SEARCH

SEARCH BY CITATION

Keywords:

  • dengue;
  • Aedes aegypti;
  • insecticide-treated material;
  • insecticide-treated curtain;
  • use;
  • vector control
  • dengue;
  • Aedes aegypti;
  • matériel traitéà l’insecticide;
  • rideaux imprégnés d’insecticide;
  • utilisation;
  • lutte antivectorielle
  • Dengue;
  • Aedes aegypti;
  • material impregnado con insecticida;
  • mosquiteras impregnadas;
  • uso;
  • control vectorial

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objective  To evaluate the acceptance and long-term use of insecticide-treated (IT) materials for dengue vector control.

Methods  In 2007, IT jar covers and/or curtains (PermaNet®) were distributed under routine conditions to 4101 households (10 clusters) in Venezuela and to 2032 households (22 clusters) in Thailand. The use of IT tools was measured at distribution (uptake), at 5/6 months (short-term use) and at 18/22 months (continued use) after distribution. Determinants of use were assessed with logistic regression analysis.

Results  The uptake of IT curtains was 76.7% in Venezuela and 92.3% in Thailand. It was associated with being a resident for >5 years (OR Venezuela 3.0 95% CI 2.0–4.4; OR Thailand 3.5 95% CI 1.7–7.3) and with pre-intervention use of ordinary curtains (OR Venezuela 2.2 95% CI 1.4–3.6). The continued use decreased significantly to 38.4% of households in Venezuela and 59.7% in Thailand and was, conditional on short-term use, only determined by the perceived effectiveness of IT curtains (OR Venezuela 13.0 95%CI 8.7–19.5; OR Thailand 4.9 95% CI 3.1–7.8). Disease knowledge and pre-intervention perception of mosquito nuisance were not associated with IT curtains’ uptake or use. The uptake of IT jar covers in Venezuela was 21.5% and essentially determined by the presence of uncovered jars in the household (OR 32.5 95% CI 14.5–72.6). Their continued use, conditional on short-time use, was positively associated with the household use of Abate® (OR 7.8 95% CI 2.1–28.9).

Conclusion  The use of IT curtains rapidly declines over time. Continued use is mainly determined by the perceived effectiveness of the tool. This poses a real challenge if IT curtains are to be introduced in dengue control programmes.

Objectif:  Evaluer l’acceptation et l’utilisation à long terme de matériaux traités à l’insecticide pour le contrôle du vecteur de la dengue.

Méthodes:  En 2007, des couvre-jarres et/ou des rideaux (PermaNet®) imprégnés d’insecticide ont été distribués dans des conditions de routine à 4101 ménages (10 groupes) au Venezuela et à 2032 ménages (22 groupes) en Thaïlande. L’utilisation des matériaux traités à l’insecticide a été mesurée à la distribution (adoption), à 5/6 mois (utilisation de court term) et à 18/22 mois (utilisation continue) après la distribution. Les déterminants de l’utilisation ont étéévalués par analyse de régression logistique.

Résultats:  L’adoption des rideaux imprégnés d’insecticide était de 76,7% au Venezuela et 92,3% en Thaïlande. Elle était associée au fait d’être un résident de plus de 5 ans (OR Venezuela: 3,0; IC95%: 2,0 4,4; OR Thaïlande: 3,5; IC95%: 1,7–7,3) et à l’utilisation pré-intervention de rideaux ordinaires (OR Venezuela: 2,2; IC95%: 1,4–3,6). L’utilisation continue a diminué de manière significative à 38,4% dans les ménages au Venezuela et 59,7% en Thaïlande et était seulement déterminée par l’efficacité perçue des rideaux imprégnés d’insecticide (OR Venezuela: 13,0; IC95%: 8,7–19,5; OR Thaïlande: 4,9; IC95%: 3,1–7,8), conditionnée par l’utilisation à court terme. Les connaissances sur la maladie et les perceptions pré-intervention de la nuisance des moustiques n’étaient pas associées à l’adoption ou à l’utilisation des rideaux imprégnés. L’adoption des couvre-jarres au Venezuela était de 21,5% et essentiellement déterminée par la présence de jarres découvertes dans le ménage (OR: 32,5; IC95%: 14,5–72,6). Leur utilisation continue était positivement associée à l’usage domestique de Abate® (OR: 7,8; IC95%: 2,1-28,9), conditionnée par l’utilisation à court terme.

Conclusion:  L’utilisation de rideaux imprégnés d’insecticide diminue rapidement au fil du temps. L’utilisation continue est principalement déterminée par l’efficacité perçue de l’outil. Cela pose un véritable défi lorsque les rideaux imprégnés doivent être introduits dans des programmes de lutte contre la dengue.

Objetivo:  Evaluar la aceptación y el uso a largo plazo de materiales impregnados con insecticida (ImIn) para el control del dengue.

Métodos:  En el 2007, se distribuyeron bajo condiciones rutinarias tapas de contenedores de agua y/o mosquiteras (PermaNet®) impregnados con insecticida en 4101 hogares (10 conglomerados) de Venezuela y en 2032 hogares (22 conglomerados) de Tailandia. El uso de herramientas ImIn se midió en el momento de la distribución (acogida), a los 5/6 meses (uso a corto plazo) y a 18/22 meses (uso continuado) tras la distribución. Los determinantes de uso fueron evaluados mediante un análisis de regresión logística.

Resultados:  La acogida de mosquiteras ImIn era del 76.7% en Venezuela y del 92.3% en Tailandia. Estaba asociada con el ser residente durante >5 años (OR Venezuela 3.0 IC 95% 2.0–4.4; OR Tailandia 3.5 IC95% 1.7–7.3) y con el uso pre-intervención de mosquiteras ordinarias (OR Venezuela 2.2 IC95% 1.4–3.6). El uso continuado disminuyó significativamente al 38.4% de los hogares en Venezuela y a 59.7% en Tailandia y estaba, condicionado sobre el uso a corto plazo, solo determinado por la percepción de la efectividad de las mosquiteras impregnadas (OR Venezuela 13.0 IC95% 8.7–19.5; OR Tailandia 4.9 IC95% 3.1–7.8). El conocimiento de la enfermedad y la percepción pre-intervención de lo molesto que es el mosquito no estaban asociados con la acogida o el uso de las mosquiteras impregnadas. La acogida de las tapas de contenedores impregnados era del 21.5% y estaba esencialmente determinada por la presencia en el hogar de contenedores sin tapa (OR 32.5 IC95%14.5–72.6). Su uso continuado, condicionado al uso a corto plazo, estaba positivamente asociado con el uso en el hogar de Abate® (OR 7.8 IC 95% 2.1–28.9).

Conclusión:  El uso de las mosquiteras impregnadas disminuye a lo largo del tiempo. El uso continuo está determinado principalmente por la percepción de la efectividad de esta herramienta. Esto plantea un reto real para las mosquiteras impregnadas para ser introducida en los programas de control del dengue.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

An estimated 40% of the world population lives at risk of contracting dengue, the most important mosquito-borne viral disease. Worldwide, it causes 24 000 deaths, 250 000–500 000 haemorrhagic fever cases and up to 50 million infections annually (Gibbons & Vaughn 2002; Farrar et al. 2007). To decrease the burden of disease, prevention of transmission is crucial. No curative treatment is available and the avoidance of lethality in severe dengue cases hinges on early case detection and supportive treatment. In the absence of a vaccine, control of Aedes aegypti, the main dengue vector, is the only available strategy to control disease transmission (Farrar et al. 2007). However, most of the existing control methods and strategies against A. aegypti do not succeed in sustaining vector infestation to levels below transmission thresholds (TDR 2006).

The most successful A. aegypti control experiences rely on integrated community-based approaches tailored to local eco-epidemiological settings (Toledo Romani et al. 2007; Erlanger et al. 2008; Sanchez et al. 2009; Vanlerberghe et al. 2009a; Kay et al. 2010). Insecticide-treated materials (ITMs) such as window curtains and jar covers have potential for reducing dengue vector densities (Kroeger et al. 2006; Seng et al. 2008; Vanlerberghe et al. 2009b). Unlike most current control strategies, ITM target the adult mosquito, the epidemiologically most important stage. ITM made from long-lasting insecticide-treated fabrics are expected to remain efficacious for longer periods than other chemical control measures (such as larvicides or insecticide spraying), and would require less frequent distribution or application. The first efficacy trials reported a high acceptance by households at distribution (Kroeger et al.2006; Seng et al. 2008) and it was suggested that maintaining high coverage might be possible without a need for obtaining major changes in user behaviour. However, whether uptake would also be high under routine programme conditions and remain high in the medium term are still unanswered questions.

The reported uptake of other A. aegypti control tools is often disappointing. The acceptability of larval control in domestic-water containers varies widely. In Thailand, only 25% of households used Abate® (Phuanukoonnon et al. 2005), and in Mexico, only 29% of containers contained Pyriproxifen 2 weeks after distribution (Kroeger et al. 2006). On the other hand, with active community involvement, Mesocyclops remained correctly applied in 80% of large containers in Northern Vietnam (Kay et al. 2010) and 88% of containers remained well covered in Cuba (Toledo Romani et al. 2007). Chemical control of adult A. aegypti by space and peri-domestic spraying has a high visibility but limited acceptability (Renganathan et al. 2003). In contrast, lethal ovitraps were well accepted, with less than 9% missing after 4 weeks in an efficacy-trial in Australia (Ritchie et al. 2009).

To assess the acceptance of IT curtains and IT jar covers for dengue control and to study their continued use and its determinants, we followed, over a period of up to 22 months, communities in Thailand and Venezuela where these tools had been distributed in 2007 under routine conditions.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study setting

We independently conducted community interventions in a provincial city in Venezuela and a port city in Thailand.

In Venezuela, the study was conducted in and around Valera (9°19′N 70°36′W; altitude 541 m), the economic capital of Trujillo State, North West Venezuela, with a population of 172 769. The climate is tropical with two rainy seasons (March/April and September/November), an average annual rainfall of 750 mm and average daily temperatures ranging from 23 to 30 °C. Dengue is endemic in Trujillo State. From 2006 to 2008, dengue case reports ranged between 203 and 396 cases/100 000 inhabitants/year. All routine A. aegypti vector control activities are carried out by a team of 24 persons from the department of environmental health of the provincial Health Ministry. Activities comprise adulticiding (indoor spraying with malathion) and larviciding (Abate®) in a radius of 200 m around a reported dengue case. When the number of clinical cases crosses a threshold, space spraying is added.

In Thailand, the study was conducted in Laem Chabang (13°06′N 100° 54′ E; altitude 25 m), Chon Buri province, 100 km South East of Bangkok. In this major port city 42 480 households were registered in July 2005, but Laem Chabang attracts many seasonal workers from other Thai regions. The climate is tropical with the heaviest rains between May and October, the average annual rainfall is 1600 mm and the average daily temperatures range from 26 to 29 °C. Dengue is endemic and from August 2006 to July 2007 a total of 90 dengue cases were reported by the local health authorities (approximately 112 cases/100 000 inhabitants). All routine A. aegypti vector control activities are carried out by a team of five persons from the Municipal Government, together with 110 village health volunteers and with occasional support of the Municipal hospital team. Every year, two to three campaigns of spatial spraying (with deltamethrin) and of household distribution of the larvicide Abate® are planned, covering the whole municipality of Laem Chabang. Additionally, all houses within a 100 m radius of a reported dengue index case are sprayed with deltamethrin.

Intervention

In Venezuela, the intervention took place in 10 study clusters of 300–600 houses (corresponding to the most decentralized administrative unit) selected in the districts that had dengue notification rates of 40/10 000 inhabitants or more in the period 2003–2005. A total of 4101 houses were covered. The inclusion criteria at cluster level were: middle or low socio-economic status (the number of high socio-economic level clusters was small and they were not representative of the area); less than 50% of the population residing in apartment blocks (for operational reasons) and no rural characteristics (dengue is not a health problem in these clusters). In Thailand, 22 clusters of 80–110 houses (defined by infrastructural boundaries), totalling 2032 households, were selected in four town districts, assuring an at least one street distance between clusters.

Both IT jar covers (depending on the number of water containers, no maximum of covers/house fixed) and IT curtains (depending on the number of windows in living and bedrooms, with a maximum of five curtains/house) were distributed free of charge between July and September 2007 to all households of the intervention clusters in Venezuela and IT curtains alone in March 2007 in Thailand. The IT curtains and IT covers were made from PermaNet® (Vestergaard-Frandsen) polyester netting treated with a long-lasting formulation of deltamethrin (55 mg/m2), coated with an unknown protectant to prevent degradation of the insecticide when exposed to UV light. The manufacturer stated that this material does not require re-treatment and its insecticidal effect is expected to last for up to 2 years or 6 ‘standard’ washes (http://www.vestergaard-frandsen.com/permanet-curtain-e-brochure.pdf, accessed 22/05/2008).

In Venezuela, the tools were implemented through two distribution channels in five clusters each: (1) the routine vector control programme followed the pattern of routine Abate® distribution, covering all five clusters in five working days with three teams of five men; (2) the local health committees (‘Comité de Salud’, an existing structure attached to local health centres) mobilized their members, local leaders and volunteers to distribute the tools, over a period of 9 weeks, through house visits during evening hours. Promotion and explanation on use and maintenance of tools was, in both channels, based on person-to-person communication during distribution. In the follow-up phase, no additional activities were organized in any of the distribution groups.

In Thailand, the village health volunteers, supervised by the municipal vector control programme and a team of Chon Buri’s regional disease control office, provided before distribution to every household in the 22 selected clusters a leaflet explaining the use and maintenance of IT curtains. Taking note of the possible side effects, inhabitants in some clusters asked for the tools to be used for at least 1 day in the local leader’s house, to ascertain their safety. This request was honored and afterwards all houses were revisited to distribute the tools. A check-up visit was made 1 week after distribution, and if IT curtains were not hanged up, households were encouraged to do so and assisted if necessary.

Data collection

In both countries we had collected pre-intervention information on the characteristics of the households in the intervention clusters. In Venezuela, we followed this up in February 2008, 6 months after distribution of the tools, with a household survey in a systematic random sample of 782 houses across all clusters (every fifth house). In June 2009, 22 months after distribution, these houses were revisited. In Thailand, the first follow-up survey was conducted in a systematic random sample of 562 houses across all clusters (every fourth house) in August 2007, 5 months after IT curtain distribution. In October 2008, 18 months after distribution, these houses were revisited. The follow-up samples had 80% power to detect a 15% decrease in the use of IT tools (from 80% initial coverage) at an alpha error level of 0.05, taking into account a design effect of 2.5 (Henderson & Sundaresan 1982).

We conducted interviews with the head of household — or, if absent, an adult family member — using semi-structured questionnaires and observation guides pre-tested and adapted to the local characteristics of each country. 75.4% and 68.0% of respondents were female in Venezuela and Thailand, respectively. Pre-intervention data included: socio-economic and demographic characteristics, history of clinical dengue cases in the household, knowledge on dengue and its transmission, use of Aedes and other vector control methods and perception of mosquito nuisance. In Venezuela, the socio-economic level of the households was classified based on the profession of the head of household, education level of the mother, main source of family income and housing conditions (Méndez-Castellano & Méndez 1994). In Thailand, three variables were used to estimate the socio-economic level: house characteristics, salary income and education level of head of household. Knowledge on clinical dengue was evaluated with three questions on signs and severity of infection; on the vector with three questions on species, biting and breeding habits of vector; on prevention by asking for effective Aedes control methods. Correct answers received one point. A variable ‘Knowledge on Dengue’ was constructed: six or more on a total maximum score of nine points was categorized as good knowledge on dengue, less than six as limited knowledge.

In the post-distribution surveys, we observed whether the IT tools were still used by the households. Acceptance, a concept involving a complex set of individual and socio-cultural factors, was defined by Severy et al. (2005) as ‘voluntary sustained use of a method in the context of alternatives’. Apart from actual use observed, we explored acceptance through various proxies that have shown their usefulness in other studies (Brieger et al. 1996; Gyapong et al. 1996; Kachur et al. 1999; Tami et al. 2004; Kroeger et al. 2006). We asked the following (hypothetical) questions: ‘If IT curtains were to be sold at the local market, would you be willing to pay for them?’; ‘Are you willing to recommend the IT curtains to your relatives/friends?’ (Venezuela); ‘Are you willing to receive other IT curtains if they were provided for free?’ (Thailand). The care for tools (washing and repairing) was recorded; their cleanliness was observed by the interviewer and assessed as ‘clean or quite clean’ and ‘dirty or very dirty’ according to local standards. The condition of IT curtains and of IT covers was categorized as ‘in bad condition’ if at least 1 hole of 5 × 5 cm2; or 1 tear of at least 5 cm was observed or, for covers, if the elastic rim was torn and hence could no longer seal hermetically the jar. Additionally, data were also collected on: perceived effectiveness (‘Are there fewer mosquitoes in your house since you use curtains?’), adverse effects attributed to the use of ITM and reasons for and barriers to the use of the tools.

Data analysis

Data were entered into an Access (Microsoft Office Access 2003) database. A random check of 5% of records, comparing paper records with electronic data, was performed by the first author in Venezuela and by the third author in Thailand. Analysis was performed with spssvs.18.0 software (SPSS Inc., Chicago, IL, USA).

Uptake and use of IT curtains and IT covers were the outcome measures and were, according to the moment of observation, defined as accepting to receive (at distribution) or observed to actually use (at 5/6 months and 18/22 months post-distribution) at least one curtain or cover per household. This is further referred to as uptake, short-term use and continued use, respectively. For analyzing the determinants of these outcomes we included, at the time points above, all surveyed households, households that received the tools and households using them at the short term, respectively.

We first performed bivariate analysis with all relevant independent variables. Odds ratios (OR) and their 95% confidence intervals (CI) were calculated. We then constructed, separately for Thailand and for Venezuela, binary logistic General Estimating Equations models with the outcome measure as dependent variable. This type of model takes into account the clustering of data (Donner & Klar 2000). We built models from the variables that were significant in bivariate analysis and these non-significant variables that were deemed important or potential confounders on a priori grounds (Kleinbaum et al. 1998). Possible interactions between variables were assessed. Adjusted OR with 95% CI were computed.

Ethical considerations

The study received clearance from the ethical committee that oversees research of the Institute of Tropical Medicine, Antwerp, from the bio-ethical committee of the Jose Witremundo Torrealba Research Institute, Trujillo and from the ethics committee of the Faculty of Tropical Medicine, Mahidol University, Bangkok. Community representatives of each participating cluster approved the intervention and written informed consent was obtained from each individual household head included in the study. The tools used were made from insecticide-treated material that is approved by the World Health Organization Pesticide Evaluation Scheme (WHOPES) for bed net use. The study was registered at ClinicalTrials.gov (number NCT 00883441).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Venezuela

The general characteristics of the study population are described in Table 1. In 29.4% of households, interviewees reported a history of dengue in the family. The majority of households had a good knowledge on dengue and reported conducting dengue control activities in the peri- and intra-domestic area.

Table 1.   General pre-intervention characteristics of the study population Valera, Venezuela and Laem Chabang, Thailand (2006)
VariablesVenezuela = 782 householdsThailand = 562 households
  1. Values denote n (%).

  2. n.a.: not available in the locally adapted version of the questionnaire.

  3. *Determined by the Graffar method (see ‘Methods’).

  4. †Jar: large water storage container (volume of >200 l).

History of dengue in the household (ever)230 (29.4)97 (17.3)
Socio-economic level*
 Stratum I (highest level)16 (2.0)n.a.
 Stratum II101 (12.9)
 Stratum III223 (28.5)
 Stratum IV417 (53.3)
 Stratum V (lowest level)25 (3.2)
Salary income in household (yes)n.a.195 (34.7)
Education level head of household
 No or primary schooln.a.246 (43.8)
 Secondary school198 (35.2)
 Higher education118 (21.0)
House characteristics
 Single housen.a.174 (31.0)
 Row house247 (44.0)
 Townhouse/apartment141 (25.1)
Urban area (yes)390 (49.9)562 (100)
Duration of residency(9 missing data)(34 missing data)
 ≤5 years115 (14.7)240 (42.7)
 >5 years658 (84.1)288 (51.2)
Number of household members
 1–2 members140 (17.9)166 (29.5)
 3–5 members428 (54.7)316 (56.2)
 6 or more members214 (27.4)80 (14.2)
Ordinary curtains used (yes)592 (75.7)390 (69.4)
Window/door screens (yes)60 (7.7)266 (47.3)
Air conditioning (yes)191 (24.4)64 (11.4)
Knowledge on dengue
 Limited knowledge175 (22.4)83 (14.8)
 Good knowledge607 (77.6)479 (85.2)
Nuisance of mosquitoes in house (yes)554 (70.8)327 (58.2)
At least two practices to prevent dengue in last year (yes)661 (84.5)353 (62.8)
Use of Abate in last month (yes)60 (7.7)n.a.
Use of Abate in last year (yes)n.a.215 (38.3)
Used mosquito control methods (all mentioned)
 Natural repellents49 (6.3)25 (4.4)
 Spraying with commercial insecticides283 (36.2)264 (47.0)
 Bed nets6 (0.8)98 (17.4)
 Cleaning of intra- and peridomestic area423 (54.1)
 Electric vaporizer/mosquito coil94 (12.0)94 (16.7)
 Cleaning water storage containers402 (51.4)
 Window screens233 (41.5)
 Electric fan284 (50.5)
 Electric stick106 (18.9)
Frequency of water supply
 Permanent or each day427 (54.6)n.a.
 Each 2 days266 (34.0)
 Less frequent89 (11.4)
Storing water in household (yes)n.a.511 (90.9)
Presence of at least one water storage jar† (yes)373 (47.7)340 (60.5)
Observed correct covering of water storage containers(yes)492 (62.7)n.a.
Presence of uncovered water storage containers (yes)148 (18.9)n.a.

Tools were distributed to 48.5% of households by the routine vector control programme and to 51.5% by local health committees.

Curtains.

At distribution, mid-2007, 76.7% (95% CI 73.7–79.7) of the households accepted to receive and intended to use the curtains (on average 4.2 curtains/house; SD 2.2); the attained coverages were independent of the distribution channel (= 0.28). Six months after reception, at least one curtain was observed hanging in 64.3% (95% CI 60.9–67.7) of houses (on average 3.6 curtains/house; SD 2.0); 22 months after reception 38.4% (95% CI 34.2–42.6) of houses had still at least one curtain in use (on average 3.0 curtains/house; SD 2.0) (Figure 1). For the last survey, 274 (35.0%) households were lost to follow up because people moved out or were not at home at the time of survey.

image

Figure 1.  Coverage of IT curtains and IT jar covers (% of households with at least one tool in use) and 95% CI. Valera, Venezuela and Laem Chabang, Thailand (2007–2009).

Download figure to PowerPoint

Households that received curtains, but where no curtains were in use 22 months after distribution, forwarded as main reasons: damaged or dirty (73.4%), stored to alternatively replace ordinary curtains (10.8%), not effective (8.9%) and minor allergy after handling (1.3%). Positive reasons for continued use were: curtains impeding the entry of mosquitoes into the house (88.4%), effect on mosquito abundance (7.6%), function as dengue control tool (2.3%) and protection from dust and/or sun (1.7%).

Of the 38.4% of households still using their curtains after 22 months, 85.5% of these remained in good condition (Table 2) and most curtains (81.3%) were clean/quite clean. After 6 and 22 months, 43.7% and 94.4% of the households had washed at least one curtain, respectively. In 6.5% of households, the curtains were sewed to fit the window size.

Table 2.   Observations on IT curtains still used by Venezuelan and Thai households for 22 and 18 months, respectively (2009)
 Venezuela = 570 curtainsThailand = 555 curtains
  1. Values denote n (%).

  2. *Bad condition: a curtain with at least 1 hole of 5 × 5 cm2 or 1 tear of at least 5 cm.

  3. †IT curtain placed over open partitions of inner walls, used to cover an open cupboard or another place where mosquitoes are resting.

Condition of IT curtains
 Good condition502 (85.5)530 (95.5)
 Slightly damaged, repaired30 (5.1)6 (1.1)
 Slightly damaged, not repaired17 (2.9)16 (2.9)
 Bad condition*38 (6.5)3 (0.5)
Place in the house
 Outside door29 (4.9)182 (32.8)
 Inside door7 (0.1)28 (5.0)
 Window534 (91.0)296 (53.3)
 Other†17 (2.9)49 (8.8)
Hanging condition
 Open465 (79.2)430 (77.5)
 In a knot/curled up/put aside122 (20.8)125 (22.5)
Dirtiness of the IT curtains
 Clean/quite clean477 (81.3)480 (86.5)
 Dirty/very dirty110 (18.7)75 (13.5)

After 6 and 22 months of use 78.8% and 59.1% of the respondents, respectively, would recommend the IT curtains to relatives/friends. In the last survey, 35.3% of the households were willing to pay for new curtains (if they were sold at the local market).

The univariate association between outcome measures and their potential determinants is quantified in Table 3. In the logistic regression models (Table 4), already having ordinary curtains, residing for more than 5 years in the area and not yet practicing two dengue prevention activities at household level are relatively weak independent determinants of IT curtain uptake. Short-term use is most strongly determined by perceived effectiveness, but also by socio-economic group and having ordinary curtains or air-conditioning. Continued use is only and strongly associated with the perceived effectiveness of the curtains.

Table 3.   Association of potential determinants with uptake, short-term and continued use of IT curtains (crude OR and 95% CI). Venezuela and Thailand (2007–2009)
Independent variablesUptakeShort-term use*Continued use†
Venezuela (= 782)Thailand (= 562)Venezuela (= 600)Thailand (= 519)Venezuela (= 343)Thailand (= 262)
  1. n.a.: not available in the locally adapted version of the questionnaire.

  2. *Conditional on uptake.

  3. †Conditional on short-term use.

Nuisance of mosquito in house (pre-intervention)
 Yes1.1 (0.8–1.6)1.4 (0.73–2.5)1.6 (1.0–2.6)1.3 (0.8–2.1)1.2 (0.7–1.9)1.9 (1.1–3.2)
 No111111
Socio-economic level
 Stratum V (lowest level)1.0 (0.2–4.5)n.a.1.3 (0.3–5.7)n.a.0.2 (0.0–2.0)n.a.
 Stratum IV1.2 (0.4–4.0)1.0 (0.3–3.5)0.1 (0.0–1.0)
 Stratum III1.0 (0.3–3.2)0.7 (0.2–2.5)0.1 (0.0–0.9)
 Stratum II0.9 (0.3–2.9)0.8 (0.2–2.9)0.1 (0.0–1.1)
 Stratum I (highest level)111
House characteristics
 Row housen.a.1.8 (0.7–4.3)n.a.1.7 (0.9–3.4)n.a.1.1 (0.6–2.0)
 Town house/apartment0.3 (0.1–0.7)0.3 (0.1–0.5)0.4 (0.2–0.6)
 Single house111
Salary income in household
 Non.a.1.7 (0.9–3.2)n.a.2.0 (1.2–3.3)n.a.1.0 (0.6–1.8)
 Yes111
Ordinary curtains used
 Yes2.2 (1.5–3.1)1.1 (0.6–2.1)2.2 (1.4–3.5)1.8 (1.1–3.0)1.6 (0.9–2.9)1.7 (1.0–3.0)
 No111111
Air conditioning
 Yes1.4 (0.9–2.0)0.8 (0.3–1.9)1.4 (0.8–2.3)0.7 (0.3–1.4)0.8 (0.5–1.3)0.9 (0.4–2.0)
 No111111
Duration of residency
 >5 years2.8 (1.8–4.2)4.1 (1.9–8.8)1.1 (0.6–2.1)1.2 (0.7–1.9)0.5 (0.2–1.1)0.9 (0.5–1.5)
 ≤5 years111111
Urbanization
 Suburban0.7 (0.5–1.0)n.a.0.8 (0.5–1.3)n.a.1.2 (0.8–1.9)n.a.
 Urban111
Knowledge on dengue
 Good1.0 (0.7–1.6)0.9 (0.4–2.5)0.8 (0.5–1.4)1.6 (0.8–3.0)1.3 (0.8–2.2)1.1 (0.5–2.3)
 Limited111111
At least two practices to prevent dengue (pre-intervention)
 Yes0.5 (0.3–0.8)1.4 (0.7–2.6)0.9 (0.5–1.6)1.1 (0.6–1.8)0.8 (0.4–1.3)1.2 (0.7–2.1)
 No111111
History of dengue prior to corresponding survey
 Yes1.3 (0.8–1.8)1.3 (0.5–3.2)0.7 (0.3–2.0)1.8 (0.4–7.8)1.1 (0.5–2.5) 
 No11111
Perceived effectiveness prior to corresponding survey
 Yes  23.1(13.1–40.8)5.5 (2.6–11.4)12.7 (7.6–21.4)4.6 (2.5–8.8)
 No1111
Allergic reaction to tools after distribution
 Yes  0.9 (0.4–2.1)0.9 (0.3–2.5)1.1 (0.5–2.6)0.7 (0.3–1.7)
 No1111
Table 4.   Determinants and confounding factors of household IT curtains uptake, short-term and continued use (adjusted OR, 95% CI), conditional on use at previous time point, Venezuela and Thailand (2007–2009)
Independent variablesUptakeShort-term use†Continued use‡
Venezuela (= 782)Thailand (= 562)Venezuela (= 600)Thailand (= 485)Venezuela (= 343)Thailand (= 262)
  1. Estimated with logistic regression models (see ‘Methods’).

  2. n.a.: not available in the locally adapted version of the questionnaire.

  3. *Significant at 0.05 level.

  4. †Conditional on uptake.

  5. ‡Conditional on short-term use.

Nuisance of mosquitoes in house (pre-intervention)
 Yes     1.8 (0.9–3.5)
 No1
Socio-economic level
 Stratum V (lowest level) n.a.17.0(1.7–173.8)*n.a. n.a.
 Stratum I (highest level)1
House characteristics
 Row housen.a. n.a. n.a.1.0 (0.5–1.9)
 Town house/apartment  0.5 (0.2–1.2)
 Single house1
Salary income in household
 Non.a.1.6 (1.0–2.4)n.a.2.2 (1.2–4.1)*n.a. 
 Yes11
Ordinary curtains used
 Yes2.2 (1.4–3.6)* 1.7 (1.0–2.9)*   
 No11
Air conditioning
 Yes  1.8 (1.0–3.3)*   
 No1
Duration of residency
 >5 years3.0 (2.0–4.4)*3.5 (1.7–7.3)*    
 ≤5 years11
Urbanization
 Suburban n.a.0.6 (0.4–1.0)n.a. n.a.
 Urban1
At least two practices to prevent dengue (pre-intervention)
 Yes0.4 (0.2–0.9)*     
 No1
Perceived effectiveness prior to corresponding survey
 Yes  25.4 (15.4–41.7)*5.3 (2.8–10.0)*13.0 (8.7–19.5)*4.9 (3.1–7.8)*
 No   1 1 11
Allergic reaction to tools after distribution
 Yes     0.3 (0.1–1.1)
 No  1
Jar covers.

Mid-2007, 21.5% (95% CI 18.6–24.4) of the households accepted IT jar covers (on average 1.6 covers/house; SD 0.9); after 6 months, 16.4% (95% CI 13.8–19.0) (on average 1.5; SD 0.8) and at 22 months 9.6% of households (95% CI 7.0–12.2) (on average 1.4; SD 0.8) still used at least one cover. The main reason evoked for non-continued use of the covers was that they became damaged or dirty (93.0%). Of the covers that were still in use 6 and 22 months after distribution 94.5% and 69.4% were observed to be in good condition, respectively.

In the logistic regression models, the uptake and short-term use of IT covers are mainly determined by the presence of uncovered jars in the house at distribution. The frequency of water supply (lower in suburban area) and being a resident for more than 5 years (Table 5) are additional determinants of short-time use. Using Abate® in the month preceding the survey was the only determinant of continued use of jar covers: of the 79 households only 32 (40.5%) not using Abate® were using IT covers, against 3 of the 4 households using Abate® (adjusted OR 7.8; 95% CI 2.1–28.9).

Table 5.   Determinants and confounding factors of household IT jar-covers uptake, short-term and continued use (adjusted OR, 95% CI), conditional on use at previous time point, Venezuela (2007–2009)
Independent variablesUptake (= 782)Short-term use† (= 171)Continued use‡ (= 83)
  1. Estimated with logistic regression models (see Methods).

  2. *Significant at 0.05 level.

  3. †Conditional on uptake.

  4. ‡Conditional on short-term use.

Presence of uncovered jars (pre-intervention)
 Yes32.5 (14.5–72.6)*74.0 (29.1–188.4)* 
 No 1 1
Nuisance of mosquitoes in house (pre-intervention)
 Yes  1.8 (0.4–7.0)
 No1
Duration of residency
 >5 years 5.3 (2.4–11.6)* 
 ≤5 years1
Socio-economic level
 Stratum V (lowest level)1.6 (0.5–5.4)  
 Stratum IV4.0 (1.3–12.0)*  
 Stratum III3.0 (0.7–13.0)  
 Stratum II2.8 (1.2–6.5)*  
 Stratum I (highest level)1
Model of distribution
 By local health committees0.7 (0.3–1.8)7.0 (3.4–14.4)* 
 By routine programme11
Urbanization
 Suburban area2.1 (1.4–3.0)*9.3 (4.0–21.2)* 
 Urban area11
 Model of distribution* Urbanization0.0 (0.0–0.1)*0.0 (0.0–0.0)*
Use of Abate month previous to corresponding survey
 Yes  7.8 (2.1–28.9)*
 No1
At least two practices to prevent dengue (pre-intervention)
 Yes0.6 (0.3–1.0)  
 No1
Frequency of water supply (pre-intervention)
 Less frequent 5.8 (1.1–32.4)* 
 Each 2 days 1.6 (1.0–2.7)* 
 Permanent or each day1

Thailand

The general characteristics of this study population are shown in Table 1. In 17.3% of the households, respondents reported a history of dengue in the family. The majority of households had a good knowledge on dengue and reported conducting dengue control activities in the peri- and intra-domestic area.

In March 2007, 92.3% (95% CI 90.1–94.5) of the households accepted to receive at least one curtain (on average 3.2 curtains/house; SD 1.3); 5 months after reception, 80.1% (95% CI 76.8–83.4) (on average 3.2; SD 1.5) and 18 months after reception 59.7% (95% CI 54.3–65.1) (on average 3.0; SD 1.8) of households continued using at least one curtain (Figure 1). For the last survey, 189 households (33.6%) were lost to follow up, mainly due to the fact that seasonal workers moved out of Laem Chabang.

Houses that received, but did not use curtains at 18 months, gave as main reasons that the curtain was damaged or dirty (35.8%), not effective (27.4%), not liking the curtains (17.0%) or blocking the air flow (15.1%). Main reasons evoked for continued use of the curtains were: fewer mosquitoes in/around the house (54.8%), not minding curtain hanging in the house (37.8%) and protection from dust or sun (3.2%).

After 18 months of use, 95.5% of the curtains remained in good condition (Table 2). Most curtains (86.5%) were clean/quite clean. After 5 and 18 months, 20.2% and 67.0% of the households had washed at least one IT curtain, respectively. In 16.2% of households, the curtains were sewed to fit the window or door size.

After 5 and 18 months of use 59.1% and 48.9% of the respondents, respectively, were willing to receive new IT curtains if provided for free. In the last survey, only 12.7% of the households were willing to pay for new curtains (if they were sold at the local market).

The univariate association between the outcome measures and their potential determinants is quantified in Table 3. In the logistic regression models (Table 4), being a resident for more than 5 years was the only determinant of uptake. Neither dengue knowledge nor previous household control practices were associated with it. Short-term use was significantly associated with perceived effectiveness and having a salary income in the household. The continued use of the IT curtains was only determined by their perceived effectiveness.

In both Venezuela and Thailand, some respondents stated that in the beginning they observed dead insects below the IT curtains, an effect that disappeared over time.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In both countries, the household uptake of IT curtains for dengue control was globally high (over 75%) and positively associated with residency for >5 years. Using ordinary curtains pre-intervention and not applying dengue control activities at household level were further determinants in Venezuela. Around 6 months after distribution, use, conditional on uptake, was in both settings predominantly determined by the perceived effectiveness of the IT curtains and, to a lesser extent, by low socio-economic status. In Venezuela short-term use remained weakly associated with pre-intervention use of ordinary curtains. Use decreased significantly to 38.4% and 59.7% of households after 22 months in Venezuela and 18 months in Thailand, respectively. It was, conditional on short-term use, strongly and exclusively determined by the perceived effectiveness. The uptake of IT covers was mainly determined by the presence of uncovered jars in the houses. Continued use was strongly and positively associated with the use of Abate® in water storage containers.

This is the first study reporting on the use of IT curtains and IT covers that were introduced for A. aegypti control into communities under ‘real life’ operational conditions. It is also the first one to document with panel surveys subsequent levels and determinants of ITMs’ short-term and continued use. The contrasting eco-bio-social environments in which the study was conducted lend robustness to our findings.

A limitation of our study is the loss to follow up, for the ‘continued use’ analysis, of the temporal workers residing for less than 6 months in the study area. However, this mainly reflects that they form a special and difficult to reach population for household-based disease control strategies, since being a relative newcomer was also negatively associated with the acceptance of IT curtains at distribution.

The achieved high initial uptake is remarkable and, apart from the observed determinants, probably also influenced by the check-up visit after 1 week (in Thailand) and by the free distribution by the disease control programmes in both countries. When bednets are sold through commercial outlets, the uptake depends on the households’ purchasing power (Vanlerberghe et al. 2010a). On the other hand, it has been observed that commercial net markets have contributed substantially to equitable and sustainable coverage of mosquito nets in Africa (Webster et al. 2005). The remarkable decline at short term in our effectiveness trial is in contrast with published findings from efficacy trials: 5 months after distribution IT curtain coverage remained 88% and 96% in Mexico and Venezuela (Kroeger et al. 2006), and IT jar-cover coverage 91% and 68% in Cambodia (Seng et al. 2008) and Venezuela (Kroeger et al. 2006), respectively. However, our findings on short-term and continued use are in line with observations in Burkina Faso, where IT curtains were tried out for malaria control. Habluetzel et al. (1999) reported 59% of door curtains still in use 15 months after distribution and Diallo et al. (2004) observed 69% coverage after 2 years.

In contrast to Atkinson’s findings (Atkinson et al. 2009) on bed nets – one of the best-studied vector control tools –, disease knowledge was not correlated with uptake or use of IT curtains and covers. This is not surprising in these specific contexts where more than 75% of people have already good knowledge on dengue. Educational messages on the subject will probably not be helpful to assure sustained high coverages and other strategies need to be developed. Studies indicating that community involvement in routine A. aegypti control programme activities increases effectiveness and sustainability (Toledo Romani et al. 2007; Toledo et al. 2008; Vanlerberghe et al. 2009a), suggest that actively engaging community members in the promotion of continued use of IT tools could be an avenue.

The observed association between the use of ordinary curtains and the uptake and short-term use of IT curtains in Venezuela is in line with the observation that the use of new disease control tools is not only determined by product attributes, but also by pre-intervention behaviours and contextual factors (Severy et al. 2005).

The association between continued use of IT curtains and perceived effectiveness is also of note but merits further study. When tested in Malawi for malaria control (Rubardt et al. 1999), households that still used curtains after 12 months, also expressed higher perceived efficacy for mosquito control then households that had discontinued their use. Observation of dead insects has been identified as a strong motivation to use IT materials (Ritmeijer et al. 2007), but Brieger et al. (1996) signal that the number of dead insects reportedly decreased after using the curtains for some weeks as was also mentioned by the households in our study. Resistance of Culex to deltamethrin, the insecticide impregnated in the PermaNet® material, has been reported in many tropical regions (Kulkarni et al. 2007) and it is unclear whether this or other phenomena are at the root of the decline in user perception of IT curtains’ effectiveness.

Our research was quantitative and could have benefited from a complementary qualitative component. It could have permitted to explore more in depth the social representations of effectiveness and to understand why perceived effectiveness is, consistently in both countries, the strongest determinant of short-term and continued IT curtain use.

Recently, it has been shown that high IT curtain coverage is required to obtain a significant effect on A. aegypti densities (Vanlerberghe et al. 2009b) and that IT curtain coverage and adulticiding effect should remain stable for around 3 years for IT curtain deployment to be, at equal relative effectiveness and actual prices, as efficient as routine vector control programmes (Baly 2010). Although the insecticidal activity of IT curtains lasts consistently for longer than 1 year (Vanlerberghe et al. 2010b), drastic decline of curtain use, observed in our study at 18/22 months, jeopardizes the effectiveness of IT curtains introduction in dengue control programmes. Further studies are needed to explore the underlying mechanisms that render the tools not acceptable for (part of) the target population and to permit the design of accompanying measures to ITM implementation that will boost its continued use.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We are grateful to the populations, the municipal authorities, the routine vector control programme and the health committees who accepted to participate in the study. We are indebted to the field research teams for the many hours they spent with the communities and for conducting the household surveys. We thank Vestergaard-Frandsen for providing the IT tools free of charge for the purpose of the study. The study reports partial results of Workpackage 4 of the project DENCO-Towards successful dengue prevention and control, financed by the European Union (INCO-CT-2004-517708).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Atkinson JA, Bobogare A, Fitzgerald L et al. (2009) A qualitative study on the acceptability and preference of three types of long-lasting insecticide-treated bed nets in Solomon Islands: implications for malaria elimination. Malaria Journal 8, 119.
  • Baly A (2010) Costo y costo-efectividad de estrategias comunitarias para el control de Aedes aegypti y la prevencion del dengue. Doctoral Thesis, Institute of Tropical Medicine ‘Pedro Kouri’, Havana, Cuba.
  • Brieger WR, Onyido AE, Sexton JD, Ezike VI, Breman JG & Ekanem OJ (1996) Monitoring community response to malaria control using insecticide-impregnated bed nets, curtains and residual spray at Nsukka, Nigeria. Health Education Research 11(2), 133145.
  • Diallo DA, Cousens SN, Cuzin-Ouattara N, Nebie I, Ilboudo-Sanogo E & Esposito F (2004) Child mortality in a West African population protected with insecticide-treated curtains for a period of up to 6 years. Bulletin of the World Health Organization 82(2), 8591.
  • Donner A & Klar N (2000) Design and Analysis of Cluster Randomization Trials in Health Research. Arnold, a member of the Hodder Headline Group.b
  • Erlanger TE, Keiser J & Utzinger J (2008) Effect of dengue vector control interventions on entomological parameters in developing countries: a systematic review and meta-analysis. Medical and Veterinary Entomology 22(3), 203221.
  • Farrar J, Focks D, Gubler D et al. (2007) Towards a global dengue research agenda. Tropical Medicine & International Health 12(6), 695699.
  • Gibbons RV & Vaughn DW (2002) Dengue: an escalating problem. British Medical Journal 324(7353), 15631566.
  • Gyapong M, Gyapong JO, Amankwa J, Asedem J & Sory E (1996) Introducing insecticide impregnated bednets in an area of low bednet usage: an exploratory study in north-east Ghana. Tropical Medicine & International Health. 1(3), 328333.
  • Habluetzel A, Cuzin N, Diallo DA et al. (1999) Insecticide-treated curtains reduce the prevalence and intensity of malaria infection in Burkina Faso. Tropical Medicine & International Health. 4(8), 557564.
  • Henderson RH & Sundaresan T (1982) Cluster sampling to assess immunization coverage: a review of experience with a simplified sampling method. Bulletin of the World Health Organization 60(2), 253260.
  • Kachur SP, Phillips-Howard PA, Odhacha AM, Ruebush TK, Oloo AJ & Nahlen BL (1999) Maintenance and sustained use of insecticide-treated bednets and curtains three years after a controlled trial in western Kenya. Tropical Medicine & International Health 4(11), 728735.
  • Kay BH, Tuyet Hanh TT, Le NH et al. (2010) Sustainability and cost of a community-based strategy against Aedes aegypti in northern and central Vietnam. American Journal of Tropical Medicine and Hygiene. 82(5), 822830.
  • Kleinbaum DG, Kupper LL, Muller KE & Nizam A (1998) Applied Regression Analysis and Other Multivariable Methods, 3rd edn. Brooks/Cole Publishing Company, California, USA.
  • Kroeger A, Lenhart A, Ochoa M et al. (2006) Effective control of dengue vectors with curtains and water container covers treated with insecticide in Mexico and Venezuela: cluster randomised trials. British Medical Journal 332(7552), 12471252.
  • Kulkarni MA, Malima R, Mosha FW et al. (2007) Efficacy of pyrethroid-treated nets against malaria vectors and nuisance-biting mosquitoes in Tanzania in areas with long-term insecticide-treated net use. Tropical Medicine & International Health 12(9), 10611073.
  • Méndez-Castellano Hernan & Méndez Maria Cristina (1994) Sociedad y estratificación (Método Graffar Méndez-Castellano). FUNDACRESA edn, Caracas, Venezuela.
  • Phuanukoonnon S, Mueller I & Bryan JH (2005) Effectiveness of dengue control practices in household water containers in Northeast Thailand. Tropical Medicine & International Health 10(8), 755763.
  • Renganathan E, Parks W, Lloyd L et al. (2003) Towards sustaining behavioural impact in dengue prevention and control. Dengue Bulletin 27, 612.
  • Ritchie SA, Rapley LP, Williams C et al. (2009) A lethal ovitrap-based mass trapping scheme for dengue control in Australia: I. Public acceptability and performance of lethal ovitraps. Medical and Veterinary Entomology 23(4), 295302.
  • Ritmeijer K, Davies C, van Zorge R et al. (2007) Evaluation of a mass distribution programme for fine-mesh impregnated bednets against visceral leishmaniasis in eastern Sudan. Tropical Medicine & International Health 12(3), 404414.
  • Rubardt M, Chikoko A, Glik D et al. (1999) Implementing a malaria curtains project in rural Malawi. Health Policy Planning 14(4), 313321.
  • Sanchez L, Perez D, Cruz G et al. (2009) Intersectoral coordination, community empowerment and dengue prevention: six years of controlled interventions in Playa Municipality, Havana, Cuba. Tropical Medicine & International Health 14(11), 13561364.
  • Seng CM, Setha T, Nealon J, Chantha N, Socheat D & Nathan MB (2008) The effect of long-lasting insecticidal water container covers on field populations of Aedes aegypti (L.) mosquitoes in Cambodia. Journal of Vector Ecology 33(2), 333341.
  • Severy LJ, Tolley E, Woodsong C & Guest G (2005) A framework for examining the sustained acceptability of microbicides. AIDS and Behaviour 9(1), 121131.
  • Tami A, Mubyazi G, Talbert A, Mshinda H, Duchon S & Lengeler C (2004) Evaluation of Olyset insecticide-treated nets distributed seven years previously in Tanzania. Malaria Journal 3, 19.
  • TDR (2006) Report of the Scientific Working Group meeting on Dengue. Geneva, 1–5 October 2006, TDR/SWG/08 edn.b
  • Toledo Romani ME, Vanlerberghe V, Perez D et al. (2007) Achieving sustainability of community-based dengue control in Santiago de Cuba. Social Science & Medicine 64(4), 976988.
  • Toledo ME, Baly A, Vanlerberghe V et al. (2008) The unbearable lightness of technocratic efforts at dengue control. Tropical Medicine & International Health 13(5), 728736.
  • Vanlerberghe V, Toledo ME, Rodriguez M et al. (2009a) Community involvement in dengue vector control: cluster randomised trial. British Medical Journal 338, b1959.
  • Vanlerberghe V, Villegas E, Oviedo M et al. (2009b) Effectiveness of insecticide treated materials for Aedes aegypti control in Venezuela. Tropical Medicine & International Health 14(Suppl 2), 169 [Abstract 4.4-029].
  • Vanlerberghe V, Singh SP, Paudel IS et al. (2010a) Determinants of bednet ownership and use in visceral leishmaniasis-endemic areas of the Indian subcontinent. Tropical Medicine & International Healt 15(1), 6067.
  • Vanlerberghe V, Trongtokit Y, Cremonini L, Jirarojwatana S, Apiwathnasorn C & Van der Stuyft P (2010b) Residual insecticidal activity of long-lasting deltamethrin-treated curtains after 1 year of household use for dengue control. Tropical Medicine & International Health 15(9), 10671071.
  • Webster J, Lines J, Bruce J, Armstrong S Jr & Hanson K (2005) Which delivery systems reach the poor? A review of equity of coverage of ever-treated nets, never-treated nets, and immunisation to reduce child mortality in Africa. The Lancet Infectious Diseases 5(11), 709717.