Assessment of long-lasting insecticidal net coverage , use and physical integrity one year after universal distribution campaign in Plateau department in South-East Benin

Assessment of long-lasting insecticidal net coverage, use and physical integrity one year after universal distribution campaign in Plateau department in SouthEast Benin Filémon T. Tokponnon *, Yolande Sissinto, Aurore Hounto Ogouyémi, Arthur Sovi, Virgile Gnanguenon, Adicath Adéola Adéothy , Abel Wapo, Bruno Aholoukpe, Patrick Makoutodé, Alioun Adéchoubou, Ossè Razaki, Raphael N’guessan, Dina Gbénou, Mariam Oke, Dorothée Kinde Gazard, Achille Massougbodji and Martin C. Akogbéto

vector-borne diseases (World Health Organisation (WHO), 2005).Several studies in malaria endemic countries have shown the usefulness of LLINs in reducing man-vector contact from malaria (Greenwood et al., 2005;Lengeler, 2004;Eisele et al., 2006;Thwing et al., 2008).It is a technology based on the slow release of pyrethroid insecticides, rendering it wash-resistant and extending insecticide residual effectiveness to at least three years without the need of re-treatment.This is why the recent objective of roll back malaria (RBM) focuses duplication of efforts including maintaining universal coverage of nets to achieve a 75% reduction compared to 2000, the number of reported malaria cases in endemic countries in the African region, 2015 (OMS, 2011; AFR/RC50/12 -WHO, 2012) .Benin, by its National Malaria Control Program, supported since 2000 the initiative RBM and conducted in July, 2011 a universal distribution campaign where approximately 5 million of nets were distributed.This distribution is intended to significantly increase the national coverage of insecticide-treated nets.Before the distribution, the proportion of households owning at least one net at national level was estimated at 52% (PNLP-AFRICARE BENIN-CRS BENIN, 2010).After the distribution of July, 2011, this proportion was around 86.4% (PNLP, 2011).In the net level analysis, factors independently associated in both surveys with reduced likelihood that a net would be used were: increasing net age, increasing damage of nets, increasing household net density (nets/person).Some studies have shown that several factors are associated with LLIN ownership and effective use (Graves et al., 2011).Factors associated with ownership were assumed to be: cultural beliefs and practices mechanisms of LLIN distribution and distance to LLIN suppliers, rumours about LLINs and social support and pressure (Wiseman et al., 2007;Baume and Marin, 2007).But factors associated with net used were reported to be: perceived benefits and disadvantages of nets, trust in health workers providing health education and LLINs, knowledge of appropriate net use/care practices, and net-hanging skills, household size and composition, the number of children under five years of age and use of other vector control measures (Toé et al., 2009;Edelu et al., 2010).Other studies conducted in Benin showed that several mosquito nets after 12 month of use had already holes (Gnanguenon, personnal communication).From these observations, questions arise: Do people effectively use the mosquito nets freely distributed?Or do people not use LLINs to their insufficient number or their poor physical condition?To clarify these questions, we conducted a study on insecticide-treated nets coverage, use and physical condition in plateau department in southern Benin one year after the distribution.This study aimed to provide useful data on the effectiveness of the distribution Tokponnon et al. 77 campaign.

Study area
Plateau is a county of Benin in West Africa.The selection of this county was based on its geographic accessibility, the high use of mosquito nets by children under 5. Entomological surveys conducted in the plateau have shown that there are both high and low pyrethroids resistance areas (Yadouleton et al., 2010) 51"E and 7° 5' 12"N; it has an area of 400 km², representing 11% of the county's area.32 rural villages were selected through all four districts (Figure 1).

Study design
Larvae prospection's for insecticides susceptibility tests on Anopheles gambiae, main malaria vector in Benin have been conducted in many villages of the targeted districts.This baseline study on the resistance of malaria vectors to deltamethrin in the department of Plateau helped to make the choice of sentinel villages where various activities were held in our work.A total of 32 clusters were selected including 17 clusters at Ifangni, 6 at Sakété, 2 at Pobè and 8 at Kétou.Each cluster (village) was composed of several hamlets and comprised a minimum of 100 Children under five years old.Household cross-sectional surveys were undertaken in each cluster in May and August, 2012, conducting during high malaria transmission period.The survey covered the targeted groups in different villages.The choice of the targeted population has followed the standards set in the collection of basic data on morbidity and mortality due to malaria in the monitoring/evaluation RBM/RBM in Benin in 2004 (Kinde-Gazard et al., 2004).The targeted persons by cluster were: 30 mothers of children or care for children less than 5 years to evaluate fever or malaria during the last two weeks, 25 Pregnant women in the third trimester of pregnancy and women who gave birth in the last 6 months to assess malaria prevention, 30 householders for the availability and use of LLINs.The households were randomly selected from each cluster.Data on bed net ownership and usage, physical condition of the nets, demographics of household members, other characters and issues related to sourcing were asked of all targets to assess the origins of LLIN found at the household level.Data were gathered using an adapted version of the standard Malaria Indicator Survey (RBM, 2013).Specific questions relating to the Universal Campaign Coverage process were asked to the householder and to the mothers of children or care for children less than 5 years.The physical condition of the nets was estimated using one of the two indicators recommended by WHO: the proportion of LLINs with any hole(s) (WHO, 2011).The main hole category in the LLINs was recorded as follows: T1: holes size < thumb (0.5 to 2 cm); T2: holes size > thumb < fist (2 to 10 cm); T3: holes size > fist < head (10 to 25 cm); T4: holes size > head (> 25 cm).

Statistical analysis
Interviews were conducted using questions.At the end of the survey, data were recorded with Epi-Info and data were transferred  2012).

Ethics approval
This study was planned and approved by the Ministry of Health.
The protocol was also reviewed and approved by National Ethics Committee for Health Research at the Ministry of Health.A briefing note indicating the objectives of the study, the advantages and disadvantages was given to the respondents in order to obtain consent.Confidentiality was respected and questionnaires were anonymous.

Study households characteristics
960 households and 4688 people were included in the survey and the response rate was 100% (  2).

Factors associated with ITN ownership
The sex of the heads of households (male and female) as well as their education level were not associated with ITN ownership (p > 0.05).Also, the composition of the household size in children under five, pregnant women and persons over five years old was not associated with ITN ownership (p > 0.05) (Table 3).

ITNs utilization
Of the total of 1746 reported LLINs, 42.96% (n = 750) LLINs were observed by the surveyors.84% of the nets were reportedly used the night preceding the survey.The people over 5 years were reported as using LLINs (Figure 3).On the 807 LLINs observed, around 16% was found with hole(s) (Table 5).No significant difference was observed between categories of holes (T1 to T4) and while comparing the physical condition of the nets by the level of instruction of their owners (p < 0.05).

DISCUSSION
ITNs coverage was increased by the universal distribution  distribution campaign with a proportion of 88.96% of households that owned at least one ITN.This proportion is higher when compared to the preceding distribution campaign evaluation (Tokponnon et al., 2013), and the average number of ITNs per household was around 2. Mosquito nets (LLINs) that were from the distribution campaign were significantly higher (87%) comparatively to those that were purchased full price (3%) or received from pregnant women antenatal consultation (9%).In our study, the average number of LLINs per household visited was 1.82 considering the 960 households.The average household size was 4.88 people visited.Thus we can say that about two nets are available for 5 people at the household level.However, the objective of this campaign is to increase the distribution level of 56% coverage (PNLP, 2012) at least 80% and have a LLIN for two people in the general population.
In the protocol of the LLIN distribution campaign of July, 2011, households of two people were covered with two LLINs and this rule has been well respected in rural areas in which the distribution has been fewer problems (Tokponnon et al., 2013).Our observations were consistent with the same requirements a year after the campaign.A disadvantage of the distribution was that 11% of households did not receive LLINs (PNLP, 2012).There were no significant differences between the availability of LLINs in different villages when considering the proportion of households with at least two LLINs.This   proportion was higher compared to those with three or four due to the fact that LLIN distribution campaign has limited the number of nets distributed for two people and a maximum of 8 LLINs for large households at the same size.This observation is a good performance and the coverage attained was similar to what has been achieved in other countries (Tanzania, Nigeria and Togo) (West et al., 2012;Stevens et al., 2013).Continuous distribution of LLINs to pregnant women and children under five is an important way to increase nets coverage and replace torn nets (RBM, 2011).An additional continuous distribution system, via NGOs for example, can also be planned for other households.78% of the net observed were reported to be used the previous night, and the proportion of nets use did not vary significantly between clusters, suggesting that awareness for net use was a success.But full net used by target group was not achieved.
76.4% of pregnants women were using ITNs while only 82.88% of children under five were using ITNs.This observation suggests that net usage could be improved in target group.
The high ITNs usage could be a consequence of the increased availability of ITNs at the household level due to the universal distribution campaign.This observation has already been noticed in Tanzania by West and colleagues (West et al., 2012).The level of ITN use could also be influenced by high temperature and mosquito density (Graves et al., 2011).But comparatively to other studies, ITN usage observed in target group in this study was higher than those observed in Tanzania and Nigeria (Ye et al., 2012;West et al., 2012).ITN ownership and use seems to reflect the general coverage and ITNs ownership in the whole population.The proportion of LLINs from the distribution campaign found in good condition was significantly higher than those observed in serviceable condition or torn out.

Conclusion
The universal distribution campaign in Benin has significantly increased ITN coverage and ownership policy in the community.Non target persons and target persons (pregnant women and children under five) had similar level in ITN ownership and usage.The level of ITN ownership and usage were also similar between villages.This is an important indicator of universal coverage goal.But additional effort must be done to fully achieve universal coverage goal, and routine distribution must be used to maintain ITN coverage.

Figure 1 .
Figure 1.Map showing the study villages in Plateau department (Benin).

Figure 3 .
Figure 3. Net usages in target and non target group.

Table 1 .
Background characteristics of the households surveyed.
into SPSS 16.0 software.The investigators had cleaned and analysed the data using the same software program.Household ownership of LLIN was calculated as a proportion of households having at least one LLIN among the total surveyed households.LLIN use was estimated as the proportion of households using at least one LLIN in the LLIN owning households.Once the LLIN ownership and use were determined; the data were filtered into a separate file of LLIN owned households.Then, households using LLINs were compared to those who did not use any to identify the factors associated with LLINs non-use.Barriers of LLIN use were determined by using enter method multivariate logistic regression model.Data on the physical integrity of LLINs and those relating to the origins of LLINs were extracted.The various holes observed were divided into four and frequencies were estimated to assess the general condition of operation of LLINs(WHO, 2011; Kilian,

Educational level of the heads of households
n = number CI = Confidence Interval.

Table 4 .
Multivariate regression on ITN use.

Table 4 .
Contd.: household who's everybody use LLIN the days before survey.M-: household who's not everybody use LLIN the days before survey.OR: odd ratio; Significant at p < 0.05 level. M+