The phenomenon of homeless/street children is global, with the streets throughout the world are home to millions of children [1]. Circumstances on the street render children and youth vulnerable to various kinds of psychological problems and health risks [2]. Their lifestyles are characterized by a high rate of sex trade [3–5] and substance abuse and misuse [6, 7] which places them at high risk for STI, including HIV infection [8, 9].

The type and prevalence rate of substances used varies remarkably according to context and geographical location. In a comprehensive review, [10] reported that street youth in high income countries usually use injecting drugs and other substances that are not commonly used by homeless adolescents from low and middle resourced countries (LMIC). A study among street children in Egypt indicated that 62% of the participants reported substance use, with the highest substance use being alcohol consumption (35%) with only 3% indicating injection drug use [4]. Similarly, [11] found that eighty-two percent of street children in the Democratic Republic of Congo (DRC) used marijuana, 63.5% used alcohol and 3.8% used cocaine [11]. A similar pattern of substance use was also reported among street youth in Kenya, where lifetime and current substance use was 74% and 83% respectively [12].

Substance use among homeless youth has been found to be influenced by factors such as gender, age, duration of homelessness and social networks (e.g. peer influence). Research has indicated that more males than females use alcohol, marijuana, cocaine and inhalants (i.e. glue) [6, 9, 11, 13, 14]. Further, evidence shows that although differences in the choice of substances exist, the prevalence of substance use disorders is similar among males and females [15]. Another factor that influences substance use is age [4, 9]. In a study of drug use among young and older homeless youth, [9] found that younger participants (less than 21 years) where more likely to be engaged in excessive drinking of alcoholic beverages, while older participants (21 years and more) were more likely to be involved in heroine, crack and injecting drug use. The number of years lived on the street influences substance use [7, 12, 16]. Similarly, [17] reported from their cross-national study that the length of homelessness was associated with higher rates of substance misuse and psychiatric disorders.

Substance use also increases the likelihood that individuals will engage in risky sexual behaviours, for example, non-condom use and multiple sexual partner patterns which put them at risk of STIs and HIV/AIDS [4, 10, 11]. A study conducted among 280 street youth in Ethiopia indicated that 77% of the sample, who indicated they were sexually active at the time of the study, did not use condoms during their last sexual activity [18]. The sexual behaviour trends of street youth show that not only are street youth not protected sexually, but that they also have multiple sexual partners [4, 8, 9, 11, 19, 20]. A study conducted by [8] on street youth in Ethiopia revealed that 62% of their sample were sexually active and 97% of the sexually active had multiple sexual partners. The same study further showed that 80.5% of the street youth used condoms inconsistently, and the length of stay on the street was also associated with inconsistent condom use [8]. Reasons for non-condom use among street youth include not been able to buy condoms, drug use and negligence as well as belief that it decreases pleasure, and difficulty in using it in the “heat of the moment” [5, 8, 11].

Closely related to having multiple sexual partners and inconsistent condom use, is transactional sex, especially for female street youth (which includes sex for money and sometime for protection on the street) [4, 20–22]. Among homeless children in Egypt, [4] found that 25% of the girls had sold sex to males, while [21] reported in a study in Tanzania that those aged between the ages of 11–17 years engaged in commercial sex.

Evidence from Ghana indicates that homeless youth is growing in cities such as Accra. Headcounts of street children ranged from 35,000 in 2009 [23], 2010) to 90,000 in 2013 [24]. Despite this increasing number of homeless children and youth, the prevalence of substance use and abuse, and the relationship between substance use and sexual risk behaviours has not been examined in Ghana. There is also sparse literature on age and gender differences regarding substance use among homeless children and youth in Ghana. Previous studies conducted among homeless children and young adults in Ghana, have focused primarily on economic, social and cultural causes of homelessness, their engagement in risky sexual behaviours and the prevalence of STI including HIV and AIDS [5, 20, 25–28].

In the context of this gap in knowledge, the aims of this study aims to determine i) the prevalence of substance use and misuse ii) age and sex differences in substance use and iii) the relationship between substance use and risky sexual behaviours among a sample of street connected children and youth in the Central Business District (CBD) of Accra, Ghana.

The prevalence of substance use in this study is comparable to figures from other studies [4, 12] but lower than the rates found in other African countries [10, 34, 35]. In the past one month the sample reported 12.0% and 16.2% daily use of alcohol and marijuana respectively. Even though, these rates are relatively low, they suggest that these drugs are easily accessible to this population. Comparing this prevalence to studies in the general population, [36] reported 12.7% alcohol use and 5.2% among school going adolescents in Ghana. The high prevalence of marijuana use could be attributed to street socializing by peers on the street [37], and as coping mechanisms to reduce emotional problems [15]. The fact that approximately one in six of the youth reporting using marijuana shows the extent of risk youth face on the street of Accra.

The study also revealed that over 54% of the homeless youth reported having exchanged sex for food, money and even a place to sleep, with females more likely to engage in such behaviours than males. The high prevalence of females engaging in survival sex is consistent with studies conducted in developed countries [3, 22, 38] and those from sub-Saharan Africa including Ghana [5, 20, 25, 39, 40]. These findings also showed the vulnerability of females to exchange sex for money and or protection in a male dominated street life. A recent study in Ghana, described that insecurity on the street and the avoidance of forced sex, compelled females to seek protection from older boys by forming sexual relationships with them [5]. Power dynamics and females’ inability to generate income through other forms of work, has also seen as contributory factors that “push” female street youth to engage in survival sex [3]. It is however, important to note that an equally high number of males in this study also reported involvement in survival sex. Since over 54% of the sample had engaged in survival sex, implies that at least one out of every 2 homeless youth living on the street of Accra, are engaging in some form of prostitution.

The findings of the study showed that more female had smoked cigarette, used alcohol and marijuana than male in the past month. This finding contradicts previous studies among homeless youth where males generally reported higher substance use more than females [6, 9, 11, 13]. The sex difference may possibly be attributed to the fact that female adolescents with sexual abuse histories do abuse drugs on the street [37]. In this study, more females reported sexual abuse as a reason for being homeless than male participants. The extent of their engagement in survival sex is possibly linked to supporting their substance use behaviours. While early adolescents were more likely to abuse alcohol than the other age groups, marijuana use increases with age. Early adolescence as a stage of development is fueled with experimentation [41], and is likely to explain their vulnerability to experiment with alcohol which is generally accessible to both minors and adults in Ghana [42]. The increase of marijuana use with age is not surprising as it is less accessible as it is against the law and homeless youth have to develop various mechanisms to access it. This finding is also suggestive of the fact that the longer youth live on the street, the more likely they are to engage in substance use including marijuana [43], as the experiences of homelessness increases the vulnerability to substance use on the street [7, 12]. Social estrangement, which occurs when street youth who move to the street, become deep-rooted in street life with time, may explain youth susceptibility to engage in high risk behaviours including substance use [44]. Substance use with increased age may be associated with increased risk for the involvement in risky sexual behaviours.

A striking result in the study was that substance use among street youth particularly alcohol use, marijuana and cigarette smoking were independently associated with having unprotected sex, multiple sexual partners and engagement in survival. This aggregation of health damaging behaviours is consistent with the theory of problem behaviour [45]. The clustering of health risk behaviours among homeless youth is more enhanced as they are predisposed to several risk factors heightening multiple vulnerabilities [4, 11, 12]. Among street children in the Democratic Republic of Congo (DRC), a constellation of health compromising behaviours were reported where a history of drug use were linked to the engagement in sexual risk behaviour [11]. Also among Egyptian, Malawian and Sudanese street children, similar evidence of co-occurrence of health compromising behaviours has been identified [4, 39, 40]. Substance use is viewed as a risk factor to other health damaging behaviours, particularly risky sexual behaviour [46]. In the present study, it is unclear whether homeless youth engaged in risky sexual behaviour because they were under the influence of substance use or whether they engage in risky sexual behaviours as a means to support substance use or some “third factor” such as the social environment (e.g. peer influence and social networks) that facilitated risky sex or both behaviours. It is however known that substance use interfere with rational behaviours [8], which could make individuals more vulnerable to unsafe behaviours.

Health compromising behaviours among homeless youth in this study shows their vulnerability for HIV infection as a result of their high levels of sexual risk behaviours i.e. multiple sexual partners, engagement in survival sex and inconsistent condom use in contexts of considerably high alcohol and drug use. This linkage between substance use vulnerability to acquiring HIV and AIDS is well established [20, 25, 47].

The findings from this study have implications for future intervention efforts and research. First, the results suggest that over 10 percent of the homeless youth abuse substance such as alcohol and marijuana, and concurrently engaged in risky sexual behaviours. It is therefore important for agencies and NGOs who provide psychosocial services to homeless youth to have access to clinical psychologists and other mental health professionals who can assess, support, and treat the varied health problems such as substance use and risky sexual behaviours that have been shown to be associated with homelessness. This could be done in collaboration with the Government of Ghana through the Ministry of Gender, Children and Social Protection, that has oversight responsibilities among others to promote the rights of women and children. Second, the findings suggest a constellation of health risk behaviours and vulnerability among homeless youth, where substance use was associated with unprotected sex, having multiple sexual partners and commercialization of sex. This thus put homeless youth at risk for several physical and mental health problems such as HIV and STIs infection and drug dependency. It is therefore important for homeless youth service providers to address the effect of multiple risk behaviours through health promotion interventions that to address both substance use and risky sexual behaviours. Third, our findings reveal that substance use increases with age, which means that as homeless youth become older, they would be more likely to engage in substance abuse and abuse. It is therefore important for NGOs and international organizations that provide psycho-social services for homeless youth to target early entrants to the street before they become familiarized with the street sub-culture which is characterized with violence, substance abuse and sexual risky behaviours. This could be done by initiating harm reduction programmes that would address the needs of this population by ensuring that they use condoms correctly and consistently when engaging in sexual intercourse, especially for those who have multiple sexual partners. This initiative could use previously successful street youth to act as peer educators, as peer-to-peer contact has proven to be one of the effective approach to reaching most-at-risk young people [48]. Fourth, cognizance of gender dimensions is important in developing interventions for homeless population. A client-centered approach could be used through the development of trust when dealing with psycho-social problems, especially in respect of females who might have been sexually abused or involved in some form of survival sex. On the contrary, since males are known to be perpetrators of sexual violence on the street [33], programmes must focus on reducing or limiting such abuses. Finally, the multiplicity of problems experienced by homeless youth, demands a responsive mental health system. However, in Ghana the mental health system is not well developed, with accessibility for vulnerable groups such as homeless youth limited. Community-based organizations and NGOs who have taken on such responsibilities of providing limited psycho-social services have workers who might not receive the adequate training in a variety of psychotherapeutic approaches.

It should be noted that there as some limitations to this study. First, because it is impossible to randomly sample homeless youth from the population of homeless adolescents in Accra, Ghana, these data are not based on a probability sample and therefore results cannot be generalize to the whole population of homeless youth in Accra. Notwithstanding this, collecting data from the different places within Central Business District of Accra increased the reliability of the findings. Further research is however needed to determine whether homeless youth in other parts of Accra and or other regions of Ghana hold similar perspectives as found in this study. Secondly, results of the study should be interpreted cautiously as the study cross-sectional nature of the data limits causal inference. Third, the measures are based on homeless youth self-reports, which might be confounded by systematic bias and social desirability. However, if there was any reporting bias, the direction was under reporting rather than over reporting of the use substance use and sexual behaviour especially for male participants. Despite these shortfalls, on a whole, however, this study fills in an important gap in literature and provides useful piece of information for harm reduction interventions among homeless youth.

This study examined the relationship between substance use and sexual risky behaviours among homeless youth. Substance use, especially alcohol and marijuana use among the sample in this study was relatively high and was used as a means of fitting in and coping with emotional distress. Substance use was found to be risk factor for risky sexual behaviours among Ghanaian homeless youth, placing them at risk for contracting STIs including HIV, with its accompanying health and social implications as well as bringing them into conflict with the laws. Harm reduction programmes targeting these youth must take in consideration the likelihood of substance use especially marijuana and alcohol use among the sexually experienced youth. Further studies are needed within the African context using stronger statistical models to explore the multiple pathways between substance use and risky sexual behaviours among homeless populations.