The study took place in the city of Cotonou, the economic capital of Benin, which makes up the Littoral County, one of the country’s twelve counties. In addition to the tertiary sector (trade and services) supported by a number of manufacturing industries, the town also benefits from the activities of the port sector (Port Autonome de Cotonou), the most important of which is that of second-hand vehicles. It is also the point of departure for large trucks carrying goods inland or to countries bordering Benin, such as Burkina-Faso and Niger [9] . A number of parking lots have been created for them in and around Cotonou. These include parking lots in Cotonou, Allada, Godonmè-gare and Sèmè Podji.

This was a cross-sectional, descriptive and analytical study held in March 2023.

The study focused on drivers of large trucks in the parking lots of Cotonou, Allada, Godonmè-gare and Sèmè Podji, whether employed by transport companies or working on their own account.

Drivers aged 18 and over who were present in the above-mentioned parking lots during the data collection period and who had given their free and informed consent were included in the study.

The sample size was calculated using the Schwartz formula, based on the prevalence of excessive drowsiness in Parakou, Benin, of 23.4% [8] and a precision of 0.05, then increased by 10%, giving a sample size of 304 drivers.

Drivers included in the study were selected for convenience. Parking lots were visited one after the other, in descending order of their capacity in terms of surface area. When all eligible and available drivers in a parking lot had been found, we moved on to the next parking lot.

The dependent variable was excessive drowsiness among large truck drivers, measured by the Epworth Sleepiness Score (ESS), developed and published by John. Drivers with Epworth scores above 10 were classified as excessive drowsiness [11] .

The independent variables are five groups of factors, namely:

Ÿ Socio-demographic factors: age, marital status and level of education;

Ÿ Health factors: history of diabetes, hypertension, musculoskeletal pathologies, vision disorders and epilepsy;

Ÿ Sleeping habits: daily sleep duration, number of naps per week and snoring or nocturnal respiratory discomfort;

Ÿ Behavioral factors: daily practice of sport, consumption of coffee, tea, tobacco, alcohol, cannabis and medication;

Ÿ Working conditions: working hours, average driving time before a rest period, average daily driving time, seniority as a truck driver, average distance covered per day, number of substantial meals per day, habitual overweight, remuneration and night shifts.

Data were collected by administering a questionnaire designed to incorporate the Epworth Drowsiness Scale. This consists of ten questions, each with four modalities, scored from 0 to 3 points (0 = never dozing, 1 = little chance of dozing, 2 = good chance of dozing and 3 = very high chance of dozing).

Collected data were recorded on an Excel 2019 sheet, processed and analyzed using Stata 15.0 software. Analysis was performed in two phases, one descriptive and the other analytical.

Parameters for central tendency and dispersion were used in the descriptive phase. The following thresholds were used for variable transformation:

Ÿ Average daily sleep duration: less than 6 hours of sleep per day puts you at risk of excessive drowsiness [12] [13] [14] ;

Ÿ Number of naps: less than 2 per week = risk of excessive drowsiness [15] ;

Ÿ Alcohol consumption: more than 2 standard glasses per day plus = excessive consumption [9] [10] ;

Ÿ Average driving time before taking a rest: less than 5 hours (4 h 30 minutes), in Benin [16] ;

Ÿ Average daily driving time: 10 hours [17] [18] ;

Ÿ Average distance covered daily: less than 500 kilometers (464 km) [3] [18] [19] .

The analytical phase consisted of a bivariate and then a multivariate analysis. For the bivariate analysis, simple logistic regression was used. The multivariate analysis was a top-down stepwise multiple logistic regression, at a significance level of 5%. Variables with a p-value less than or equal to 20% at the end of the bivariate analysis were included in the initial logistic regression model. Baseline modalities were those with the lowest risk.

The fit of the final model was tested, and we concluded that there was a "good fit" because the p-value of the Hosmer-Lemshow test was greater than 5%.

Once the factors associated with drowsiness had been identified, we investigated the link between drowsiness as an exposure and accident history (occurrence of accidents in drivers’ past) as an event. To this end, a comparison of proportions was made, using Pearson’s test, with a threshold of 5%.

Authorization was granted by the administrative authorities in charge of transport. An information note was presented to drivers prior to the abstention of their consent. Data were collected and processed with respect for anonymity and confidentiality.

Altogether 304 truck drivers were surveyed, all male, aged 35.98 ± 8.42 years. Average daily sleep duration on the working day was 4.28 ± 1.79 hours. There was one driver with epilepsy and one with impaired vision. The socio-demographic features of the drivers are presented in Table 1.

The mean Epworth drowsiness test score was 9.89 ± 2.83, with a range of 2 to 21. Of the 304 large truck drivers surveyed, 89 (29.2%) had an Epworth drowsiness scale score of over 10. The prevalence of excessive drowsiness among long-haul truck drivers in Cotonou in 2023 was therefore 29.2%.

These results are shown in Table 2 and Table 3, which demonstrate that factors associated with drowsiness were: daily sleep duration on the working day, high blood pressure, lack of daily exercise and average distance travelled per day.

The following Table 4 presents the final model of the multivariate analysis.

The final model revealed three variables associated with drowsiness among Benin’s large truck drivers in 2023: absence of sport, distance traveled per day by drivers and daily sleep duration on the working day.

Thus, knowing the fact of sleeping less than 6 hours on the working day and the distance traveled per day, drivers who did not practice sport were twice as likely to experience sleepiness as those who practiced it. Those who slept less than 6 hours per day were 3 times more likely to be sleepy than those who slept more than 6 hours per day, taking into account sports practice and distance traveled per day. Drivers who traveled more than 500 kilometers per day were 1.82 times more likely to have excessive sleepiness than those who traveled less than 500 kilometers, taking into account sleeping less than 6 hours per day and playing sports.

The Hosmer-Lemshow test had a p-value of 0.60, greater than 0.05, indicating that the model was adequate.

The findings are given in the contingency Table 5.

OR: 1.73; CI 95% [1.04; 2.87]; p = 0.03.

Drivers of large trucks suffering from excessive drowsiness were 1.73 times more likely to have a road accident, compared to those without drowsiness. A significant association exists between drowsiness and the occurrence of accidents among drivers of large trucks in Cotonou in 2023.