Study area
The study was undertaken during September 2006 to August 2007 in 16 villages of district Kanker in Chhattisgarh state central India (Figure 1). Durability surveys were carried out after 18, 24, 30 and 36 months of net distribution (April 2008 to October 2009). District Kanker (pop. 763,549) is one of the malaria endemic districts in the state. It has an area of 5,285 km2 lying between latitudes 19°09’ and 20°06’ North and longitudes 80°30’ to 81°15’ East. About 60% of the area is hilly and forested. The district has a tropical climate with an average annual rainfall of 1,100 mm received between June and October in 60–70 rainy days. The villages are scattered and mostly inhabited by tribals belonging to Gond tribe of marginal socio-economic status. The houses are brick built with tiled or thatched roof and mud plastered walls and flooring. There are about 4–5 bedrooms including 1–2 separate rooms for cattle and fodder within the compound. Mud plastering is done generally for about six times in a year or more due to festivals and family occasions or celebrations. People generally sleep in rooms during all seasons and quite often the elder members sleep in verandah within the compound. Subsistence agriculture is the main occupation and rain-fed paddy is the principal crop. The Annual Parasite Index (API) of the district during 2002 to 2006 ranged from 23.5–39.1 cases/1,000 population per year. Proportion of P. falciparum infections during the corresponding period ranged between 89.4 and 95.3%. No death attributable to malaria was reported by the malaria control programme during the study period.
For the purpose of the study, villages under Amoda Community Health Centre (CHC) were selected. CHC Amoda provides health services to a population of about 0.13 million. The CHC has 36 sections each comprising of about 4,000 population living in three to five villages. Malaria surveillance is carried out by a male and a female Multi-Purpose Health Worker (MPHW-M/F) in each section by undertaking fortnightly door-to-door visits. Surveillance in the study villages was strengthened by deploying two trained local volunteers in each of the study arm. Blood slides were prepared from febrile cases and anti-malarials were administered to all microscopically confirmed malaria positive cases as per the national drug policy. Based on previous year’s malaria incidence data, villages with >2 API qualify to receive two rounds of indoor residual spraying (IRS) of insecticide. Under the national programme in district Kanker, synthetic pyrethroids have been used for IRS since mid-1990. No IRS was undertaken in villages selected for the trial during the first phase of the study (September 2006 to August 2007) and also during 2008. However, two rounds of IRS with alpha-cypermethrin 5% WP were given in study villages with >2 API during 2009.
Selection of study villages and distribution of nets
Village level malaria information for the year 2006 was retrieved from the CHC records and after careful consideration, 16 villages with varying degree of endemicity were identified for the study. Baseline data on malaria incidence and mosquito density were generated for three months from September to November 2006 from all villages before randomizing them into three arms, viz. Interceptor net (LN), untreated polyester net (UN) and no net. A population census was carried out in every village during the pre-intervention period and all the houses in the selected villages identified for net distribution were numbered. The census also revealed that about 1.13 nets/house were owned by the community. Interceptor nets, rectangular netting consisting of 75 denier multifilament, 100% polyester, with minimum 250 kPa bursting strength, mesh: 24 holes/cm2; density: 30 ± 2 g/m2; active ingredient: 200 mg/m2 alpha-cypermethrin coated on polyester fibers, warp knitting and untreated polyester nets (100% polyester, multifilament yarn: 100 denier strength with a minimum 405 kPa bursting strength, mesh: 24 holes/cm2; density: 30 ± 2 g/m2) were distributed free of cost during November 2006. A total of 932 Interceptor nets (one net per house) were distributed in five villages (population 5316; parasite incidence (PI)/1000 population-14.3); similarly, 909 untreated polyester nets were given in six villages (population 4802; PI-6.6) and another group of five villages (population 3865; PI-3.1) was kept as no net control area. Investigators ensured to cover all the target population in the villages for distribution of nets. Active surveillance was strengthened in all study villages by ensuring the adequate supply of materials with the MPHWs for better management of the disease during the trial period. Inhabitants were educated about the proper usage, storage, washing and up-keep of nets.
Ethical approval and informed consent
The study was approved by the Ethics Committee of the National Institute of Malaria Research, New Delhi, India. Villagers were informed about the aims and objectives of the study through house-to-house contact and group meetings. The written consent of each household was obtained for his/her willingness to participate in the study before distributing the nets. Information on various aspects of the study was read in the local language for obtaining their written consent.
Mosquito collections
Mosquito collections were made from four fixed dwelling rooms during the morning hours (0600–0800 hrs) in two villages in each of the three study arms once every month. Resting mosquitoes were caught using an aspirator and flash light for fifteen minutes each employing a trained person on mosquito collection. Employing two trained personnel, the room was then sprayed with pyrethrum solution (0.2% in kerosene) using hand held atomizers and after 15 min all mosquitoes found dead on the floor-sheets were collected and transferred into paper cups with moist cotton pad at the bottom to prevent desiccation. Mosquito species were identified and their abdominal condition was recorded. Unfed An. culicifacies were dissected for parity using the WHO procedure [12]. Baseline mosquito collection was carried out for three months before setting up sentinel rooms for organizing routine monthly monitoring. WHO tube test using 0.1% alpha-cypermethrin impregnated papers (diagnostic dose) was conducted to determine the susceptibility of wild caught An. culicifacies in the beginning of the study.
Assessment of usage pattern, net washing behaviour and physical aspects of nets
Using a structured pre-tested questionnaire, adult households were interviewed to assess the net usage pattern, net washing behaviour and physical integrity of the net. In all, four surveys were carried out at six-monthly intervals at 18, 24, 30 and 36 months post-net distribution in Interceptor villages. In the first three surveys, questionnaire was administered to about 10% of the households selected randomly. Houses once surveyed were excluded from the subsequent surveys. However, in the survey carried out after three years of net distribution, all houses which were initially given the Interceptor LN were surveyed. The houses were checked for physical possession of the nets.
All the nets were examined for the presence, size (by measuring diameter in cm) and location of holes, tears, burns, open seams, stitches and knots. Based on the size, holes were grouped into four categories, viz. Size 1: smaller than a thumb (0.5–2 cm), Size 2: larger than a thumb but smaller than a fist (2–10 cm), Size 3: larger than a fist but smaller than a head (10–25 cm) and Size 4: larger than a head (>25 cm) and respectively weighed as 1, 23, 196 and 576 [13]. Hole index for each net was calculated as: (1 × no. of size − 1holes) + (23 × no. of size − 2holes) + (196 × no. of size − 3holes) + (576 × no. of size − 4holes).
Wash resistance and bio-efficacy of interceptor LNs
Four Interceptor LNs and two untreated nets were washed 20 times in the laboratory. Nets were washed separately and individually in plastic tubs. A teaspoon full of Surf Excel® (washing powder, M/s. Hindustan Unilever Ltd, Mumbai, India) was dissolved in five litres of tap water (26 ± 2°C) and nets were rinsed four to five times and, thereafter two to three times in clean water before hanging on laundry line under shade for drying for 24 h. Nets were stored under the optimal room temperature and humidity between the washes. Bioassays were conducted after two to three days post washing. On each net a cone was fixed and held in place using a rubber band. Laboratory-reared An. culicifacies mosquitoes (sugar fed, 2–5 day old), susceptible to alpha-cypermethrin were introduced into each cone and exposed for 3 min [14]. This was repeated twice on each of the four cones fixed on four nets. Thus, a total of 60 and 30 mosquitoes were exposed on Interceptor LNs and untreated polyester nets respectively. Number of mosquitoes knocked-down after one hour and mortality after 24 hour of exposure were recorded. The WHO criteria for efficacy is ≥95% knock-down after one hour and ≥80% mortality after 24 hour. Abbott’s formula was applied for correcting the mortality in test replicates if the mortality in control replicates was between 5 and 20%. If the mortality in control replicates was >20%, the tests were discarded.
Sampling for chemical analysis
After 36 months of household use in the field, 30 Interceptor LNs were retrieved from thirty households randomly selected from all the study villages. New Interceptor LN was provided in lieu of withdrawn net. From each of the 30 LNs, four pieces of 30 cm × 30 cm size were cut from positions 2 to 5 as per WHO protocol using sharp scissors. Four sub-samples were assembled as one sample, rolled up and placed in a new, clean aluminium foil, labeled and stored at +4°C prior to dispatch to Walloon Agricultural Research Centre (CRA-W), Gembloux, Belgium for chemical assay which is a WHO collaborating centre. Sample from position 1 was not cut since netting fabric at this position is subjected to excessive abrasion in routine use (this portion of net is frequently manipulated while tucking the nets under the bed/mattress). Analytical method RESSM016 of the Walloon Agricultural Research Centre (CRA-W) was used. The analytical method was already validated and is ISO 17025 accredited for alpha-cypermethrin in coated LNs. Results were expressed as g alpha-cypermethrin (cis II) / kg and converted to mg alpha-cypermethrin (cis II)/m² using the density [15].
Statistical analysis
All the data were entered in MS Excel and statistical analysis was performed using Excel programme. Categorical variables were compared using Chi-square test and continuous discrete variables were tested using Analysis of variance (ANOVA) to test the significance among the three study areas. A p-value <0.05 was considered as significant.