- Open Access
Housing gaps, mosquitoes and public viewpoints: a mixed methods assessment of relationships between house characteristics, malaria vector biting risk and community perspectives in rural Tanzania
© The Author(s) 2018
- Received: 11 July 2018
- Accepted: 9 August 2018
- Published: 17 August 2018
The Correction to this article has been published in Malaria Journal 2018 17:327
House improvement and environmental management can significantly improve malaria transmission control in endemic communities. This study assessed the influence of physical characteristics of houses and surrounding environments on mosquito biting risk in rural Tanzanian villages, and examined knowledge and perceptions of residents on relationships between these factors and malaria transmission. The study further assessed whether people worried about these risks and how they coped.
Entomological surveys of indoor mosquito densities were conducted across four villages in Ulanga district, south-eastern Tanzania. The survey involved 48 sentinel houses sampled monthly and other sets of 48 houses randomly recruited each month for one-off sampling over 12 months. Physical characteristics of the houses and surrounding environments were recorded. Questionnaire surveys were administered to 200 household heads to assess their knowledge and concerns regarding the observed housing and environmental features, and whether they considered these features when constructing houses. Focus group discussions, were conducted to clarify emergent themes on people’s perceptions on relationships between housing or environmental factors and malaria transmission.
The entomological surveys showed statistically higher indoor densities of the malaria vectors (Anopheles arabiensis and Anopheles funestus) in houses with mud walls compared to plastered or brick walls, open eaves compared to closed eaves and unscreened windows compared to screened windows. Most respondents reported that their houses allowed mosquito entry, at least partially. Participants were aware that house structure and environmental characteristics influenced indoor mosquito densities and consequently malaria transmission. They were concerned about living in poorly-constructed houses with gaps on eaves, walls, windows and doors but were constrained by low income.
In rural south-eastern Tanzania, significant proportions of people still live in houses with open eaves, unscreened windows and gaps on doors. Though they are fully aware of associated mosquito biting and pathogen transmission risks, they are constrained by low-income levels and competing household priorities. The study proposes that community-based house improvement initiatives combined with targeted subsidies could lower the financial barriers, improve access to essential construction materials or designs, and significantly accelerate malaria transmission control in these communities.
- Community knowledge
- Malaria transmission
- Housing characteristics
- Environmental features
The WHO 2017 World Malaria Report showed significant successes in malaria control in various countries, but with an overall slight increase in malaria cases and deaths relative to the previous year . Over the past decades, LLINs and IRS have been widely used alongside improved diagnosis and treatment, primarily with artemisinin-based combination therapy. These tools have contributed most of the gains accrued against malaria in the past decade [1, 2], but evidence suggests that new interventions will be required to finally move to zero transmission [3, 4]. Unfortunately, novel strategies, such as improved housing  and environmental management for malaria control, have only rarely considered [6, 7]. Tanzania has experienced more than 50% reduction in malaria prevalence in children aged 2–10 years across the country since 2000 . It is estimated that less than 2% of people in mainland Tanzania now live in places considered as having intense year-round transmission, while more than 60% of Tanzanians live in areas considered as having low transmission . The 2015–2016 malaria indicator surveys reported an increase in average malaria prevalence to 14.8% prevalence in children under 5 years , up from 9.5% in 2012 , but the most recent survey has shown much reduced prevalence of 7.3% . Despite these gains, there are still numerous challenges, among them, the rise of insecticide resistance , sub-optimal net use each night , and the high costs. Greater vigilance and additional interventions are therefore required to consolidate past gains and accelerate efforts towards the current national malaria control targets and eventual elimination.
Community involvement in vector control programmes is essential for successful malaria transmission reduction [13–16]. Existing evidence suggests that acceptance and optimal use of interventions, such as indoor residual spraying (IRS) and long-lasting insecticide-treated nets (LLINs) depends on the level of education and awareness in target communities , as well as grass-root organization of such communities to support the health initiatives . Similar successes have been observed with other vector control approaches such as larval source management. For example, in Tanzania, a large-scale community-based larvicide application programme in the city of Dar es Salaam, significantly reduced malaria infection prevalence among city residents , despite significant logistical and coverage challenges. In that programme, community owned resource persons supported specific efforts such as mapping, larvicide application and vector surveillance [19, 20]. Elsewhere, in an interesting integration of farmers’ experiences, van den Berg and Knols  highlighted the importance and potential of using farmer field schools as a method for improving community participation in malaria control. They suggested that malaria control can benefit from complementarities between rural development, farming and pest management and argued for a form of education that uses experiential learning methods to build farmers experiences, thereby achieving improvements in both health and development .
Unfortunately, despite the substantial evidence on benefits of community engagement in malaria prevention and various innovations on how to achieve this goal, community participation is increasingly relegated, as malaria-endemic countries adopt vertical campaigns such as universal LLIN distributions and mass drug administration in some settings. Instead, the knowledge and experiences of recipient communities is rarely sought even where such additional knowledge would significantly improve outcomes. In rural Ulanga district, south-eastern Tanzania, recent studies have demonstrated that community members were either at par or more knowledgeable than experts in identifying locations where mosquitoes are most abundant  and locations where male mosquitoes regularly swarm . Both examples emphasize the potential benefits of relying on communities for improved vector control.
Another important question is whether individuals understand how malaria transmission risk varies across their communities and how it is affected by factors such as housing characteristics and the environment. Evidence suggests that transmission intensities are rarely uniform between districts and villages or even between households in the same village. Instead some households are always disproportionally more exposed to biting risk and malaria infections than community averages . In a recent study, it was found that there is strong correlations between household occupancy and indoor densities of common malaria vectors (Anopheles arabiensis and Anopheles funestus), as well as spatial correlations of the two variables within and between villages in rural Tanzania . That study also showed that high indoor vector densities cluster in locations where houses with highest occupancy are also clustered, suggesting the possibility of relying on regular census data to predict malaria transmission patterns. Similarly, Smith et al.  illustrated that even though biting risk from malaria vectors may be highest at the edge of the villages near aquatic habitats, infectious proportions of these mosquito populations (tending to be the older ones) are far more likely to be clustered in the middle of the villages, with peak populations of the infectious ones occurring when overall vector densities start declining .
Despite the above observations, and other evidence suggesting that house design, geographical location of the house within settlements, and characteristics of surrounding environments all influence malaria transmission [16–19], several questions remain unanswered, particularly with regard to community awareness of the same factors. For example, it is unclear whether community members fully understand the importance of such factors, whether they have any concerns about them, and what their counter-efforts are during site selection and house construction. Yet, as malaria transmission declines, control efforts aiming at elimination should target locations and possibly households with greatest risk [27–29], while also actively engaging members of those communities and households.
The aims of this study were therefore: (a) to identify common house characteristics that may be associated with increased vector biting risk, and (b) to assess community knowledge and experiences regarding housing and environmental factors that influence malaria transmission in four rural Tanzanian villages and how these experiences can be incorporated in ongoing and future intervention options.
The primary malaria vectors are An. funestus and An. arabiensis, with minor contributions from Anopheles rivulorum . Malaria transmission intensities, i.e. entomological inoculation rate (EIR) was last estimated at 4.2 and 11.7 infectious bites/person/year (ib/p/yr) by An. arabiensis and An. funestus, respectively, totalling 15.9 ib/p/year . The main vector control method is LLINs, usually distributed through mass campaigns done every 3–4 years, and keep-up campaigns done through reproductive health clinics. The main malaria vectors are resistant to pyrethroids used in the LLINs, but still susceptible to organophosphates .
Longitudinal surveys of indoor mosquito densities
Random sampling with replacement was done using household listings originally obtained from the Ifakara Health & Demographic Surveillance Systems unit (2433 houses) covering the study area. An initial set of 48 houses were selected and recruited upon consent of household heads, and were set as fixed sentinel houses where vector surveillance was conducted once every month for 12 months. Each month, another set of 48 houses was selected randomly from the same villages (with approximately similar spatial distribution as sentinel houses) for a one-off mosquito collection that month. The following month a new set of 48 houses were selected without repeating those that had previously been sampled. In the end, a total 624 households were studied over the 12-month period, with the 48 sentinels sampled 12 times each.
Since the three study villages were different in size and number of households, the actual selection and distribution of the sampling houses was done in such a way as to cover the full geographical extents of the entire study area. Then stratified random sampling was used. First, 1600 households were randomly identified and spatially assigned to 16 geographical clusters, each having 100 households. The sampling clusters were defined based on household latitudes so that clusters 1–16 were on a North–South direction. On the first month of study, the geographical clusters were visited and six households selected randomly per cluster. The six household heads were requested to volunteer in the study, and half of them requested for permission so their houses could be enrolled as fixed sentinel sampling stations for repeated monthly mosquito collections. Whenever a household head did not consent, the next household in the random listing was selected, so that there were always six households per cluster (three of them being fixed sentinels and another three being random households for one-off sampling). This way, there were 48 fixed sentinels throughout the study period and another 48 new houses newly selected each month across the clusters.
Thereafter, mosquito collections were conducted monthly in each of the study households, each time selecting a new set of three random households from each cluster. A weekly schedule was followed, working for 4 weeks/month. Each week, mosquitoes were sampled in four of the 16 geographical clusters, by visiting all six households per cluster each night, working for four nights per week. This way, all the 16 clusters were covered once every month. In each household, one occupied room was selected for assessing indoor mosquito densities. CDC light traps set near occupied bed nets were used , and were operated from 18:00 h to 06:00 h. Backpack aspirators (CDC model 1412-12 VDC 17 AmpHr Gel-Cell battery) were also used to sample resting mosquitoes each subsequent morning. The collected mosquitoes were killed in a closed container using petroleum fumes, then morphologically identified using taxonomic keys for Anopheles Africa, south of Sahara, developed by Gillies and Coetzee . All samples were then sorted by sex, taxa and physiological status. Sub-samples of Anopheles gambiae complex were further identified by multiplex polymerase chain reaction (PCR), to identify sibling species using protocols developed by Scott et al. . Also a sub samples of An. funestus were identified by using PCR, following the techniques developed by Cohuet et al. , and Koekemoer et al. .
Characterization of the houses and their environments
At each initial visit, all the 624 human houses were characterized from where mosquito collections had been done over the 1-year period. Geo-positions (latitudes and longitudes) of all the households were captured using hand-held GPS receivers (Magellan eXplorist 110) and recorded. In each house, the condition of windows, doors and walls were examined, and whether the houses had closed or open eaves. The surroundings of the households were examined and recorded any nearby water bodies, cattle sheds or other animal sheds. Other environmental variables recorded included whether the houses were located at the edge or in the middle of the villages and whether people kept animals, such as cattle, pigs or chicken.
Survey questionnaires: quantitative assessment of community knowledge, opinions and concerns regarding house designs and environmental characteristics, and how these factors influence malaria transmission
Description of the main themes addressed in the survey to assess community knowledge, opinions and concerns regarding house designs and environmental characteristics, and how these factors influenced malaria transmission
Specific questions asked by the interviewer
Relevance of the concepts
Knowledge and perception about house characteristics, mosquito entry and malaria transmission risks
Do you know if the house design can influence mosquito entry?
Assessment of knowledge and perception of malaria transmission risks in relation to house characteristics
Does your house allow mosquito entry?
If your house does not allow mosquito entry, how do you prevent mosquito entry?
Why does your house allow mosquito entry?
How do mosquitoes enter your house?
When was your house constructed?
Why did you decide to construct this kind of house?
What did you consider during construction?
Knowledge and perception about environmental variables influencing mosquito density
Do you know if the environments surrounding your house influence mosquito density?
Assessment of knowledge and perception of malaria transmission risks in relation to environmental characteristics
How does the environments surrounding your house influence mosquito density?
Mention the common mosquito breeding sites in your area
What do you do to prevent mosquito bites?
Knowledge and perception about settlements, mosquito density and malaria transmission risks
Do you think the number of houses in an area can influence mosquitoes and malaria transmission?
Assessment of knowledge and perception of malaria transmission risks in relation to settlement patterns
Do you think constructing houses near other houses or far from other houses is an important factor in regard to mosquito biting risk and malaria transmission?
Why do you think close house have many mosquitoes?
What can be done to control mosquitoes in such kind of environment?
The questionnaires were translated and administered in Kiswahili, which is the common language spoken by communities across the study area. After pre-testing and correcting all issues arising, the questionnaire was administered by trained research officers to consenting household heads across three study villages where the entomological surveys had been done and in one village where there was no entomological surveillance. A total of 200 participants responded to the survey.
Focus group discussions: qualitative assessment of community knowledge, opinions and concerns regarding house designs and environmental characteristics, and how these factors influence malaria transmission
Study participants were selected from among the households participating in the entomological survey and recruited upon consent with help from the community leaders. A total of 12 people were recruited for each focus group, ensuring all participants were from the same village.
Eight focus group discussions (FGD) were conducted across the four study villages, each FGD comprised 12 participants recruited from the same village. Overall the study had a total of 96 participants aged between 18 and 70 years. There was an equal representation of men and women in each of the FGD groups (6 men and 6 women). The discussions were held between January and March 2017 and were conducted at local primary schools within the respective villages.
A study guide for the FGD was developed and pretested. The FGD guide followed the themes described in Table 1 for the questionnaire surveys. Adults and youths were separated during the interviews to avoid any chance of intimidation during the discussions. The purpose of the discussions was explained and verbal and written informed consents were obtained from the participants. The discussions were semi-structured: the facilitator asked a question and the participants discussed their knowledge, views and comments. The discussions were facilitated by two social scientists; a note taker and an interviewer, and were tape recorded to ensure that all information was captured. Discussions were all conducted in the local language (Swahili), and lasted between 40 and 60 min each.
The study assessed whether people think that the house designs matter in terms of the risk they are exposed to. The study then assessed if people care about environmental variables and if they worry about these factors when they construct their houses. In cases where people were concerned, the study also assessed how they cope with these concerns. Another concept investigated during the FGDs was whether participants were aware of any relationship between settlement patterns and malaria transmission. For example, the study examined whether the participants thought constructing houses near other houses or far from other houses was an important factor in regard to mosquito biting risk and malaria transmission.
Analysis of quantitative data
Data were analysed using open source software, R version 3.1.0, using the lme4 package . Associations between house characteristics and indoor mosquito densities were examined by generalized linear mixed effects models (glmer) , by fitting the data in log-linked Poisson error distributions. The indoor densities of An. arabiensis, An. funestus, Culex and Mansonia species were modelled as a function of different house and environmental characteristics. This study assessed effects of (a) wall type, (b) roof type, (c) window type, (d) whether chickens were kept inside or outside the house, and (e) whether the eave spaces were closed or open. Date of mosquito collection and house identification numbers were incorporated as random variables in the glmer models. Estimated mean indoor mosquito densities per house per night, the relative rates (RR) of observing these mosquito catches, and associated 95% confidence intervals (CI), were computed from exponentials of coefficients generated from the glmer models.
The monthly temporal patterns of biting risk by the two malaria vector species and also the culicine genera, i.e. Culex spp. and Mansonia spp. mosquitoes were presented graphically.
Data generated through the questionnaire surveys were summarized and presented descriptively in tables to capture: (a) knowledge about risk factors, (b) whether people worry about house and environmental characteristics and (c) whether they consider such variables when constructing their houses.
Processing and analysis of qualitative data
After the FGD session, all recordings were retrieved from the tape recorder and archived on a password-protected computer. A trained behavioural scientist then transcribed and translated the data from Kiswahili to English for further analysis. The scripts were then imported into NVivo software version 13 for organizing and indexing the data. Emergent themes were organized to assess: (a) whether people worry about the characteristics of the house and if they consider such variables when constructing their houses, (b) whether people care about environmental variables and if they consider such variables when constructing their houses, (c) whether people know effects of seasonality on malaria transmission, (d) whether people know the relationship of settlement patterns and human population on malaria transmission, and (e) how the people cope with these risk factors if they are actually aware of relationships with malaria transmission. Direct quotes from the participants are presented to support their responses on the themes.
Indoor densities of malaria and non-malaria mosquito vectors
Physical characteristics and microclimate of sampled houses
Physical characteristics and microclimatic conditions in sampled houses
Plastered brick walls
Un-plastered brick walls
Corrugated iron sheet
With netting screen
Without netting screen
Door (observed from 6 pm to 7 pm)
Household risk factors associated with indoor mosquito densities
Summary statistics for mosquitoes caught in houses with different characteristics. The mean nightly catches of An. arabiensis and An. funestus mosquitoes, relative rates and the associated significance levels were calculated from the generalized linear mixed models (GLMMs) at 95% confidence interval
RR (95% CI)
RR (95% CI)
2.0 (1.8 –2.1)
Summary statistics for the number of mosquitoes caught in houses with different characteristics
RR (95% CI)
RR (95% CI)
51.7 (45.2 – 86.9)
0.6 (0.3 – 1.5)
66.2 (43.9 – 88.5)
2.3 (2.1– 2.6)
1.5 (0.4 – 2.5)
2.5 (1.7 – 3.8)
57.9 (27.7 – 88.1)
0.3 (0.1 – 0.6)
67.2 (48.7 – 85.8)
1.9 (1.8 – 2.0)
1.5 (0.7 – 2.3)
1.5 (1.4 –1.8)
56.7 (33.9 – 79.4)
0.7 (0.1 – 1.4)
70.2 (48.7 – 91.7)
2.6 (2.4 – 2.7)
1.5 (0.6 – 2.3)
2.5 (1.8 – 2.4)
29.5 (0.0 – 79.9)
0.2 (0.0 – 0.9)
56.2 (43.9 – 88.5)
2.3 (2.1– 2.6)
1.5 (0.4 – 2.5)
2.5 (1.7 – 3.8)
55.4 (23.2 – 86.6)
0.9 (0.0 – 2.0)
68.4 (50.3 – 86.5)
2.5 (2.4 – 2.6)
1.3 (0.6 – 2.0)
1.8 (1.7 – 1.9)
63.8 (46.2 – 81.5)
1.1 (0.5 –1.8)
67.4 (34.6 – 100.0)
1.2 (1.1– 1.2)
1.3 (0.0 – 2.7)
2.6 (2.4 – 2.7)
Socio-demographic characteristics of survey participants
Socio-demographic characteristics of participants who responded to the survey questionnaire to assess people’s awareness of how house characteristics affect malaria transmission
Level of education
No formal education
Peasant (self-employed in agriculture)
Small scale business
Results of the questionnaire survey: participants’ knowledge regarding indoor mosquito density and how this relates to housing characteristics and environmental variables
Summary of community members’ knowledge about indoor mosquito density and how these relate to certain housing and environmental characteristics
Whether participants believe their house let in mosquitoes
How participants believed mosquitoes entered their houses
Through open doors
Through holes in the wall
Through the eave space
Through both open windows and doors
Through open window doors, holes in walls and eaves
Age of house (date of construction)
< 4 years ago
5–8 years ago
> 9 years ago
Main reasons for constructing the kind of house
Because it is permanent
Because it prevents animals from entering the house
Because it prevents insects
Whether people consider mosquito prevention as a key factor when constructing houses
Does not know
Specific practices considered by participants (during construction of their houses) for preventing mosquitoes
Netting on the window
Blocking the eaves
Using bricks on the wall
Using cement on the wall
Does not know
Whether participants knew open eave spaces let in mosquitoes
Whether participants knew that surrounding environments influence vector densities in their houses
Focus group discussions: participants’ knowledge regarding indoor mosquito density and how this relates to housing characteristics and environmental variables
“Many mosquitoes are found in those houses with smaller windows, in which there is less light because mosquitoes like dark places. Then mosquitoes enter the house and won’t leave the area”. (46 years old male, Milola village).
“It is important to make sure that houses are built as far away from the mosquito aquatic habitats as possible. It is good to know the distance that mosquitoes can fly from their aquatic habitats so that we can know where to build houses”. (37 years old male, Minepa village).
“It is necessary to build a house that makes it hard for mosquitoes to get inside. I know that houses that are built lower let mosquitoes in than the houses that have larger height”. (26 years old male Milola village).
“If the environment is dirty mosquitoes increase in number, because they are able to rest in trash. That means if we clean our surroundings then there will be very few mosquitoes. But if our environment is dirty mosquitoes reproduce and increase in numbers” (58 years old male, Minepa village).
“Mosquitoes are found in dirty water around the house, like in water that is in broken containers or cups. They are also found in the ponds, shrubs and in dark places. I believe that is where they accumulate” (34 years old female, Mavimba village).
“When there are many houses together, it means that there is a lot of trash and other things in which mosquitoes can rest. For example, I may keep my surroundings clean, but if my neighbours do not clean their surroundings, and they live close to me, then mosquitoes can easily come from their houses to mine, it will make it easy for mosquitoes to bite me”. (37 years old female, Milola village).
“If there are many houses close together then mosquitoes can move from one house to another. But if houses are few and far apart then the mosquitoes cannot really get from one house to another easily, because there can be many obstacles for them on the way, like the wind and rain” (40 years old female, Milola village).
“When there are a lot of houses together then it is a problem for us because this way many mosquitoes come, because they know that they can get a lot of people in close proximity. It is easy for the mosquitoes to transmit malaria” (23 years old male, Kivukoni village).
“If houses are scattered, then the wind will blow mosquitoes to other places where they cannot easily get people” (51 years old male, Milola village).
“One thing that most of the people in the village can afford is to use bed nets, because everyone has bed nets here. But in order for these nets to be useful they have to be in a good shape. Sadly not everyone has new nets. I sleep with my children in the same bed because we have one net, but our net is not in a very good state, it has a lot of holes that sometime let mosquitoes in. But it still helps; it is not the same as if we did not use anything” (31 years old female, Kivukoni village).
“We make sure that children are protected from mosquito bites by making sure they go to bed early when it gets dark, and then we make sure they are covered by bed nets. Sometimes when they are sitting outside then we use a piece of cloth to chase mosquitoes away” (54 years old male, Milola village).
“Mosquitoes are found here throughout the year. However, their population is the highest during the rainy season because there is a lot of water around, hence many aquatic habitat. During the dry season there are fewer places where mosquitoes accumulate, so their population is lower” (28 years old male, Milola village).
“There are a lot of mosquitoes from February to April because this is the time where there is a lot of water everywhere. There are many pools, ponds and bushes where mosquitoes can accumulate. The mosquito population starts to go down from May and June when the weather starts to get dry and cold” (24 years old male, Minepa village).
Housing design has been shown in different studies to significantly influence malaria transmission [30–32]. Most people in rural and peri-urban areas live in poorly finished houses with open eaves, doors and windows, hence exposure to malaria transmission is very high. Mosquitoes enter the houses easily through eaves and other openings, thus increasing likelihood of biting and pathogen transmission [29, 30]. This study found that the number of mosquitoes were significantly higher in houses with open eaves, grass roofs, Mud walls and unscreened windows. Furthermore, keeping chickens inside the house was also associated with high number of mosquitoes. Unfortunately, these communities cannot afford essential house construction and mosquito-proofing materials. Greater sensitization and enabling financing initiatives to promote better housing should therefore be considered in vector control initiatives to accelerate malaria elimination efforts. This study found statistically higher indoor densities of the main malaria vectors (An. arabiensis and An. funestus) in houses with mud-walls as opposed to plastered or brick walls, open as opposed to closed eaves, and also in unscreened as opposed to screened windows. Keeping chickens inside the house was similarly associated with higher indoor densities.
There was already very high levels of awareness about how housing characteristics and the surrounding environment influence mosquito densities, and consequently malaria transmission. Another factor identified was the small size of houses (area and height). In additional to poorly-constructed houses with gaps in the eaves, walls, windows and doors. However, people were constrained by low income and could not readily afford construction of properly ventilated and mosquito-proof houses. Nearly all respondents reported that their houses allowed mosquito entry, but they could not afford modern building materials. They expressed that they are also being bitten by mosquitoes while indoors when they are not under bed nets. The relationship between poor housing and low-income is expected in such settings. However, although many households in rural areas are unable to afford modern housing, there are cheaper and simple ways of modifying or screening houses which could readily prevent mosquito entry, especially if the community members are adequately sensitized and financial barriers removed, e.g. through subsidies. Unfortunately, malaria control authorities in sub-Saharan Africa have not invested in house improvement programmes despite the significant available evidence [39, 40].
Community knowledge on mosquito aquatic habitat is another important factor to consider for successful community-based vector control. In this study, participants of the FGD were identified various sources of mosquitoes in their surroundings, such as stagnant water pools, dense vegetation and rice fields. Studies elsewhere have yielded similar findings [41, 42], and such high levels of community awareness can be harnessed to improve disease prevention and control initiatives. In rural south-eastern Tanzania, the nucleated settlement patterns, combined with the previously described spatial distribution of vector densities across the villages, indicate that targeted interventions if supported by local community champions could indeed be highly effective. A study done by Mwangungulu et al. in the same villages yielded evidence that community members are capable of identifying places with low, medium and high mosquito densities . Mosquitoes accumulate in shallow pools of water and puddles [43, 44], which are numerous in rural areas especially during the rainy season. Draining pools of water, levelling land, construction of drains and proper waste management could eliminate mosquito aquatic habitat around homes [45, 46], and subsequently reduce malaria transmission . Given the existing high-levels of awareness, further control initiatives involving environmental management could be improved if community members are considered the main stake holders. Such community-based approaches could also be used to scale-up better housing designs and access to commodities necessary for screening windows, eaves and doors even in rural, low-income areas.
Community knowledge, perception and practices are very important for designing or improving disease control programmes and identifying factors for effectiveness of the interventions . Promoting community based programmes can help to achieve population level change in risk behaviours which in turn results into positive changes . Unless communities recognize the importance of changing behaviours which influence disease transmission, the best designed interventions might not achieve the intended goals. More broadly in primary health care, it has been demonstrated in places such as Tigray, Ethiopia, that teaching mothers about home-based malaria treatment and management can vastly improve treatment outcomes and reduces child mortality . In this particular programme in Ethiopia, reduction of up to 40% in under-five mortality was achieved in holoendemic malaria areas, by simply training and relying upon local mothers to teach each other how to dispense anti-malarial medicines to their children.
Another important point that emerged from this study was participants’ impression that if the houses are close to each other, malaria transmission is likely to be higher because mosquitoes can easily move between houses, and spread pathogens. Though participants were not able to describe settlement patterns, they perceived that closer and congested household are at higher risk of malaria transmission, and were concerned with the distances that a mosquito can move from one house to another. Previous study suggested strong correlations between household occupancy and malaria vector densities , but no specific analyses have been done on effects of distance between households on malaria transmission. By combining the new findings from this current study with the observations previously described by Mwangungulu et al.  and Kaindoa et al. , community driven intervention could be designed, which relies on trained locals to support environmental management and house improvement initiatives for mosquito control . For example, community-based artisans, if provided minimal government support, could create local businesses that support sustainable house-improvement programs with potential benefits against vector-borne diseases.
To improve malaria control and elimination efforts, it is important to develop an understanding of community perspectives on persistent malaria transmission within and around their villages, then engage these communities in identifying and addressing the key factors contributing to the persistent transmission. This study has shown that in rural south-eastern Tanzania, significant proportions of people still live in houses with open eaves, unscreened windows and gaps on doors. Though the people are fully aware of associated mosquito biting and pathogen transmission risks, they are constrained by low-income levels and cannot readily afford better housing or house improvement (e.g. through screening of eaves and windows), alongside other competing priorities. This study concludes that community-based house improvement initiatives combined with targeted subsidies could lower the financial barriers, improve access to essential construction materials or designs, and significantly accelerate malaria transmission control in these communities.
EK and FO conceived the study and developed study protocols and supervised data collection. EK, MF, JK, conducted data analysis, EK, MF, JK, AN, MC and FO wrote the manuscript. All authors read and approved the final manuscript.
We thank the local communities and participants from Ulanga district for attending and contributing to the focus group discussions. We thank Ms. Irene Ngonja, Tumpe Mwandyala and Mr. Hamis Bwanary for their assistance during the FGDs. We also Ms. April Monroe for assisting us in data analysis plan and reviewing the first draft of this manuscript. We also thank the village leaders for providing us with a venue for conducting FGDs. We thank Mr. Deogratius Philemon for translating the scripts.
The authors declare that they have no competing interests.
Availability of data and materials
The dataset generated by this study is available from the corresponding author upon reasonable request.
Ethics approval and consent to participate
Meetings with local leaders were held in the study areas and the main aims of the study were explained by the research team. A written and signed informed consent was obtained from each study participant. All information was given in Kiswahili, the local language. Ethical approval for the study was obtained from Ifakara Health Institute Institutional Review Board (IHI/IRB/No.: 029-2016), and from the Medical Research Coordinating Committee (MRCC) at the National Institutes of Medical Research (NIMR) (Ref: NIMR/HQ/R.8a/Vol.IX/2428). Approval was also obtained from the Human Research Ethics Committee (Medical) at University of the Witwatersrand (Approval Certificate No.: M160806). Study household were provided with new LLINs as previously distributed by the universal bednets coverage campaign. All personal information that could identify the study participants was omitted from the final report so as to protect identification of respondents. The participants were assured anonymity in the presentation and publishing of the data. Permission to publish this manuscript was obtained from the National Institutes of Medical Research (NIMR) (Ref: NIMR/HQ/P.12/VOLXXV/).
This work was supported by Wellcome Trust Intermediate Fellowship in Public Health and Tropical Medicine awarded to FO (Grant Number: WT102350/Z/13). EK, LF and JK were supported through a Consortium for Advanced Research Training in Africa (CARTA), which is jointly led by the African Population and Health Research Centre (APHRC) and the University of the Witwatersrand and funded by the Wellcome Trust (UK) (Grant No.: 087547/Z/08/Z), the Carnegie Corporation of New York (Grant No.: B 8606.R02), Sida (Grant No.: 54100029). MC is supported by a DST/NRF South African Research Chairs Initiative Grant (Grant No.: 64763).
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 2017. Geneva: World Health Organization; 2017. http://www.who.int/malaria/publications/world-malaria-report-2017/report/en/. Accessed July 2018.
- Bhatt S, Weiss DJ, Cameron E, Bisanzio D, Mappin B, Dalrymple U, et al. The effect of malaria control on Plasmodium falciparum in Africa between 2000 and 2015. Nature. 2015;526:207–11.View ArticlePubMedPubMed CentralGoogle Scholar
- malERA Refresh Consultative Panel on Tools for Malaria. Elimination. malERA: an updated research agenda for diagnostics, drugs, vaccines, and vector control in malaria elimination and eradication. PLoS Med. 2017;14:e1002455.View ArticleGoogle Scholar
- malERA Refresh Consultative Panel on Characterising the Reservoir and Measuring Transmission. malERA: an updated research agenda for characterising the reservoir and measuring transmission in malaria elimination and eradication. PLoS Med. 2017;14:e1002452.View ArticleGoogle Scholar
- Lindsay SW, Jawara M, Paine K, Pinder M, Walraven GEL, Emerson PM. Changes in house design reduce exposure to malaria mosquitoes. Trop Med Int Health. 2003;8:512–7.View ArticlePubMedGoogle Scholar
- Fillinger U, Lindsay SW. Larval source management for malaria control in Africa: myths and reality. Malar J. 2011;10:353.View ArticlePubMedPubMed CentralGoogle Scholar
- Utzinger J, Tozan Y, Singer BH. Efficacy and cost-effectiveness of environmental management for malaria control. Trop Med Int Health. 2001;6:677–87.View ArticlePubMedGoogle Scholar
- Mandike R. Tanzanian National Malaria Control Programme (NMCP) (2014) Strategy 2014–2020.Google Scholar
- United Republic of Tanzania Tanzania. Demographic and Health Survey and Malaria Indicator Survey (DHS-MIS) 2015-2016. NBS, MoHCDGEC, ICF. International. 2016. https://doi.org/10.1017/CBO9781107415324.004.Google Scholar
- Tanzania Commission for AIDS (TACAIDS). Tanzania-2011–12 HIV/AIDS and Malaria Indicator Survey 2011–12: Key Findings. Tanzania Comm AIDS (ZAC), Zanzibar AIDS Comm (NBS), Natl Bur Stat (OCGS), Off Chief Gov Stat ICF Int. 2013; 16.Google Scholar
- National Bureau of Statistics. Tanzania Malaria Indicator Survey; Key indicators. 2017. https://dhsprogram.com/pubs/pdf/PR103/PR103.pdf. Accessed July 2018.
- Kisinza WN, Nkya TE, Kabula B, Overgaard HJ, Massue DJ, Mageni Z, et al. Multiple insecticide resistance in Anopheles gambiae from Tanzania: a major concern for malaria vector control. Malar J. 2017;16:439.View ArticlePubMedPubMed CentralGoogle Scholar
- Musoke D, Karani G, Ssempebwa JC, Etajak S, Guwatudde D, Musoke MB. Knowledge and practices on malaria prevention in two rural communities in Wakiso district, Uganda. Afr Health Sci. 2015;15:401–12.View ArticlePubMedPubMed CentralGoogle Scholar
- Tynan A, Atkinson J-A, Toaliu H, Taleo G, Fitzgerald L, Whittaker M, et al. Community participation for malaria elimination in Tafea Province, Vanuatu: part II. Social and cultural aspects of treatment-seeking behaviour. Malar J. 2011;10:204.View ArticlePubMedPubMed CentralGoogle Scholar
- Ruebush TK, Zeissig R, Koplan JP, Klein RE, Godoy HA. Community participation in malaria surveillance and treatment III. An evaluation of modifications in the volunteer collaborator network of Guatemala. Am J Trop Med Hyg. 1994;50:85–98.View ArticlePubMedGoogle Scholar
- Ghebreyesus TA, Alemayehu T, Bosman A, Witten KH, Teklehaimanot A. Community participation in malaria control in Tigray region Ethiopia. Acta Trop. 1996;61:145–56.View ArticlePubMedGoogle Scholar
- Atkinson J, Vallely A, Fitzgerald L, Whittaker M, Tanner M. The architecture and effect of participation: a systematic review of community participation for communicable disease control and elimination. Implications for malaria elimination. Malar J. 2011;10:225.View ArticlePubMedPubMed CentralGoogle Scholar
- Geissbühler Y, Kannady K, Chaki PP, Emidi B, Govella NJ, Mayagaya V, et al. Microbial larvicide application by a large-scale, community-based program reduces malaria infection prevalence in urban Dar Es Salaam, Tanzania. PLoS One. 2009;4:e0005107.View ArticleGoogle Scholar
- Vanek MJ, Shoo B, Mtasiwa D, Kiama M, Lindsay SW, Fillinger U, et al. Community-based surveillance of malaria vector larval habitats: a baseline study in urban Dar es Salaam, Tanzania. BMC Public Health. 2006;6:154.View ArticlePubMedPubMed CentralGoogle Scholar
- Maheu-Giroux M, Castro MC. Impact of community-based larviciding on the prevalence of malaria infection in Dar es Salaam, Tanzania. PLoS One. 2013;8:e0071638.View ArticleGoogle Scholar
- van Den Berg H, Knols BGJ. The farmer field school: a method for enhancing the role of rural communities in malaria control? Malar J. 2006;5:3.View ArticlePubMedPubMed CentralGoogle Scholar
- Mwangungulu SP, Sumaye RD, Limwagu AJ, Siria DJ, Kaindoa EW, Okumu FO. Crowdsourcing vector surveillance: using community knowledge and experiences to predict densities and distribution of outdoor-biting mosquitoes in rural Tanzania. PLoS ONE. 2016;11:e0156388.View ArticlePubMedPubMed CentralGoogle Scholar
- Kaindoa EW, Ngowo HS, Limwagu A, Mkandawile G, Kihonda J, Masalu JP, et al. New evidence of mating swarms of the malaria vector, Anopheles arabiensis in Tanzania. Wellcome Open Res. 2017;2:88.View ArticlePubMedPubMed CentralGoogle Scholar
- Woolhouse ME, Dye C, Etard JF, Smith T, Charlwood JD, Garnett GP, et al. Heterogeneities in the transmission of infectious agents: implications for the design of control programs. Proc Natl Acad Sci USA. 1997;94:338–42.View ArticlePubMedGoogle Scholar
- Kaindoa EW, Mkandawile G, Ligamba G, Kelly-Hope LA, Okumu FO. Correlations between household occupancy and malaria vector biting risk in rural Tanzanian villages: implications for high-resolution spatial targeting of control interventions. Malar J. 2016;15:199.View ArticlePubMedPubMed CentralGoogle Scholar
- Smith DL, Dushoff J, McKenzie FE. The risk of a mosquito-borne infection in a heterogeneous environment. PLoS Biol. 2004;2:e368.View ArticlePubMedPubMed CentralGoogle Scholar
- Konradsen F, Amerasinghe P, Van Der Hoek W, Amerasinghe F, Perera D, Piyaratne M. Strong association between house characteristics and malaria vectors in Sri Lanka. Am J Trop Med Hyg. 2003;68:177–81.View ArticlePubMedGoogle Scholar
- Kirby MJ, Green C, Milligan PM. Risk factors for house entry by malaria vectors in a rural town and satellite villages in The Gambia. Malar J. 2008;7:2.View ArticlePubMedPubMed CentralGoogle Scholar
- Russell TL, Lwetoijera DW, Knols BG, Takken W, Killeen GF, Kelly-Hope LA. Geographic coincidence of increased malaria transmission hazard and vulnerability occurring at the periphery of two Tanzanian villages. Malar J. 2013;12:24.View ArticlePubMedPubMed CentralGoogle Scholar
- Geubbels E, Amri S, Levira F, Schellenberg J, Masanja H, Nathan R. Health & Demographic Surveillance System Profile: the Ifakara Rural and Urban Health and Demographic Surveillance System (Ifakara HDSS). Int J Epidemiol. 2015;44:848–61.View ArticlePubMedGoogle Scholar
- Kaindoa EW, Matowo NS, Ngowo HS, Mkandawile G, Mmbando A, Finda M, et al. Interventions that effectively target Anopheles funestus mosquitoes could significantly improve control of persistent malaria transmission in south-eastern Tanzania. PLoS ONE. 2017;12:e0177807.View ArticlePubMedPubMed CentralGoogle Scholar
- Mboera LEG, Kihonda J, Braks MAH, Knols BGJ. Influence of centers for disease control light trap position, relative to a human-baited bed net, on catches of Anopheles gambiae and Culex quinquefasciatus in Tanzania. Am J Trop Med Hyg. 1998;59:595–6.View ArticlePubMedGoogle Scholar
- Gillies MT, Coetzee M. A Supplement to the Anophelinae of the South of the Sahara (Afrotropical Region). Publications of the South African Institute for Medical Research; 1987. p. 1–143.Google Scholar
- Scott JA, Brogdon WG, Collins FH. Identification of single specimens of the Anopheles gambiae complex by the polymerase chain reaction. Am J Trop Med Hyg. 1993;49:520–9.View ArticlePubMedGoogle Scholar
- Cohuet A, Simard F, Toto JC, Kengne P, Coetzee M, Fontenille D. Species identification within the Anopheles funestus group of malaria vectors in Cameroun and evidence for a new species. Am J Trop Med Hyg. 2003;69:200–5.View ArticlePubMedGoogle Scholar
- Koekemoer LL, Kamau L, Hunt RH, Coetzee M. A cocktail polymerase chain reaction assay to identify members of the Anopheles funestus (Diptera: Culicidae) group. Am J Trop Med Hyg. 2002;66:804–11.View ArticlePubMedGoogle Scholar
- Development R, Core Team R. R: a Language and environment for statistical computing. R foundation for statistical. Computing. 2011. https://doi.org/10.1007/978-3-540-74686-7.Google Scholar
- Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. J Stat Softw. 2015;67:51.View ArticleGoogle Scholar
- Tusting LS, Ippolito MM, Willey BA, Kleinschmidt I, Dorsey G, Gosling RD, et al. The evidence for improving housing to reduce malaria: a systematic review and meta-analysis. Malar J. 2015;14:209.View ArticlePubMedPubMed CentralGoogle Scholar
- Kirby MJ, Ameh D, Bottomley C, Green C, Jawara M, Milligan PJ, et al. Effect of two different house screening interventions on exposure to malaria vectors and on anaemia in children in the Gambia: a randomised controlled trial. Lancet. 2009;374:998–1009.View ArticlePubMedPubMed CentralGoogle Scholar
- Liu X, Wan F, Cirendunzhu Cirenwangla, Bai L, et al. Community knowledge and experience of mosquitoes and personal prevention and control practices in Lhasa, Tibet. Int J Environ Res Public Health. 2014;11:9919–37.View ArticlePubMedPubMed CentralGoogle Scholar
- Opiyo P, Mukabana WR, Kiche I, Mathenge E, Killeen GF, Fillinger U. An exploratory study of community factors relevant for participatory malaria control on Rusinga Island, western Kenya. Malar J. 2007;6:48.View ArticlePubMedPubMed CentralGoogle Scholar
- Norris DE. Mosquito-borne diseases as a consequence of land use change. EcoHealth. 2004;1:19–24.View ArticleGoogle Scholar
- Sattler M, Mtasiwa D, Kiama M, Premji Z, Tanner M, Killeen GF, et al. Habitat characterization and spatial distribution of Anopheles sp. mosquito larvae in Dar es Salaam (Tanzania) during an extended dry period. Malar J. 2005;4:4.View ArticlePubMedPubMed CentralGoogle Scholar
- Keiser J, Singer BH, Utzinger J. Reducing the burden of malaria in different eco-epidemiological settings with environmental management: a systematic review. Lancet Infect Dis. 2005;5:695–708.View ArticlePubMedGoogle Scholar
- WHO. Manual on environmental management for mosquito control. Geneva: World Health Organization Offset Publication; 1982. p. 284.Google Scholar
- Yohannes M, Haile M, Ghebreyesus TA, Witten KH, Getachew A, Byass P, et al. Can source reduction of mosquito larval habitat reduce malaria transmission in Tigray, Ethiopia? Trop Med Int Health. 2005;10:1274–85.View ArticlePubMedGoogle Scholar
- Merzel C, D’Afflitti J. Reconsidering community-based health promotion: promise, performance, and potential. Am J Public Health. 2003;93:557–74.View ArticlePubMedPubMed CentralGoogle Scholar
- Koenker H, Keating J, Alilio M, Acosta A, Lynch M, Nafo-Traore F. Strategic roles for behaviour change communication in a changing malaria landscape. Malar J. 2014;13:1.View ArticlePubMedPubMed CentralGoogle Scholar
- Kidane G, Morrow RH. Teaching mothers to provide home treatment of malaria in Tigray, Ethiopia: a randomised trial. Lancet. 2000;356:550–5.View ArticlePubMedGoogle Scholar
- WHO. Global vector control response 2017–2030. Geneva: World Health Organization; 2017.Google Scholar