Study area
This study was conducted from 2019 to 2021 across three regions of Benin: Southern, Central, and Northern. The Southern region has a sub-equatorial climate, with two dry seasons (August-September and December-March) and two rainy seasons (April-July and October-November). The Central and Northern regions experience a tropical climate with one rainy season (May to October) and one dry season (November to April), with a more pronounced contrast between the rainy and dry seasons in the Northern compared to the Central region [3]. Study sites were chosen in collaboration with the National Malaria Control Program of Benin and included both public and private health centers in rural and urban areas within these regions. Health districts were randomly selected in the Southern, Central, and Northern parts of Benin. The selection involved a random draw of stratified maternity clinics, requiring a minimum of 20 ANC visits per month. A total of 14 maternity clinics were stratified and selected based on their location in the Southern (five clinics), Central (five clinics), and Northern (four clinics) regions, as well as their status as public or private and their setting as rural or urban (see Fig. 1).
Data collection
For this study, data were collected using the KoBo Toolbox platform. The survey was designed in XLSForm, a flexible tool that enabled us to create a structured, intuitive questionnaire incorporating skip logic and validation rules to minimize errors. Field teams were trained in using this data collection tool and subsequently collected data electronically with the KoBo Collect mobile app. This app facilitated real-time data entry, even in offline settings, and simplified secure data transfer for analysis.
The sample size was determined based on a previous study [26], which included approximately 240 women in the southern region of Benin. For this study, we targeted the same number of participants, around 240, for each of the three regions (Southern, Central, Northern), aiming for a total of 720 women. By the end of the recruitment period, 718 participants had been enrolled. Pregnant women were recruited at their first antenatal care visit at the maternity clinic and visited twice at home. All women attending their first antenatal care visit (ANC1) were invited to participate in the study. Those who agreed and signed the consent form were included in the study. The first home visit occurred 48 h after recruitment, and the second visit took place one month later. During the first home visit, data was collected on LLIN ownership by pregnant women prior to their first antenatal care (ANC) visit, as well as data on usage, including whether the pregnant woman slept under a bed net the previous night and the age of these LLINs if applicable. Information on LLIN distribution during ANC1 was also gathered. If the pregnant woman had received a LLIN at ANC1, it was noted whether she had installed it on her bed. Additionally, pregnant women were asked if they received sensitization during their ANC1 visit, which included instructions on proper LLIN use and the importance of using LLINs, especially during pregnancy. At the second home visit, the LLINs found on the sleeping area of each pregnant woman were collected for laboratory evaluation of physical integrity and bio-efficacy. These LLINs were then replaced with new ones provided by the study.
Physical integrity
The physical integrity of LLINs was assessed by counting and categorizing holes according to size on each side of the net:
Size 1: Smaller than a thumb (0.5–2 cm).
Size 2: Larger than a thumb but smaller than a fist (2–10 cm).
Size 3: Larger than a fist but smaller than a head (10–25 cm).
Size 4: Larger than a head (> 25 cm).
The area of each hole was calculated assuming circular holes, with the diameter equal to the midpoint of the size category (except for the largest category, which uses an arbitrary diameter of 30 cm). Estimated areas were weighted using the size 1 area as a reference. The hole index (HI) was calculated for each LLIN using:
HI = (A x no. of size-1 holes) + (B x no. of size-2 holes) + (C x no. of size-3 holes) + (D x no. size-4 holes) with A = 1, B = 23, C = 196, and D = 576 [27].
An LLIN with an HI between 0 and 64 is considered “good,” between 65 and 642 “acceptable,” and greater than 643 “bad” [28].
Bio-efficacy
Bio-efficacy of LLINs was evaluated using the standard WHO cone test. Twenty 2–5 day-old, non-blood-fed female Anopheles gambiae (Kisumu strain) mosquitoes were exposed to five pieces (25 cm x 25 cm each) of each LLIN for 3 min. As negative controls, 100 mosquitoes were exposed to non-insecticide-treated net pieces. Post-exposure, mosquitoes had access to a 10% sugar solution and were kept at 27 ± 2 °C with 75 ± 10% relative humidity. Knock-down (Kd) and mortality rates of mosquito were recorded at 1 and 24 h post-exposure, respectively. According to WHO criteria, an LLIN is bio-effective if the Kd rate is ≥ 95% or the mortality rate is ≥ 80% [27].
Statistical analysis
Statistical analysis was performed to estimate proportions and confidence intervals for each indicator, including LLIN ownership, distribution, use, physical integrity, Kd, and mortality rates. Chi-square tests compared these indicators across stratified variables: region, urban/rural status, and health center public/private status. STATA version 15 was used for all statistical analyses. All statistical analyses accounted for clustering induced by recruiting women from different maternity wards via STATA’s survey data analysis module, utilizing the svyset command to declare the survey design and svy commands to estimate robust standard errors, adjusted confidence intervals, and p-values.