Factors associated with utilization of LLINs among women of child-bearing age in Igabi, Kaduna State, Nigeria

Igabi Local Government Area (LGA) with headquarters at Turunku, is one of the 23 LGAs in Kaduna State Nigeria. It has a population of 557,624 and is sub-divided into five urban districts, which are, Afaka, Birnin-yero, Kwarau, Rigachikun and Rigasa, and seven rural districts namely; Fanshanu, Gwaraji, Igabi, Kerawa, Sabo-Birnin, Turunku, and Zangon-aya. The LGA is located 650 m above sea level, between latitude 10° 47′ 0″ N and longitude 7° 46′ 0″ E in the tropical Sahel to Sudan Savannah with annual rainfall varying from 1000 to 1500 mm and highest precipitation of 72% in August. The rainy season is usually from May to October. The annual mean temperature is 34 °C. This can rise to 41 °C at the peak of the dry season in April and drop to as low as 12 °C in January during severe harmattan. This climatic condition influences the breeding and survival of malaria mosquitoes, and transmissions occur throughout the year. The incidence [20] and prevalence of malaria cases among under-five children at Igabi LGA is 13.6% and 55%, respectively. Also, 92% of households in Kaduna State have at least one LLIN and 62% slept under LLIN the previous night [6]. This study was conducted in October 2015 at the end of the rainy season. Six months before this study, the LGA benefitted from a mass household LLIN distribution campaign conducted in Kaduna state [20].

This study employed a cross-sectional design.

Women of child-bearing age, 15 to 49 years (WCBA) who reside in the selected households in the community for 12 months or more without cognitive deficits, or any chronic and debilitating illness that hinders participation in an effective interview were the participants.

Using the formula, n = z²pq/d² at 95% confidence interval (1.96), prevalence of WCBA in North-Western Nigeria who slept under any net a night before, 43% [21] with a confidence limit of ± 5%, and a non-response rate of 10%. To adjust the required sample size for cluster sampling design, a correction factor of 1.5 is used as a design effect, as a less heterogenous community in this district was assumed. Therefore, the calculated minimum sample size was 630.

A three-stage cluster sampling technique was used to select the participants. Ten wards were randomly selected from the Igabi LGA. This included six rural wards—Fanshanu, Gwaraji, Igabi, Sabo-Birnin, Turunku, Zangon-aya and four urban wards—Afaka, Birnin-yero, Rigachikun, and Rigasa. These were selected at first stage making 2638 villages or settlements and 161,241 households. For a cluster or village/settlement selection, the household enumeration data generated by WHO/UNICEF during micro-planning for the mass LLIN campaign in these settlements were used as the sampling frame. With a sample size of 630, to improve the validity and precision of estimates and considering a wide community representation (population variance), using a probability proportional to size, 63 villages or clusters were randomly selected with a cluster size of 10.

Using the list of the households in each village/settlement generated during microplanning for mass LLIN distribution, ten households were systematically selected from each village/settlement. After the community entry at the traditional leader’s house, the first selected household was located and thereafter every Kth household was recruited into the study until all the 10 households. In a household with multiple eligible WCBA, one was selected by the ballot method and recruited into the study. Any household without eligible participants is an ineligible household.

Data was collected from September to October 2015, by ten trained female community health workers who speak English and Hausa languages fluently and reside within the district. They were supervised by five undergraduate medical students. The data collectors used a pretested, structured questionnaire adapted from the Malaria Indicator Survey [22, 23] and other literature [24, 25] to collect information from the respondents. Information collected from the respondents included their demographic profile, knowledge of LLIN as a malaria prevention strategy, LLIN ownership and utilization, and influencers of LLIN utilization. The utilization of LLIN was verified by asking if the LLIN was used and who among the household members had slept under a LLIN the night before the survey. The interviewers also inspected and documented the conditions of all nets in the household to determine if they were used, unused, torn, hung, new or old. The LLIN knowledge items were questions from the literature, Malaria Indicator Survey and Demographic Health Survey. The adapted questions were reviewed by two malaria subject matter experts for face and content validity. They focused on the item’s accuracy, relevance and clarity. However, face validity was assessed by five malaria researchers and 15 WCBA for relevance to the WCBA. Following the review and corrections, each correct response was scored [9] “1” and incorrect “0”. The aggregate score for each respondent was expressed as percentage. Knowledge was categorized using percentile scores of 75th percentile or more for good knowledge, average knowledge between 50th and 74th percentile, and poor knowledge as < 50th percentile.

The outcome or dependent variable for this study was LLIN utilization. LLIN ownership was defined as having at least one LLIN within the household that may be used for sleeping and utilization is to sleep inside hung LLIN a night before the survey. Independent and explanatory variables are respondents’ demographic characteristics, LLIN ownership, and knowledge items.

Data was entered and analysed using Epi-Info version 7 statistical software. Frequency and proportions for sociodemographic characteristics, knowledge of LLIN, ownership and LLIN utilization were calculated. Ownership was the proportion of WCBA living in a household with any net for sleeping; its denominator was the total number of households sampled. Utilization referred to the proportion of respondents that slept under LLIN a night before the survey. Therefore, the denominator is the number of WCBA with LLIN. The mean (± standard deviation) or median for continuous descriptive variables, were also calculated. Bivariate analysis using odd ratio was done to determine the association between categorical outcome and independent variables and any associated factors with p-value ≤ 0.2 selected for a multivariable logistic regression model with an upward loading approach to identify factors independently influencing LLIN ownership and LLIN utilization among WCBA. For all statistical analyses, the significance level was set at a p-value of < 0.05 and a 95% confidence interval.

The response rate was 99.8%. Table 1 shows the sociodemographic characteristics of the respondents. A total of 629 WCBA interviewed, their mean age (± standard deviation) was 29.3 (± 6.2) years, about 22.0% were currently pregnant, 40.5% had no formal education, 41.1% were employed, 47.7% were living in rural communities, 57.4% were housewife and 84.3% were Hausa by tribe.

LLIN awareness among the women was 96.3% and 36.9% WCBA were able to differentiates LLIN from other bed nets by its factory impregnated insecticide (Table 2). LLIN was described by 45.0% WCBA as a free net given by the government, 29.6% as nets with impregnated insecticide that last longer more than 3 years, kill and prevent mosquito bite, and 21.1% of respondents said it is used for sleeping. The replacement period for LLIN was said to be > 2 years by 44.4% WCBA. On the use of LLIN, 67.6% responded that LLIN can be used to prevent mosquito bite and 36.9% knew that factory impregnated insecticide differentiates LLIN from other household nets. Overall, 24.0% (151/629) WCBA had good knowledge of LLIN (Table 3).

In overall, 68.6% (341/497) of WCBA got LLIN in the last 7 months and ownership of at least one LLIN for sleeping by WCBA was 79% (497/629). The mean (± SD) household residence per LLIN was 2.6 (± 1.7) not different from the household with pregnant WCBA (Tables 4 and 5). Despite having at least one LLIN for sleeping, 6% (30/497) did not hang the available LLIN because they have no place to hang it and 13.1% (65/497) hanged more than enough bed nets for the use of household members. But 2.4% (12/497) did not hang the available LLIN because they did not know how to do it.

Generally, LLIN utilization was 69.9% (440/629), but in a household with at least 1 LLIN for sleeping, 75% (373/497) WCBA utilized it. Among the 137 currently pregnant WCBA, 82% (112/137) owns at least one LLIN and 74% (101/137) utilized it. However, 79% (89/113) of pregnant WCBA with at least one LLIN slept under LLIN a night before.

The reasons for not having at least 1 LLIN by 132 WCBA were dislike to LLIN (56.8%), not knowing where to get LLIN (40.2%), using other malaria preventive measures (34.8%), no money to buy (25.8%), no voucher to collect free LLIN during mass distribution campaign (17.4%), stock out of LLIN during mass distribution campaign (8.3%), and loss the voucher during free LLIN mass distribution campaign (6.0%). Also, 11.5% (57/497) WCBA with at least one LLIN for sleeping in their household did not utilized it because of heat and discomforts (66.7%); poor perception that mosquitoes still bite while sleeping under LLIN (38.6%) and that no or low mosquito density in the sleeping room (22.8%); have a damaged or torn LLIN for sleeping (10.5%), LLIN in the household is inadequate as visitors used the LLIN (8.8%), and feeling of breathlessness or choking while sleeping under LLIN (7.0%).

Table 6 shows the characteristic of WCBA associated with LLIN ownership. Women who lived in in rural communities were four times more likely to own LLINs compared to their urban counterpart (OR 3.8, CI 2.5–6.0, p < 0.01). Younger women (aged < 30 years) were less likely to own LLINs compared to the older women (OR 0.6, CI 0.4–0.6, p = 0.01). In Table 6, other respondents characteristics like having formal education (OR 0.6, CI 0.4–0.9, p = 0.03), and living in a household with less than three sleeping rooms (OR 0.4, CI 0.2–0.8, p = 0.01) were less likely factors associated with LLIN ownership.

Women who are currently pregnant were more likely to own LLIN compared to other non-pregnant WCBA, but it was not significant (OR 1.3; CI 0.8–2.2, p = 0.3). The main predictor of LLIN ownership is rural residence as women who lived in rural communities were more likely to own LLINs when compare to the urban women (aOR 3.4; 95% CI 2.2–5.4).

Generally, rural WCBA is strongly associated with LLIN utilization. Women who lived in rural communities were more likely to utilize LLINs than the urban women (OR 3.4 (95% CI 2.3–4.9) (Table 7). Also, younger women (aged < 30 years) were less likely to utilize LLINs compared to older women (OR 0.7; 95% CI 0.5–0.9, p = 0.03). Finally, after adjusting for these parameters age, currently pregnant, formal education and number of rooms for sleeping, women who lived in rural communities were more likely to utilize LLINs compared to their urban counterparts (aOR 3.9, CI 2.5–6.0, p < 0.01). Also, the number of rooms for sleeping in the households positively influences the utilization of LLINs among the respondents. Women who lived in a household with 2 or 3 rooms for sleeping were more likely to use LLIN compared to those living in 1 room (aOR 1.6, CI 1.1–2.4, p = 0.01).

The study was done at the peak of a malaria survey and about the same time when data collection for the 2015 Malaria Indicator Survey was ongoing. Nigeria is presently amongst the leading countries with a high burden of malaria and is making slow progress in efforts to reduce the malaria burden globally. The information on LLIN utilization was based on self-reported information from the participants. Data collectors observed the available LLIN in the household to know if they were hung, not hung, and if kept in its original packaging. The data collectors asked for who slept under a hung LLIN a night before the survey among the household members. Therefore, information bias like recall and social desirability are possible limitations. Efforts were made to reduce some of these limitations by piloting the questionnaire, training female data collectors to ensure standardized data collection processes, and participants were encouraged to be truthful with their responses and were assured of their confidentiality. However, it was not possible to verify actual LLIN utilization the previous night as well as confirm other self-reported data.