Free versus purchased mosquito net ownership and use in Budondo sub-county, Uganda

This is one of few studies to date that attempted to elucidate the factors influencing bed nets usage status (e.g., bed nets ownership, adequacy and purpose of bed nets use) in Uganda that may inform future intervention strategies. Notably, freely-obtained bed nets (e.g., obtained via national free net distribution campaign, or distributed via public and private clinics or NGOs) were significantly more likely to be misused or used as an alternative purpose rather than to sleep under. As anticipated, lack of awareness of any benefits that can be provided by bed nets was associated with inadequate use of bed nets and decreased bed nets ownership in this study population. However, interestingly, having knowledge that mosquitoes are vectors of malaria was not predictive of increased ownership or decreased inadequate use of bed nets.

This study also demonstrated that only 40% of the households had at least one net among the 1815 households visited in Uganda; and only 27% individuals among the 8011 included in this study correctly have bed nets available to them for sleeping purpose. Despite the relatively low household bed net ownership compared to previously reported high coverage rates elsewhere including Sierra Leone (87.6%), Togo (96.7%) and Ethiopia (91.0%) [27‐29], the results from this present study in Budondo subcounty in Uganda showed a similar pattern of lower bed net use than bed net coverage as in those other countries. Relatedly, coverage in these three other settings varied from 65.0% in Ethiopia to 68.3% in Togo and 76.5% in Sierra Leone [12]. Interestingly, household ownership of at least one net in the Ministry of Uganda Malaria Indicator survey (2014) were reported to be 84.6% for East Central region which encompasses the area covered within this paper [1]. This may suggest a global discrepancy of reported rates.

The consistent gap between household possession of bed nets and individual correct usage draws attention to an important aspect of human behaviour [30]. This study found that neither the complementary distribution of insecticidal nets nor the knowledge on malaria transmission automatically translated into correct usage. Therefore, it may be inferred that information disseminated needs to be culturally-sensitive and based on existing positive beliefs and behaviour if it is to be acceptable by the community.

The finding in this study that having a perceived benefit of net-use was significantly associated with a higher likelihood of the household owning at least one net, as well as individuals having a net to sleep under, is particularly important given its influence on aspects of preventative measures against malaria [31]. The interesting disconnect between the knowledge that mosquitoes propagate malaria and the corresponding benefit of using mosquito nets to prevent malaria, was showcased through the analyses at all levels; the household-, net- and individual-levels. The household-representative’s knowledge that malaria comes from mosquitoes was not statistically significant with households owning a net, nets being used correctly, or individuals sleeping under a net. The perceived benefits provided by respondents in the study were largely stated to be protection from malaria, but non-health benefits were also valued in this study. Other studies have also highlighted the non-malaria benefits in using nets as perceived by communities. For example, a study in Gambia showed that privacy obtained by sleeping under a net to be a motivating factor [32], and in Zanzibar, aspects of comfort appeared to play a key role in personal decisions [33]. The importance of emphasizing the non-health advantages to sleeping under nets have also been suggested in response to seasonal fluctuations in mosquito numbers which can affect the perceived threat of malaria and, by association, net use [8]. This evidence suggests future educational campaigns could benefit from advocating non-malaria benefits of net use, in addition to malaria-prevention, to provide a long-term rationale for consistent use.

This study also suggests that being a child ≤ 5 years was significantly associated with sleeping under a net, and noteworthy, also associated with the household owning a net. These promising results indicate that those most vulnerable to malaria are being protected [4]. Interestingly, among children ≤ 5 years not sleeping under a net, 79% lived in a household that did not own a bed net. This result, that the largest category of non-use is directly related to household bed net ownership, indicates that accessibility is still a barrier to bed net use for this population. On the other hand, children ≤ 5 years of age living in households owning a bed net made up a relatively small proportion (21%) of not sleeping under a net, which suggests that caretakers recognize the importance of protecting young children with bed nets.

Similarly, being pregnant/having pregnant women in household was significantly associated with sleeping under a net and increased odds a net is used correctly. Of the 122 pregnant women who did not sleep under a net, 86% lived in a household that did not own a bed net. This also suggests that households owning bed nets recognize the importance of protecting pregnant women with bed nets. Use of bed nets among pregnant women is associated with lower prevalence of malaria infection, fewer premature births and significant reductions in all-cause maternal anaemia [29]. Given the priorities for children ≤ 5 years of age and for women of reproductive age, the popular belief that the man of the household often takes the net for his own use has been rejected in this study and others [34].

However, bed net ownership will have little impact on relieving the burden of malaria unless people sleep under bed nets. Many large-scale programmes have encountered challenges in consistent use of bed nets [29]. Once households acquire a bed net, there are still other considerable factors that determine their actual use [16]. Reports of misuse of subsidized nets complicate the efforts of government and non-governmental organization (NGO) malaria prevention programmes seeking to promote bed net coverage and use in poor countries [16]. This study found that households who obtained nets for free had lower odds of using them correctly. Comparably, there are documented cases of nets being used for other purposes in Ethiopia [29], for drying fish near Lake Victoria [35], and for fishing in Lake Tanganyika [36]. A multi-country analysis across four countries found that campaign nets (obtained free) were nearly six times more likely to be given away than non-campaign nets [33].

Another important net-use issue is that of nets that go unused at all. One study that investigated intra-household net use in Ethiopia, Ghana, Mali, Nigeria, Senegal and Zambia noted that nets that were paid for were used more than free nets [34]. Another study showed that charging for nets was associated with a high level of usage among the poorest quintile (up to 50%), though with aggressive social marketing of the nets [37]. One proposed explanation has been that of economic theory—a negative price (i.e. paying consumer to use an item in the form of an incentive; free) may provide a negative quality signal and potentially devalue the item in the view of some consumers [16]. As shown in a study in Mozambique where demographic variables were adjusted in the analyses, it was found that compared to non-pregnant women, pregnant women were 27 times more likely to sleep under a mosquito bed net if all bed nets were purchased, and 13 times more likely if all bed nets were received as a donation [38]. Given that Mozambique’s population is considered multi-dimensionally poor, the finding that more bed nets were purchased than received through donations, highlights that Mozambican strategies to achieve universal coverage through mass donation strategies have not yet been reached saturation—in order to maximize household mosquito bed net ownership and correct usage, a combination of both free distribution strategies as well as commercial options will likely be needed [38]. A mix of target subsidy programmes that combine discount vouchers and free distribution showed promising results in Nigeria, Senegal, Uganda and Zambia [29]. The tremendous gains achieved in all four countries suggested that free distribution to poorer families should complement rather than replace giving those with more resources the option to buy a net and will result in optimal ownership [39].

The recently proposed framework that designates households of bed net non-use (no net, having net but not hanging, and hanging but not slept under), could similarly be applied at the individual-level. The findings of this study suggest that 72.5% of individuals lacked access to a sleeping net, and 9.3% had a net in their household but it was not hanging. The third category was unattained due to a lack of individual data on bed net use the previous night. Identifying specific areas of non-use may enable malaria programmes to make informed decisions about which intervention strategies may improve bed net use. Outreach, informed by non-use data, could be aimed at (1) improving bed net access, (2) encouraging hanging bed nets and (3) targeting individuals to sleep under existing bed nets [40].

While this study provides informative and significant results with analyses in a reasonably large sample size, some caveats must also be acknowledged. First, the cross-sectional nature of this study is able to show associations but do not imply causal relationships; thus, future longitudinal or prospective studies are needed to elucidate causal pathways for bed net use and non-use as well as sustainability. Second, households undertook the interview but refused research team’s spot-observations of nets in the home may introduce recall bias. However, it is likely to be non-differential misclassification as the participants were not informed about the purpose of the study and thus misclassification of the outcome is less likely to be dependent on exposure of interest. On the other hand, the findings from this study may not be generalizable to the population that usually do not have any adults in the household during the day time (i.e., typical time of home visits by research team). For example, households with multiple individuals working outside the home may be more likely to be educated or wealthier and thus more likely to use nets. Third, some of the factors investigated may be an indicator for poverty in the study population and hence, may be confounded by socio-economic variables that were not available in this study. Fourth, accurate information on the age of the net was not obtained, which could be another explanatory factor, in addition to source of obtainment of the net, that is related to decreased correct usage of net due to wear and tear of the nets and being used for alternative purposes. Yet, given that literature suggested that the estimated useful life of nets are about 3–4 years [41], and given that the last preceding governmental free mass mosquito net distribution campaign in Jinja district occurred about 2.5 years before data collection of this study, the lifetime of the nets at time of study is not likely to hinder the ownership or condition of nets significantly. Future studies including an analysis of the age of nets would be advantageous to determine the likelihood of confounding by age of the net on the relationship between source of net obtainment and correct net use. Finally, missing values (~ 8%) of potential predictors in the individual-level data slightly reduced the sample size of the analyses for examining the predictors of whether an individual has a net to sleep under. Nonetheless, the distributions of basic characteristics such as gender and age were similar in those included and excluded in the analyses (p > 0.05).