Effect of user preferences on ITN use: a review of literature and data
© The Author(s) 2017
Received: 21 March 2017
Accepted: 26 May 2017
Published: 1 June 2017
Insecticide-treated bed nets (ITNs) are the primary tool for vector control, and optimizing ITN use is a key concern of national programmes. Available evidence indicates that bed net users often have preferences for shape, colour, size, and other attributes, but it is unclear whether these preferences are strong enough to have any significant effect on bed net use, and whether countries and donors should invest in more expensive attributes in order to maximize ITN use. The link between bed net attributes, preferences, and use was investigated using a literature review and review of publicly available, nationally representative household surveys from sub-Saharan Africa.
A literature search was conducted to identify publications with data on preferences for net attributes and on associations between net attributes and use. Publicly available DHS and MIS datasets were screened for variables on net preferences and net attributes. Wald tests were run to obtain odds ratios and confidence intervals for the use of nets of various attributes in univariate analysis. A multilevel logistic regression was constructed to assess the odds of a net’s use, controlling for background variables and adding random effects variables at the household and cluster level.
Preferences for certain net attributes exist, but do not impede high rates of net use in countries where data were available. Stated preferences for shape and colour do not significantly influence net use to degrees that would require action by programme planners. By and large, people are using the nets they receive, and when they do not, it is for reasons unrelated to shape and size (primarily perceived mosquito density, heat or an excess of nets). Households in higher wealth quintiles tend to own greater numbers of conical nets, indicating that they have the ability to obtain or purchase these nets on their own, and individuals resident in higher wealth quintile households also use conical nets preferentially.
The increased manufacturing costs for conical nets are not outweighed by the very small, often non-existent, increases in use rates in sub-Saharan Africa. Programmes that wish to explore the relationship between net attributes, preferences and use rates should include these questions in nationally representative household surveys to be able to capture trends across geographic and socio-economic groups.
While it is well known that consumer preferences influence spending habits and use of commercial products, very little is known about the influence of preferences for mosquito bed nets on their use. Many studies have investigated determinants of use and barriers to use, with strong evidence that insecticide-treated net (ITN) access is the primary driver of ITN use [1–4], and that the primary barriers to use when ITNs are available are discomfort (heat) and low perceived mosquito density .
During the early days of treated bed nets, formative research identified preferred net attributes in an effort to maximize use by the target populations. From about 1980–2005, nets were treated and retreated with treatment kits, making it possible for consumers to purchase their preferred untreated net, and turn it into a treated net. With the introduction of mass campaigns and routine distribution, first in 2004/2005 to children under five and pregnant women, and then universal coverage distributions beginning in 2009, large-scale procurements of long-lasting insecticidal nets (LLINs), where treatment was done at the factory level, significantly reduced local net markets in sub-Saharan Africa . Beneficiaries then began to receive nets of a particular colour, shape and size during mass campaigns and were not able to select a campaign net that might meet their preferences.
The logistics of large-scale procurement and distribution make it impossible to satisfy the variety of net preferences (for shape and colour, for example) that exist among consumers, as this would involve stocking enough nets of various types at all distribution points for people to be able to choose among them, requiring overly excessive procurement, or direct to beneficiary distribution system. On the other hand, programmes want to procure nets that will be most used and avoid procuring nets that may go unused for sleeping, due to procurement specification decisions. Programmes and donors also want to know whether differences in size, colour and shape, particularly, are worth additional investment in order to maximize net use. Previous work on the issue of value for money in LLIN procurement is summarized in a 2012 report from the Results for Development Institute, which examined the costs of colour, size, packaging, labels, and hanging materials . Its conclusions: recommending increased standardization in height and width, bulk packaging, simplified packaging logos, standard labels, and not procuring additional hanging materials, have contributed to the development of standard LLIN specification in tenders amongst the major procurers. It did not, however, address shape, given the minimal data available, although it did note that in Mozambique, users who did not have their preferred net were no less likely to have used it than those who did have their preferred net. A 2013 study on bed net preferences in the Peruvian Amazon  called for net preference to be brought into the value for money equation, and for appropriate methods for studying and measuring net preference to be developed.
Current procurement practices of major donors (the Global Fund, US President’s Malaria Initiative (PMI), UNICEF) allow for some level of preference to be included in tenders, in varying ways. At the Global Fund, interim guidance has been developed  for all LLIN procurement (pooled or direct procurement), noting that all LLINs procured with Global Fund resources should have an interim or full WHOPES recommendation, have a maximum height of 180 cm, and be rectangular. They note that both polyester and polyethylene should be included to allow fair and open competition. For PMI, their 2015 technical guidance  refers to a recently defined ‘standard specifications’, and requests countries to provide justification for procuring non-standard LLINs. Manufacturers are asked in these cases to provide quotes for both the standard and non-standard options. UNICEF procures on behalf of governments and donors, meaning that once the specifications are agreed upon with PMI or Global Fund, UNICEF then issues tenders. They also have standard size options for rectangular nets in white, blue or green. If requests for conical or non-standard nets are received, these are discussed with the Ministry of Health, and may be accepted if the donor also agrees. In all cases, brand-specific requests are not allowed.
Since the advent of universal coverage campaigns, it has been difficult to even measure the influence of net attributes on net use, as the majority of nets in a given country are usually the same product. Few Demographic and Health Surveys (DHS) or Malaria Indicator Surveys (MIS) include even basic information on the net shape, size and colour that allow such analysis; even fewer include questions on the respondent’s preferences for various attributes.
This paper reviews the published literature on net use and user preferences, and examines the available data to assess whether shape, colour and size are influencing rates of net use. The extent to which these differences would affect programmatic decision-making is discussed.
For the literature search, a PubMed search was done on the following terms: net, prefer*, malaria; net, malaria colour/colour; net malaria size; net malaria shape. Resulting publications were then screened, and included if they contained data on user preferences for nets, or data on net attributes as determinants of net use. References lists were then scanned for additional publications. The resulting 33 references were then reviewed and categorized (Additional file 1) as ‘acceptability’ studies, ‘crossover acceptability’ studies, ‘simple preference’ (reporting solely on stated preferences without including net attribute information), and ‘determinants of use’ (where net attributes were included as an independent variable, but preferences were not recorded). Full text records were obtained.
Net attributes and preference variables included in DHS and MIS
Burkina Faso 2014
Information on basic demographics and net characteristics for each net in the household was extracted and reshaped to form a ‘net file’ to permit analysis by net. Using information about the net brand, a ‘textile’ variable was created for selected datasets. A variable for household supply of nets was generated and coded as ‘not enough’ if the household owned fewer than one net for every two people; ‘universal coverage’ if the household owned at least one net for every two people, but fewer than one net per person; and ‘too many’ if the household owned at least one net for every household member. Single person households owning one net were classified as having ‘universal coverage’. Additional information was included in the 2010 and 2014 Malawi MIS. The household respondent’s preference for shape and colour was recorded, along with the reasons they preferred a particular shape in the 2014 MIS. A preference variable was created to assess whether the net being the preferred colour, shape, or both, was associated with increased use. The 2012 Malawi MIS recorded the preferred shape only among households that had no nets at all, making analysis of the relationship between the shape of the household’s nets, their use, and preferences impossible.
Using the ‘svy’ family of commands in Stata 14, the proportions of nets of various attributes were calculated, along with the per cent of nets that were used, disaggregated by attribute. Wald tests were run to obtain odds ratios and confidence intervals for the use of nets of various attributes in univariate analysis. Finally, building on work by Baume et al. [12, 13], other predictors of net use including basic demographic information (household supply of nets, age of head of household, sex of head of household, wealth quintile, urban/rural status, whether household was sprayed in the previous 12 months) and other net characteristics (age of net, net treatment status, presence of holes, shape, colour if available) were included in a multilevel logistic regression to assess whether shape remained a significant factor in the net’s use, controlling for these other variables. To correct for clustering, random effects at the household and cluster level were included in the model, and marginal effects were computed to obtain average marginal effects for net shape and for stated preferences where included in the datasets.
The 33 articles were divided into two categories: those that reported on stated preferences (preferences stated by study participants, whether through survey questionnaires or focus group discussions), and determinants of use (preferences demonstrated by increased use of specific types of nets). Twenty-six articles reported stated preferences, and seven reported determinants of use.
Early acceptability studies in The Gambia in the 1980s [14, 15] reported that the study population was already sleeping under traditional nets at very high rates (95% or greater), and assessed preferences for different textiles, with the majority preferring opaque sheeting to open netting, for privacy and durability. White was the preferred colour in these small studies (90%), washed two to three times a month to keep them clean . With this information, a bed net trial using white nets made of sheeting fabrics was undertaken and reported on in 1993 , noting that 86% of subjects were bed net users. Use of nets during acceptability trials in northern Ghana was also high (99% in rainy and 20% in dry season), where 73% of respondents felt that the size of the nets was adequate, while the rest preferred a larger size [18, 19].
In the early 2000s, Tami et al. asked participants whether they preferred the Olyset or polyester nets when offered at the same price. Preferences were evenly split (51% preferred Olyset, 95% CI 41–61), and participants stated stronger preferences for green nets, to mask dirt . An acceptability trial in India and Nepal also compared polyester and polyethylene nets, using a sequential design in India and a crossover design in Nepal (households used each net for 7 days total). There was a slight preference for the softer quality of polyester. Qualitative data noted varying colour preferences between men (green) and women (blue) .
In the late 2000s, a study in Sri Lanka explored preferences for shape among various ethnic/religious groups . The majority preferred conical nets, although disaggregation by ethnic/religious group showed that Tamils liked the shapes they owned, 78% of Muslims liked their rectangular nets (none owned conical), and Sinhalese with conical nets strongly liked them, while only 37% of Sinhalese with rectangular nets liked them. The majority (82 and 76%) of Sinhalese and Muslims used their nets daily, while Tamil used them mainly in periods of high mosquito density.
Bed net acceptability trials in the Solomon Islands began with a qualitative study in 2009 to elicit preferences; respondents preferred darker (green) nets and wider nets to accommodate sleeping mats made of coconut leaves . The ensuing crossover acceptability trial compared Olyset, PermaNet and DuraNet in a three-stage crossover design, with participants using each net for 10 days. Olyset were deemed less acceptable due to wrinkling and shrinking after washing; larger nets and darker colours were preferred, and shape preference was split equally between conical and rectangular . A similar study in Vanuatu noted preferences for larger mesh and wider nets , while a qualitative study in Timor Leste, despite aiming to elucidate preferences, found that distinctions between attributes were muddied due to different brands having the same colour, and because participants did not have enough experience with nets to make preferences revealing .
In East Africa, a 2009 study explored preferences among 400 households in Central Kenya; 63% preferred rectangular nets, although this varied among the four villages in the sample. Green nets were the most preferred (51%) versus blue (21%), once again due to green’s dirt-concealing abilities. In the population, 96% of those with access to a net used one . In Liberia, an acceptability trial of Interceptor nets noted that perceived insecticidal activity was an important factor for acceptability and use, and that despite half the respondents preferring blue, white ITNs were still used .
During the universal coverage campaign era, a study in Chipinge District, Zimbabwe specifically explored shape preferences. Respondents preferred conical ITNs (84%) compared to rectangular ITNs (15%), primarily due to easier hanging. The authors noted that conical nets were introduced 10 years prior to the rectangular nets, likely influencing overall preferences. In the study population of 380 matriarchs, 93% had received an ITN 2 years prior, while only 33% used one the previous night. No information on shape of nets in these households was gathered .
Two qualitative studies from the Peruvian Amazon found that LLINs were used less frequently in ‘open’ houses (with limited walls, no ceilings, open eaves), because they provided less privacy than traditional opaque nets, did not protect from the cold, and did not prevent debris falling onto beds from the roofing material [8, 30].
A 2014 study in Amhara, Ethiopia found that shape preference was more or less evenly split, with ease of use the main reason for preferring conical nets, and better fit with the bed the main reason for preferring rectangular nets. Blue colour and medium size were the most preferred . However, in southwestern Ethiopia, conical nets were perceived to fit beds and houses better . Qualitative work in Senegal found that while participants stated preferences for shape, size and colour, they did not feel that not having these attributes prevented use. Between 72 and 86% of nets in the study households were used the previous night . An acceptability trial of Dumuria nets in Garissa, Kenya, confirming earlier qualitative findings from southern Sudan , found that non-mesh, opaque, bed-sheet-like fabric nets were accepted by 95% of participants from nomadic communities (81% if missing values are considered negative answers), and that darker colours were preferred to hide dirt .
A study from Cambodia reported that respondents preferred soft, colorful market nets with small mesh over coarse, small, LLINs with larger mesh . These findings were echoed in a willingness-to-pay study in Laos  and in Madagascar .
Reasons for preferring net attributes from qualitative data
Fit more sleepers (Ng’ang’a , Kenya)
Reduces body contact with the net (Sande , Zimbabwe)
Easy to hang (Aleme , Ethiopia)
More compatible with sleeping arrangements and house style (Birhanu , Ethiopia)
Looks clean (MacCormack , Gambia)
Dark colour masks dirt and smoke (Tami , Tanzania; Ng’ang’a , Kenya; Atkinson , Solomon Islands; Banek , Liberia; Gore-Langton , Kenya; Harvey , Peruvian Amazon; Gobena , Ethiopia; Gyapong , Ghana)
Less movement restriction (Atkinson , Vanuatu)
Better ventilation (Atkinson , Vanuatu)
Polyethylene/larger mesh size
Polyester/smaller mesh size
Warmer (Mattern , Madagascar)
Attractive (MacCormack , Gambia)
Open netting fabric
Attractive (MacCormack , Gambia)
Cooler in hot weather (MacCormack , Gambia)
Determinants of net use
In 2009, Baume et al. reported on a cross-sectional survey among 857 households in Oromia and Amhara States in Ethiopia. Net use with the net as unit of analysis was the outcome indicator. Conical ITNs were more likely to be used the previous night compared to rectangular ITNs [OR = 2.27 (95% CI 1.10–4.68) p < 0.05]. Colour of ITN was significant in the multivariate analysis, but it was removed from the final model because of collinearity with price (free/not free) and with shape (all ITNs distributed freely in the prior 2 years were blue rectangular). The researchers also noted that conical nets were easier to hang in the regions’ round houses. There were 1177 rectangular nets in the sample, with only 126 conical nets. Overall, 65% of nets were used the previous night . A similar study in Ghana among 1796 households found that shape was not significant, while light blue colour (n = 332) was predictive of use compared to white colour (n = 1176), controlling for other net characteristics. Light blue nets were distributed for free in some areas and available in markets in others .
In Zanzibar, cross-sectional surveys were implemented in two districts among 509 caretakers of young children. Liking the net’s colour, size and shape of the LLIN were not significant factors in use of the net the previous night by children under-five. Over 85% of nets were used the previous night by a child under five. Qualitative data collected found that the community liked the light blue colour, but disliked the large mesh size and short height of the nets. Caretakers liking the mesh size was associated with increased use of the nets in one district, but not the other .
Two studies did not contain enough data to draw conclusions. A study in two villages (142 households) outside Kinshasa attempted to analyse net use by colour, but since only one net in the sample was not white, no conclusions could be drawn. ITN use by children age five to 15 was 90%, and just under 50% for adults. Shape was not recorded . A study from eastern Ethiopia used a cross-sectional design among 2867 households and multivariate regression to determine reasons for non-use of LLINs . ‘ITN colour preference’ was found to be a significant predictor, but was undefined in the publication. The authors stated that blue and cylindrical-shaped LLINs were preferable to white and rectangular, but showed no data for this.
In Zambia, a small cross-sectional survey using an MIS-style protocol among 483 households investigated the determinants of hanging and use of nets after a mass distribution . Net characteristics, including shape, colour and whether or not the ITN was purchased, were not associated with net deployment, controlling for background variables.
A community-based, cross-sectional survey among 530 LLIN-owning households in Sri Lanka found that if the shape was conical, the odds of LLIN use were 5.6 times higher than if the shape was rectangular. The nets were evenly split between Yorkool and Olyset nets, with 505 conical and 155 rectangular. The main reason for not using a net was due to heat or lack of mosquitoes; 79% of the nets were used the previous night. Other variables associated with use were more residents, fewer untreated nets, participants reporting practical benefits of LLINs over untreated nets (e.g., greater durability or ventilation), newer nets and lack of side effects .
Summary of literature review
The nineteen studies reporting on stated preferences do not link actual net use to the stated preferences, either by design (e.g., an acceptability trial providing only one type of net, or directing net use for a certain time period) or by oversight. Stated preferences can be useful in gauging likes and dislikes, but only determinants of use tell how net use is affected (or not) by these preferences.
Among the seven studies that reported on determinants of use, five included adequate information to compare shape and/or colour as dependent variables in multivariate analysis. Of these five, two found a significant association between shape and net use, and three found no effect. Regarding colour, only one study found a significant effect of colour on net use, and three found no effect. Neither overall size nor height of net were studied.
Secondary data analysis
Household stated preferences for shape
Households owning no nets
% of households who prefer rectangular nets
% of households who prefer conical nets
% no preference
% of households who prefer rectangular nets
% of households who prefer conical nets
% no preference
Guinea DHS 2012
Kenya MIS 2015
Madagascar MIS 2013*
Malawi DHS 2010
Malawi MIS 2012
Malawi MIS 2014
In Malawi 2014, participants were asked why they preferred specific shapes. These match closely with findings from other literature noted in Table 2: 49% reported that conical nets were easier to hang, compared to rectangular (23%); 29% reported that rectangular nets allowed more people to sleep under them, compared to conical (6%). Similar percentages (15%) of respondents reported that conical and rectangular nets fit beds better, look nicer (13–16%), and are stronger (9–11%).
When shape preferences were analysed sub-nationally, for stated preferences, three of eight regions in Guinea preferred rectangular nets, while four regions preferred conical, and one (Mamou) preferred conical and ‘don’t care’ equally. In Malawi in 2010, all three regions preferred conical nets roughly 60:40, and this continued in 2014, increasing to a roughly 70:30 ratio. In Kenya (2015), the Coast and Northeastern regions had strong preferences (>2:1) for conical nets, while the remaining provinces were evenly split between the two shapes, and households in Nairobi preferred rectangular nets.
Female survey respondents were slightly more likely to prefer conical nets than male respondents. Shape preferences were not correlated with housing type. These findings are described further in Additional file 2.
Size and textile preferences
In Madagascar 2013, of the 5154 respondents who said they did have preferences about nets, 92% preferred taller nets (180 cm height) versus shorter nets (150 cm height); 85% preferred softer textile (polyester) over ‘hard texture’ polyethylene nets.
Reasons for non-use or non-hanging of nets related to net attributes
Percent of nets used the previous night, by shape
% of nets in the sample that are
% of nets that were used the previous night
Crude OR of rectangular net being used (vs conical)
Difference (% points)
Burkina Faso MIS 2014
Gambia DHS 2013
Malawi DHS 2010
Malawi MIS 2012
Malawi MIS 2014
Mali MIS 2015
Nigeria MIS 2015
Rwanda DHS 2010
Rwanda MIS 2013
Rwanda DHS 2014
Senegal MIS 2008
Senegal DHS 2010
Modification of nets
Three surveys contained data on whether a net had been modified from its original shape. In Senegal 2012, 4.6% of nets were modified, increasing to 5.8% in the 2014 survey. The nature of the modification was not recorded. In Malawi 2014, 5.9% of the 588 conical nets had been altered from their original rectangular form to be conical. For three of these 35 nets, two nets were used to do the alteration; the remainder were altered using only the original net.
Determinants of net use
Twelve datasets from seven countries include information on net shape (Table 4). In Malawi and Senegal, the majority of nets are rectangular in all surveys. Rwanda is more or less evenly split between the two shapes; The Gambia has primarily conical nets (2:1). Burkina Faso, Mali, and Nigeria had nearly all rectangular nets (96–99%).
Per cent of nets used the previous night by shape, among households owning both shapes
Difference (% points)
Burkina Faso 2014
To confirm that ownership of particular shapes of nets was not restricted to certain regions or wealth quintiles, the nine datasets were assessed. The overall percentage of net-owning households that owned both conical and rectangular nets ranged from 12.4% (Malawi 2012) to 37.9% (Gambia 2013), with some households owning both shapes in each area of the country (Additional file 3). Owning both shapes was positively associated with increasing wealth quintile (Additional file 3), with the exception of The Gambia.
Except for the Rwanda 2013 MIS, the per cent of all conical and rectangular nets that were used the previous night are programmatically equivalent, with differences of between 1 and 7 percentage points. However, in The Gambia, Malawi and Rwanda DHS datasets containing 20,000–30,000 nets, these small percentage point differences result in statistically significant differences in net use.
Mean users per net for conical and rectangular nets
Mean users per net, all households
Mean users per net in households owning both shapes
Burkina Faso 2014
In Table 6 the mean number of users per net is higher for conical nets in The Gambia 2013, Nigeria 2015, Rwanda 2013 and Senegal 2008 when looking at all households. However, when the sample is restricted to households that own both shapes, the mean number of users per conical net is higher in all surveys, except Rwanda 2010, Mali 2015, and Nigeria 2015. Given the higher ownership of conical nets in wealthier quintiles (Fig. 3), this may indicate preferential use of conical nets by households that can afford to pay for their preferred net.
In Malawi, 2010 and 2014 data on preferences were recorded for each household. In 2010, the use rate for nets that did not match the user preference for shape and colour was the same as the use rate for nets that did match shape and colour preference, both at 64%. Nor did use rates differ for nets that matched (64%) or did not match (63%) shape preference only.
In 2014, the use rate for nets that did not match the user preference for shape and colour was 83%, versus 87% for nets that did match both shape and colour preference. If the net matched only the shape preference, use rate was 86%, compared to 82% for nets that did not match the shape preference. Neither of these differences was statistically significant at the p < 0.05 level.
Percentage of nets used the previous night for different net textile types
Burkina Faso 2014
Net use and size of nets in Senegal 2008 and 2010
Mean users per net
% of nets used
Determinants of net use: multivariate analysis
Adjusted odds ratios for outcome ‘net was used the previous night’ in 12 surveys across seven countries in sub-Saharan Africa
Burkina 2014 (14,086)
Gambia 2013 (13,486)
Malawi 2010 (30,035
Malawi 2012 (3314)
Malawi 2014 (4750)
Mali 2015 (13,942)
Nigeria 2015 (12,601)
Rwanda 2010 (20,786)
Rwanda 2013 (7923)
Rwanda 2014 (20,313)
Senegal 2008 (21,951)
Senegal 2010 (22,193)
Household-level fixed effects
Household net supply (ref: owns 1 net for 2 people)
Not enough (<1 net for 2 people)
Too many (>1 net for 2 people)
Household is urban (vs rural)
Age of household head (ref: <5)
Household head is female
Wealth quintiles (ref: poorest)
Net-level fixed effects
Net is conical (vs rectangular)
Net age (ref: <1 year)
Is an ITN (vs untreated)
Green (vs other colour)
Net matches stated preferences (ref: does not match)
Both shape and colour
Net was purchased
Average marginal effects (percentage points by which proportion of nets used the previous night increase if all nets are conical, or all nets match shape/colour preferences
Net is conical
Net matches both preferences (vs neither)
Matches only shape (vs neither)
Matches only colour (vs neither)
In the 2010 and 2014 Malawi MIS, information on a household’s preferred shape and colour of net were recorded. Nets were then categorized into four groups: neither the preferred shape nor colour; both the preferred shape and the preferred colour; just the preferred shape (but not colour); and, just the preferred colour. In 2010, net use was not significantly different whether the net type matched the user preferences or not. In 2014, a net that was both the preferred colour and shape had 1.78 greater odds of being used the previous night. Average marginal effects were computed for each country. For net shape the results were statistically significant only in the Malawi and Rwanda datasets. In those datasets, the overall percentage of nets used the previous night have significantly improved if all nets in the dataset were switched to conical shape, by 5.9 and 4.5 percentage points in Malawi 2010 and 2012, respectively, and by 13.1 and 5.3% points in Rwanda 2013 and 2014, respectively. Average marginal effects were also computed for the preferences variables in the Malawi datasets, but were generally not statistically significant. The results for the preference variables were statistically significant in the Malawi 2014 dataset, where the proportion of nets used the previous night would have increased by 5% points if all nets had been the user’s preferred shape and colour.
Costs for ITNs came down significantly in 2014–2015 as oil prices fell and competition among suppliers increased. Different net attributes have different cost implications. Standard colours (white, blue, green) have minimal cost differences, although white polyester nets are slightly cheaper than coloured nets, given the dyes and dying process involved. For polyethylene nets there is generally no price premium for colour, as additives for colour are needed even for white nets. The Value for Money report noted a price premium of US$0.79 for nets over 170 cm in height, and recommended that only at least double size (>130 cm wide) nets be procured to avoid limiting the number of sleepers .
Standard long-lasting insecticidal net reference prices from the Global Fund and from UNICEF Supply Division in 2016
Net type and dimensions
(cm: L × W × H)
Global Fund standard LLIN* reference price 2016 (US$)
UNICEF Supply Division June 2016
180 × 160 × 150
180 × 160 × 170
180 × 160 × 180
180 × 160 × 200
180 × 190 × 150
180 × 190 × 160
190 × 180 × 170
190 × 180 × 180
195 × 160 × 200
200 × 140 × 150
210 × 190 × 180
Hammock 140 × 240
+Net 240 × 65 × 120
1000 × 65 × 250
1050 × 65 × 220
1250 × 65 × 250
Conical nets are more expensive to manufacture as they require more skilled labourers for both cutting and sewing the trapezoidal panels, fixing the ring. Adjusting production lines for conical nets also has cost implications and sometimes implications for meeting production timelines. Conical nets require more fabric and raw materials (an additional 2–3 sq m of fabric, compared to rectangular nets, and the supporting ring). Due to their increased packing size and weight per net, conical nets can also be more expensive to transport and warehouse, as they can be up to twice the volume of rectangular nets . For calendar year 2015, two of the 104 orders listed in the Global Fund PQR Database were for conical ITNs, comprising 2% of total ITNs purchased.
Cost and utility functions
For most countries, gap analyses provide the number of ITNs required and volumes and, often, specifications are written into concept notes (in the case of the Global Fund) or Malaria Operational Plans (in the case of PMI). While there are not always fixed budgets from the outset countries are nonetheless faced with decisions about which net types to purchase. This can result in trade-offs whereby a more preferable net, which may be more likely to be used, is more expensive and therefore, given a fixed budget, fewer of these nets would be bought overall (or, alternatively, other components of the programme would be reduced). It is difficult currently to evaluate these trade-offs. However, this decision can be viewed as a basic optimization problem. For simplicity, a fixed budget is assumed. Two functions are needed: a cost function and a utility function. Here utility is defined as the total number of people using a net, using mean users per net.
If this last equation is evaluated at a given set of inputs, programmers can determine whether to switch or not: if the answer is positive, it is worthwhile to buy alternative nets, as the number of overall net users increases as a result. If the answer is negative, countries should procure the standard nets to maximize net users. If the answer is zero, there is no difference either way.
The final two equations here have been used in an Excel tool (Additional file 4) to further illustrate how programmes may decide to make evidence-based decisions about net procurement at different net costs. The tool provides space to make the inputs of mean users per net, cost and budget, and provides both the ‘yes/no’ answer about whether to switch to alternative nets, as well as the total number of net users expected with each type of net, and the expected loss or gain in net users if alternative or standard nets are procured.
Case study for shape: Rwanda
To illustrate the above equations, Rwanda provides a case study. From the data presented in Table 6, The Gambia 2013, Nigeria 2015, Rwanda 2013 and Senegal 2008 have higher mean users per net for conical nets than for rectangular, while the remaining countries have more users per rectangular net. The average cost for standard nets of various widths obtained from Global Fund in March 2016 (pers. comm.) was US$2.35 for rectangular and US$4.21 for conical (both CIF (Cost, Insurance and Freight). Rwanda’s population is roughly 12 million people. The country would need to procure about 6.6 million ITNs for a universal coverage campaign, using the ‘population divided by 1.8′ algorithm recommended by WHO [46, 47]. These parameters, inputted into the Excel tool, assume that the budget is fixed at 6.6 million ITNs x the cost of the standard net. The tool estimates that these 6.6 million standard nets would protect 10.2 million users, based on mean users per net, and that if alternative (conical) nets are procured, there would only be budget enough to procure 3.7 million of them instead of the needed 6.6 million. Despite higher mean users per conical net, total users if conical nets are purchased would be expected to be 6.1 million, an overall loss of 4.1 million users, due to the overall reduction in nets distributed. If the differential in mean users is calculated only in households that own both shapes (1.53), still just over 1.1 million users would be lost. In other words: conical nets would need to be less than 6% more expensive before any population-level gains in net users would be expected.
Case study for size: Senegal
While size data are limited, the mean users per two-place net and three-place net from Senegal can be used to assess whether procuring three-place nets (the alternative net in this scenario) would be cost effective. Using dimensions of 180H × 160L × 150W as two-place or double, and 180H × 160L × 170W as three-place or queen size, with a cost differential of US$0.19, in Senegal’s case (population roughly 15 million, 7.5 million nets required), one would expect to see a net gain of about 750,000 net users if the three-place nets were procured, using the Excel tool. The increased mean users per larger net offsets an overall reduction of about 610,000 nets that could be procured.
While data are scarce on ITN attributes such as shape and size, the available data, taken together, indicate that user preferences for certain types of nets does not impede net use at the population level. It is worth noting that overall proportions of nets used the previous night are reasonably high (60–80%) and that the proportion of the population using nets if they have access to them is similarly high, with a mean use of ITNs given access of 81% across 81 surveys, and a median of 87%, as described in the ITN Access and Use Report .
It is telling that in only one survey were respondents asked about both their shape and colour preferences and the shapes and colours of the nets they own. It is not enough to simply ask respondents for their stated preferences, as this has no strong correlation with the rates at which the nets are used. What is observed is that people tend to use the nets they have, and that wealthier quintiles may seek out additional nets that meet their preferences. But preferences for shape and colour are not strong enough to impact net use rates overall, especially when taking into account the increased cost of conical ITNs.
In the multilevel model for Malawi 2014, having one’s preferred shape and colour net contributes to increased odds (1.78 OR) of using that net, but this association is not stronger than household supply of nets (having not enough, or more than enough), nor is it stronger than the net being an ITN. Average marginal effects were low (4.5%). Indeed, having only a net of one’s preferred shape (but not colour), or vice versa, was not significant in the model, demonstrating a relatively weak influence of preferences overall. Conical net shape improved the odds of the net being used by up to 2.26 times in Rwanda 2013, Rwanda 2014, Malawi 2010 and Malawi 2012. By 2014, however, shape alone ceased to be a significant predictor of net use in Malawi. Age of net was highly correlated (p < 0.001) with shapes and colours, but generally, net characteristics are aligned with distribution cohorts, and significant age results are likely a reflection of whenever the last campaign(s) occurred. Presence of holes and colour were not significant predictors of use.
The overall difference in the rates of use of nets of different shapes is minimal, ranging from 1 to 7 percentage points difference, and an average of 3.6 percentage points. The exception is Rwanda, where conical nets were used 14.3 percentage points higher than rectangular nets in 2013. While several of these small percentage point differences were statistically significant, due to the very large sample size of over 20,000 nets, these statistical differences in use rates should not be interpreted as programmatically significant.
In addition, when mean users under each type of net is calculated, and costs of procurement are taken into account, using the Excel tool, it is clear that even in Rwanda, given a fixed budget for nets, procuring the more expensive net results in a direct loss of overall ITN coverage which is not made up by the increased use of conical nets. Therefore, decisions to procure conical nets do not provide value for money under a fixed budget unless the price of conical nets is considerably reduced.
There is a significant influence of wealth quintile on both ownership and preferences for conical nets in particular. Wealthier households are more likely to own conical nets, more likely to own both conical and rectangular nets, and more likely to express preferences for conical nets. Procuring more expensive conical nets with public funds to satisfy the preferences of wealthier households goes against public health principles, especially when these are the very groups that have the means to purchase their preferred net if unsatisfied with the free rectangular option. Likewise, there is bias inherent in the observations in households owning both shapes where preferential use of conical nets over rectangular nets is seen (except in The Gambia), and the increased use of conical nets cannot therefore be extrapolated to all households.
The reasons for non-use of nets are well known and well described in Pulford et al. The primary reason for a net going unused the previous night are that it is not needed (extra), or because the perceived mosquito density is low, or because of heat . These findings are corroborated with the data presented here. Among the many possible answers to the question ‘why did no one sleep under this net the previous night’, only a tiny proportion of respondents cite shape or size as the key problem. This again points to preferences being present, perhaps, but not key factors that determine net use. Likewise, making the effort to transform a net indicates a relatively strong level of shape preference. The fact that only 5% of nets in Senegal and 5% of conical nets in Malawi had been transformed indicates that the majority of the population in these settings are content to sleep under the net they have. Whether or not these types of alterations have any negative or positive impact on overall ITN durability remains to be studied.
Minimal data have been collected on preferences for textiles, although these seem relatively strong in Madagascar. There are no programmatically significant differences observed between textiles. There are relatively strong and uniform colour preferences in the available datasets, and as colour is not a key cost driver for ITNs , the current standard options of white, blue and green are adequate.
There are almost no data to assess the influence of size on net use. Anecdotally, taller nets are preferred as they provide more room to move around, more room for sexual activity and are easier to tuck in (Some of the preference for conical nets may in fact be due to their significantly taller height). Wider nets accommodate more sleepers. Indeed, given the increased mean users for wider nets, procuring these nets may be cost effective, given the relatively small increase in cost. However, procuring wider nets will not always result in increased users in each household, as a family of four, for example, may use one two-place net for the parents and one two-place net for the two children, and receiving three-place nets would not lead to additional individuals being protected. Likewise, several studies have noted that children age 6–15 years are the least likely to sleep under nets [49–51], both because they are less prioritized than small children and adults when nets are insufficient, and because they may not always be able to share sleeping spaces with siblings due to gender differences and associated cultural norms. Whether or not procuring wider nets can resolve these issues remains to be seen.
The variability of results over time in Malawi and in Rwanda demonstrate that preferences are not static. While stated preferences for conical nets in Malawi increased from 2010 to 2014, the strength of the association between conical shape and net use the previous night decreased over the same period. In Rwanda, conical nets were less likely to be used than rectangular nets in 2010, then significantly more likely than rectangular in 2013, and then less strongly in 2014. While it is possible that survey timing or sampling contributes to these changes, it is also likely that preferences do change over time, as households become accustomed to different types of nets. Both Guinea and Zimbabwe have recently procured rectangular nets after several rounds of conical nets (PMI, pers. comm.), and it will be important to assess use patterns after this transition.
The available data come mainly from sub-Saharan Africa, which has experienced in some countries, several rounds of mass ITN campaigns, and where retail markets and ability to purchase nets of one’s preference have been quite limited compared to Southeast Asia. These findings should not be extrapolated to Southeast Asia, where markets for nets are more vibrant and consumer preferences may have stronger influence over net use. Further research is needed to assess rates of net use for different net attributes in this region.
Additionally, the analysis of individual and household level data related to the effect of preferences for net type on the use of nets may be somewhat compromised by self-selection of households into ownership of the type of nets they are most likely to use. However, the limited market availability of varied types of nets may restrict this especially in rural areas of sub-Saharan Africa.
These analyses indicate that overall, at current price differentials, rectangular nets are a ‘best buy’ for the majority of the population and provide the most value for money. Given the trends in conical net preference (and purchase) by wealthier households, it is likely that the retail market is better suited to provide conical nets to these households than are donor organizations and Ministries of Health.
Recommendations for measuring net preferences and their impact
Small scale qualitative research on preferences: these studies, generally implemented using focus group discussions, with a limited sample in a restricted study area, have a tendency to focus largely on barriers to use when ITN use rates are relatively strong. These can make issues such as difficulty hanging nets or not having just the right shape and colour and size appear more important than they really are. Berthe et al. illustrate the concept nicely whereby focus group discussions often elicit vigorous complaints about nets despite high rates of actual use .
Small scale acceptability study: these studies tend to distribute ITNs of one or multiple types to a small number of households in a particular study area, usually at the initial stages of production or testing in order to confirm that the product is acceptable. However, the limited study area prevents extrapolation to the rest of the country, which becomes problematic for mass procurement, where the preferences of a small number of users would be applied to a nation as a whole. These types of studies may be better suited for particular vulnerable niche groups, or in Trials of Improved Practices (TIPS), for example among nomadic groups in sub-Saharan Africa, forest goers in Southeast Asia , and in South America . Cluster-randomized, crossover acceptability studies, such as Atkinson  in the Solomon Islands, are a better approach than single acceptability studies, but have the same limitations in terms of national representativeness, and costs of ensuring national representation are likely to be prohibitive.
Household survey asking only about preferences: the results from this type of survey are able to assess net preferences across the country (if the survey is nationwide), but tell programmes absolutely nothing about whether or not these stated preferences have any influence on overall use of ITNs.
Household survey asking only about net shape, size: the results from these surveys provide valuable information on whether nets of various attributes are used differently, but cannot be linked to preferences which would show whether or not having a net of one’s preference is associated with an increased probability of being used. However, this type of data can be used to generate mean users per net of each type, which can then be used in the Excel Tool to calculate whether switching to the alternative net under a fixed budget would result in increased or decreased overall net use.
Household survey recording both preferences and matching net attributes: for countries that wish to justify procurement based on preferences, both these variables must be recorded and analysed. The implications of this are the addition of questions in the net roster on net shape and net size, and in the household questionnaire on key attributes (preferred shape, preferred size of net, preferred textile while it is easy to record shape and colour, width and length present challenges for measurement in the course of household interviews, as nets are not always available for measurement (e.g., not hanging, or tied up), and may not be easily categorized into single, double, queen, king, and other sizes (particularly for conical nets). Requesting that data collectors measure each net’s height length and width with a tape measure is likely too time consuming and consistently accurate measurements would be difficult. These studies could also be compromised by selection of households to net types that are preferred, which might bias estimates of the effect of preferences on net use.
These findings are limited by the lack of widespread data on net shape, width, colour, and length, all of which are key cost drivers, and by lack of nationally representative data on preferences for these attributes, most notably in Southeast Asia. In addition, despite massive scale-up of ITN distribution in the last decade, access to ITNs is still below target. Practically, this means that most people do not have a choice of nets to use within their household; they do not choose to sleep under a conical net or a square net on any given night; they choose to sleep or not sleep under the net (or nets) that they have access to. This limits household flexibility in using any preferred net, which this study has attempted to illustrate in the analyses on the households owning both shapes.
Stated preferences for shape and colour do not significantly influence net use to degrees that would require action by programme planners. By and large, people are using the nets they receive, and when they do not, it is for reasons unrelated to shape and size (primarily perceived mosquito density, heat or an excess of nets). Wealthier quintiles tend to own greater numbers of conical nets, indicating that they have the ability to obtain or purchase these nets on their own, and wealthier quintiles also use conical nets preferentially. However, the preferences and use patterns of a sub-set of wealthier households should not drive procurement policies, as the increased manufacturing costs for conical nets are not outweighed by the very small, often non-existent, increases in use rates. Programmes that wish to explore the relationship between net attributes, preferences and use rates should include these questions in nationally representative household surveys to be able to capture trends across geographic and socio-economic groups.
HK reviewed the literature and conducted the secondary analysis and drafted the paper; JOY developed the cost functions and drafted portions of the paper. Both authors read and approved the final manuscript.
The authors are grateful to staff at PMI, the Global Fund, WHO Global Malaria Programme, UNICEF, and several LLIN manufacturers for their inputs on this topic.
The authors declare that they have no competing interests.
Availability of data and materials
The data used in this study are publicly available from dhsprograms.com.
This report is made possible by the generous support of the American people through the US Agency for International Development (USAID) under the terms of USAID/JHU Cooperative Agreement No: AID-OAA-A-14-00057. The contents do not necessarily reflect the views of USAID or the US Government.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- WHO. World Malaria Report 2015. Geneva: World Health Organization; 2015. p. 1–280.Google Scholar
- Hetzel MW, Gideon G, Lote N, Makita L, Siba PM, Mueller I. Ownership and usage of mosquito nets after four years of large-scale free distribution in Papua New Guinea. Malar J. 2012;11:192.View ArticlePubMedPubMed CentralGoogle Scholar
- Eisele TP, Keating J, Littrell M, Larsen D, Macintyre K. Assessment of insecticide-treated bednet use among children and pregnant women across 15 countries using standardized national surveys. Am J Trop Med Hyg. 2009;80:209–14.PubMedGoogle Scholar
- Koenker H, Kilian A. Recalculating the net use gap: a multi-country comparison of ITN use versus ITN access. PLoS ONE. 2014;9:e97496.View ArticlePubMedPubMed CentralGoogle Scholar
- Pulford J, Hetzel MW, Bryant M, Siba PM, Mueller I. Reported reasons for not using a mosquito net when one is available: a review of the published literature. Malar J. 2011;10:83.View ArticlePubMedPubMed CentralGoogle Scholar
- Koenker HM, Yukich JO, Mkindi A, Mandike R, Brown N, Kilian A, et al. Analysing and recommending options for maintaining universal coverage with long-lasting insecticidal nets: the case of Tanzania in 2011. Malar J. 2013;12:150.View ArticlePubMedPubMed CentralGoogle Scholar
- Results for Development Institute. Value for Money in LLIN Specifications Guidance. Washington DC: Results for Development Institute; 2012.Google Scholar
- Grietens KP, Muela Ribera J, Soto V, Tenorio A, Hoibak S, Aguirre AR, et al. Traditional nets interfere with the uptake of long-lasting insecticidal nets in the Peruvian Amazon: the relevance of net preference for achieving high coverage and use. PLoS ONE. 2013;8:e50294.View ArticlePubMedPubMed CentralGoogle Scholar
- Interim Standard Procurement Specifications for LLINs purchased with Global Fund grant funds 17 December 2015. 2016.Google Scholar
- President’s Malaria Initiative, USAID Global Health. President’s Malaria Initiative Technical Guidance. 2015:1–154.Google Scholar
- The DHS Program—quality information to plan, monitor and improve population, health, and nutrition programs http://www.dhsprogram.com.
- Baume CA, Reithinger R, Woldehanna S. Factors associated with use and non-use of mosquito nets owned in Oromia and Amhara regional states, Ethiopia. Malar J. 2009;8:264.View ArticlePubMedPubMed CentralGoogle Scholar
- Baume CA, Koh ACF. Predictors of mosquito net use in Ghana. Malar J. 2011;10:265.View ArticlePubMedPubMed CentralGoogle Scholar
- MacCormack CP, Snow RW. Gambian cultural preferences in the use of insecticide-impregnated bed nets. J Trop Med Hyg. 1986;89:295–302.PubMedGoogle Scholar
- MacCormack CP, Snow RW. Use of insecticide-impregnated bed nets in Gambian primary health care: economic aspects. Bull World Health Organ. 1989;67:209–14.PubMedPubMed CentralGoogle Scholar
- MacCormack CP. The human host as active agent in malaria epidemiology. Trop Med Parasitol. 1987;38:233–5.PubMedGoogle Scholar
- Aikins MK, Pickering H, Alonso PL. A malaria control trial using insecticide-treated bed nets and targeted chemoprophylaxis in a rural area of The Gambia, West Africa 4. Perceptions of the causes of malaria and its treatment and prevention in the study area. Trans R Soc Trop Med Hyg. 1993;87:25–30.View ArticlePubMedGoogle Scholar
- Binka FN, Adongo P. Acceptability and use of insecticide impregnated bednets in northern Ghana. Trop Med Int Health. 1997;2:499–507.View ArticlePubMedGoogle Scholar
- Gyapong M, Gyapong JO, Amankwa J. Introducing insecticide impregnated bednets in an area of low bednet usage: an exploratory study in north-east Ghana. Trop Med Int Health. 1996;1:328–33.View ArticlePubMedGoogle Scholar
- Tami A, Mubyazi G, Talbert A, Mshinda H, Duchon S, Lengeler C. Evaluation of Olyset insecticide-treated nets distributed seven years previously in Tanzania. Malar J. 2004;3:19.View ArticlePubMedPubMed CentralGoogle Scholar
- Das ML, Singh SP, Vanlerberghe V, Rijal S, Rai M, Karki P, et al. Population preference of net texture prior to bed net trial in Kala-Azar-endemic areas. PLoS Negl Trop Dis. 2007;1:e100.View ArticlePubMedPubMed CentralGoogle Scholar
- Fernando SD, Abeyasinghe RR, Galappaththy GNL, Gunawardena N, Rajapakse LC. Community factors affecting long-lasting impregnated mosquito net use for malaria control in Sri Lanka. Trans R Soc Trop Med Hyg. 2008;102:1081–8.View ArticlePubMedGoogle Scholar
- Atkinson J-A, Bobogare A, Fitzgerald L, Boaz L, Appleyard B, Toaliu H, et al. 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. Malar J. 2009;8:119.View ArticlePubMedPubMed CentralGoogle Scholar
- Atkinson J-A, Bobogare A, Vallely A, Boaz L, Kelly G, Basifiri W, et al. A cluster randomized controlled cross-over bed net acceptability and preference trial in Solomon Islands: community participation in shaping policy for malaria elimination. Malar J. 2009;8:298.View ArticlePubMedPubMed CentralGoogle Scholar
- Atkinson J-A, Fitzgerald L, Toaliu H, Taleo G, Tynan A, Whittaker M, et al. Community participation for malaria elimination in Tafea Province, Vanuatu: part I. Maintaining motivation for prevention practices in the context of disappearing disease. Malar J. 2010;9:93.View ArticlePubMedPubMed CentralGoogle Scholar
- Lover AA, Sutton BA, Asy AJ, Wilder-Smith A. An exploratory study of treated-bed nets in Timor-Leste: patterns of intended and alternative usage. Malar J. 2011;10:199.View ArticlePubMedPubMed CentralGoogle Scholar
- Ng’ang’a PN, Jayasinghe G, Kimani V, Shililu J, Kabutha C, Kabuage L, et al. Bed net use and associated factors in a rice farming community in Central Kenya. Malar J. 2009;8:64.View ArticlePubMedPubMed CentralGoogle Scholar
- Banek K, Kilian A, Allan R. Evaluation of Interceptor long-lasting insecticidal nets in eight communities in Liberia. Malar J. 2010;9:84.View ArticlePubMedPubMed CentralGoogle Scholar
- Sande S, Jagals P, Mupeta B. An investigation of the use of rectangular insecticide-treated nets for malaria control in Chipinge District, Zimbabwe: a descriptive study. Pan Afr Med J. 2012;13:5.PubMedPubMed CentralGoogle Scholar
- Harvey SA, Olórtegui MP, Leontsini E, Pezo CB. The whole world will be able to see us: determining the characteristics of a culturally appropriate bed net among communities of the Peruvian Amazon. Am J Trop Med Hyg. 2008;79:834–8.PubMedGoogle Scholar
- Aleme A, Girma E, Fentahun N. Willingness to pay for insecticide-treated nets in Berehet District, Amhara Region, Northern Ethiopia: implication of social marketing. Ethiop J Health Sci. 2014;24:75–84.View ArticlePubMedPubMed CentralGoogle Scholar
- Birhanu Z, Abebe L, Sudhakar M, Dissanayake G, Yihdego Y, Alemayehu G, et al. Access to and use gaps of insecticide-treated nets among communities in Jimma Zone, southwestern Ethiopia: baseline results from malaria education interventions. BMC Public Health. 2015;15:1–11.View ArticleGoogle Scholar
- Berthe S, Loll D, Faye SL, Wone I, Koenker H, Arnold B, et al. “When I sleep under the net, nothing bothers me; I sleep well and I‘m happy”: Senegal’s culture of net use and how inconveniences to net use do not translate to net abandonment. Malar J. 2014;13:357.View ArticlePubMedPubMed CentralGoogle Scholar
- Bean J. OXFAM GB Malaria Control Assessment Upper Nile–South Sudan 16 Apr–31 May 2001. 2001:1–19.Google Scholar
- Gore-Langton GR, Mungai J, Alenwi N, Abagira A, Bicknell OM, Harrison R, et al. Investigating the acceptability of non-mesh, long-lasting insecticidal nets amongst nomadic communities in Garissa County, Kenya using a prospective, longitudinal study design and cross-sectional household surveys. Malar J. 2015;14:52.View ArticlePubMedPubMed CentralGoogle Scholar
- Gryseels C, Durnez L, Gerrets R, Uk S, Suon S, Set S, et al. Re-imagining malaria: heterogeneity of human and mosquito behaviour in relation to residual malaria transmission in Cambodia. Malar J. 2015;14:330–42.View ArticleGoogle Scholar
- Shirayama Y, Phompida S, Kuroiwa C, Miyoshi M, Okumura J, Kobayashi J. Maintenance behaviour and long-lasting insecticide-treated nets (LLITNs) previously introduced into Bourapar district, Khammouane province, Lao PDR. Public Health. 2007;121:122–9.View ArticlePubMedGoogle Scholar
- Mattern C, Pourette D, Raboanary E, Kesteman T, Piola P, Randrianarivelojosia M, et al. “Tazomoka is not a problem”. Local perspectives on malaria, fever case management and bed net use in Madagascar. PLoS ONE. 2016;11:e0151068.View ArticlePubMedPubMed CentralGoogle Scholar
- Beer N, Ali AS, de Savigny D, Al-Mafazy A-WH, Ramsan M, Abass AK, et al. System effectiveness of a targeted free mass distribution of long lasting insecticidal nets in Zanzibar, Tanzania. Malar J. 2010;9:173.View ArticlePubMedPubMed CentralGoogle Scholar
- Ndjinga JK, Minakawa N. The importance of education to increase the use of bed nets in villages outside of Kinshasa, Democratic Republic of the Congo. Malar J. 2010;9:279.View ArticlePubMedPubMed CentralGoogle Scholar
- Gobena T, Berhane Y, Worku A. Low long-lasting insecticide nets (LLINs) use among household members for protection against mosquito bite in Eastern Ethiopia. BMC Public Health. 2012;12:914.View ArticlePubMedPubMed CentralGoogle Scholar
- Macintyre K, Littrell M, Keating J, Hamainza B, Miller J, Eisele TP. Determinants of hanging and use of ITNs in the context of near universal coverage in Zambia. Health Policy Plan. 2012;27:316–25.View ArticlePubMedGoogle Scholar
- Whidden CE, Premaratne RG, Jayanetti SR, Fernando SD. Patterns and predictive factors of long-lasting insecticidal net usage in a previously high malaria endemic area in Sri Lanka: a cross-sectional survey. Trans R Soc Trop Med Hyg. 2015;109:553–62.View ArticlePubMedGoogle Scholar
- Ferguson AF. Long-lasting insecticide-treated net (LLIN) market and data analysis 2011 Addendum. Arlington: USAID; 2012.Google Scholar
- Alliance for Malaria Prevention. Toolkit for Mass ITN Distribution Campaigns 2.0.Google Scholar
- WHO. Recommendations for achieving universal coverage with long-lasting insecticidal nets in malaria control. Geneva: World Health Organization; 2013.Google Scholar
- Kilian A, Boulay M, Koenker H, Lynch M. How many mosquito nets are needed to achieve universal coverage? Recommendations for the quantification and allocation of long-lasting insecticidal nets for mass campaigns. Malar J. 2010;9:330.PubMedPubMed CentralGoogle Scholar
- ITN Access and Use—VectorWorks. 2016. http://www.vector-works.org/resources/itn-access-and-use/. p. 1–107.
- Ricotta E, Koenker H, Kilian A, Lynch M. Are pregnant women prioritized for bed nets? An assessment using survey data from 10 African countries. Glob Health Sci Pract. 2014;2:165–72.View ArticlePubMedPubMed CentralGoogle Scholar
- Kilian A, Koenker H, Baba E, Onyefunafoa EO, Selby RA, Lokko K, et al. Universal coverage with insecticide-treated nets–applying the revised indicators for ownership and use to the Nigeria 2010 malaria indicator survey data. Malar J. 2013;12:314.View ArticlePubMedPubMed CentralGoogle Scholar
- Buchwald AG, Walldorf JA, Cohee LM, Coalson JE, Chimbiya N, Bauleni A, et al. Bed net use among school-aged children after a universal bed net campaign in Malawi. Malar J. 2016;15:127.View ArticlePubMedPubMed CentralGoogle Scholar
- Grietens KP, Xuan XN, Ribera J, Duc TN, Bortel WV, Ba NT, et al. Social determinants of long lasting insecticidal hammock use among the Ra-glai ethnic minority in Vietnam: implications for forest malaria control. PLoS ONE. 2012;7:e29991.View ArticlePubMedGoogle Scholar
- Harvey SA, Paredes Olórtegui M, Leontsini E, Ramal Asayag C, Scott K, Winch PJ. Trials of improved practices (TIPs): a strategy for making long-lasting nets last longer? Am J Trop Med Hyg. 2013;88:1109–15.View ArticlePubMedPubMed CentralGoogle Scholar