In Eastern province, children with fever are not taken to the clinic due to beliefs that adult relatives are having sex and touching them with unclean hands (bamutyola mwana chifuba). Among the Bembas in the Northern and Luapula provinces this is known as Ichifuba chantanda bwanga. Fever among >2-year-old children are also believed to be caused by the same belief. Another belief is that the chemicals in LLINs cause allergy and suffocation to the users (Hamooya et al., 2015). Other causes include living in swampy and dambo areas that encourage mosquito breeding in the environment. Poor utilization of LLINs by households as well as poor adherence to other malaria preventive measures like refusal to allow in door residual spraying (IRS) in homes (Hamooya et al., 2015).

The utilisation of LLINs depends on a variety of inter-related factors that include ownership of the LLINs, socio-economic and cultural factors, seasons, perceptions on the use of LLINs, education and age, large family size, income, habitation and community participation in environmental and case management. Overall, ITN utilization among pregnant women and under-five children in sub-Saharan countries is very low. Poor educational and awareness level, and ITN related factors (that is, ITN accessibility, sufficiency, quality, physical condition, maintenance, replacement, and effectiveness) contribute to poor ITN utilization (Ozims and Eberendu, 2014). However, these factors could vary across settings and over time, and hence the relevance of understanding the situation in the Zambian context. The use of LLINs is highly recommended, as they greatly reduce the cost and operational difficulties associated with retreatment of nets (Ntuku et al., 2017).

The current generation of LLINs lasts three to five years, after which they should be replaced. So, this reduces the cost of acquiring the nets especially that in Zambia, most mosquito nets are provided free  of  charge with the goal of ensuring at least one net per sleeping space. However, longitudinal studies (WHO, 2011) have observed physical deterioration in LLINs well before 3 years and maintenance of insecticidal activity may be affected by varying conditions and handling practices (e.g. washing). The target for the NMSP 2011-2015 is 100% ownership of targeted households and 80% utilisation of LLINs by 2015 (Zambia Ministry of Health [MoH], 2011). The majority of these LLINs are long-lasting insecticidal nets; 67% of households own at least one LLINs. There has been an increase in household ownership of any type of net and of LLINs over the last six years, from 64 percent and 53%, respectively, in the 2007 ZDHS to 73 and 68%, respectively (Central Statistical Office [CSO] et al, 2015). With all the information, education and communication (IEC) on malaria prevention and control that are conducted by health care professionals (nurses) at Antenatal and under-five children’s clinics in Lusaka, there should be little or no confirmed cases of malaria among the pregnant mothers and their under-five children in Lusaka province and surrounding compounds but on the contrary, between 2013-2015, the reverse is the prevailing situation. Therefore, there is an urgent need to bring about immediate interventions based on the study findings that would not only contribute to the prevention of malaria in human beings but also bring about vector–control. There is paucity of data in evidence-based studies on factors that are associated with the use of the LLINs among the vulnerable groups in the high density areas of Lusaka. Therefore, the study aimed at assessing the utilisation of LLINs among pregnant women and under-five children living in high malaria endemic areas of Lusaka, Zambia.

This was a descriptive non interventional cross- sectional study that Matero and Chainda first level Hospitals as these are among the areas with the high malaria endermicity in Lusaka. Data on use of the LLINs was collected using a semi-structured interview schedule over a 15 months period beginning from October 2019 to January 2021. The proportion of pregnant women and children under the age of five who sleep under mosquito nets is unknown because the Health information management system and Zambia Demographic Health Survey standard collection do not collect this indicator. Therefore, the sample was calculated by a proportion of 50% plus or minus 5%. Based on a 10% non-response rate, the predicted sample size was increased to 420. Therefore, the interviews were done with 420 women with children under the age of five.

The pair of pregnant women and under-five children attending antenatal clinic between 08:30 and 12:30 h in each of the health facilities was sampled using a 1 in 2 systematic random sampling technique until the requisite sample size in each location was reached. The participants were identified through the health centre in-charges, and a private room for the interviews was provided with their assistance. Then, for women who had more than one child, a child was chosen at random and enrolled in the study with the mother.

Ethical approval was sought from ERES converge. The  research team  was  granted   permission   by   the   Lusaka   District   Health Management to visit the study's health facilities. Access to the participants was granted by the respective health facility managers. The study's purpose was described to the participants, and all of them were given information about how they might participate and what the study's outcomes would be. The participants' anonymity and confidentiality were maintained, but it was made clear that the information they provided would be shared with medical personnel if they were found to be in danger. Participants were informed that they had the option to withdraw from the study at any time without facing any penalty. Those who agreed to take part in the interview signed the consent forms before the interview started.

The data were analyzed using SPSS version 22 and associations between quantitative variables were established using the Yate’s corrected Chi-squared tests and Fisher’s exact test. The logistic regression analysis was run in-order to adjust for confounding factors. The Odds Ratio and the 95% confidence interval were used to measure the magnitude of association. The level of significance was set at 5% level. Frequency tabulations described the problem under study while analytical cross tabulations compared groups so that the degree of impact of each associated factor on the use of the ITNs where measured.

Socio-demographic characteristics

The study comprised of 420 participants of which only 0.5% (2) of them were males and the rest (418) were females. Only 25% (104) of the participants were aged above 50years while the majority 75% (316) was within the child bearing age of 15 to 49years old.   Most of the respondents 36% (152) were from Chipata compound and majority 78% (326) lived within 12 km radius of the health facility. Three quarters (314) of the respondents were married. More than a quarter, 36% (152) of the respondents had been to secondary school level of education (Table 1).

Utilisation of long-lasting-insecticide-treated nets

This study showed that more than half,  68%, (286) of the pregnant mothers and their children utilised the LLIN (Figure 1). This number is still low as it is expected that everyone should utilise LLINs. The low utlisation can be attributed to lack of mosquito nets in most homes or because of preferences on whom should sleep under the LLIN. For instance, a little more than a quarter 27%, (113) of the respondents stated that their husbands slept alone under the LLIN depriving the mother and under 5 children. Similarly, a study by Taremwa et al. (2017), to determine knowledge, attitude and behaviour towards the use of insecticide treated mosquito nets among pregnant women and children in rural South-western Uganda reported that only 66.1% were using LLITNs regularly despite most of the respondents (84%) owning LLINs and over 90% understanding the value of LLINS use as a preventive measure against malaria. Similarly, another study by Aderibigbe (2014), to determine the ownership and   utilization  of  long  lasting  insecticide  treated  nets following free distribution campaign in South-west Nigeria, reported that the overall utilization rate was low at 58.5%. Coverage and ownership of LLINs increased significantly following the free distribution campaign but the rate of use was lower than possession. On the contrary, a study conducted in Mbarara, Uganda by Nuwamanya et al. (2018), reported utilisation of LLINs among 91.1% of adult respondents and among 91.9% of children under-five. This is more likely because the participants were given mosquito nets six months prior to the study. Further, another study conducted by Ugwu et al (2013) to determine the utilization of insecticide treated nets among pregnant women in Enugu, South Eastern Nigeria reported that the use of ITNs was poor (39.1%) among pregnant women in Enugu.

Socio-cultural and traditional beliefs of the participants

This study revealed that almost all 99.5%, (418)  of the  respondents believed that the use of LLINs causes respiratory and skin problems. However, almost all 98%, (412) of the respondents denied knowing any cultural beliefs associated with fever in pregnant women (Table 2). Similarly, Taremwa et al. (2017) in a study to determine knowledge, attitude and behaviour towards the use of LLINs among pregnant women and children in rural South-western Uganda reported that almost 41% of respondents who were not using LLINs stated that they were concerned about side effects of the LLINs or had beliefs that chemicals used could cause cancer. Another study by Fokam et al. (2017), to determine the predictive factors of long-lasting insecticide-treated net ownership and utilisation in the Bamenda Health District of Cameroon showed that negligence, heat, discomfort/suffocation and torn nets were among other reassons for non-usage of bednets.

Utilisation of long-lasting-insecticide-treated nets by demographic characteristics of the participants

The chi-square test showed that age, number of adults in a household older than 18 years old, sex, participants distance from house to clinic, total number of children under-five year old children, number of women aged 15 to 45 years old and source of income were not significantly associated with utilisation of LLINs. However, marital status, total number of family members, residential area, and family income were associated with utilisation of LLINs.

On univariate binary logistic regression, husband sleeping alone under the LLITN and number of wives in a family were significantly associated with utilisation of LLITNs with a p-value of 0.01 and 0.037 respectively (Table 5).

Multivariate logistic regression model revealed that only the total number of family members was significantly associated with utilisation of LLINs. This could be due to accessibility that family members have to nets. If a household has one net but more members, only a few of them would utilise the net. The pattern of sleeping also affects the utilisation of nets in a household. If mothers sleep separately from under-five children in  a  household with only one net, then one of the two would not have access to the mosquito net (Tables 4 and 6).

Contrary to findings of this study, a systematic review of studies in 18 selected African countries conducted by Noor et al. (2009) showed that age was significantly associated with use of LLINs. The utilisation was higher among children under the age of 5 years and pregnant women as compared to school going children. This is most likely because of the fact that the distribution of LLINs  focuses on children under the age of 5 years and pregnant women during under-five clinic and antenatal visits and also during mass-catch-up immunization campaigns programs for under-five years old children. Additionally, the high utilisation among the under 5 and the pregnant women could be related to the sleeping patterns where nursing and younger children share sleeping structures in a household with their mothers and/or both parents, while older children sleep on separate beds or mats elsewhere in the household (Tables 3 and 4).

The utilisation of LLINs was low among pregnant women and under 5 year old children because of  low  ownership of mosquito nets by the study respondents and also large number of members in households against the available mosquito nets. Health facilities must lobby and distribute LLINs to the community if the utilisation can be scaled up in order to achieve the desired level of prevention of malaria among pregnant women and children under the age of 5 years old. Further, More awareness should be created on the importance of sleeping under Long Lasting Insecticide Treated Mosquito Nets.

The authors declare no conflicts of interest.

The authors would like to thank Dr Dorothy Chanda the PI for her efforts in seeing to it that this study is conducted. Many thanks also to the Directorate of Research and Graduate Studies- University of Zambia for the financial support that facilitated conducting of this study.