquitoes
The mean abundance of immature Aedes was significantly influenced by the selected climatic variables. Total rainfall (Fig 2A) and mean temperature (Fig 2B) in previous 30 days of entomological sampling had a significant negative effect on Aedes abundance. Each mm increase in total rainfall decreased the mean number of immature Aedes per positive container by 0.04 (95% CI: −0.08, −0.01; p = 0.024). Each degree centigrade rise in mean temperature was associated with a decrease in the mean Aedes number by 6.49 (95% CI −12.40, −0.58; p = 0.032). Relative humidity (Fig 2C) in previous 30 days of survey had a significant positive impact on Aedes abundance, as the mean number of mosquitoes increased by the rise of relative humidity (β = 2.73; 95% CI 0.59, 4.88; p = 0.014).
The present study investigated the abundance and oviposition habitats of Aedes mosquitoes in Dhaka during three consecutive monsoon seasons (2011, 2012, and 2013) and one dry season (2012). A broad spectrum of container types within high, medium and low urban socio-economic zones were inspected and Aedes larvae and pupae were identified and recorded from several key container types.
Traditionally, Aedes abundance is determined by the search for immatures within and around households and these data are used to calculate Stegomyia indices such as, HI, BI and CI . These indices not only measure the success of vector control strategy but also help understanding the vector ecology. However, in the contemporary literature, these indices are considered poor methods for predicting vector abundance because of their failure to effectively associate with and explain abundance of adult female mosquitoes and the potential for dengue virus transmission. The determination of dengue risk by interpreting such indices has been questioned by numerous analysts [27,28]. In this regard, Focks (2003) suggested that PI and identifying key and effective containers could be better predictors for disease transmission [27,29]. In the present study, BI, CI, and HI were observed to be lower in 2013 than in 2011 and they remained high in the LSESZ relative to other urban zones. In contrast, PI, did not appear to have any changing pattern during in the four surveys. It remained high in the HSESZ compared to other zones, indicating the “better-offs” in Dhaka are potentially more exposed to dengue risk. The prevailing local proverb that “dengue is a disease of the rich” was supported by the findings of the present study . Although the indices recorded in our investigation were higher than the indices recorded during the dengue outbreak of 2000, the number of reported dengue cases during the study period (2011–2013) was low , which can be due to under-reporting and passive nature of the surveillance for human cases of dengue virus infection.
More than one third of Wet Containers (WCs) harboured Aedes immatures during the first two wet seasons (2011, 2012). However, fewer WCs were positive in the third survey (2012 monsoon). The ratio of PCs to WCs was observed to be low during the 2012 dry season. Reduced mosquito abundance during the dry season might be due to the effects of lower environmental temperature, very low rainfall, lower relative humidity, and fewer residents storing water during the winter (dry) months. Gradual decreases of the proportion of PCs observed over the monsoon survey periods might be attributed to increased study participant’s awareness regarding dengue vector mosquitoes and their tendency to develop within selected containers within and around HH. Significant numbers of outdoor containers were reported to be positive with immature Aedes in the previous studies conducted in urban areas [17,31–33]. As urban dwellers (and the participants in our study) become more aware of the possible oviposition sites of Aedes mosquitoes, they apt first to check and clean indoor containers located inside their HHs. Consequently, the vectors tend to shift to the outdoor container, which was clearly observed in the present study. This finding is consistent with several previous studies [17,34–36]. Such adaptive behavior of Aedes mosquitoes poses serious challenge to the vector control efforts .
Different types of WCs produced variable numbers of Aedes immatures during the present study. In all four surveys, plastic drums, plastic buckets, water tanks, clay pots, and flower tubs were observed to producing large number of Aedes larvae and pupae. City dwellers of Dhaka use relatively smaller plastic drums, plastic buckets, and larger water tanks to store water from supplied piped water, as municipal water supply is not reliably available. As reported previously, if the water stored within containers is not emptied weekly, they become ideal oviposition sites for Aedes mosquitoes . Residents of Dhaka use flower tubs in and around houses to plant relatively small ornamental trees, and refrigerator trays located underneath refrigerator compartment remain out of sight and unnoticed. Tires and clay pots are usually located in outdoor locations where water remains stagnant, become suitable for Aedes oviposition. We observed that during the wet seasons, tires and refrigerator trays were mostly responsible for Aedes production. These Aedes habitats need to be the focus of vector control efforts. It is worth noting here that, novel methods utilizing outdoor containers like tires to downsize Aedes mosquito population are being evaluated in comparable countries . In a recent study in Lahore (Pakistan), Malik et al.  also noted a distinct seasonality to vector abundance. Periods of high intensity of vector mosquitoes development (i.e., larval observed in standing water environments) occurred during the wetter periods of the season with large numbers of immatures developing within water tanks, bottles, jugs and pitchers. However, during (or shortly after) substantial rainfall events, the occurrence of larvae was low. The increased vector abundance was correlated with periods when the resident human population was more prone to disease outbreaks. Entomological survey results of our study conform to these patterns of optimal range of climatic factors and human behavioural adjustments to various seasonality (such as storing water during dry and/or hot seasons). Also consist with the outcomes of our study, Banerjee et al.  in a study of pupal productivity in Kolkata (India), observed that plastic containers were the most productive habitats and tires were least productive sites for Aedes mosquitoes. Arduino  in the context of coastal urban Brazil, distinguished between movable and fixed containers, and concluded that only three container categories made greater contribution to risk of dengue transmission: metallic and plastic items among the roving containers, and boats/ships among the fixed containers. The potential influence of mobile vs. fixed containers deserves further research attention in various social and ecological contexts. In addition, some previous reports indicate that the shift of container productivity from one type to another is common. For example, in Malaysia, earthenware jars and miscellaneous containers superseded ant traps, buckets, basins, bowls, and concrete tanks as major Aedes production site over time [43–46]. Unlike these studies, previously identified key containers in Dhaka including tires, tanks, drums and flowers pots were observed to be preferred oviposition sites for Aedes .
In the present study, we found that small plastic reservoirs were most abundant and discarded vehicle and construction materials were most efficient immature habitats. Although wide-mouthed large water storage containers like concrete tanks, wells and drums are well-known key oviposition sites in urban areas [47,48] as well as in rural areas , we documented that these larger reservoirs were moderately efficient as Aedes mosquitoes habitats. Due to adequacy of potential habitats, Aedes abundance remains high in Dhaka, like most tropical cities in the developing world, which in turn put people at risk to arboviral infections [50,51]. While Dhaka is presently expanding very rapidly, the required planning for providing healthy environment to the population in the newer areas is generally absent. As a result, the risk of mosquito-borne viral diseases remains high.
In our study, wet containers located in the LSESZ were significantly more likely to be positive for Ae. aegypti immatures than wet containers located in other SESZs but the distribution of Ae. albopictus immatures was similar in all SESZ. The higher number of positive wet containers recorded in the LSESZ might be due to inadequate municipal services, such as water supply provisions and garbage collection methods. In addition, in this socioeconomically “worst-off” urban zone, there is a general lack of awareness among the community members about vector breeding sources . However, as stated above, in terms of PI, it has been consistently high in the HSESZ relative to others, indicating that adult Aedes survival rate may be much higher in the HSESZ due to supportive habitat (for both immature and adult mosquitoes) in communities with build-up infrastructure. As reported in our study of 2011 data , “mosquitoes are everywhere”  in Dhaka, and differences among urban zones by socioeconomic status did in fact have nominal effects upon variations in Aedes abundance.
In our study, containers filled with rainwater were significantly more likely to be PCs for either Ae. aegypti or Ae. albopictus, this conforms to the findings of some previous studies [52–54]. Rainfall is one of the most important environmental factors influencing the production of Aedes larvae. Not surprisingly, more Ae. aegypti and Ae. albopictus larvae was collected from containers holding rain water. Unsheltered containers that fill with rain water may serve as excellent larval development sites for Aedes mosquitoes. Vegetation near these larval production sites might also affect the abundance of mosquito larvae. Trees and shrubs may indirectly enhance mosquito abundance by providing important sugar feeding resources and resting sites. Vegetation may also promote gravid females to lay eggs in locally available containers. Organic detritus of vegetation may alter the aquatic chemistry of the containers, potentially influencing the attractiveness of such containers to gravid females. Moreover, since larvae feed on microorganisms in the water, the organic components in the containers might promote algal and microbe growth, and hence, support mosquito production [55,56]. This is concordant with the findings of the present study, as larval production of any species had been significantly affected by the presence of vegetation near the containers. Effect of shade on the containers did not differ significantly when variables such as SESZ, place of the container and shelter of the container from rain were adjusted. However, high numbers of larvae in the partially shaded and exposed containers was likely due to random factors or as a consequence of rain water storage in those containers made them more attractive for egg-laying females.
In our study, increase in total rainfall was associated with a reduction in mosquito abundance that might have been related to “flushing” of immature forms, especially from containers exposed to rainfall outdoors. The negative impact of rainfall on production of Aedes mosquitoes has been reported in several previous studies [57,58]. Furthermore, when coupled with strong winds, heavy rainfall can limit the dispersal and reproduction of the mosquitoes by reducing flight activity, leading to difficulties in finding mates, hosts and suitable larval development sites . During the monsoon surveys, mean temperature varied between 28 and 30°C. In this connection, we observed that an increase in mean temperature was associated with reduction in the abundance of Aedes immatures. Warmer temperatures (20–35°C) generally increase the rate of development of dengue vectors, improve or enhance feeding frequency of adult mosquitoes and the rate of dengue virus replication in infected females, so it was surprising that warmer temperatures led to fewer immature Aedes mosquitoes. However, several studies have revealed lag times [59–61] for these relationships. Further investigations are therefore required to determine more precisely how temperature impacts Aedes abundance in the Bangladesh context. We observed a positive effect of relative humidity on the abundance of vector mosquitoes in the range between 77–82% relative humidity. It was assumed by previous studies that a higher relative humidity is associated with an increase in hatching of Aedes eggs leading to the increased abundance of immature mosquitoes; findings of our investigation conform to this [62,63].
The present study investigated and revealed the degree and dynamics of the abundance of Aedes mosquitoes in different urban SESZs in Dhaka and identified the key containers responsible for harbouring Aedes immatures. These containers are chiefly generated and/or possessed by the city dwellers and produce sizeable populations of adult female mosquitoes which increases the risk of pathogen transmission. Proper use, disposal and recycling of the containers by the city dwellers and responsible authorities are necessary for environmental and mosquito habitat management. Community education regarding the biology of Aedes mosquitoes, the need to clean HH premises including inverting/discarding unused containers in appropriate ways is anticipated to increase the uptake of these simple but effective intervention strategies. Because the traditional vector indices are not reliable predictors of dengue virus outbreaks, annual mosquito surveillance is necessary to gain a better understanding of how vector populations change over time and how this relates to risk of pathogens that are transmitted by Aedes mosquitoes in Bangladesh. Surveillance along with sound prevention and source reduction programs implemented by and within at-risk communities are seen as crucial elements for reducing the risk of dengue and other arboviral infections.
The study was subject to multiple limitations. Firstly, because of accessibility issues, we did not survey containers located on rooftops or potential cryptic larval developmental sites, which may have been important sources of Aedes mosquitoes. Secondly, due to a very high rate of change of residence/relocation among Dhaka city dwellers, we could not enrol some HHs in the subsequent surveys. Thirdly, the study collected only immature (larvae and pupae) mosquitoes and not adults, which, if available, may have provided additional valuable information about population structure of the resident vector species. Use of alternative collection methods, such as lethal oviposition traps may have also helped to provide a more robust picture of the vector population structure in this study. Fourthly, species determinations were performed only on a subset of the immature mosquitoes collected in this study; however, the sub-sampling was performed randomly across the different surveys and as such as likely representative of the overall population of mosquitoes encountered. Finally, we did not perform any time series analysis to explore association between climatic factors and mean number of Aedes, so the observed association might be true for limited ranges of climatic variables.
icddr,b acknowledges with gratitude the commitment of IDRC to its research efforts. We are thankful to the HH members who gave their kind consent for participation in the study. We also want to acknowledge the supportive role of the Dept. of Public Health, North South University, Bangladesh and the students of the department who worked hard for data collection. icddr,b is also grateful to the Government of Bangladesh, Canada, Sweden and the UK for providing core/unrestricted funding support.
- 1. Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, Moyes CL, et al. (2013) The global distribution and burden of dengue. Nature 496: 504–507. pmid:23563266
- 2. Brady OJ, Gething PW, Bhatt S, Messina JP, Brownstein JS, Hoen AG, et al. (2012) Refining the global spatial limits of dengue virus transmission by evidence-based consensus. PLoS Negl Trop Dis 6: e1760. pmid:22880140
- 3. Russell PK, Buescher EL, McCown JM, Ordonez J (1966) Recovery of dengue viruses from patients during epidemics in Puerto Rico and East Pakistan. Am J Trop Med Hyg 15: 573–579. pmid:4957424
- 4. Bang DYH, Shah DNK (1987) Review of DHF Situation and Control of Aedes aegypti in South-East Asia.
- 5. Yunus EB, Bangali AM, Mahmood M, Rahman MM, Chowdhury AR, Talukder KR (2001) Dengue Outbreak 2000 in Bangladesh: From Speculation to Reality and Exercises.
- 6. Hossain MA, Khatun M, Arjumand F, Nisaluk A, Breiman RF (2003) Serologic evidence of dengue infection before onset of epidemic, Bangladesh. Emerg Infect Dis 9: 1411–1414. pmid:14718084
- 7. Sharmin S, Viennet E, Glass K, Harley D (2015) The emergence of dengue in Bangladesh: epidemiology, challenges and future disease risk. Trans R Soc Trop Med Hyg 109: 619–627. pmid:26333430
- 8. Khatun S, Chakraborty A, Rahman M, Nasreen Banu N, Rahman MM, Hasan SM, et al. (2015) An Outbreak of Chikungunya in Rural Bangladesh, 2011. PLoS Negl Trop Dis 9: e0003907. pmid:26161995
- 9. Salje H, Lessler J, Paul KK, Azman AS, Rahman MW, Rahman M, et al. (2016) How social structures, space, and behaviors shape the spread of infectious diseases using chikungunya as a case study. Proc Natl Acad Sci U S A 113: 13420–13425. pmid:27821727
- 10. Anonymous (2016) First Case of Zika found in 2014. The Daily Star. Dhaka: The Daily Star.
- 11. Dhar-Chowdhury P, Paul KK, Haque CE, Hossain S, Lindsay LR, Dibernardo A, et al. (2017) Dengue seroprevalence, seroconversion and risk factors in Dhaka, Bangladesh. PLoS Negl Trop Dis 11: e0005475. pmid:28333935
- 12. Kraemer MU, Sinka ME, Duda KA, Mylne AQ, Shearer FM, Barker CM, et al. (2015) The global distribution of the arbovirus vectors Aedes aegypti and Ae. albopictus. Elife 4: e08347. pmid:26126267
- 13. Scott TW, Takken W (2012) Feeding strategies of anthropophilic mosquitoes result in increased risk of pathogen transmission. Trends Parasitol 28: 114–121. pmid:22300806
- 14. Asir-Ud-Din M (1952) Mosquitoes breeding in tree-holes and bamboo stumps in Dacca (East Pakistan). Pakistan Journal of Health 2: 110–112.
- 15. Khan M, Ahmed T (1986) Dengue status in Bangladesh. Dengue News Letter 12.
- 16. Knudsen AB (1997) An assignment report on Dengue. World Health Organizaion: 35.
- 17. Ferdousi F, Yoshimatsu S, Ma E, Sohel N, Wagatsuma Y (2015) Identification of Essential Containers for Aedes Larval Breeding to Control Dengue in Dhaka, Bangladesh. Trop Med Health 43: 253–264. pmid:26865829
- 18. Regis LN, Acioli RV, Silveira JC Jr., de Melo-Santos MA, da Cunha MC, Souza F, et al. (2014) Characterization of the spatial and temporal dynamics of the dengue vector population established in urban areas of Fernando de Noronha, a Brazilian oceanic island. Acta Trop 137: 80–87. pmid:24832009
- 19. Su MD, Chang NT (1994) Framework for application of geographic information system to the monitoring of dengue vectors. Gaoxiong Yi Xue Ke Xue Za Zhi 10 Suppl: S94–101.
- 20. Honorio NA, Castro MG, Barros FS, Magalhaes Mde A, Sabroza PC (2009) The spatial distribution of Aedes aegypti and Aedes albopictus in a transition zone, Rio de Janeiro, Brazil. Cad Saude Publica 25: 1203–1214. pmid:19503951
- 21. Dhar-Chowdhury P, Haque CE, Lindsay R, Hossain S (2016) Socioeconomic and Ecological Factors Influencing Aedes aegypti Prevalence, Abundance, and Distribution in Dhaka, Bangladesh. Am J Trop Med Hyg 94: 1223–1233. pmid:27022149
- 22. Needham RD, Loe RC (1990) The policy Delphi: purpose, structure, and application. The Canadian Geographer 34: 133–142.
- 23. Jeffery JA, Clements AC, Nguyen YT, Nguyen le H, Tran SH, Le NT, et al. (2012) Water level flux in household containers in Vietnam—a key determinant of Aedes aegypti population dynamics. PLoS One 7: e39067. pmid:22911683
- 24. Tun-Lin W, Kay B, Barnes A (1995) The premise condition index: a tool for streamlining surveys of Aedes aegypti. The American journal of tropical medicine and hygiene 53: 591–594. pmid:8561259
- 25. Hammond SN, Gordon AL, Lugo Edel C, Moreno G, Kuan GM, Lopez MM, et al. (2007) Characterization of Aedes aegypti (Diptera: Culcidae) production sites in urban Nicaragua. J Med Entomol 44: 851–860. pmid:17915519
- 26. Tsuzuki A, Huynh T, Tsunoda T, Luu L, Kawada H, Takagi M (2009) Effect of existing practices on reducing Aedes aegypti pre-adults in key breeding containers in Ho Chi Minh City, Vietnam. Am J Trop Med Hyg 80: 752–757. pmid:19407119
- 27. Focks DA (2003) A review of entomological sampling methods and indicators for dengue vectors. Geneva: World Health Organization TDR/IDE/DEN/03.1.
- 28. Bowman LR, Runge-Ranzinger S, McCall PJ (2014) Assessing the relationship between vector indices and dengue transmission: a systematic review of the evidence. PLoS Negl Trop Dis 8: e2848. pmid:24810901
- 29. Focks D, Alexander N, Villegas E (2006) Multicountry study of Aedes aegypti pupal productivity survey methodology: findings and recommendations. Geneva, World Health Organization TDR/IRM/DEN/06.1.
- 30. Ullah M (2015) Deadly Dengue, City corporations too big to spot tiny mosquitoes. The Daily Sun. Dhaka: Maynal Hossain Chowdhury on behalf of East West Media Group Limited.
- 31. Barrera R, Amador M, Clark GG (2006) Use of the pupal survey technique for measuring Aedes aegypti (Diptera: Culicidae) productivity in Puerto Rico. Am J Trop Med Hyg 74: 290–302. pmid:16474086
- 32. Seng CM, Setha T, Nealon J, Socheat D (2009) Pupal sampling for Aedes aegypti (L.) surveillance and potential stratification of dengue high-risk areas in Cambodia. Trop Med Int Health 14: 1233–1240. pmid:19708900
- 33. Ibarra AMS, Ryan SJ, Beltran E, Mejia R, Silva M, Munoz A (2013) Dengue vector dynamics (Aedes aegypti) influenced by climate and social factors in Ecuador: implications for targeted control. PloS one 8: e78263. pmid:24324542
- 34. Ahmed TU, Rahman GMS, Bashar K, Shamsuzzaman M, Samajpati S, Sultana S, et al. (2007) Seasonal prevalence of dengue vector mosquitoes in Dhaka City, Bangladesh. Bangladesh J Zool 35: 205–212.
- 35. Chadee DD, Huntley S, Focks DA, Chen AA (2009) Aedes aegypti in Jamaica, West Indies: container productivity profiles to inform control strategies. Trop Med Int Health 14: 220–227. pmid:19236668
- 36. Dhar-Chowdhury P, Emdad Haque C, Michelle Driedger S, Hossain S (2014) Community perspectives on dengue transmission in the city of Dhaka, Bangladesh. Int Health 6: 306–316. pmid:24981443
- 37. Saifur RG, Dieng H, Hassan AA, Salmah MR, Satho T, Miake F, et al. (2012) Changing domesticity of Aedes aegypti in northern peninsular Malaysia: reproductive consequences and potential epidemiological implications. PLoS One 7: e30919. pmid:22363516
- 38. Phuanukoonnon S, Mueller I, Bryan JH (2005) Effectiveness of dengue control practices in household water containers in Northeast Thailand. Trop Med Int Health 10: 755–763. pmid:16045462
- 39. Ulibarri G, Betanzos A, Betanzos M, Rojas JJ (2016) Control of Aedes aegypti in a remote Guatemalan community vulnerable to dengue, chikungunya and Zika virus: Prospective evaluation of an integrated intervention of web-based health worker training in vector control, low-cost ecological ovillantas, and community engagement [version 1; referees: 1 approved with reservations]. F1000Research 5: 598. pmid:28105304
- 40. Malik A, Yasar A, Tabinda AB, Zaheer IE, Malik K, Batool A, et al. (2017) Assessing spatio-temporal trend of vector breeding and dengue fever incidence in association with meteorological conditions. Environ Monit Assess 189: 189. pmid:28353206
- 41. Banerjee S, Aditya G, Saha GK (2013) Pupal productivity of dengue vectors in Kolkata, India: implications for vector management. Indian J Med Res 137: 549–559. pmid:23640563
- 42. de Brito Arduino M (2014) Assessment of Aedes aegypti Pupal Productivity during the Dengue Vector Control Program in a Costal Urban Centre of São Paulo State, Brazil. Journal of Insects 2014: 9.
- 43. Cheong W, Rudnick A, Lin T (1986) The vectors of dengue and dengue hemorrhagic fevers in Malaysia. Inst Med Res Malays Bull 23: 155–167.
- 44. Lee H, Cheong W (1987) A preliminary Aedes aegypti larval survey in the suburbans of Kuala Lumpur city. Trop Biomed 4: 111–118.
- 45. Lee H (1990) Breeding habitats and factors affecting breeding of Aedes larvae in urban towns of Peninsular Malaysia. J Biosci 1: 107–112.
- 46. Lee H (1991) A nationwide resurvey of the factors affecting the breeding of Aedes aegypti (L.) and Aedes albopictus (Skuse)(Diptera: Culicidae) in urban town of peninsular Malaysia–1988–1989. Trop Biomed 8: 157–160.
- 47. Maciel-de-Freitas R, Marques WA, Peres RC, Cunha SP, de Oliveira RL (2007) Variation in Aedes aegypti (Diptera: Culicidae) container productivity in a slum and a suburban district of Rio de Janeiro during dry and wet seasons. Mem Inst Oswaldo Cruz 102: 489–496. pmid:17612770
- 48. Garelli FM, Espinosa MO, Weinberg D, Coto HD, Gaspe MS, Gurtler RE (2009) Patterns of Aedes aegypti (Diptera: Culicidae) infestation and container productivity measured using pupal and Stegomyia indices in northern Argentina. J Med Entomol 46: 1176–1186. pmid:19769052
- 49. Nguyen le AP, Clements AC, Jeffery JA, Yen NT, Nam VS, Vaughan G, et al. (2011) Abundance and prevalence of Aedes aegypti immatures and relationships with household water storage in rural areas in southern Viet Nam. Int Health 3: 115–125. pmid:24038184
- 50. Gubler DJ (2011) Dengue, Urbanization and Globalization: The Unholy Trinity of the 21(st) Century. Trop Med Health 39: 3–11.
- 51. Li Y, Kamara F, Zhou G, Puthiyakunnon S, Li C, Liu Y, et al. (2014) Urbanization increases Aedes albopictus larval habitats and accelerates mosquito development and survivorship. PLoS Negl Trop Dis 8: e3301. pmid:25393814
- 52. 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. pmid:8510120
- 53. 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. pmid:12641413
- 54. Morrison AC, Gray K, Getis A, Astete H, Sihuincha M, Focks D, et al. (2004) Temporal and geographic patterns of Aedes aegypti (Diptera: Culicidae) production in Iquitos, Peru. J Med Entomol 41: 1123–1142. pmid:15605653
- 55. Merritt RW, Dadd RH, Walker ED (1992) Feeding behavior, natural food, and nutritional relationships of larval mosquitoes. Annu Rev Entomol 37: 349–376. pmid:1347208
- 56. Chaves LF, Keogh CL, Nguyen AM, Decker GM, Vazquez-Prokopec GM, Kitron UD (2010) Combined sewage overflow accelerates immature development and increases body size in the urban mosquito Culex quinquefasciatus. J Appl Entomol 135: 611–620.
- 57. Paaijmans KP, Wandago MO, Githeko AK, Takken W (2007) Unexpected high losses of Anopheles gambiae larvae due to rainfall. PLoS One 2: e1146. pmid:17987125
- 58. Dhimal M, Gautam I, Joshi HD, O'Hara RB, Ahrens B, Kuch U (2015) Risk factors for the presence of chikungunya and dengue vectors (Aedes aegypti and Aedes albopictus), their altitudinal distribution and climatic determinants of their abundance in central Nepal. PLoS Negl Trop Dis 9: e0003545. pmid:25774518
- 59. Hsieh YH, Chen CW (2009) Turning points, reproduction number, and impact of climatological events for multi-wave dengue outbreaks. Trop Med Int Health 14: 628–638. pmid:19392743
- 60. Hii YL, Rocklov J, Ng N, Tang CS, Pang FY, Sauerborn R (2009) Climate variability and increase in intensity and magnitude of dengue incidence in Singapore. Glob Health Action 2.
- 61. Wu PC, Guo HR, Lung SC, Lin CY, Su HJ (2007) Weather as an effective predictor for occurrence of dengue fever in Taiwan. Acta Trop 103: 50–57. pmid:17612499
- 62. Dickerson CZ (2007) The effects of temperature and humidity on the eggs of Aedes aegypti (L.) and Aedes albopictus (Skuse) in Texas [Dissertation]: Texas A&M University. 119 p.
- 63. Waldock J, Chandra NL, Lelieveld J, Proestos Y, Michael E, Christophides G, et al. (2013) The role of environmental variables on Aedes albopictus biology and chikungunya epidemiology. Pathog Glob Health 107: 224–241. pmid:23916332