SYSTEMATIC REVIEWS

Strategies for delivering insecticide-treated nets at scale for malaria control: a systematic review

Stratégies de distribution de moustiquaires imprégnées adaptées à la lutte contre le paludisme: revue systématique

Estrategias para la distribución a escala de mosquiteros tratados con insecticida para controlar la malaria: revisión sistemática

Barbara A Willey^I,*; Lucy Smith Paintain^II; Lindsay Mangham^I; Josip Car^III; Joanna Armstrong Schellenberg^II

^IFaculty of Public Health and Policy, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, England
^IIFaculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, England
^IIISchool of Public Health, Imperial College London, London, England

ABSTRACT

OBJECTIVE: To synthesize findings from recent studies of strategies to deliver insecticide-treated nets (ITNs) at scale in malaria-endemic areas.
METHODS: Databases were searched for studies published between January 2000 and December 2010 in which: subjects resided in areas with endemicity for Plasmodium falciparum and Plasmodium vivax malaria; ITN delivery at scale was evaluated; ITN ownership among households, receipt by pregnant women and/or use among children aged < 5 years was evaluated; and the study design was an individual or cluster-randomized controlled design, nonrandomized, quasi-experimental, before-and-after, interrupted time series or cross-sectional without temporal or geographical controls. Papers describing qualitative studies, case studies, process evaluations and cost-effectiveness studies linked to an eligible paper were also included. Study quality was assessed using the Cochrane risk of bias checklist and GRADE criteria. Important influences on scaling up were identified and assessed across delivery strategies.
FINDINGS: A total of 32 papers describing 20 African studies were reviewed. Many delivery strategies involved health sectors and retail outlets (partial subsidy), antenatal care clinics (full subsidy) and campaigns (full subsidy). Strategies achieving high ownership among households and use among children < 5 delivered ITNs free through campaigns. Costs were largely comparable across strategies; ITNs were the main cost. Cost-effectiveness estimates were most sensitive to the assumed net lifespan and leakage. Common barriers to delivery included cost, stock-outs and poor logistics. Common facilitators were staff training and supervision, cooperation across departments or ministries and stakeholder involvement.
CONCLUSION: There is a broad taxonomy of strategies for delivering ITNs at scale.

RÉSUMÉ

OBJECTIF: Réaliser une synthèse des études récentes menées sur les stratégies permettant de fournir des moustiquaires imprégnées d'insecticide (MMI) à grande échelle dans les zones où le paludisme est endémique.
MÉTHODES: À partir de bases de données, on a recherché les études publiées entre janvier 2000 et décembre 2010, dans lesquelles: les sujets résidaient dans des zones où le paludisme à Plasmodium falciparum et à Plasmodium vivax était endémique; une distribution à grande échelle de MMI a été évaluée; la détention de MMI dans les foyers, la réception par les femmes enceintes et/ou l'utilisation chez les enfants âgés de moins de 5 ans a été évaluée; la conception de l'étude impliquait un contrôle individuel ou en grappes, était quasi expérimentale, avant et après, en séries temporelles interrompues, ou transversale sans contrôle temporel ou géographique. Les documents de travail décrivant les études qualitatives, les études de cas et les études d'évaluation des processus et de rentabilité, associés à un document de travail éligible, ont également été inclus. La qualité des études a été appréciée à l'aide de la liste de vérification des risques Cochrane et des critères de l'approche GRADE. On a relevé et évalué d'importantes influences sur l'augmentation de la distribution dans les différentes stratégies.
RÉSULTATS: Un total de 32 documents de travail décrivant 20 études africaines a été étudié. Bon nombre des stratégies de distribution impliquaient différents secteurs de la santé, ainsi que le réseau du commerce de détail (partiellement subventionné), les maternités (intégralement subventionnées) et les campagnes (intégralement subventionnées). Les stratégies qui ont obtenu une meilleure détention dans les foyers et une plus grande utilisation chez les enfants âgés de moins de 5 ans étaient les campagnes de distribution gratuite des MMI. Les coûts étaient largement comparables dans les stratégies étudiées, les MMI constituant le principal coût. Les estimations de rentabilité variaient surtout en fonction de la durée de vie et de la résistance présumée de la moustiquaire. Parmi les inconvénients les plus courants figuraient le coût, la rupture de stock et une mauvaise logistique. Les facteurs favorables les plus courants étaient la formation et la supervision du personnel, la coopération interdépartementale ou interministérielle, ainsi que l'implication des intervenants.
CONCLUSION: Il existe une vaste taxonomie de stratégies pour une distribution à grande échelle des MMI.

RESUMEN

OBJETIVO: Sintetizar los resultados de estudios recientes acerca de las estrategias para distribuir a escala mosquiteros tratados con insecticida (RTI) en zonas con malaria endémica.
MÉTODOS: Se examinaron bases de datos en busca de estudios publicados entre enero de 2000 y diciembre de 2010 en los que: los sujetos residían en áreas en las que la malaria por Plasmodium falciparum y Plasmodium vivax es endémica; se evaluó la entrega de RTI a escala; se evaluó la propiedad de RTI en hogares, la recepción por parte de mujeres embarazadas y/o el uso por parte de niños menores de 5 años; y cuyo diseño del estudio era un estudio controlado individual o aleatorio sobre grupos, no aleatorio, cuasiexperimental, antes y después, de series de tiempo interrumpido o transversal sin controles temporales o geográficos. También se incluyeron artículos que describían estudios cualitativos, estudios de caso, evaluaciones de proceso y estudios de efectividad de costes vinculados a un artículo que cumplía con las condiciones. La calidad del estudio fue evaluada por medio de la herramienta Cochrane de riesgo de sesgo y los criterios GRADE. Se identificaron y evaluaron importantes influencias sobre el aumento progresivo en las estrategias de distribución.
RESULTADOS: Se revisaron un total de 32 artículos que describían 20 estudios africanos. En muchas de las estrategias de distribución participaron sectores sanitarios y establecimientos de venta al por menor (subsidio parcial), clínicas de atención prenatal (subsidio completo) y campañas (subsidio completo). Las estrategias que consiguieron un grado de participación entre los hogares y un uso entre niños menores de 5 años elevados distribuyeron RTI de forma gratuita mediante campañas. Los costes de las diversas estrategias fueron en gran medida comparables; las RTI supusieron el coste principal. Los cálculos de efectividad de costes fueron sensibles sobre todo a la vida útil esperada del mosquitero y a las fugas. Entre las barreras frecuentes a la distribución figuraron el coste, la falta de existencias y una logística deficiente. Los facilitadores comunes fueron la formación y supervisión del personal, la cooperación entre departamentos o ministerios y la implicación de las partes implicadas.
CONCLUSIÓN: Hay una amplia taxonomía de estrategias para distribuir RTI a escala.

Introduction

Malaria continues to represent a major public health problem in areas of endemicity, with an estimated 225 million cases worldwide in 2009.¹ The 2015 goals of the World Health Organization's (WHO's) Roll Back Malaria Partnership are to reduce global malaria cases by 75% from 2000 levels and to reduce malaria deaths to near zero through universal coverage by effective prevention and treatment interventions.¹ Among other preventive interventions, WHO recommends the use of insecticide-treated nets (ITNs), particularly long-lasting insecticidal nets, which have been shown to be cost-effective,^2–4 to reduce malaria episodes among children < 5 years of age (hereafter, "children under 5") by approximately 50% and all-cause mortality by 17%.^5,6 Universal coverage with ITNs is defined as use by > 80% of individuals in populations at risk.⁶ WHO recommends supplying ITNs without charge or with a high subsidy and using a combination of periodic mass campaigns and routine delivery channels to deliver ITNs at scale.⁶ Other strategies include supporting the existing commercial sector and distributing vouchers exchangeable for partially subsidized ITNs through retailers.⁷

In response to the Roll Back Malaria Partnership's targets for universal coverage, considerable efforts have been made recently to scale up ITN delivery. However, there is still low coverage in many countries and a need to understand the lessons learnt from experiences of scaling up ITN delivery. We therefore conducted a systematic review to synthesize recent evidence on the delivery of ITNs (including long-lasting insecticidal nets) at scale in malaria-endemic areas by documenting and characterizing the strategies for delivering ITNs at scale (at the district level or higher); summarizing ITN ownership among households and ITN use among children under 5, stratified by measures of equity when possible; summarizing the reported cost or cost-effectiveness of different strategies; and synthesizing information on reported factors influencing delivery of ITNs at scale.

Methods

A systematic review was conducted to identify studies that reported on the delivery of ITNs (including long-lasting insecticidal nets) at scale. The findings reported here form part of a larger systematic review on the scale-up of WHO-recommended malaria control interventions.⁸ We used a definition of "scaling up" that characterized this activity as the expansion of a health intervention beyond the initial geographical area or population group covered.^9,10 We considered "at scale" to be ITN delivery in at least one district or the equivalent lowest level of health service administration in a given country.

Search strategy

Medline (Ovid), EMBASE, CAB Abstracts, Global Health and Africa Wide databases were searched using subject heading classification terms and free-text words. The following categories were combined using the AND Boolean logic operator: malaria terms, ITN and long-lasting insecticidal net terms and scaling-up terms (Box 1, available at: http://www.who.int/bulletin/volumes/90/9/11-094771). Filters were used to limit the search to humans and to publication dates from January 2000 to December 2010. Relevant papers from the grey literature were identified by searching Eldis and WHOLIS databases and Roll Back Malaria, Malaria Consortium, Africa Malaria Network Trust, and The Global Fund to Fight AIDS, Tuberculosis and Malaria web sites. Citation data for identified papers were exported to EndNote (Thomson Reuters, Carlsbad, USA), where duplicates were removed.

Eligibility criteria

Screening was a two-stage process. First, two authors (BW and LSP) independently screened titles and abstracts to determine which papers should undergo full-text assessment for eligibility. Retained papers underwent full-text review (performed independently by BW and LSP) to determine whether they described studies that satisfied the following criteria: subjects resided in areas where Plasmodium falciparum and Plasmodium vivax are endemic; ITN delivery at scale was evaluated; ITN ownership among households, receipt by pregnant women and/or use among children under 5 was evaluated; and an individual or cluster-randomized controlled design, a nonrandomized design, a quasi-experimental design, a before-and-after design, an interrupted time series design or a cross-sectional design without temporal or geographical controls was used.^11–13 Papers meeting these criteria were termed "index papers". In addition to documenting and characterizing the strategies for delivering ITNs at scale and summarizing ITN ownership among households and ITN use among children under 5, this review also aimed to summarize the reported cost or cost-effectiveness of different strategies and to synthesize information on reported factors influencing delivery of ITNs at scale. As such, we also included papers that described qualitative studies, case studies, process evaluations and cost-effectiveness studies that were linked to an index paper.

The reference lists from eligible papers were hand-searched for additional relevant citations. All data relevant to the review were extracted from final included papers into an Access database (Microsoft, Redmond, United States of America).

Analysis

The first objective was to document and characterize the strategies for delivering ITNs at scale and was guided by a framework adapted from Kilian et al.¹⁴ Strategies were characterized by target population, implementation scale, implementer type, user cost and implementation duration (Fig. 1).

The effectiveness of ITN delivery strategies was not compared using meta-analysis because study designs were too variable.¹⁵ Rather, narrative synthesis with a Best Evidence Synthesis approach was used to summarize findings and compare results across the different delivery strategies.^16,17

The extent to which ITN ownership or use changed over time and whether such changes were attributable to the delivery strategy were assessed according to study quality. The quality of studies with a randomized or nonrandomized control group and of those using an interrupted time-series design was assessed using the Cochrane risk of bias checklist¹⁵ and Grading of Recommendations Assessment, Development and Evaluation (GRADE) criteria.¹⁸

All reported costs were adjusted for inflation by two authors (LSP and LM) and are presented as 2010 United States dollars (US$) using the consumer price indices available from the International Monetary Fund.¹⁹ When possible, costs are reported separately as financial (i.e. monetary) costs or economic costs (including opportunity costs and costs of donated goods and services).

Content analysis and narrative synthesis were used to identify important influences on delivering ITNs at scale and themes were assessed across the different ITN delivery strategies.^16,17

Results

Fig. 2 details the literature search and screening process, performed according to guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Group.²⁰ We included 32 papers that described 20 studies from 12 African nations (Burkina Faso, Eritrea, Ghana, Kenya, Madagascar, Malawi, Niger, Nigeria, Togo, Uganda, the United Republic of Tanzania and Zambia) and one partially autonomous region (Zanzibar). Six studies were implemented on a national level, two on a regional scale and 12 at the district level (of which three took place in only one district). Fourteen studies delivered ITNs only to children under 5 and/or pregnant women (Table 1 and Table 2, both available at: http://www.who.int/bulletin/volumes/90/9/11-094771).

Strategies for delivering ITNs at scale

Fig. 3 summarizes the characteristics of the strategies used to deliver ITNs at scale using the categories presented by Kilian et al.¹⁴ Routine health services, retailers and community-based agents were used to deliver ITNs on a continuous basis. Time-limited strategies either integrated the distribution of ITNs with a public health campaign or delivered ITNs through a stand-alone campaign. Most continuous strategies partially subsidized the delivery of ITNs, whereas all time-limited strategies fully subsidized delivery of ITNs. Most strategies that used routine health services targeted pregnant women or children under 5. All strategies involving time-limited integrated campaigns and stand-alone campaigns targeted children under 5, whereas strategies using retailers and community-based delivery provided ITNs to the general population. Seven studies used a combination of strategies.

Studies with high ITN ownership or use

Eighteen studies reported ITN ownership among households and/or ITN use among children under 5 (Table 2). ITN ownership among households ranged from 1.3% to 94% and ITN use among children under 5, which is typically lower than the prevalence of household ITN ownership, ranged from 12% to 94%. Ten studies reported a high prevalence of ITN ownership or use during at least one survey conducted after initiation of the ITN delivery strategy. Six reported ownership by > 60% of households,^{25,42–44,47,48,50,51} two reported ownership by > 80% of households^29,30,49 and two reported use by > 87% of children under 5.^27,41

Of the six studies reporting ownership by > 60% of households, four used an uncontrolled cross-sectional survey design, surveying 300–3000 households 1–3 years after delivery began.^25,44,50,51 The other two used a before-and-after design in which approximately 2500 households were surveyed before and one year after ITN delivery during campaigns integrated with measles vaccination.^42,43,47,48 During the 1–2-year period between baseline and endline surveys, ITN ownership among households increased from 24.5% to 79% in one study and from < 1% to 55–70% in the other.

The two studies reporting ITN ownership by > 80% of households were uncontrolled cross-sectional surveys. A total of 475 households in Ghana^29,30 and 2074 households in Zambia⁴⁹ were surveyed five months and six months, respectively, after ITN delivery campaigns. ITN ownership in Ghana was 90%, whereas ownership in Zambia was 88% in rural areas and 82% in urban areas. In Ghana, a follow-up survey conducted 38 months after the initial survey revealed that ownership among households had decreased by 18%, to 74%.

Both studies reporting a high prevalence of ITN use among children under 5 also had an uncontrolled cross-sectional design. A total of 378 households in the Adjumani district of Uganda were surveyed 5–7 months after distribution of partially subsidized ITNs to pregnant women through antenatal care clinics²⁷ and 264 households in the North A district of Zanzibar were surveyed 5 months after ITN delivery during a stand-alone ITN campaign.⁴¹ Responses revealed use by 94% of children under 5 in households surveyed in the Adjumani district and by 87% of children under 5 in the North A district.

All 10 studies that reported a high prevalence of ITN ownership or use provided fully subsidized ITNs through at least one component of their delivery strategy (Fig. 3 and Table 2). Seven studies provided fully subsidized ITNs through a stand-alone campaign only (in one⁴¹) or through an integrated campaign only (in six^{42–44,47–51}). One study considered the continuous delivery of free ITNs through antenatal clinics.²⁵ Two studies evaluated combined strategies.^27,29,30 In one, ITNs were delivered to pregnant women through antenatal clinics on a continuous basis by use of a partially subsidized voucher system and to children under 5 through a campaign integrated with measles vaccination, at full subsidy.^29,30 In the other, ITNs were delivered under a full subsidy to pregnant women through antenatal clinics on a continuous basis and for free to children under 5 during a stand-alone campaign on a time-limited basis.²⁷

Equity of ITN ownership and use

Thirteen studies reported coverage stratified according to socioeconomic status as a measure of equity (Table 2). One study evaluated equity on the basis of urban and rural residence and twelve studies evaluated it on the basis of a household asset index. Of the latter studies, three reported a concentration index and nine reported an equity ratio. A concentration index ranges from – 1 to 1, with a value of 0 indicating equitable distribution and values > 0 indicating inequitable distribution benefiting the least poor group. An equity ratio measures the equity of distribution in the poorest quintile relative to that in the least poor quintile, with a value of 1 indicating equitable distribution and values between 0 and 1 indicating inequitable distribution benefiting the least poor group.

The study that evaluated equity in terms of urban and rural residence was based on data from a national survey performed after partially subsidized delivery of ITNs to pregnant women and children under five at health centres.²⁸ The survey found greater use among children under 5 in urban areas, compared with those in rural areas (51% versus 17%).

Three studies presented the concentration index of ITN ownership among households or ITN use among children under 5. The concentration index in each revealed higher ITN ownership or use among the least poor groups. One study had a quasi-experimental design and evaluated continuous delivery of partially subsidized ITNs through health care facilities.²¹ The other two used a cross-sectional design to assess the fully subsidized delivery of ITNs during a stand-alone campaign²⁷ or during a campaign integrated with measles vaccination.⁴⁶

Nine studies presented the equity ratio, or sufficient data for its calculation, of ITN ownership among households or ITN use among children under 5 (Fig. 3). The highest ownership was reported in the poorest quintile in four campaigns that integrated the delivery of free ITNs with measles vaccination. Two of the four used a cross-sectional design to evaluate strategies at either the national or district levels.^44,49 The other two used a before-and-after design and also reviewed delivery at the district or national levels.^42,43,47,48 The change in equity index was available only for one of the before-and-after studies and involved a decrease from 1.2 to 1.1.^42,43

ITN use was similar across quintiles in two studies, both of which used an uncontrolled cross-sectional survey design of delivery at the district level. The strategy evaluated in one delivered ITNs during a stand-alone campaign.⁴¹ The other investigated a combined strategy involving delivery of fully subsidized ITNs to children under 5 through a campaign integrated with measles vaccination and partially subsidized ITNs to pregnant women through antenatal clinics.^29,30

In five studies, ITN ownership or use was higher in the least poor quintile. Three studies evaluated the delivery of free ITNs to children under 5 through a campaign integrated with polio or measles vaccination in Niger, in Lindi region of the United Republic of Tanzania, and four rural districts of Zambia (Chilubi, Kaputa, Mambwe and Nyimba).^45,49,50 In the fourth study, the delivery of partially subsidized ITNs to pregnant women via antenatal care clinics in the United Republic of Tanzania was examined.^32–35 The fifth study reviewed a stand-alone ITN campaign involving distribution of fully subsidized nets to children under 5 in the Micheweni district of Zanzibar.⁴¹

Study quality

Table 2 shows the variety of study designs used to assess ITN delivery strategies. Of the 18 studies reporting data on ITN ownership among households and ITN use among children under 5, the study design in two (a cluster-randomized controlled trial^22–24 and a quasi-experimental study without randomization²¹) involved comparison areas, and the study design in four involved a temporal comparison. Two of the studies with a temporal comparison evaluated time-limited delivery of fully subsidized ITNs^42,43,47,48 and two analysed continuous delivery of partially subsidized ITNs.^3,36,38–40 As such, the interpretation of ITN ownership among households and ITN use among children under 5 between survey years varies by study design and delivery strategy.

Only the cluster-randomized controlled trial directly compared different delivery strategies.^22–24 One strategy involved subsidized sale, promoted by social marketing, of ITNs to the general population plus free distribution of long-lasting insecticidal nets to pregnant women at antenatal care clinics. The other strategy involved only subsidized sale, promoted by social marketing, of ITNs to the general population through retailers. Ownership of ITNs was 35% in the dual-intervention arm and 23% in the retail-only arm (P < 0.001). Although the risk of bias was low in this study, the quality of the evidence was downgraded from high to moderate on the basis of the GRADE criteria because it was unclear whether analyses adjusted for the clustered design and because no relative measure of effect was provided.

One study described the delivery of partially subsidized ITNs at the district level through sales by health facility staff.²¹ ITN ownership was 14% in three intervention districts, compared with 1.3% in two comparison districts (P < 0.001). The risk of bias in this study was moderate principally because of the lack of randomization. The quality of evidence was very low on the basis of the GRADE criteria because there were important differences between intervention and comparison areas at baseline (e.g. socioeconomic status) that were not adjusted for in the analysis and because no relative measure of effect was provided.

In nonrandomized studies, identification of the channel through which the ITN is delivered (i.e. antenatal clinics or retail shops) may help determine whether the change in coverage achieved can be allocated to the delivery strategy.¹² Studies in three countries did not stratify ITN ownership by delivery channel.^{3,36–40,47,48} However, elsewhere, a decline in the proportions of unsubsidized ITNs sourced from retailers and partially subsidized ITNs sourced from maternal and child health clinics was seen among children under 5.^42,43 Both decreases occurred after initiation of an integrated campaign in 2006 to distribute fully subsidized ITNs, with the campaign contributing almost half of the ITNs used by children under 5 surveyed during 2006–2007.

Costs

Ten studies reported on the cost or cost-effectiveness of ITNs (Table 3). Of these, seven described only cost per ITN delivered or cost per treated-net–year. The remaining three were cost-effectiveness studies that also presented cost per death or per disability-adjusted life year averted. All except one of the economic evaluation studies conducted sensitivity analyses around the major cost and outcome parameters.

Four studies investigated the cost of delivering free ITNs through antenatal care clinics, with three at the district level and one at the national level. In the district-level studies, financial costs ranged from US$ 8.20 to US$ 10.54 per ITN delivered^22,26,27 and economic costs ranged from US$ 5.47 to US$ 5.89 per ITN delivered.^22,27 The study at the national scale reported an economic cost of US$ 10.77 per ITN delivered.^25,35

Of the four studies that evaluated the delivery cost of partially subsidized ITNs, three investigated delivery through the retail sector and one investigated voucher use. Studies of retail-based delivery reported financial costs of US$ 5.47 and US$ 11.16 per ITN delivered in Burkina Faso and Malawi, respectively, and of US$ 12.57 and US$ 18.72 per treated-net–year in the United Republic of Tanzania and Malawi, respectively.^3,22,36 The studies in Burkina Faso and the United Republic of Tanzania were at the district level and the study in Malawi was at the national level; the length of protection afforded by ITNs in calculations of cost per ITN delivered was assumed to be 12 months in Burkina Faso and 6 months in Malawi. The fourth study investigated the Tanzanian National Voucher Scheme and found economic costs of US$ 10.77 per ITN delivered and US$ 6.02 per treated-net–year, with the latter calculation assuming 12-month protection from a treated net.³⁵

The four studies that evaluated fully subsidized campaigns found financial costs per ITN delivered of US$ 3.71 to US$ 11.79 for those integrated with vaccination campaigns^30,45,49 and US$ 9.48 for a stand-alone campaign.²⁷ The stand-alone campaign considered in one of the studies had an economic cost per ITN delivered of US$ 4.76.²⁷

Three studies presented some measure of health impact. The economic cost per child death averted was US$ 1242 for a national voucher scheme³⁵ and US$ 2924 for a retail sector programme involving partially subsidized delivery.³⁶ The economic cost per disability-adjusted life year averted was similar, at US$ 100 and US$ 107.^25,36

Cost or cost-effectiveness estimates were most sensitive to the assumed ITN lifespan (i.e. physical viability and duration of insecticide protection) and the proportion of ITNs actually used (leakage). The main cost associated with ITN delivery programmes was the ITNs themselves, most often followed by staff and transport.

Factors influencing ITN delivery

Information on factors influencing delivery of ITNs at scale was available for 12 of 20 studies (Table 4). Important perceived influences on the delivery of ITNs at scale, from the perspective of actors involved, were categorized into those at the user level, the implementer or health system level and the policy level.⁵²

Facilitators at the implementation level included provision of training and appropriate supervision and support. At the policy level, facilitators included involvement of relevant stakeholders during planning and implementation and cooperation across ministries, departments and sectors (e.g. health and retail). Several barriers were identified, including costs to users for partially subsidized strategies, variation in implementation due to insufficient supplies of ITNs and vouchers and to poor communication and adherence to distribution procedures, and, at the policy level, financial resources to sustain current and future distribution strategies.

Discussion

Strategies frequently used to deliver ITNs at scale reported in the published and grey literature include continuous delivery of partially subsidized ITNs through the health sector and retail outlets, continuous delivery of free ITNs though antenatal care clinics and time-limited delivery of free ITNs, either alongside other public health goods (usually vaccines) during integrated campaigns or through stand-alone ITN campaigns. Few experiences with continuous delivery by community-based agents were recorded. The majority of strategies delivered to a targeted population of children under 5 or pregnant women. Seven studies from six countries described multiple concurrent or sequential delivery strategies, particularly continuous strategies in combination with a time-limited campaign.

These studies showed wide variability in ITN ownership among households and ITN use among children under 5. Although findings of high ownership or use were largely drawn from uncontrolled studies, strategies reviewed in the majority of studies included at least one component that delivered ITNs at a full subsidy. The majority of equity evidence was from uncontrolled studies: in general, strategies that used time-limited delivery of fully subsidized ITNs were equitable or pro-poor, in contrast to strategies that used continuous delivery of partially subsidized ITNs. No equity evidence from fully subsidized continuous strategies was available.

Comparisons of costs and cost-effectiveness across these strategies are challenging because of variations in the methods of economic analysis used and in the scale of delivery, as emphasized previously.⁵³ Nonetheless, the cost of delivering ITNs across the strategies was reasonably comparable. The main cost was the ITNs themselves, a cost frequently supported by donor funding, and all of the cost-effectiveness estimates were most sensitive to ITN lifespan and proportion of ITNs actually used.

This review aimed to synthesize details on the context of, barriers to and facilitators of strategies to deliver ITNs at scale, some of which were implemented under near-programmatic conditions. Important factors influencing the delivery of ITNs at scale were similar across delivery strategies. Barriers involving cost were common at the user level, whereas barriers involving stock-outs and poor logistics for ITN procurement and transport were common at the implementer level. Training and supervision of staff was often highlighted as a facilitator at the implementer level and cooperation across departments or ministries and stakeholder involvement were highlighted at the policy level.

The evaluation of large-scale health programmes has been highlighted as a "top priority in global health"⁵⁴ and researchers have emphasized that the use of randomized designs for such evaluation may be inappropriate because of low external validity.^11,55 Therefore, to characterize the full breadth of ITN delivery strategies and to synthesize evidence that corresponded to the conditions under which large-scale ITN delivery may occur in practice, we included a variety of study designs.⁵⁶ However, this made interpretation of findings challenging, particularly because a before-and-after study of a campaign conducted at a single time point is qualitatively different from annual surveys conducted during a continuous distribution strategy.

The Medical Research Council recommends that the evaluation of complex interventions include information on the context and implementation of interventions. Our experience in conducting this review suggests that future synthesis of evidence involving large-scale delivery of complex public health interventions would benefit from improved consistency of reporting of the implementation process by included studies.^57,58 Recommendations for reporting are available from the Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) statement.⁵⁹

It is simplistic to interpret the findings of this review as providing a single recommendation to policy-makers on which ITN delivery strategy to adopt. Rather, the review highlights that choosing among alternatives depends on contextual factors, such as the epidemiologic characteristics of malaria, attributes of health systems and contextual constraints. Moreover, the review demonstrates how a framework for characterizing delivery strategies can prove useful in synthesizing evidence, which may help policy-makers formulate implementation strategies to deliver ITNs to populations in their local settings.

Acknowledgements

We thank Mark Petticrew and Neil Spicer (Faculty of Public Health and Policy, London School of Hygiene and Tropical Medicine) and Rifat Atun (Imperial College London, formerly with The Global Fund to Fight AIDS, Tuberculosis and Malaria) for helpful comments.

Funding: This review was supported by the Alliance for Health Services and Policy Research, World Health Organization, which commissioned this work as a background paper for the First Global Symposium on Health Systems Research.

Competing interests: None declared.

References

1. World malaria report. Geneva: World Health Organization; 2010.         

2. Mueller DH, Wiseman V, Bakusa D, Morgah K, Dare A, Tchamdja P. Cost-effectiveness analysis of insecticide-treated net distribution as part of the Togo Integrated Child Health Campaign. Malar J 2008;7:73. doi: 10.1186/1475-2875-7-73 PMID: 18445255         

3. Stevens W, Wiseman V, Ortiz J, Chavasse D. The costs and effects of a nationwide insecticide-treated net programme: the case of Malawi. Malar J 2005;4:22. doi: 10.1186/1475-2875-4-22 PMID: 15885143         

4. Wiseman V, Hawley WA, ter Kuile FO, Phillips-Howard PA, Vulule JM, Nahlen BL et al. The cost-effectiveness of permethrin-treated bed nets in an area of intense malaria transmission in western Kenya. Am J Trop Med Hyg 2003;68(Suppl):161–7. PMID: 12749500         

5. Lengeler C. Insecticide-treated bed nets and curtains for preventing malaria. Cochrane Database Syst Rev 2004;2:CD000363. PMID: 15106149         

6. Global strategic plan: roll back malaria 2005–2015. Geneva: Roll Back Malaria Partnership; 2005.         

7. Targeted subsidy strategies for national scaling up of insecticide-treated netting programmes – principles and approaches. Geneva: Global Malaria Programme, World Health Organization; 2005.         

8. Willey B, Smith L, Armstrong Schellenberg J. How to scale up delivery of malaria control interventions: a systematic review using insecticide-treated nets, intermittent preventive treatment in pregnancy, and artemisinin combination treatment as tracer interventions. Geneva: World Health Organization; 2010 (background paper for the First Global symposium on Health Systems Research). Available from: http://www.hsr-symposium.org/images/stories/5scale_up_delivery_malaria_control.pdf [accessed 4 July 2012]          .

9. Mangham LJ, Hanson K. Scaling up in international health: what are the key issues? Health Policy Plan 2010;25:85–96. doi: 10.1093/heapol/czp066 PMID: 20071454         

10. Scaling up health service delivery: from pilot innovations to policies. Geneva: World Health Organization; 2007.         

11. Habicht JP, Victora C, Vaughan J. Evaluation designs for adequacy, plausibility and probability of public health programme performance and impact. Int J Epidemiol 1999;28:10–8. doi: 10.1093/ije/28.1.10 PMID: 10195658         

12. Webster J, Chandramohan D, Hanson K. Methods for evaluating delivery systems for scaling-up malaria control intervention. BMC Health Serv Res 2010;10:S8. doi: 10.1186/1472-6963-10-S1-S8 PMID: 20594374         

13. Webster J, Kweku M, Dedzo M, Tinkorang K, Bruce J, Lines J et al. Evaluating delivery systems: complex evaluations and plausibility inference. Am J Trop Med Hyg 2010;82:672–7. doi: 10.4269/ajtmh.2010.09-0473 PMID: 20348517         

14. Kilian A, Wijayanandana N, Ssekitoleeko J. Review of delivery strategies for insecticide treated mosquito nets: are we ready for the next phase of malaria control efforts? TropIKA Net J 2010;1:1.         

15. Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions. The Cochrane Collaboration; 2011. Available from: http://www.cochrane-handbook.org/ [accessed 20 August 2012]          .

16. Popay J, Roberts H, Sowden A, Petticrew M, Britten N, Arai L et al. Guidance on the conduct of narrative synthesis in systematic reviews. J Epidemiol Community Health 2005;59:A7.         

17. Pope M. Synthesising qualitative research. In: Pope C, Mays N, Popay J, editors. Synthesising qualitative and quantitative health evidence: a guide to methods. Maidenhead: Open University Press; 2007. pp 142-152.         

18. Guyatt GH, Oxman AD, Vist G, Kunz R, Falck-Ytter Y, Alonso-Coello P et al. Rating quality of evidence and strength of recommendations GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 2008;336:924–6. doi: 10.1136/bmj.39489.470347.AD PMID: 18436948         

19. International Monetary Fund [Internet]. World Economic Outlook Database April 2011. Washington: IMF; 2012. Available from: http://www.imf.org/external/pubs/ft/weo/2011/01/index.htm [accessed 16 August 2012]          .

20. Liberati A, Altman DG, Tetzlaff JGAD; The PRISMA Group. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 2009;6:e1000100. doi: 10.1371/journal.pmed.1000100 PMID: 19621070         

21. Agha S, Van Rossem R, Stallworthy G, Kusanthan T. The impact of a hybrid social marketing intervention on inequities in access, ownership and use of insecticide-treated nets. Malar J 2007;6:13. doi: 10.1186/1475-2875-6-13 PMID: 17261185         

22. De Allegri M, Marschall P, Flessa S, Tiendrebeogo J, Kouyate B, Jahn A et al. Comparative cost analysis of insecticide-treated net delivery strategies: sales supported by social marketing and free distribution through antenatal care. Health Policy Plan 2010;25:28–38. doi: 10.1093/heapol/czp031 PMID: 19752178         

23. Beiersmann C, De Allegri M, Sanon M, Tiendrebeogo J, Jahn A, Mueller O. Community perceptions on different delivery mechanisms for insecticide-treated bed nets in rural Burkina Faso. Open Public Health J 2008;1:17–24. doi: 10.2174/1874944500801010017         

24. Müller O, De Allegri M, Becher H, Tiendrebogo J, Beiersmann C, Ye M et al. Distribution systems of insecticide-treated bed nets for malaria control in rural Burkina Faso: cluster-randomized controlled trial. PLoS ONE 2008;3:e3182. doi: 10.1371/journal.pone.0003182 PMID: 18784840         

25. Yukich JO, Zerom M, Ghebremeskel T, Tediosi F, Lengeler C. Costs and cost-effectiveness of vector control in Eritrea using insecticide-treated bed nets. Malar J 2009;8:51. doi: 10.1186/1475-2875-8-51 PMID: 19331664         

26. Guyatt HL, Gotink MH, Ochola SA, Snow RW. Free bednets to pregnant women through antenatal clinics in Kenya: a cheap, simple and equitable approach to delivery. Trop Med Int Health 2002;7:409–20. doi: 10.1046/j.1365-3156.2002.00879.x PMID: 12000650         

27. Kolaczinski JH, Kolaczinski K, Kyabayinze D, Strachan D, Temperley M, Wijayanandana N et al. Costs and effects of two public sector delivery channels for long-lasting insecticidal nets in Uganda. Malar J 2010;9:102. doi: 10.1186/1475-2875-9-102 PMID: 20406448         

28. Tilson D. The social marketing of insecticide-treated nets (ITNs) in Kenya. Cases Public Health Comm Mark 2007;1:1-20. Available from: http://www.gwumc.edu/sphhs/departments/pch/phcm/casesjournal/volume1/commissioned/cases_1_12.pdf [accessed 26 June 2012]         

29. Grabowsky M, Nobiya T, Selanikio J. Sustained high coverage of insecticide-treated bednets through combined catch-up and keep-up strategies. Trop Med Int Health 2007;12:815–22. doi: 10.1111/j.1365-3156.2007.01862.x PMID: 17596247         

30. Grabowsky M, Nobiya T, Ahun M, Donna R, Lengor M, Zimmerman D et al. Distributing insecticide-treated bednets during measles vaccination: a low-cost means of achieving high and equitable coverage. Bull World Health Organ 2005;83:195–201. PMID: 15798843         

31. Kweku M, Webster J, Taylor I, Burns S, Dedzo M. Public-private delivery of insecticide-treated nets: a voucher scheme in Volta Region, Ghana. Malar J 2007;6:14. doi: 10.1186/1475-2875-6-14 PMID: 17274810         

32. Hanson K, Marchant T, Nathan R, Mponda H, Jones C, Bruce J et al. Household ownership and use of insecticide treated nets among target groups after implementation of a national voucher programme in the United Republic of Tanzania: plausibility study using three annual cross sectional household surveys. BMJ 2009;339:b2434. doi: 10.1136/bmj.b2434 PMID: 19574316         

33. Magesa SM, Lengeler C, deSavigny D, Miller JE, Njau RJA, Kramer K et al. Creating an "enabling environment" for taking insecticide treated nets to national scale: the Tanzanian experience. Malar J 2005;4:34. doi: 10.1186/1475-2875-4-34 PMID: 16042780         

34. Marchant T, Schellenberg D, Nathan R, Armstrong-Schellenberg J, Mponda H, Jones C et al. Assessment of a national voucher scheme to deliver insecticide-treated mosquito nets to pregnant women. CMAJ 2010;182:152–6. doi: 10.1503/cmaj.090268 PMID: 20064944         

35. Mulligan JA, Yukich J, Hanson K. Costs and effects of the Tanzanian national voucher scheme for insecticide-treated nets. Malar J 2008;7:32. doi: 10.1186/1475-2875-7-32 PMID: 18279509         

36. Hanson K, Kikumbih N, Armstrong Schellenberg J, Mponda H, Nathan R, Lake S et al. Cost-effectiveness of social marketing of insecticide-treated nets for malaria control in the United Republic of Tanzania. Bull World Health Organ 2003;81:269–76. PMID: 12764493         

37. Kikumbih N, Hanson K, Mills A, Mponda H, Schellenberg JA. The economics of social marketing: the case of mosquito nets in Tanzania. Soc Sci Med 2005;60:369–81. doi: 10.1016/j.socscimed.2004.05.005 PMID: 15522492         

38. Schellenberg JR, Abdulla S, Nathan R, Mukasa O, Marchant TJ, Kikumbih N et al. Effect of large-scale social marketing of insecticide-treated nets on child survival in rural Tanzania. Lancet 2001;357:1241–7. doi: 10.1016/S0140- 6736(00)04404-4 PMID: 11418148         

39. Chavasse D, Kolwicz C, Smith B. Preventing malaria in Malawi. Essent Drugs Monit 2001;3:2–3.         

40. The Malawi ITN delivery model. Washington: Population Services International; 2005.         

41. Beer N, Ali A, de Savigny D, Al-Mafazy AW, Ramsan M, Abass A et al. System effectiveness of a targeted free mass distribution of long lasting insecticidal nets in Zanzibar, Tanzania. Malar J 2010;9:173. doi: 10.1186/1475-2875-9-173 PMID: 20565860         

42. Hightower A, Kiptui R, Manya A, Wolkon A, Vanden Eng J, Hamel M et al. Bed net ownership in Kenya: the impact of 3.4 million free bed nets. Malar J 2010;9:183. doi: 10.1186/1475-2875-9-183 PMID: 20576145         

43. Noor AM, Amin AA, Akhwale WS, Snow RW. Increasing coverage and decreasing inequity in insecticide-treated bed net use among rural Kenyan children. PLoS Med 2007;4:e255. doi: 10.1371/journal.pmed.0040255 PMID: 17713981         

44. Kulkarni MA, Vanden Eng J, Desrochers RE, Cotte AH, Goodson JL, Johnston A et al. Contribution of integrated campaign distribution of long-lasting insecticidal nets to coverage of target groups and total populations in malaria-endemic areas in Madagascar. Am J Trop Med Hyg 2010;82:420–5. doi: 10.4269/ajtmh.2010.09-0597 PMID: 20207867         

45. Skarbinski J, Massaga JJ, Rowe AK, Kachur SP. Distribution of free untreated bednets bundled with insecticide via an integrated child health campaign in Lindi Region, Tanzania: lessons for future campaigns. Am J Trop Med Hyg 2007;76:1100–6. PMID: 17556618         

46. Khatib RA, Killeen GF, Abdulla S, Kahigwa E, McElroy PD, Gerrets RPM et al. Markets, voucher subsidies and free nets combine to achieve high bed net coverage in rural Tanzania. Malar J 2008;7:98. doi: 10.1186/1475-2875-7-98 PMID: 18518956         

47. Centers for Disease Control and Prevention (CDC). Distribution of insecticide-treated bednets during an integrated nationwide immunization campaign – Togo, West Africa, December 2004. MMWR Morb Mortal Wkly Rep 2005;54:994–6. PMID: 16208313         

48. Terlouw DJ, Morgah K, Wolkon A, Dare A, Dorkenoo A, Eliades MJ et al. Impact of mass distribution of free long-lasting insecticidal nets on childhood malaria morbidity: the Togo National Integrated Child Health Campaign. Malar J 2010;9:199. doi: 10.1186/1475-2875-9-199         

49. Grabowsky M, Farrell N, Hawley W, Chimumbwa J, Hoyer S, Wolkon A et al. Integrating insecticide-treated bednets into a measles vaccination campaign achieves high, rapid and equitable coverage with direct and voucher-based methods. Trop Med Int Health 2005;10:1151–60. doi: 10.1111/j.1365-3156.2005.01502.x PMID: 16262740         

50. Thwing J, Hochberg N, Vanden Eng J, Issifi S, Eliades MJ, Minkoulou E et al. Insecticide-treated net ownership and usage in Niger after a nationwide integrated campaign. Trop Med Int Health 2008;13:827–34. doi: 10.1111/j.1365-3156.2008.02070.x PMID: 18384476         

51. Blackburn BG, Eigege A, Gotau H, Gerlong G, Miri E, Hawley WA et al. Successful integration of insecticide-treated bed net distribution with mass drug administration in Central Nigeria. Am J Trop Med Hyg 2006;75:650–5. PMID: 17038688         

52. Hanson K, Ranson MK, Oliveira-Cruz V, Mills A. Expanding access to priority health interventions: a framework for understanding the constraints to scaling-up. J Int Dev 2003;15:1–14. doi: 10.1002/jid.963         

53. Kolaczinski J, Hanson K. Costing the distribution of insecticide-treated nets: a review of cost and cost-effectiveness studies to provide guidance on standardization of costing methodology. Malar J 2006;5:37. doi: 10.1186/1475-2875-5-37 PMID: 16681856         

54. Evaluation: the top priority for global health. Lancet 2010;375:526. doi: 10.1016/S0140-6736(10)60056-6 PMID: 20079530         

55. Victora CG, Black RE, Boerma JT, Bryce J. Measuring impact in the Millennium Development Goal era and beyond: a new approach to large-scale effectiveness evaluations. Lancet 2011;377:85–95. doi: 10.1016/S0140-6736(10)60810-0 PMID: 20619886         

56. Shepperd S, Lewin S, Straus S, Clarke M, Eccles MP, Fitzpatrick R et al. Can we systematically review studies that evaluate complex interventions? PLoS Med 2009;6:e1000086. doi: 10.1371/journal.pmed.1000086 PMID: 19668360         

57. Oakley A, Bonell C, Allen E, Stephenson J. Process evaluation in randomised controlled trials of complex interventions. BMJ 2006;332:413–6. doi: 10.1136/bmj.332.7538.413 PMID: 16484270         

58. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M.. Developing and evaluating complex interventions: the new Medical Research Council guidance. BMJ 2008;337:a1655. doi: 10.1136/bmj.a1655 PMID: 18824488         

59. Des Jarlais DC, Lyles C, Crepaz N. Improving the reporting quality of nonrandomized evaluations of behavioral and public health interventions: the TREND statement. Am J Public Health 2004;94:361–6. doi: 10.2105/AJPH.94.3.361 PMID: 14998794         

(Submitted: 24 August 2011 – Revised version received: 31 January 2012 – Accepted: 2 February 2012 – Published online: 6 July 2012)

* Correspondence to Barbara Willey (e-mail: barbara.willey@lshtm.ac.uk).