Data collection took place from September to October 2020.

During this study, we interviewed the inhabitants of the seven districts of the commune of Sô-Ava. The sample size was calculated using the Schwartz formula adapted to cross-sectional studies [6] : N = (Z)² × p × (1 − p)/d².

The confidence interval for Z is 95% and the margin of error is 5%. According to the Communal Development Plan [7] , the value of p (proportion of the population of fishermen using prohibited techniques) is estimated to be 76%. The corresponding N value was 280 volunteer inhabitants.

Subjects between the ages of 18 and 60 years, having an income-generating activity and having freely accepted to participate in the study, were included in the study. Those who could not meet any of these inclusion criteria were not included in our study. The random selection procedure for our study was adapted from WHO field survey methods [8] . The two-stage cluster survey was adopted on the assumption that the study population is homogeneous. The sampling frame was made up of the lists of the 7 districts and 69 villages of the municipality.

The data collection focused on three groups of variables, namely

· Socio-demographic and economic variables: age, level of education, district of residence, ethnicity, religion, type of housing,

· occupational variables: main current activities, years of experience,

· Behavioral variables: net use for fishing, source of nets, types of nets used, number of fishermen using nets for fishing, number of people using acadja, nets, hooks.

Data were collected by a group of four interviewers. An individual form was used. The questions were designed on the basis of the different variables listed above. A translation of the questionnaire was made into the local language “Fongbé” for the participants who are mostly uneducated. This questionnaire was tested in Sotchanhoué, a lake locality whose populations have similar practices to those of the Sô-Ava populations. The necessary corrections were then made and the questionnaire was validated.

Several steps were taken to process the data. All data recorded on the collection forms were checked and corrected. They were then entered into Excel 2013 and analyzed using XLSTAT 2021.1.1 and Stata/SE 11.0 statistical analysis software (Stata Corporation, College Station, USA). The study population was described according to sociodemographic, behavioral, and occupational characteristics. Categorical variables were described using proportions. To identify factors associated with the use of impregnated mosquito nets as fishing tools, a univariate analysis followed by a multivariate analysis was performed.

· Univariate analysis

Univariate analyses were performed between each of the independent variables and the dependent variable. Proportions were compared using the Pearson chi-square test or the Fisher exact test, as appropriate. This analysis allowed us to select the variables with a p-value less than or equal to 20% for inclusion in the multivariate analysis model.

· Multivariate analysis

We introduced the variables thus retained in the univariate analysis into a logistic regression model using a “stepwise descending” approach. Independent variables with a p-value above the 5% significance level were eliminated to obtain the final model.

The goodness-of-fit of the model was tested using the Hosmer and Lemeshow test.

Data collection for this component of the study took place in August 2022. This allowed an inventory of the perceptions that the populations of Sô-Ava have on the social uses of impregnated mosquito nets in order to identify the determinants that underlie the detour of these nets to fishing practices.

To better understand the phenomenon and gather reliable information, we interviewed forty (40) social actors selected in a reasoned manner. Four focus groups composed of eight (08) people each (women mothers or guardians of children aged 0 to 5 years, community relays, elders and notables and fishermen) were conducted. The rest of the respondents composed of resource persons (health agents and municipal agents) participated in individual semi-directive interviews. The collection tools were an interview guide and a tape recorder. The empirical data collected were interpreted using theoretical models such as Boudon’s methodological individualism [9] , and the comprehensive theory of Weber [10] . Given that social life for Weber and Boudon starts from individual actions embedded in the field of an environment dominated by a permanent competition for survival, we have been able to position ourselves in relation to individual actions in order to achieve a holistic analysis of these social and sociological realities.

The data were manually processed and entered into Word version 10 and analyzed by triangulating the information by source.

The results of our study reveal that fishing is the main activity of the people of Sô-Ava. It occupies 64.29% of the respondents, followed by agriculture (12.50%) and trade (11.07%). These results are similar to those found in a study in 2012 which reported that fishing is the activity most practiced by the natives while agriculture is practiced by lake dwellers [7] .

Another study carried out in an area with the same topography showed that all the inhabitants practiced fishing, and of these 80% combined it with other activities [11] . This might be explained by the geographical location of the inhabitants of these localities, which divine providence wanted to settle in predominantly lake environments, with water bodies rich in fish products. The fishing contributes then to provide for the vital needs of the majority of these inhabitants. In order to increase the volume of catches, the fishermen of the Sô-Ava municipality used to use different practices commonly known as acadja (parks made of branches stuck in the bottom of the lake to serve as traps for fish) and still nets that do not respect the regulatory norms. This situation would be at the origin of the rarefaction of the halieutic species as reported in a study on the overuse of fisheries resources by the populations [12] . Our results also show that fishermen, in search of finer mesh nets to catch more fish and raise fry in the lake, resort to using nets. In spite of all the sensitization campaigns carried out by the State and non-governmental organizations on mosquito control measures, 67% of nets are used for fishing practices. These findings are similar to those found in a study conducted in a community of fishermen of the lake and rivers of the Congo Basin where the majority of impregnated mosquito nets distributed by international organizations within the framework of the National Malaria Control Program are used in the form of beach seines with mosquito nets to capture fish products [1] . In Sô-Ava, there is a serious public health problem linked to the detour of impregnated mosquito nets intended for the fight against malaria to fishing practices. However, local health agents supported by community relays should increase the awareness of the population on the benefits of the appropriate use of mosquito nets against mosquito bites.

Fishing as a main professional activity and residency in the Sô-Ava area reduce the risk of using IMNs as fishing tools. The professional fishermen would therefore use less impregnated mosquito nets as fishing tools compared to the other trades. This attitude could be explained by the fact that professional fishermen have adequate fishing tools and techniques to increase the volume of their catches, compared to occasional fishermen for example. In fact, professional fishermen buy large areas of water on which they install acadja systems in order to increase their income [13] .

Within the municipality of Sô-Ava, the township of Sô-Ava is home to the town hall as well as most of the socio-community and administrative infrastructures. With these assets that make it one of the most frequented districts, it also receives many tourists. This would explain the position of the communal authorities who reprimand the fishermen who engage in such practices.

The practice of seasonal fishing by farmers and the fact of having more than 35 years of professional experience in fishing practices are the main risk factors associated with the use of impregnated mosquito nets as fishing tools in the commune of Sô-Ava. In fact, farmers who also fish take advantage of the lean season to use the treated nets received during the distribution campaigns for fishing.

In addition, individuals who have lived in the community for several years are much more likely to copy the poor fishing practices observed, in response to the diminishing yields, as described elsewhere [14] . However, the consequences of the pollution of the lake by LLINs do not spare any of the districts of the commune and could also impact all the surrounding communes through which the lake passes, since the lake is a continuum. As a proof, the fish which are caught there are consumed by the populations of the locality and elsewhere. We therefore recommend that the political and administrative authorities extend repressive measures to the entire commune in order to force occasional fishermen to stop this bad practice. As these fishermen and women are of a slightly advanced age, it would also be desirable that technical and financial partners or non-governmental organizations contribute to help them retrain in other sectors of income-generating activities during the lean season so that they can continue to meet their vital needs.

Several historical reasons could explain the low frequency of the use of impregnated mosquito nets against malaria in the municipality of Sô-Ava. According to some participants in the various focus groups, impregnated nets allegedly contain substances that are harmful to human health. The respondents reported having experienced eye and skin irritations as well as breathing complaints following the use of these nets. The toxicity of impregnated nets for children has been described [15] . Similarly, it has been reported that IMNs make it difficult to breathe and make people feel hot at night [16] .

The translucency of the new nets, compared to the old ones, was also mentioned as a reason for their non-use against mosquitoes, thus explaining the low frequency of 33% of net use against malaria in the commune of Sô-Ava, as also observed in the municipality of Abomey-Calavi, still in Benin 38.69%. In the same municipality, 47.62% of the respondents considered that the insecticide-treated net interfered with breathing and 16.67% considered it to be like a shroud [15] . Thus, a connection exists between the reasons given by the various social actors for the detour of LLINs to fishing and the non-use of LLINs against malaria in the area. The distribution of nets in the commune should be equitable, taking into account the size of each household during distribution. People selling their nets for profit should also be informed of the possible repressive measures they may face. Last but not least, health workers should inform the population about the need to spread the nets in the shade for the required time before use, to avoid skin and eye irritation issues faced by net users.