The impact of mosquito proof lids of underground tanks “tanka” on the breeding of Anopheles stephensi in a village in western Rajasthan, India

Background Western Rajasthan in India has a typical desert climate. Until the introduction of the canal water irrigation system, malaria was an unstable and seasonal occurrence. Due to the scarcity of water, the community practised having one large underground tank (locally known as the tanka) in their house to collect rainwater for long-term household use. Anopheles stephensi, one of the major malaria vectors, breeds in improperly covered "tankas” if not properly covered and harbours a vector population throughout the year. Methods Two villages, Ajasar (intervention) and Tota (control), with similar ecological features, were selected for the study. A pre-intervention survey was carried out in both villages to assess the presence and quality of lids of tankas, and mosquito breeding and adult mosquito density. Awareness of the community about malaria and mosquitoes was also assessed during the pre-intervention period. In the intervention village, damaged or improper lids were replaced with improved mosquito proof polyvinyl chloride plastic (PVC) lids and lasted longer than the conventional lids. The fitness of the lids was assessed one year after the pre-intervention survey. The entomological assessment was carried out in both intervention and non-intervention villages. The level of community awareness about malaria, mosquitoes, their breeding places, and the role of tankas in malaria transmission was assessed both during pre- and post-intervention. Results During the pre-intervention survey, Anopheles breeding was found in 22.1% (58/262) of tankas in the intervention village and 27.1% (19/70) in tankas in the control village. Mosquito breeding was observed in the tankas with iron lids in the intervention village (48.3%) and the control village (42.1%). In the intervention village, out of 262 tankas in the village, 200 lids were replaced, resulting in the complete absence of mosquito breeding. In the pre-intervention survey conducted in May 2018, Anopheles stephensi consisted of 46% of adult mosquitoes in the intervention village and 55% in the control village. Its density was significantly reduced to 0.55 per man-hour (94.95%) and 0.22 per man-hour (97.8%) in the post-intervention survey in June 2018 and a follow-up survey in May 2019, respectively, in the intervention village. Discussion The density of Anopheles stephensi adults was reduced significantly (97.8%) in the intervention village due to complete prevention of breeding in the underground tankas in the intervention village as compared to the control with no density reduction. The awareness level of the community was also improved due to their involvement in the study. Conclusion Provision of proper metal lids or replacement of damaged lids on underground water storage tanks as an environmental management approach prevented the breeding of the malaria vector, Anopheles stephensi, in a desert village in western Rajasthan.


Background
The arid and semi-arid regions of western India are prone to malaria outbreaks [1]. Major malaria epidemics have been reported in these regions where malaria incidence is influenced by the amount of rainfall [2,3]. However, after the establishment of the Indira Gandhi Canal for catering water in western Rajasthan in 1970, malaria became a major problem [4,5]. The ecological scenario in terms of cropland and vegetated areas increased by ~ 68% due to increased water availability [6] The breeding of malaria vectors in western Rajasthan was also aided by erratic rainfalls, migration and community water storage activities. The underground water tanks (locally known as the tankas) are the primary water storage containers in almost every house in the districts of Jodhpur, Jaisalmer, Barmer and Bikaner in Western Rajasthan. Tanka is an underground rainwater collection usually covered and impermeable cistern on shallow ground. Tankas collect rainwater from rooftops, courtyards, and artificially constructed catchments [7].
Malaria transmission in the Thar desert of India is primarily caused by Anopheles stephensi, which breeds almost exclusively in the underground water reservoirs, such as 'tanka' and 'beri' (step wells) [8,9]. According to some studies, the tankas become potential breeding grounds for the main malaria vector Anopheles stephensi due to damaged lids with holes due to rusting, broken lids, dents and gaps between lids and frames due to bends [9,10]. During the summer season (April-June), in harsh climatic conditions, Anopheles larvae are confined to the tankas with no or damaged lids. In monsoon and post-monsoon seasons, Anopheles mosquitoes cater in the rainwater collections, while the tankas serve as the mother foci for mosquito breeding in lean seasons [7].
The present study aimed to assess the impact on the prevention of mosquito breeding by improving the lid design from traditionally used materials to polyvinyl chloride (PVC) plastic and installing proper lids on open tankas. This improved lid type was expected to prevent mosquito entry into tankas and to prevent the breeding of mosquitoes, especially of the vector of malaria, Anopheles stephensi, in the study village.

Study site
Jaisalmer district is located in the Thar desert of India between 26°4' and 28°23′ North Latitude and 69°20′ and 72°42′ East Longitude, is the largest district of Rajasthan and the third largest in India. It shares an international boundary in the west with Pakistan and shares district boundaries with Bikaner, Jodhpur and Barmer districts. Among the four blocks of Jaisalmer district i.e. Jaisalmer, Pokaran, Fatehgarh, and Bhaniyana, Pokaran was most affected by malaria [11]. Therefore, Ajasar village in the Pokaran block, with a population of ~ 1500 people, was chosen for the study having malaria incidence of > 5 cases/1000 (Annual Parasite Index) after consulting with the Medical Officer of the Community Health Centre, Pokhran in August 2017. A second village, "Tota", was included as a non-intervention village having a population of ~ 750, and situated ~ 7 km away from village Ajasar and with similar geographical, ecological and malarious conditions (Fig. 1).

Entomological monitoring
To evaluate adult and larval mosquito breeding of Anopheles stephensi, entomological collections were made during the pre-intervention (first fortnight of May 2018), post-intervention (second fortnight of June 2018), and follow-up (May 2019) surveys. Adult mosquitoes were collected using two methods: hand catch and total catch, as per the standard WHO manual [12].

Aspirator collections
Mosquitoes resting inside human dwellings, cattle sheds and mixed dwellings in the two villages were collected by insect collectors using a flash light and mouth aspirators during the early morning hours (6 to 8 AM). Standard taxonomic keys were used to identify mosquitoes. The man-hour density (PMHD) of each Anopheles species was calculated (PMHD = Total no. of mosquitoes collected / No. of person x Time spent in hours).

Discussion:
The density of Anopheles stephensi adults was reduced significantly (97.8%) in the intervention village due to complete prevention of breeding in the underground tankas in the intervention village as compared to the control with no density reduction. The awareness level of the community was also improved due to their involvement in the study.

Conclusion:
Provision of proper metal lids or replacement of damaged lids on underground water storage tanks as an environmental management approach prevented the breeding of the malaria vector, Anopheles stephensi, in a desert village in western Rajasthan.

Total catches
Mosquitoes resting inside dwelling structures were also collected by the spray sheet collection method, by application of pyrethrum extract using a hand atomizer. Mosquitoes knocked down on white cotton sheets spread out on the floor of the room were picked up 10 min after the spray and transferred to the petri dishes lined with wet cotton or filter paper from sheets spread on the floor and transported to the laboratory.

Larval collections
Breeding site surveys were conducted in all aquatic habitats within a 1 km radius perimeter of the villages using standard World Health Organization (WHO) larval collection methods. Sampling was done with a dipper of white enamel bowl from tankas, clay pot/bird pots, under construction tankas, cemented cattle tanks, ground cement tanks and seepage waters. Larvae were identified using standard identification keys after emergence at the adult stage.
The breeding sites, such as open cattle tanks and under construction cement tanks, were regularly monitored by the Auxiliary Nurse Midwife for mosquito breeding.

Intervention with lids
At the time of pre-intervention, there were 262 and 70 tankas present in Ajasar and Tota village, respectively. Metal lids (dimension 45 cm × 45 cm): a total of 200 (76%) tankas were found to be defective due to holes caused by rusting and the gaps between the lids and their frames due to bending or denting. Damaged lids were replaced with the help of villagers and trained skilled persons (the 'mistries') over a period of 12 days. Consent from house owners in Ajasar village for replacement of lids was taken. Instead of conventional lids made up of iron, aluminium or wood, PVC was used in the modified lids as PVC was considered to be a more durable, rust-proof and economical material. Further steel screws in the hinge joint were used to make the joints rust-free and long-lasting. The frame of the lid was also modified for better binding with the cement properly (Fig. 2).

Knowledge, attitude and practices (KAP) survey
A door-to-door survey was conducted in 120 households (88 in Ajasar and 32 in Tota village) to assess the knowledge of the community about malaria, mosquitoes, their breeding places, and the role of tankas in malaria transmission. The assessment was done before and after the intervention using pre-defined questions. Consent was taken from each participant.

Information, education and communication (IEC) activities
During the pre-intervention phase, IEC activities included conducting a door-to-door survey, meeting with the elected village head (Sarpanch), Block Development Officer, and community representatives to inform the community about the study's goal, benefits, and to obtain communal consent for the interventions. Skilled workers "Mistries" were also selected from the Ajasar village and trained on the process of replacing lids in the presence of the Sarpanch and other important community representatives (Fig. 3).

Assessing the fitness of tankas
Tankas with damaged or missing lids were labelled and geo-referenced using the Global Positioning System (GPS) during the pre-intervention survey for future reference in Ajasar village only. The fitness of the lids, as well as their ability to prevent mosquito breeding, was examined during the follow-up survey after one year. A limitation of the study was the inability to do seasonal studies due to the inadequacy of funds allocated to the study.

Pre-intervention survey in study and control villages
In The proportion of tankas with mosquito breeding in these villages was 19.6% and 21.6%, respectively. Tankas under construction (60% in the intervention village, 100% in the non-intervention), tankas with wooden lids (75% in intervention) and tankas without lids (33.3%-intervention, 61.5%-non-intervention) were found to be prone to mosquito breeding (Table 1).
In the pre-intervention survey, 22.1% (58/262) tankas in the intervention village (Ajasar) and 27.1% (19/70) of tankas in the non-intervention village (Tota) were found to have mosquito breeding. The difference between the proportion of the tankas with mosquito larvae in the intervention village and the control village was statistically non-significant (z = -0.8814, p = 0.37886 > 0.05, not significant), indicating that there was no difference in larval positivity until the lids were replaced ( Table 1).

Impact of replacement of lids on mosquito breeding
In addition to tankas, mosquito larvae were present in seepage water pools, pipeline leakage pools, small pots with water for birds, plastic tanks, desert coolers and cement tanks. In pre-assessment, Anopheles stephensi was the predominant species breeding in both intervention and non-intervention villages in the tankas, while during both post-intervention and follow-up, breeding of Anopheles stephensi was recorded in tankas in non-intervention village only. There was no mosquito breeding in other habitats such as clay pots and bird pots in Ajasar village.

Impact of replacing lids on the density of adult mosquitoes
Anopheles stephensi had the highest density per manhour (PMHD) (10.9 and 11.8 in intervention and nonintervention villages, respectively) in pre-intervention assessment, followed by Anopheles culicifacies ( Table 2). The PMHD of Anopheles stephensi in the intervention village reduced significantly to 0.55 (94.95%) in the postintervention survey and to 0.22 (97.8%) in the further follow-up survey. The reduction of the density of Anopheles stephensi in the non-intervention village was moderate at 4.6 (61.1%) and 5.9 (50%) in post-intervention and follow-up surveys, respectively.

Results of the KAP study in the intervention village
A house-to-house survey was conducted before and after the lids' replacement, with questions about malaria breeding sites, mosquito larvae detection, and other topics. A total of 120 people from different households took part in the survey. The pre-intervention results showed that 67% (80/120) respondents were aware of malaria disease and 37% (44/120) knew that malaria spreads through the bites of female mosquitoes. Only 11% (13/120) respondents were aware that larvae were a part of a mosquito's life cycle, whereas 20% (24/120) of the respondents knew that mosquitoes breed in tankas. In follow-up conducted after one year, there was a significant improvement in the knowledge of respondents (n = 120). The awareness about malaria increased from 67 to 89% (Ҳ 2 = 17.65, significant at p < 0.001); knowledge about malaria spread through the bite of female mosquitoes improved from 37 to 68% (Ҳ 2 = 24.13, significant at p < 0.001); knowledge about larval form as a part of mosquito life cycle improved from 11 to 78% (Ҳ 2 = 110.65, significant at p < 0.001); and the awareness about the breeding of mosquitoes in tankas from 20 to 74% (Ҳ 2 = 70.66, significant at p < 0.001).

Discussion
As per the National Framework of Malaria Elimination in India 2016-2030, the state of Rajasthan is classified in category 1 because of the overall low API of the state being < 1/1000 [13]. Western Rajasthan, which was an unstable malaria zone before the establishment of the Indira Gandhi Canal in 1970, malaria has become a regular feature in its command areas, indicating a change in the malariological scenario [14].
Since the local population faced scarcity of potable water in this arid zone, they practised stringing rainwater in tankas for year-round use. Anopheles stephensi had adapted for breeding in tankas and beris (step wells) throughout the year [7]. Almost all houses in western Rajasthan use tankas for their daily water needs and long-term storage. Tanka is an important feature that improves the self-reliance of rural people, stabilizes rural employment skills, and is cost-effective [6]. With the increase in water supply through the Indira Gandhi Canal, the resident population has increased in just a few decades, and the number of tankas increased, providing enhanced mosquito breeding potential and increased risk of malaria transmission. Availability of water and improper lids allowed mosquito breeding in these tankas.
The PVC lids had a low cost (~ INR 400-500) as compared with other lids made up of iron (~ INR 700), aluminium (~ INR 1000) and wood (~ INR 600-1000). Normally, the wooden lids are used as a temporary arrangement to cover the open holes in the tankas and their cost may vary according to the quality used.
Anopheles stephensi, which is the major malaria vector of western Rajasthan, predominantly breeds in these tanka [7]. Community pre-awareness study (KAP) revealed that about 67% of individuals were aware of malaria as a disease spread by mosquitoes and about 44% knew it spread due to infective bites of female Anopheles. This shows a fair awareness of the community about the disease and its transmitting agent. However, 13% of the community did not know much about mosquito larvae and only 20% knew that breeding occurs in tankas (Fig. 5). Similar results were observed in a study conducted by Yadav et al. [15], which also showed that about 20% of respondents know that tanka is the main source of mosquito breeding and its proper coverage is essential in the prevention of mosquito breeding [16]. The results showed a significant improvement in the knowledge of the community (22% to 67%) about malaria, its transmission and breeding sites. The significant improvement may be due to the active engagement of villagers in the lid replacement process during the study. The knowledge gained by the community will help understand the importance of proper lids and mosquito proofing of tankas.
Damaged lids provide an opportunity for Anopheles stephensi to breed in these tankas. During hot summer seasons, when outdoor temperatures exceed 45 °C, the tankas become resting places for the mosquitoes to avoid desiccation. During the monsoon and post-monsoon seasons, Anopheles stephensi breeds profusely and boosts malaria transmission along with another vector species, i.e. Anopheles culicifacies. Anopheles culicifacies is a seasonal malaria vector of the Western Rajasthan that appears in the post-monsoon and winter seasons only. A study by Joshi et al. showed that Anopheles stephensi is a persistent malaria vector in the region and is supported by Anopheles culicifacies in the post-monsoon season [17].
Installation of improved polyvinyl lids has completely stopped breeding in these tankas. As a result, there was a drastic reduction in the density of Anopheles stephensi (94.9% in the post-intervention and 97.9% in the followup survey after one year) in the intervention village, Ajasar, as compared to the non-intervention village, Tota.
After one year of installation of improved lids their fitness was checked in May 2019. A total of 98% (196/200) replaced lids were found fit and tankas were found free from breeding. Furthermore, they had no rusting at hinge joints except for 4 lids, where two lids were found detached from cement binding, probably due to the use of lids before cement settling, and two had mosquito breeding, possibly due to holding lids open for a long period. There was some reduction in Anopheles stephensi density (11.8 to 5.9 PMHD) in the non-intervention village (Tota) as well during the post-intervention and follow-up surveys, which may be attributed to the impact of IEC on villagers about the importance of lids and mosquito breeding. According to the World Malaria Report 2019, almost 85% of the global malaria burden is distributed in nineteen countries in Sub-Saharan Africa and India [18]. India accounted for 47% of the Plasmodium vivax burden. Seven states of India viz. Jharkhand, West Bengal, Uttar Pradesh, Chhattisgarh, Odisha, Gujarat and Madhya Pradesh contributed a combined 90% of malaria cases. In Rajasthan, the incidence of Plasmodium falciparum was very low i.e. 8% of total malaria cases in Rajasthan in 2016, which increased to 13% in 2020. Although malaria cases have declined from 12,741 in 2016 to 674 in 2020 (up to September), the Plasmodium falciparum percentage increased [19]. The availability of vectors like Anopheles stephensi during the immigration of such people coming to the village either post-monsoon for crop purposes or for developmental activities might introduce parasites into the village system for transmission [17].

Conclusion
The current study has shown that small local interventions can efficiently solve an emerging problem of malaria. A significant reduction in the density of Anopheles stephensi was observed using mosquito-proof lids, addressing the problem of persistent mosquito breeding in underground tanks. The modified polyvinyl plastic lids are of low cost and have a life expectancy of approximately 2 to 5 years if used properly. Lids not only prevent the entry of mosquitoes in tankas, but also of other opportunistic insects and reptiles such as lizards, snakes, and rodents.
Rajasthan is a low malaria transmission area, falling under category 1 of the WHO's malaria elimination target for 2030. These novel measures could help to curb perennial malaria transmission by reducing the responsible vector, which may be responsible for year-round malaria transmission. Such simple long-term interventions are critical because they reduce routine intervention costs, reduce insecticide waste, and protect the environment.