Factors associated with buying sex and the knowledge that condoms prevent HIV among long-distance truck drivers at Kazungula weighbridge terminal, Chobe District, Botswana

Background: Long-distance truck drivers (LDTDs) have a higher rate of HIV infection compared to the general population. This is due to their living and working conditions which predispose them to riskier sexual behaviours. Inadequate knowledge of HIV and AIDS, coupled with risky sexual behaviours such as unprotected sex with commercial sex workers (CSWs), predisposes LDTDs to contract and propagate HIV. This study aims to determine the factors associated with buying sex and the knowledge that condoms prevent HIV transmission among long-distance truck drivers at Kazungula weighbridge terminal in the Chobe District of Botswana. Methods: A cross-sectional descriptive survey was employed and 399 LDTDs participated. A questionnaire was used to collect sociodemographic data and information on knowledge of HIV and AIDS and the sexual behaviours of the LDTDs. Results: The results reveal that more than half of the LDTDs (56.9%; n = 227) had paid for sex with CSWs at some point in their travels, and 27.1% (n = 108) reported having had unprotected sex with CSWs. The LDTDs who preferred to have sex with CSWs without a condom had about five times the odds of buying sex with CSWs than those who preferred to use a condom (AOR 4.9; 95% CI 2.85–8.46). Disliking condom use was a factor contributing to less knowledge of condom use preventing HIV among the LDTDs (AOR 0.4; CI 0.17–0.97). Conclusion: It can be concluded from the results of this study that the LDTDs engage in considerable risky sexual behaviours, and associated factors were found to be multidimensional. This population remains of concern in HIV acquisition and transmission. Therefore, there is a need for a robust public health response to deal with the problem of both new infection and re-infection with HIV in this population.


Introduction
Long-distance truck drivers (LDTDs) take protracted periods away from their families and immediate environments, which exposes them to the temptation of gratifying their sexual needs through buying sex with commercial sex workers (CSWs) at truck terminals. A corpus of evidence is consistent in pointing out that LDTDs pay a high amount of money to CSWs for unprotected sex, which puts them at high risk of acquiring HIV (Sorensen et al., 2007;Matovu and Ssebadduka, 2013). According to the literature, this problem is further heightened by the LDTDs' lack knowledge of the risks associated with unprotected sex, and either consider themselves invulnerable to HIV, or they believe that washing their genitals after unprotected sex prevents them from acquiring HIV (Agha, 2010). Moreover, their living and working conditions put them at risk of infection and transmission of HIV. The International Organization for Migration (IOM) has identified the following working conditions as factors that contribute to the risk of HIV infection among LDTDs: long-term separation from a regular partner, subject to stress, attractive CSWs and their easy access in so-called hot spots (i.e. truck stops), and drivers usually not having access to adequate medical services, including treatment for sexually transmitted infections (IOM, 2005;Tiang et al., 2010). Mobility is one of the significant drivers of the HIV epidemic, as it aids transmission by perpetuating riskier sexual behaviours than that of the residents of the countries travelled through, who have been found in some studies to have positive sexual behaviours (Deane et al., 2010;Nasir et al., 2015). A positive behaviour change observed in local truckers can also be attributed to the implementation of HIV-prevention programmes for this population. For example, the North Star Alliance, a non-governmental organisation formed to unite the transport sector in its response to the AIDS pandemic in southern Africa, provide the LDTDs with access to primary health care and HIV-related services through roadside wellness clinics (Lalla-Edward et al., 2017).
According to the world-wide UNAIDS report, 690 000 people died of AIDS-related illnesses in 2020 compared to 1.2 million in 2010, and the new HIV infections remain as high as 1.5 million, with a 30% decline since 2010 (UNAIDS, 2020). A study conducted on groups that are at high risk of contracting HIV indicates that LDTDs are one of those groups. The risk for LDTDs is increased by their sexual practices, including regular sex with commercial sex workers (CSWs), with some reporting engaging in unprotected sex with CSWs, having multiple sexual partners, and with some reporting as high as two or more sexual partners (Mishra et al., 2012;Sawal et al., 2016). In many parts of the world, LDTDs are regarded as one key population that is at high risk of acquiring and transmitting HIV, with buying sex from CSWs and a lack of knowledge that condoms prevent HIV transmission being ranked high as some of the main causes of the high prevalence rate of HIV among this population (Atilola et al., 2010;Azuonwu et al., 2011;Delany-Moretlwe et al., 2014;Maarefvand et al., 2016;Ishtiaq et al., 2017).
This article presents findings from a study conducted among a relatively large sample (N = 399) of LDTDs, the research site being a weighbridge terminal in Botswana located close to the country's borders with Zambia and Zimbabwe. The study was investigating factors associated with buying sex and the knowledge that condoms prevent HIV transmission among long-distance truck drivers at this terminal. Kazungula weighbridge terminal is one of the major truck junctions in southern Africa and it is important to compare the findings of this study to similar studies around the world. Moreover, there is a lack of relatively large scale studies of this nature in southern Africa and none has been conducted on this topic in Botswana. To this end, this study will add to the body of knowledge about this sub-population at Kazungula weighbridge terminal.

The sociodemographics and sexual behaviours of and condom use by LDTDs
LDTDs constitute a heterogeneous population. Therefore, their sociodemographic characteristics such as age and marital status provide an important relationship with the buying of sex and the knowledge that condoms prevent HIV transmission (Yusuf et al., 2014). Studies carried out on LDTDs reveal that the significant demographic predictors of condom use are educational and marital status and age. Sociodemographic factors such as the age at which one started the trucking occupation also play a significant role in their sexual behaviours, with adolescent entrants exhibiting higher risk sexual behaviours like buying sex compared to the adult entrants (Mishra et al., 2012). A study conducted in Nigeria found age as a demographic factor related to risky sexual behaviour, with younger truck drivers being more likely to use condoms; while, in a study conducted in Bolivia, older truck drivers were found to use condoms less, and those with a higher level of education had significantly higher levels of condom use (Sorensen et al., 2007;Aniebue and Aniebue, 2009). Yaya et al. (2016) suggest that the link between education and risky sexual behaviour may indicate that education inspires an understanding of HIV-prevention information, which is provided as a form of risk reduction intervention for LDTDs. It can therefore be concluded with evidence from the literature that demographic characteristics of LDTDs such as literacy, current age and the specific routes on which the LDTDs usually travel correlates with different levels of risky sexual behaviour and the corresponding prevalence of HIV among this population (Mishra et al., 2012). However, no studies have been conducted in Botswana to establish if a connection exists between the sociodemographics of the LDTDs and buying sex, or knowledge that condoms prevent HIV transmission. This study closes this gap.
Notwithstanding that some of the sociodemographic characteristics of LDTDs are factors that contribute to buying sex and knowledge that condoms prevent HIV transmission, there is evidence that LDTDs visit or get visited by CSWs, making them both vulnerable to HIV infection. Though there is no study that investigates LDTDs' exchange of money for sex with CSWs in Botswana, there is evidence across the globe that LDTDs buy sex and truck terminals are common places for them to have sex with CSWs in exchange for money; hence, it was important to conduct this study at Kazungula weighbridge terminal. A study conducted in Hong Kong reveals that one quarter of LDTDs visited CSWs and 80.4% reported indulging in sexually risky behaviour (Wong et al., 2007). These results connect well with those of a study conducted in India in which about one quarter of LDTDs visited CSWs and about 58% to 74% reported inconsistent condom use (Mishra et al., 2012). In other studies, some LDTDs are reported to propose higher payments to be paid for non-condom use with CSWs. For example, in a study conducted in South Africa, some LDTDs paid more for sex without a condom due to the CSWs' desperate economic problems (Makhakhe et al., 2017). Nevertheless, literature shows that in spite of the availability of condoms in the condom dispensers at the rest rooms at the truck stops, their use is not necessarily associated with their availability, as demonstrated in a study that reveals that some LDTDs make commercial sex workers choose between sex without a condom or losing a sale.
The attitude by LDTDs of disliking condom use is of great public health concern as it increases the chance of HIV transmission (Sawal et al., 2016). It is clear from extant literature that LDTDs have negative attitudes towards condom use and thus indulge in riskier sexually encounters, which predisposes them to acquiring HIV and other sexual transmitted infections (Matovu and Ssebadduka, 2013). Their dislike of condom use has negative consequences for their sexual behaviours. In some studies, LDTDs put the blame on condoms, saying that they destroy the mood for sex (Sorensen et al., 2007;Matovu and Ssebadduka, 2013). But generally, studies are inconsistent on the LDTDs' level of knowledge about HIV and there is less literature that focuses on factors associated with the knowledge that condoms prevent HIV transmission. Moreover, no studies have been conducted in Botswana on factors associated with buying sex and the knowledge that condoms prevent HIV transmission among long-distance truck drivers at Kazungula weighbridge terminal. To this end, it was important to conduct a study on the factors associated with buying sex and the knowledge that condoms prevent HIV transmission among LDTDs at Kazungula weighbridge terminal to serve as a reference point in planning public health programmes for LDTDs.

Long-distance truck drivers' knowledge of HIV and AIDS
A very important aspect to consider when taking into account LDTDs' sexual behaviours is their knowledge of HIV and AIDS. In some studies conducted among truck drivers in Iran and Morocco, there was a low level of knowledge about HIV and AIDS, and the HIV knowledge index was low among the semiliterate (Tehrani and Malek-Afzali, 2008;Himmich et al., 2015). Some studies, however, reveal high levels of knowledge about HIV and AIDS. For instance, in a study conducted in India, 95% of LDTDs knew that condom use can reduce HIV transmission, but low levels of knowledge were reported in relation to the importance of refraining from having multiple concurrent sexual partners and avoiding high-risk groups like CSWs (Singh and Joshi, 2012;Nasir et al., 2015). In some studies, the level of knowledge that non-condom use during sex is a mode of HIV transmission was as high as 88.8% (Aniebue and Aniebue, 2009). The study by Aniebue and Aniebue (2009) used a cross-sectional design and the questionnare was researcher administered, while Nasir et al. (2015) carried out a mixed methods analysis of descriptive and inferential points, followed by interviews. From the different study designs and methods used in different studies, it is clear that even though LDTDs have high levels of knowledge about HIV and its prevention, they harbour some misconceptions about the causes and transmission of HIV. Empirical evidence suggests that the misconceptions harboured by the LDTDs about the causes and transmission of HIV include mosquito bites, shaking of hands and sharing of rooms (Singh and Joshi, 2012;Poda and Sanon, 2015).
With this in mind, it is necessary to examine LDTDs' HIV knowledge as a component of information, attitudes towards condoms, perceived vulnerability to HIV and AIDS and the factors associated with sexual transactions in exchange for money at critical points such as the weighbridge of Kazungula terminal in the Chobe district of Botswana.

Study design and population
The study employed a quantitative cross-sectional descriptive approach to assess the level of HIV and AIDS knowledge and related sexual behaviours among long distance truck drivers. Cross-sectional descriptive design was suitable for this study as it involved collecting data from a large sample and subsequently describing, analysing and interpreting factors associated with buying sex and the knowledge that condoms prevent HIV transmission among LDTDs at a single point in time (Polit & Beck, 2017). This study was conducted at Kazungula weighbridge terminal in the Chobe district of Botswana. Kazungula village is located on the northern side of Botswana. The village lies to the south of the confluence of two big perennial rivers, the Chobe and the Zambezi. Kazungula village has two border posts, one for crossing into Zimbabwe and the other into Zambia. Trucks crossing through Botswana to Zambia use the north-south transport corridor, which forms part of the Southern African Development Community (SADC) regional truck network that passes through Kazungula, thus providing an important link for regional economic integration, with voluminous trucking activity on this route. Kazungula weighbridge terminus lies about two kilometres from the Botswanan and Zambian borders, and trucks stop at the weighbridge for customs duty and also for a break from long-distance travel. Trucks remain at the weighbridge for long periods, making the place a lucrative rendezvous for commercial sex work. Approximately 3 000 LDTDs use the Kazungula weighbridge in a month and most of them are from neighbouring South Africa. The study population was comprised of both the international and local LDTDs available at Kazungula weighbridge terminal, who were in transit and had stopped there. The LDTDs take a day or more, queuing to have their trucks weighed to pay the required customs duty and also to have a break from a long distance drive before proceeding to their various destinations. Most of them spend their nights in their trucks.
As for the road network, Kazungula village forms a vital link with other SADC countries, which is why there is so much trucking activity in this area, making it one of the major trucking corridors on the subcontinent, with a lot of trucks parking at this terminal in transit. Nonetheless, there are other routes that also service the north-south corridor, for example, the Beitbridge route that passes between South Africa and Zimbabwe. Compared to the Kazungula route, the Beitbridge route is relatively shorter for truckers travelling from most SADC countries to more northern sub-Saharan Africa, making it more favourable. However, according to a report by Nkala (2021), the recently opened Kazungula bridge that crosses over the intersection of the Chobe and Zambezi rivers, replacing the ferry crossing, gives the Kazungula route an advantage over the Beitbridge border crossing which regularly experiences transit bottlenecks and other problems on the Zimbabwe side. These issues include the dilapidated road network, many taxes and multiple vehicle inspection roadblocks. Furthermore, Kazungula benefits from better infrastructural development initiatives, such as the Botswana-Zambia one-stop border post, which now makes this route a better alternative in facilitating intra-regional freight compared to other routes (Nkala, 2021).

Sampling procedure
A simple random sampling technique was used to select the participants. The study opted to use a population of unknown size based on the transit nature of the border posts in Botswana. Nevertheless, the sampling relied on the reported estimate of 3 000 LDTDs that pass through Kazungula weighbridge terminal each month as a basis for sampling. The sample size was calculated using the Raosoft sample size calculator at a confidence level of 95% and a 5% accepted margin of error at a response distribution of 50%. A 5% buffer of participants was included in the sample as a contingency and also to improve the extrapolation of the results, culminating in a sample size of 399 .Inclusion criteria were: • Only truck drivers who understood either English or Setswana were included in the study; • LDTDs who had truck driving as their occupation, who drove heavy trucks with three or more axles, and who had registered for custom duty services at Kazungula weighbridge terminal; and • Truck drivers that were a day or more away from home and had stopped at Kazungula weighbridge terminal.
All truck drivers who worked in Kazungula village or its vicinity and those who had not been away from their homes for a day or more were excluded.

Data collection
The data collection method was administered through structured questionnaires. The questionnaires were administered by WG and a trained research assistant, using adapted validated questions from a previous study and it was prepared in English and translated into Setswana (Madiba and Mokgatle, 2015). The questionnaire was composed of closed-ended questions to collect sociodemographic data of the LDTDs and their sexual behaviours such as exchange of money for sex with CSWs and questions to ascertain the LDTDs' knowledge that condoms prevent HIV transmission to allow a comparison of findings with previous studies which used such questions. On each day of data collection, the available LDTDs gathered at the weighbridge terminal offices in collaboration with the facility management for data collection. The questionnaires were administered to the LDTDs who voluntarily consented to participate, and the exercise took about 30 minutes per participant to complete. The questionnaires were administered when the LDTDs were less busy and waiting for their turn to be assisted at the weighbridge terminal. Data collection took about seven months, from February to August 2017.

Data analysis
Statistical analysis covered the sociodemographic profile of the study population, established the extent of the LDTDs' knowledge about HIV and AIDS, the nature of their sexual behaviours and factors associated with buying sex and the knowledge that condoms prevent HIV transmission. Data was analysed using Stata 4.0 software (StataCorp). In the univariate analysis, sociodemographic profiles, sexual behaviours and condom use of the LDTDs were calculated. The Pearson's chi-square test was used in bivariate analysis for the relationship between knowledge that condoms prevent HIV transmission, the sociodemographic characteristics and the sexual behaviours, and the relationship of sex in exchange for money to sociodemographics and risky sexual behaviours and condom use. Multivariable logistic regression analysis was performed to identify independent risk factors for the dichotomous outcomes: sex with CSW in exchange for money in the past 12 months (yes or no); and knowledge or lack of knowledge that condoms prevent HIV transmission. The variables that were statistically significant during bivariate analysis at a p-value less than 0.05 were then selected for the logistic regression model to assess among LDTDs the variables' contribution to sex in exchange for money among the LDTDs and to the knowledge that condoms prevent HIV transmission by estimating the adjusted odds ratios (ORs) at 95% confidence intervals (CI).

Ethical considerations
The Research Ethics Committee of Sefako Makgatho Health Sciences University (SMUREC) reviewed the protocol and gave ethical clearance for the study to be conducted (SMUREC/H/286/2016 of 09/2016). The letter of clearance from the SMUREC was used to get permission from the Ministry of Health and Wellness IRB (Ref: HPDME 13/18/1 X (855). Permission was also sought from the Kazungula weighbridge management to conduct the data collection at their facility (Ref: KR 5/20 III (149). Data was collected after the participants provided informed written consent. Participation was voluntary, including the right to withdraw from the study without any preconditions. For anonymity, no identifying information was collected and the data file was password protected, with access limited to both the researchers in this study.

Characteristics of LDTDs
A sample of 399 LDTDs participated in this cross-sectional study and 327 were international drivers. The ages of the LDTDs who took part in the study ranged from 23 to 63 years, with a mean age of 39.7 years (SD 8.89). The minimum age for becoming a long-distance driver was 19 and the maximum was 45. Most of the drivers (49.8%; n = 199) joined long-distance truck driving between the ages of 26 and 29 years. The minimum period that a LDTD had spent on truck driving was one year, while the maximum was 33 years. Of the 399 participants, 9.5% (n = 24) did not have a formal education and of those who had a formal education, a high proportion of 63.4% (n = 253) had secondary education, with the remaining 30.6% (n = 146) consisting of those with tertiary and primary education at 15.0% and 15.5% respectively. Of the 262 LDTDs who had intimate sexual partners, 53.1% (n = 139) had had two or more in the past 12 months. The remaining 34.3% (n = 137) had no committed intimate sexual partnerships, but had casual sex involving CSWs. Generally, the LDTDs who did not have committed sexual partners and those who had committed sexual partners all had casual sex involving CSWs. The sociodemographics of the participants are sumarised in Table 1.

LDTDs' sexual behaviours
According to the results of this study (Table 2), more than half of the LDTDs (56.9%; n = 227) had paid for sex with commercial sex workers at some point in their career, and 27.1% (n = 108) reported to have had unprotected sex with CSWs. As many as 34.1% (n = 136) of the LDTDs reported to have paid for sex with CSWs in the past 12 months. A significantly high proportion of the LDTDs (97.2%; n = 388) knew where to get condoms when needed. Most LDTDs had scores above 90% on the following: on their ability to remember to carry a condom on their trips (93.2%; n = 372); their ability to buy condoms without feeling embarrassed (90.5%; n = 361); their confidence in their ability to put on a condom (93.5%; n = 373); their confidence in their ability to remove and dispose of a condom after sexual intercourse (93.2; n = 372); their ability to get a condom from a public place without feeling embarrassed (91.5%; n = 365). The results also indicate that 29.1% (n = 116) of LDTDs indicated that they would not feel confident to propose using a condom to a new intimate sexual partner because of the fear that the partner would think they have a sexually transmitted disease.
While 37.6% of the LDTDs (n = 150) in the study were of the view that a condom creates doubt between partners, just above half of participants (51.4%; n = 205) had an opposite view, and the rest (11.0%; n = 44) were not sure. Table 2 also shows that 75.9% (n = 303) were worried about contracting HIV, and 80.2% (n = 320) of the LDTDs indicated that there was a high possibility that they might refuse coitus if their partner refused to let them use a condom. A significantly high number of 87.0% (n = 347) indicated that there was a high chance that they might have an HIV test with their partners (Table 2).

Knowledge about HIV and AIDS
The findings of this study shows that all the LDTD who participated in this study had heard about HIV and AIDS (100%; N = 399) ( Table 3). The media was reported as the main source of information about HIV and AIDS at 65.9% (n = 263), while those who had learnt about HIV and AIDS at school stood at 62.9% (n = 251). 53.6% (n = 214) had heard about it from a clinic, while 20% (n = 81) heard about HIV and AIDS at church. 5.0% (n = 20) had heard of it through unspecified ways. The overall results of the univariate analysis shows that the LDTDs had a high level of knowledge about HIV and AIDS, with scores as high as 97.0% (n = 387) on the knowledge that HIV causes AIDS. Only 12.0% of the LDTDs believed that HIV and AIDS is caused by witchcraft, and most of the participants knew that currently there is no cure for HIV (76.7%; n = 306). A high proportion of LDTDs (97.0%; n = 387) had knowledge that having multiple partners increases the chance of contracting HIV, and that HIV transmission can be prevented by using a condom during coitus (90.2%; n = 360). All the percentage scores for knowledge about HIV and AIDS were above 75% (Table 3). Table 4 presents a bivariate analysis of the knowledge that condoms prevent HIV transmission and the participants' sociodemographic characteristics. From the 90.2% (n = 360) of the LDTDs who knew that HIV can be prevented using a condom, those with a secondary education had the highest proportion with 58.7%, followed by tertiary with 14.5%, while those with no formal education or only primary education accounted for 5.8% and 11.3% respectively. This rendered a p-value of <0.001, which is statistically significant. The analysis further reveals that age was not a predictor of knowledge that condoms prevent HIV transmission, with p-value of 0.035. Knowledge that condoms can prevent HIV transmission is associated with confidence to carry a condom on trips (p = 0.003), confidence to suggest a condom to a new partner (p < 0.001), attitude of disliking  condoms (p = 0.002) and not using a condom consistently because of unplanned sex (p = 0.005).

Relationship of sex in exchange for money to sociodemographics and risky sexual behaviours and condom use
A bivariate analysis was conducted to establish the association between paying for sex in the past 12 months and sociodemographics, sexual behaviour and condom use (Table 5). Of the 136 drivers who paid for sex during the past 12 months, 31.08% (n = 124) were international LDTDs, while the local LDTDs accounted for 3.01% (n = 12), and the association rendered a p-value of 0.001. Table 5 also shows that, though age was not a predictor in paying for sex with CSWs in the past 12 months during long-distance travels, the proportion of LDTDs paying for sex with CSWs was 34.1% (n = 136). The number of sexual partners that a LDTD had an association with buying sex in the past 12 months (p < 0.001). Paying for sex with CSWs in the past 12 months had an association to the reporting of being confident in the ability to wear a condom (p = 0.002), confidence to refuse sex with a condom (p < 0.001), and an attitude of disliking condom use (p < 0.001).

Factors associated with the exchange of sex for money with CSWs among LDTDs
Factors associated with sex in exchange for money with CSWs in the past 12 months were investigated using multivariable logistic regression analysis (Table 6). This indicated that international LDTDs were about three times more likely to have bought sex in the past 12 months than the local ones (AOR 2.9; 95% CI 1.31-6.56). Furthermore, the multivariable logistic regression model analysis revealed that having more than two sexual partners had a strong association with buying sex in the past 12 months, with a statistically significant p-value of 0.001 (AOR 3.2; 95% CI 1.65-6.33). In comparison, the LDTDs who had more than two sexual relationships had about three times the odds of buying sex in the adjusted odd ratio analysis than those with only two relationships or those who only had casual sex with no committed relationship. The LDTDs who had high confidence in refusing sex with CSWs with a condom had five times the odds of buying sex with CSWs compared to those who lacked confidence (AOR 4.9; 95% CI 2. 85-8.46).

Factors associated with the knowledge that condom use prevents HIV among the LDTDs
The results of a logistic regression analysis in Table 7 shows that the LDTDs who had confidence to carry condoms on trips had about twice the odds of knowledge that condoms can prevent HIV transmission compared to those who had less confidence (AOR 2.3; 95% CI 0.78-6.47). The analysis further revealed the level of education as a factor contributing to the knowledge that condom use prevents HIV transmission among LDTDs, with those who had no formal education and those who had primary education being less likely to have knowledge that condoms use prevents HIV transmission (AOR 0.5; 95% CI 0.65-3.24; AOR 0.1; 95% CI 0.02-0.44 respectively). Disliking condom use was also a factor contributing to less knowledge that condom use prevents HIV among the LDTDs (AOR 0.4; 95% CI 0.17-0.97).

Discussion
This study took place at Kazungula weighbridge terminal to investigate factors associated with buying sex and the knowledge that condoms prevent HIV transmission among long-distance truck drivers. A number of factors including the sociodemographics such as the level of education, the number of committed sexual partners that the LDTDs had and having truck driving as an occupation especially an international LDTD were identified in this subpopulation group.

Sociodemographics and LDTDs' sexual behaviours
This study shows that international LDTDs have a significantly higher risk of contracting HIV than the local drivers. As indicated by Delany-Moretlwe et al. (2014) on the topic of LDTDs and mobility, international LDTDs generally spend more time on the road and this exposes them to acquiring and transmitting HIV. This suggests that cross-border long-distance trucking is one of the key drivers of the HIV epidemic and aids transmission through risky sexual behaviours. The number of sexual partners that an LDTD has also exhibits a significant association with a higher proportion of paying for sex with CSWs in the past 12 months, especially among those who have more than two committed sexual partners. This finding is consistent with those of other research in various settings that determined that the number of committed sexual partners that a LDTD has is linked to high-risk HIV behaviour (Nasir et al., 2015).

Factors associated with sex in exchange for money and the knowledge that condoms prevent HIV transmisison
The results of our study indicate that being an international truck driver is a factor associated with sex in exchange for money. The higher vulnerability of international LDTDs compared with the local ones may also be explained, at least partly, by their likelihood to engage in riskier sexual behaviours, including sex in exchange for money with CSWs due to the environmental factors associated with the trucking industry, such as high mobility and easy access to CSWs. This study found that as high as 56.9% of the LDTDs had paid for sex with CSWs at some point during their travels, 34.1% had paid for sex with CSWs in the past 12 months, Ref ( Nasir et al. (2015) and Wong et al. (2007), in which 54% of the truck drivers paid for sex with CSWs. In comparison and as a matter of concern, this study shows that international LDTDs were three times more likely to buy sex with CSWs than the local ones. In spite of this important finding, there is a paucity of literature that compares the risky sexual behaviours such as sex in exchange for money with CSWs between the international and local LDTDs in Botswana and this study closes the gap. It can thus be concluded from empirical evidence from other studies that the international LDTDs' riskier sexual behaviours could be attributed to staying away from their homes for lengthy periods (Nasir et al., 2015;Atilola et al., 2010 ). This conclusion resonates well with the findings of a study conducted in South Africa in which international drivers generally spent more time on the road, getting exposed to acquiring and spreading HIV through paying for sex with CSWs (Delany- Moretlwe et al., 2014). It is evident from the results of this study that LDTDs use the desperate economic situation of the CSWs to exploit them into having unprotected sex with them in exchange for money (Ntseane, 2004). In this study, the LDTDs who indicated that they would refuse sex with CSWs with a condom had five times the odds of buying sex with CSWs compared to their counterparts who would not have unprotected sex with CSWs. Despite having such risky sexual behaviours, this sub-population has not received the attention it deserves by other researchers or the government to embark on a national behavioural surveillance of the LDTDs. Even though some efforts at various governmental and non-governmental organisations have been made to achieve behaviour change through community sensitisation and mass-media exposure to promote safe sexual practices in the general population, the adoption of tailor-made preventive behaviours targeting LDTDs is relatively low. LDTDs still engage in unprotected sexual activities with CSWs. Public health initiatives geared towards the control of HIV and AIDS among LDTDs is a challenge due to limited information on LDTDs' sexual behaviours in Botswana and this consequently leads to a failure to adequately reach out to this high-risk group as an important sub-population for intervention. Similar findings on risky sexual behaviours have been reported in other studies, e.g. a study conducted in Uganda reveals that LDTDs had unprotected sex with CSWs in exchange for money (Matovu & Ssebadduka, 2013). That study shows that some LDTDs proposed a charge of 20 000 Ugandan shillings for non-condom use and 2 000 Ugandan shillings for condom use, with the former being the choice of the CSWs due to their desperate economic problems. These results are consistent with those of our study where LDTDs who indicated that they would refuse sex with CSWs with a condom had five times the odds of buying sex with CSWs compared to those who lacked the confidence to refuse protected sex. Though the reasons for non-condom use in some studies included their unavailability and having to buy them for themselves, for this study, the reasons for non-condom use included the attitude of LDTDs of disliking condoms and inconsistency in condom use because of unplanned sex (Aniebue and Aniebue, 2009).
The number of sexual partners that the LDTDs have is also a cause for concern in the fight against HIV as it is a predictor of the exchange of money for sex with CSWs. This study reveals that the number of sexual partners that the LDTDs had was positively associated with the exchange of sex for money with CSWs in the past 12 months. The LDTDs who had more than two sexual relationships had about three times the odds of buying sex with CSWs in the adjusted odd ratio analysis than those with only two or those who only had casual sex with no committed relationship. This finding is consistent with that of a study conducted by Nasir et al. (2015) in Pakistan in which a quarter of LDTDs had at least two sexual partners, and up to 54.2% had paid for sex with CSWs.
Furthermore, the confidence to carry a condom is also positively associated with the knowledge that a condom can prevent HIV transmission. This factor is important and making condoms available at truck terminals is not enough because LDTDs could avoid using them for the same reasons that they do not carry the condoms, which could be from the lack of knowledge that condoms prevent HIV transmission. In this study, the LDTDs who had the confidence to carry condoms on trips had about twice the odds of having the knowledge that condoms can prevent HIV transmission compared to those who had less confidence. Even though this finding is important in the fight against HIV, there is a dearth of literature that supports it. But that still being the case, the available literature indicates that LDTDs place the responsibility to carry condoms primarily with CSWs, and/or do not carry condoms because they prefer to engage in unprotected sex. Due to their inadequate knowledge of the importance of condoms as an HIV preventive measure, they engage in unprotected sex and use what they view as alternative safety measures such as washing their genitals with battery water or urine after sex (Makhakhe et al., 2017). The use of condoms during sex is a key preventive strategy since rates of HIV and other sexually transmitted infections continue to escalate. To this end, the level of knowledge about condoms as a preventive measure for HIV and their use is important to this sub-population.
Literature shows that adults who have attained a higher education level have better health and longer lifespans compared to their less-educated peers (Raghupath and Raghupath, 2020). So, there is no doubt that the level of education that one has attained in life is a fundamental vehicle, not only in possessing knowledge, but also for self-preservation (Bhardwaj, 2016). Thus, the level of education that the LDTDs attain can have a bearing on how they comprehend and use the information for self-preservation against HIV. The results of our study also reveal the level of education as a factor that contributes to the knowledge that condom use prevents HIV transmission among LDTDs. The LDTDs who had no formal education and those who had primary education were less likely to have knowledge that condom use prevents HIV transmission (AOR 0.5 and AOR 0.1 respectively). It can be concluded from the findings of this study that the LDTDs' level of education has an influence on their knowledge about HIV and AIDS, with those with secondary and tertiary education displaying more knowledge than those who had no formal or primary education, and this is consistent with the findings of a study conducted in Pakistan which produced similar results (Nasir et al., 2015). This finding can assist in aligning health education activities to accommodate the LDTDs' diverse levels of education for effectiveness in raising the awareness about condoms as a preventive measure against HIV. With the findings of this study, the government can intensify mass media campaigns and other public health measures aimed at stimulating appropriate risk perception, discouraging unsafe sex practices and promoting consistent condom use by LDTDs.
Although there are no studies that were conducted in Botswana to gauge the historical progress made in regard to LDTDs' sexual behaviours, a comparison of the current study with similar ones conducted elsewhere shows positive progress in their sexual behaviours. For instance, a survey conducted by Singh and Malaviya in 1994 revealed that 78% of LDTDs reported CSW exposure and 77% occasionally engaged in unprotected sex. In 1998, a study by Mishra et al. (2012) found that 80% of truck drivers visited CSWs and 75% had either used a condom inconsistently, or had never used a condom. The current study shows that 56.9% of the LDTDs visited CSWs at some point during their long-distance travels with a truck, and as little as 27.1% reported to have had sex with CSWs without a condom. That still being the case, more effort needs to be geared towards changing LDTDs' behaviours to promote the use of condoms, particularly with CSWs and casual partners.

Limitations
The study was cross-sectional by nature and thus could not establish changes in the LDTDs' behaviours over time, nor the reasons associated with their behaviours. Another limitation is participant reliability due to the nature of the topic being sensitive and controversial. Participants may be untruthful for many reasons, such as giving "moral" or "correct" answers that reflect what is socially acceptable, especially as it was a researcher-administered questionnaire. During consent negotiation, the LDTDs were informed that they could only participate in the study once, but it is possibile that some participants might have participated more than once during the study period.

Conclusion
This study documents factors associated with buying sex and the knowledge that condoms prevent HIV transmission among long-distance truck drivers at Kazungula weighbridge terminal in the Chobe district of Botswana. LDTDs remain an at-risk population by exchanging sex for money and engaging in unprotected sex with CSWs on their long-distance travels. Being in the truck driving occupation, refusing sex with CSWs with a condom and the number of sexual partners that a LDTD has are positively associated with sex in exchange for money. Factors associated with the knowledge that condoms prevent HIV transmission among LDTDs include the confidence to carry condoms on trips. The results of our study further reveal that the level of education is a factor contributing to the knowledge that condom use prevents HIV transmission among LDTDs. It can therefore be concluded with empirical evidence from our study that LDTDs remain an at-risk group for acquiring and propagating HIV, and the risk is more with international LDTDs compared to those travelling locally. Other studies ascribe the risky behaviour to being away from home for protracted periods. The findings of our study demonstrate the need for HIV and AIDS control programmes to strengthen HIV-prevention interventions among LDTDs. Making condoms available at truck stops and the promotion of safe sex practices such as consistent condom use requires policy decisions from those who wish to promote safe sex among LDTDs. Furthermore, there is a need to develop policies that embrace critical HIV-prevention strategies with high success rate for key populations, e.g. policies that support the provision of post-exposure prophylaxis (PEP) and pre-exposure prophylaxis (PrEP) to vulnerable populations such as LDTDs, including the non-citizen international truckers.