Background
As of 2016, sub-Saharan Africa carried the vast majority of the global burden of malaria cases (90%) and deaths (92%); 70% of the deaths were in children under five [
1]. Use of long-lasting insecticide-treated nets (LLINs) and indoor residual spraying (IRS) significantly contributed to a 60% decline in malaria mortality between 2000 and 2015 [
1,
2]. However, the effectiveness of IRS and LLIN strategies is threatened by widespread development of insecticide resistance among vectors in most of the malaria-affected countries [
3‐
5]. Consequently, the 2017 WHO report indicates that in the year 2016, an estimated 216 million cases of malaria were reported, which represents an increase of about 5 million cases over 2015 and that the deaths reached 445,000, a similar number to the previous year [
3]. This demonstrates a stagnation of progress in the fight against malaria. For this reason, new mosquito control tools and strategies are needed to reinforce currently used methods in order to consolidate the gains made in the fight against malaria. In spite of current malaria vector control strategies focusing mostly on LLINs and IRS, a number of other measures can be implemented at household level to significantly reduce mosquito bites in humans. These strategies include integration and installation of house screening on doors, windows and eaves in order to prevent entry of adult mosquitoes in addition to environmental management aimed at eliminating mosquito breeding places near houses [
6,
7]. Installation of house screening is necessary considering that more than 80% of malaria transmission occurs indoors, primarily at night [
8]. This is, therefore, the most vulnerable time for people to be bitten and infected with malaria or other mosquito-borne pathogens. Thus, closing windows or screening of windows and open eaves can reduce the chances of mosquito bites, hence potentially lowering the occurrence of malaria, where mosquitoes usually feed on people indoors [
6,
9,
10]. Recently, clinical trials have shown that both full house screening and ceilings provide valuable protection against anemia and exposure to malaria transmission in rural parts of The Gambia and Equatorial Guinea [
6,
9,
11,
12]. Other studies in the Gambia, Guinea, Tanzania and Ethiopia have shown that house screening is a potential, sustainable and effective intervention in preventing mosquito entry into houses [
5,
10,
12‐
16]. In related studies in Western Kenya, house modifications involving insect screen ceilings made from locally available materials and use of insecticide-treated eave tubes resulted in significant reductions in human exposure to malaria vectors representing the first and promising results with this novel control tool against malaria vectors [
13,
17].
Improvement in housing design have contributed to reduction of malaria in many parts of the world [
18,
19]. Preventing mosquitoes from entering houses has other additional advantages such as protecting all household members equally and at all times whilst indoors and offering protection against other vector borne diseases through integrated vector control [
12].
While it might seem obvious that screening of houses can protect people against mosquito bites and malaria, this intervention is not common in many African rural communities [
9,
11,
20]. The objective of the study was to determine the status of house screening in a community living in a malaria endemic area of western Kenya and to evaluate the community’s knowledge and perceptions on available interventions.
Discussion
Nyabondo is one of the malaria endemic areas in western Kenya [
28,
29]. At the time of the study, malaria was reported to be one of the leading causes of morbidity and mortality in the study area with a reported 92 students from a nearby girl’s boarding school being hospitalised in a local mission hospital presenting with malaria signs and symptoms (local dispensary records). The morbidity was attributed to lack of appropriate personal protection measures in most boarding schools at night and also probably due to migration of students from areas where malaria is not endemic.
In sub-Saharan Africa, 80% of malaria transmission occurs indoors primarily at night, with dusk through dawn being the most vulnerable times for humans to be bitten by mosquitoes [
8]. In one of the FGDs it was noted that local people are knowledgeable about the peak biting time of the local vectors as confirmed from one of the focus group discussions (Siatok widows) where participants reported that female anopheles mosquitoes normally bite from 10 PM up to around 4 AM in the early morning. Field studies have shown that various species of
Anopheles have become more efficient vectors and have developed behaviours that allow them to adapt to indoor feeding, targeting humans in homes and other built environments before dusk [
8,
30‐
32].
Anopheles gambiae s.l., a common and dominant species in the study area is able to find openings in the dwelling, such as windows, cracks, or open eaves, while outdoor feeders have not adapted this behavior [
33,
34]. This vector species behaviour corresponds with the common characteristic feature of houses in the study area where 93.8% have walls made of mud & poles. Over 95% of the windows and eaves were unscreened thereby increasing the risk for house-entry by malaria vectors and that of other insect vectors. A modern, well-built housing can protect residents in many tropical countries by reducing house entry by mosquito vectors [
19,
33‐
36]. In the Gambia, houses with walls made of mud blocks showed increased number of indoors mosquitoes as opposed to concrete, this was most likely due to presence of more cracks and aging of mud blocks [
9,
11]. Open eaves were shown to be the main routes of entry by
An. gambiae s.l which is also the main mosquito species in the study area where almost all windows and eaves were unscreened.
Previous studies have shown that, for interventions to be effective in reducing malaria morbidity they need to be based on specific risk factors that contribute to disease transmission [
10]. These risk factors for malaria transmission include but not limited to environmental parameters like topography, proximity of dwelling places to mosquito breeding sites, house design, density of human populations, knowledge of local vector and use of vector control as well as their socio-economic status [
15,
37]. All these were potential risk factor in the study area, which is host to an estimated 34,000 people with a high population density of nearly 368 persons per square km., characterized by a host of various economic activities like brick-making with hot spots of active abandoned brick making pits and poorly managed or abandoned fish ponds not far from dwelling places [
21]. In other related studies in Africa malaria risk and transmissions has been shown to be strongly influenced by socio-demographic factors [
9,
11]. In Nyabondo, 75% of the respondents were subsistence farmers and more than half reported to have low level of education (53.8%) during this study.
On knowledge and use of malaria vector prevention measures, majority of respondents (97.4%) reported being aware of use of treated mosquito nets while less than 22% of the respondents reported other personal protection including use of fire and coils [
26,
27,
38]. This was also observed when respondents were interviewed on application of the same methods at household level. Despite most respondents confirmed use of treated nets, other personal protection methods were far less reportedly mentioned. In addition to lack of awareness or information on use of personal protection methods and socio-economic factors, participants in one of the FGDs mentioned ignorance and negligence as other reasons hampering both use of other personal protection methods and eaves screening at household level as shown in Fig.
2. To address this, there is need for regular community advocacy, communication, mobilization, and social promotion of available personal protection methods in the community through development of socially accepted Information, Education and Communication (IEC/BCC) strategies [
39,
40]. The approach should focus on awareness creation, education and enabling local people to optimize use of available vector control options and enhance use of house screening in the community. This could be done through use of appropriate and suitable user-friendly approaches such as interpersonal communication, mass media, schools, mobile teams, community field days or public forums. Such approaches should be used to create awareness through drama, health dialogue, role plays and songs, during annual malaria and mosquito day’s activities in the area prior to onset of heavy rains and peak malaria transmission seasons.
Comparatively, there was relatively high awareness and use of environmental management practices at household level. This was attributed to dissemination of information by three main sources i.e. ICIPE (87.5%) radio (62.5%) and ministry of health CHVs (37.5%). However, environmental management interventions are normally not very effective on their own unless appropriately integrated with other control measures [
41]. They do not have immediate effect in reducing the number of biting vectors and may take longer before reduction in vector population can be achieved. During FGDs, it was evident that local people rarely and actively participate in environmental management at village level unless there is a motivating factor to bring them together. This being common in most African rural settings, it is prudent for the community to understand both the vector ecology and the accrued short and long term benefits of environmental management [
39,
42,
43]. During the FGDs participants confirmed that they understood the direct benefits of participating in environmental management and if given opportunity, they were much willing to participate in malaria control and house screening.
However, 57.5% of the respondents acknowledged not to have heard any prior information on house screening. Major perceived reasons given for screening doors, windows and eaves were to prevent entry of mosquito and other insects into houses with considerable number of respondents mentioning preventing people from contracting malaria. Understanding community knowledge and perception on house screening and the changing self-protection behaviors and the interactions of humans is essential in ensuring success and sustainability of community based vector control interventions in Africa. Despite house screening being not common in Nyabondo, large number of respondents perceived that prevention of mosquito entry and other insects is the main reason for house screening doors, windows and eaves (> 85%). This indicates that the community may need to be informed on the need to protect their household against mosquito bites through house screening which is expected to significantly translate into preventing residents from contracting malaria and also the overall social and economic benefits accrued at household level. In related studies in Dar es Salaam, Tanzania, similar results were observed where many respondents understood that installation of ceilings protects them from mosquitoes and some respondents associate this with protection against malaria infection [
15,
36]. This indicates that, considering other socioeconomic determinants, house screening can be easily promoted based on its multiple benefits particularly in rural areas.
Another potential advantage of house screening is the equity with which it protects all members of the household at all times while indoors unlike LLINs which primarily give protection to those with a net during sleeping hours only [
44]. It can offer protection from other vector borne diseases like Lymphatic Filariasis, Rift Valley Fever, O’Nyong Nyong, Kala-azar, Dengue, Zika virus as well as malaria [
13]. Perhaps the greatest benefit to house eaves screening would be its potential for integration with other vector control interventions including environmental management. Lack of knowledge on how to screen houses (78%) was one of the major reasons given for not screening doors, windows and eaves in the study area. The second major reason reported was economic/affordability factors at household level. Affordability of screening was rated at nearly 40% which implies that cost is an important consideration for this technology and ultimate assimilation by communities will depend on the netting material and user-care dependent durability of the screens. On screening materials colour preference, grey was the most preferred colour and the reasons given during FGD was that most houses in the area are semi-permanent and are constructed using locally available materials except for the roofing, therefore the colour matched with the walls of the house and it didn’t display or get dirty quickly/easily. All questionnaire respondents confirmed that they were willing to screen their houses if given information and skills. House screening has also been shown to have high levels of acceptability in Africa when properly implemented, which could also make it an effective intervention in areas with high malaria transmissions [
20,
37,
41].
Acknowledgements
We would like to thank all the respondents for their hospitality, cooperation and consent to participate throughout the course of this study. We are also grateful to the household survey team, Key informants and the Focus group discussion participants, for their enthusiasm, perseverance and commitment during the survey. Many thanks also go to the field staffs Polycarp Aduogo, Arthur Sune and Simon Cheruiyot for their commitment and participation during the entire period of this study. Not forgetting the rigorous work done by the amazing reviewer (Hannah Koenker, Sarah Jane Moore and Teshome Gebre Kanno) who gave important comments that greatly helped improve and shape the manuscript. Final vote of thanks goes to Biovision and ICIPE for making this study possible.