Insecticide-treated net (ITN) ownership, usage, and malaria transmission in the highlands of western Kenya

Background

Insecticide-treated mosquito nets (ITNs) used for protection against mosquito bites have proven to be a practical, highly effective, and cost-effective intervention against malaria [1]. The evidence of the public health impact of ITNs, supporting their wide-scale use in Africa, is drawn from areas of stable malaria transmission where Plasmodium falciparum infection prevalence in the community is often over 40% [1,2]. Community-based randomized controlled trials (RCT) in these regions have documented average reductions of 20% in all causes of mortality in children under 5 years old within 2 years of increasing ITN use from 0 to 50-70% [3-8]. Scaling up ITN coverage and use by young children and pregnant women has been made a consensus target of the Millennium Development Goals (MDGs), the Roll Back Malaria Partnership (RBM), and the US President's Malaria Initiative (PMI) [9-11]. Targeting individual protection to these vulnerable groups [12-14] is a well-founded and explicitly accepted priority of all three initiatives because these groups bear the highest risk of morbidity and mortality from malaria. However, this strategy ignores the potentially greater community-wide benefits of broader population coverage [15] in areas of moderate and seasonal transmissions, like the highlands, where all age groups are equally vulnerable and no explicit resources, targets, or strategies have been proposed to achieve protection.

Unlike the lowlands where perennial malaria transmission is attributed to high vector densities throughout the year [16], transmission intensity and vector densities in the highlands are relatively low and seasonal (17). Because of this, human populations across age groups have a poorly developed immunity to malaria because exposures are infrequent [17], and they are therefore vulnerable to severe clinical illness and complications from Plasmodium infection [18,19]. In addition, research indicates that the mechanisms leading to malaria transmission in the highlands are complex and are probably due to the combined effects of factors such as topography [20], hydrology [21], climate variability [22], land-use/land-cover change [23], and drug resistance [22]. Given these characteristics, ITNs are considered one of the most effective intervention strategies for the control of malaria across all age groups in the highland regions [15].

Although ITN distribution has been massively expanded in most parts of malaria endemic sub-Saharan countries since 2005, there is limited information on community based actual use of nets owned, area specific reasons for non-use, and the possible impact of the variations in use on malaria vector densities and transmission in either Kenyan highlands or other countries where malaria is seasonal and unstable. Most studies of ITN use are intervention trials that attempt to explain why vulnerable groups, such as children under five or pregnant women, are or are not using an ITN [24,25], or describe which household members use the owned ITN(s) [26,27]. Earlier studies on ITN coverage, use, and impact on malaria transmission have different approaches, with most having common conclusions, such as higher ITN use in the high transmission season when mosquito densities are high, and indoor temperatures are low, with the main focus on children under the age of 5 years and pregnant women [8,26,28-30]. As far as could be ascertained, there is inadequate vectorial and parasitological information, or data on the actual use of owned ITNs and their health impacts in areas of Africa that support low or unstable transmission since the inception of the scaling up of ITN coverage. The objective of this study was to examine the utilization of ITNs delivered by the Kenya Ministry of Health (MOH) facilities under routine operational conditions in areas of moderate to unstable/seasonal malaria transmission in two highlands sites in western Kenya.

Materials and methods

Results

Discussion

Most parts of Kenya have approached or met the RBM household ITN coverage target of at least 60% as reported by previous studies [36]. Similarly, after almost six years of MOH efforts towards distribution of nets on a national scale, this study has shown that despite current high net ownership (>60%) within the highlands, actual usage remains low, with only approximately 53% of the residents who own nets reporting net use. This scenario may be a set-back on the intended and expected role and impact of distributed ITNs. As observed, most of the unused nets were either new and kept safely from damage or hung but not used. Low education level of the head of the household is one of the reasons why such a high percentage of the population own but do not use nets. Compared with heads of household with no education at all, the majority of heads of household with at least a primary level education know that mosquitoes transmit malaria and they eventually tend to acquire and use nets. Another reason mentioned by majority respondents was seasonality, with the onset of the colder/rainy season improving use of nets among those who own them, either because of the cold weather or because of noted presence of biting mosquitoes. This observation is consistent with previous studies conducted earlier in Ethiopia, where 35% of nets owned were not being used [37]. In contrast, studies in Sierra Leone and Madagascar recorded very low percentages of owned but unused nets (5.1% and 5.6% respectively) [38]. There is therefore need for targeted sensitization of the less educated group on importance of ITN ownership and usage so as to scale-up utilization. Likewise, there is need to sensitize the community on the importance of ITN adherence throughout the seasons with emphasis that malaria transmission occurs year round thus need for protection regularly. One of the probable ways to approach this may be through routine household visit by trained community health workers.

Even though ITN ownership was similarly high across seasons, seasonality in use was remarkable during this survey in both regions of study. Net use was considerably higher in the rainy, cooler season of April-May than in the dry, hot season of January-February. During the rainy season, the non-compliance rate was lower compared with the dry season. Similar findings have also been reported in surveys in ITN intervention trials in the lowlands of western Kenya and northern Ghana [26,39]. As in this highland region, the rainy months are characterized by the highest incidence of mosquito biting and malaria transmission. Unlike most parts of Africa where malaria is perennial and not limited to the rainy season [40], this area experiences seasonality in malaria transmission, but with "stable" transmission in areas along the valley bottoms [22]. Because of stable vector breeding habitats in the valley bottoms in the highlands [41], malaria transmission occurs year round, but with low intensity, and thus the need for continuous use of ITNs. Lower use of ITNs during the dry season with the assumption of no vectors and therefore no transmission can be detrimental as experienced in Ouagadougou in Burkina Faso, where most malaria occurs in the hotter months just after the rains [42]. Remarkable seasonality in net use reported in this area highlights the need for education to promote year-round use. For accurate monitoring and evaluation of ITN impact on malaria transmission, it would be preferable to conduct daily surveys on household ITN use year round to capture both low and high malaria transmission seasons. Unfortunately, this study could not accomplish this because of logistic difficulties. However, this study was able to undertake a two season survey in two similar geographic highland regions. This made it more appropriate to compare between surveys conducted in different seasons with less bias.

The proportion of households possessing mosquito net(s) and the proportion of children less than 5 years of age who slept under a net the preceding night are two of the key RBM bed net ownership indicators used to investigate the strengths and weaknesses of monitoring malaria control [10]. In this study, the comparative difference in use by children and adults did not depend on any characteristic of the surveys, such as region or season. Usage of young children below age of 5 years with nets was often as high as or higher than for adults above 15 years old. Children between ages of 5-14 years had significantly lower usage during the dry and rainy seasons in both the regions. This phenomenon has also been found elsewhere in studies in Uganda [43], in Ghana and The Gambia, but in contrast with earlier findings from, Kenya, Rwanda, Zimbabwe and Burkina Faso, where the pattern was inverse [24], with adult ITN use being higher than that for children under 5 years of age. Another contrasting finding was observed in rural south central Somalia, where net use in younger children, older children, and adults was not different [44]. It thus appears that young children are not at a disadvantage in the allocation of scarce nets and eventual use within households in either of the regions in any season.

The protective effect of nets on infection prevalence during the rainy season among sampled participants was consistent across age-groups except for ages under 5 years in Iguhu. Unlike in the rainy season, the nets' protection role was not clearly observed in older children and adults in the dry season. Children under 5 years old in both regions had the largest protection margin with net use. Overall, during the dry season infection prevalence among net users was lower than in non-users by 56.7% in Emutete and by 26.2% in Iguhu among this age group. During this season, the odds of malaria infection among this age group were 2.3 fold and 1.4 fold higher among non-net users in Emutete and Iguhu respectively. While the approach of targeting interventions to protect at-risk individuals is based on solid scientific grounds [14,45,46], and is widely accepted [12], this strategy should not preclude efforts to maximize communal protection through less selective delivery mechanisms, more so in areas where all age groups are vulnerable. Targeting limited subsidies to maximize personal protection of the most vulnerable should remain a priority, mostly in malaria endemic zones, but more equitable and effective suppression of risk for entire populations in hypo-meso endemic areas where all ages are more susceptible to malaria infections should be encouraged. Likewise, given the relatively higher prevalence of infection through older childhood and into adulthood, during high transmission seasons, it is important to recognize the need to provide ITNs to all members of a community, and not to focus only on young children in areas of low transmission. This resonates with recent calls for high coverage among all community members across the range of transmission settings [47] where it is also recognized that individuals older than five years contribute to transmission.

The impacts of ITNs very much depend on their excito-repellent and insecticidal properties [48,49]. Furthermore, the interaction of these two properties, to yield maximum levels of personal and communal protection, is complex and has crucial implications for ITN programmes across Africa [29]. Households with a functional net during high transmission season, when densities of malaria vectors are high had significantly fewer mosquitoes than those without a functional net. These household vector reduction outcomes concur with findings from earlier studies on the properties of ITNs where it has been clearly shown that ITNs reduce malaria risk among unprotected individuals by suppressing the density, survival [50], and feeding frequency [51] of malaria vector populations. With both insecticidal and excito-repellent properties, ITNs can protect not only the individuals and households that use them, but also members of the surrounding community [15]. This is because they kill adult mosquitoes directly or force them to undertake longer, more hazardous foraging expeditions in search of vertebrate blood and aquatic habits. They repel mosquitoes to reduce the frequency with which they successfully acquire blood, often diverting them to feed on other mammals that do not host the malaria parasite, resulting in greatly reduced prevalence of sporozoite infection [29]. These two major properties add to the effectiveness of ITNs for personal protection because they constitute the major motivating force behind ITN uptake and use at the individual and subsequently the community level.

Conclusion

This study demonstrates that in hypo-meso endemic highlands of western Kenya, despite high mosquito net ownership, actual usage is still remarkably low. To track progress and draw inferences about this principal malaria intervention tool, both ownership and actual use indicators must be taken into account. Regular, region-specific rapid assessments of household possession, use of nets, and respondent's knowledge should complement ongoing distribution, and findings should be incorporated into programme policy. Based on these findings, ITN use in this region could be increased, if IEC messages on malaria transmission were strengthened either through community health workers or net provider programs to encourage participants to use nets they already possess. Since the analyses found that there were disparities in net use across age groups, distribution of free or subsidized nets to the entire population may optimize coverage as well as help increase ITN use.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

HA carried out the field surveys, assembled data, analyzed and drafted the manuscript. GZ performed statistical analysis and revised the manuscript. MCL participated in writing and revised the manuscript, YA and IM participated in study coordination and revised the manuscript. AG and GY designed the study and helped with manuscript preparation. All authors read and approved the final manuscript.

Acknowledgements

We thank all the participants in the study and the technical assistance of A. Ouko, T. Otieno, S. Mongoi and J. Maritim. This article is published with permission of the Director of Kenya Medical Research Institute. The work was supported by NIH grant R01 A1050243. Mr H. E. Atieli is a PhD student at Center for Global Health Research, Kenya Medical Research Institute, Kisumu, Kenya. His research focuses on epidemiology of malaria and vector ecology.