Egypt is thought to have a low-level HIV/AIDS epidemic with 9200 adults and children estimated to be living with HIV/AIDS  out of a total population of greater than 75 million. Over the last two decades, the country has witnessed a slow but steady increase in the number of HIV infections . HIV infection appears to be spreading more quickly within high-risk populations, such as men who have sex with men (MSM), sex workers, and injection drug users (IDU); however, accurate data from these populations are scant [2,3]. The country may, therefore, be heading towards a concentrated epidemic status with vulnerable groups bearing a disproportionate burden of infection.
Street children are a highly vulnerable population for many adverse health and social outcomes and have also been prioritized internationally for HIV prevention [4,5]. Although there have been no accurate estimates of street children in Egypt, their numbers may be staggeringly high in urban areas. According to a rapid assessment, their numbers may be as high as one million in Cairo and Alexandria . A few reports from Egypt have described the sociodemographic characteristics of street children, their group dynamics, and their exposure to violence, drugs, and crime in convenience samples [6–8]. Although their safety and daily survival needs are manifestly pressing, little is known about their HIV/AIDS related risk behaviors. Moreover, the available studies on street children in Egypt have been based on small, nonprobability samples. There are strong needs for accurate, systematically collected data on the sexual and drug use behaviors of street children. Filling these needs faces many challenges, including ethical concerns in conducting research with minors where a legal guardian is usually not accessible, lack of a complete sampling frame, and access to this hard-to-reach population.
In this context, we undertook a probability-based survey of street children in the two most populous cities of Egypt, Cairo (the capital and largest city including its greater metropolitan area) and Alexandria. Because street children are visible in a wide range of public locations, and because the environments in which street children are found affect their risk behavior, we chose time–location sampling (TLS) as an appropriate method to locate and recruit participants in our survey .
The target population of our survey was street children, boys and girls, aged 12–17 years living in Greater Cairo and Alexandria, Egypt in the summer of 2007. For the purpose of our study, street children were defined as any child living on the street most of the time regardless of whether they still maintain some contact with their families or not. We excluded street children who were using the boarding facilities of nongovernmental organizations (NGOs), as they were deemed systematically different from other street children not receiving such services.
A time–location sampling (TLS) methodology, which was developed earlier to study homeless persons [10,11], was employed to construct a probability-based cross-sectional survey of street children present in the venues where the target population visibly congregated. TLS methods have been used around the world to sample hidden and hard-to-reach populations and have been described in detail elsewhere . In brief, the method begins with a formative phase to identify the locations where street children congregate in numbers sufficient for efficient recruitment. Initial locations were identified from governmental sources, NGOs providing services to street children, and interviews with street children themselves on possible places to find street children in Greater Cairo and Alexandria. Following these consultations, a comprehensive list of locations where street children could possibly be present was compiled for each city. Next, within the identified locations, field team enumerators verified the locations, mapped the venues, and counted the numbers of street children present over three shifts across the day. Social workers from local NGOs providing services to street children accompanied the field team to help them identify street children and avoid confusing them with other children present on the street. In Greater Cairo, 88 locations were identified and observed over three shifts making a total of 264 time-location units. In Alexandria, 27 locations were identified and observed making a total of 81 time–location units. Locations were mostly crowded public places, such as major transportation centers like bus and metro stations, market places, and parks. Over the three shifts in all locations, 3357 street children were counted in Greater Cairo, and 889 in Alexandria. From the list of all time-location units, a random sample of time–location units was selected to be self-weighted according to their attendance with an approximate balance of the sample between the two cities. At the randomly sampled time-location units, field team interviewers consecutively approached children, assessed their eligibility, and invited them to participate in the survey. Interviewers were matched to the sex of the child. The numbers of street children participating were stratified by city, sex, and age to track the representation of the sample against the enumeration data.
Questionnaire development included a 6-month period of formative assessment, participant observation, and in-depth interviews with street children to help identify the topics to be included, how sensitive they were, terminologies, and possible responses. The questionnaire was developed in English then translated into Arabic. The Arabic version was substantially modified based on findings from the formative phase and in-depth interviews. The questionnaires were then pretested in the target population, refined in Arabic, and back translated to English to assess reliability prior to training the interviewer/data collectors.
The questionnaire included sections on socio-demographic characteristics, events of life on the street including violence and physical abuse, contact with family, knowledge of HIV/AIDS including its means of transmission and prevention, the use of drugs and alcohol, and risky sexual behaviors. The development stage and ethical considerations determined that more detailed sexual behavior questions were to be asked only of the children aged 15–17 years and not those aged 12–14 years. Sexual behavior questions included age of sexual debut, number of partners, same-sex partners, condom use, commercial sex, and forced sex. To ensure confidentiality, the questionnaire did not include any identifying information.
Data were entered on Epi-Info version 3.3.2 and SPSS version 12 was used for analysis. The primary aim of the study was to gauge the prevalence of multiple risky exposures and behaviors in the population of street children in Egypt's two largest metropolitan areas. In keeping with the study design, data were analyzed by strata of city, sex, and age and are self-weighted to the population enumerated at the locations within each city.
Assent of the study participants was sought after ensuring comprehension of the study's nature, procedures, risks, and benefits in the presence of a representative of one of the NGOs. In accordance with Egyptian standards and with the advocacy function of the NGOs serving the street children in their areas of operation, the social workers of the NGOs acted on behalf of the children in the absence of a legal guardian. Verbal assent was obtained from each child prior to participation in the survey and the NGO representative signed a witnessed consent form. Children who agreed to be interviewed were not compensated and were referred to social services available in the area for any needs arising at the time of the interview.
A total of 857 street children participated in the survey, including 463 in Greater Cairo and 394 in Alexandria. Our recruitment reflected the field observations at the locations and achieved the diversity in terms of sex and age groups; the number of boys enrolled was several-fold higher than girls in both cities, the two age groups were fairly balanced by city, and recruitment was higher from 3 pm to 10 pm in Greater Cairo and from 10 pm to 2 am in Alexandria.
Overall, 60% of the children attended school in the past, 35% never attended school, whereas 5% were still attending school while on the street (Table 1). The majority (72%) had been living on the street for a year or more; however, all girls interviewed in Alexandria were on the streets for less than 1 year. Although 44% of all street children reported maintaining contact with their families, percentages were higher for street children in Greater Cairo compared to their sex and age group counterparts in Alexandria.
Table 2 presents street children's alcohol and drug use, exposure to violence and abuse, and their knowledge of HIV/AIDS. Over one-third of the children (35%) reported current alcohol consumption, with use being more common among boys than among girls and more common among older than younger children. Ever use of drugs was reported by nearly two-thirds (62%) of all street-children, and use of drugs at the time of data collection was reported by just over half (51%). Higher proportions of male street children reported ever or currently consuming drugs than females in both age groups and in both cities, but drug use was notably higher in all sex-age categories in Alexandria compared with Greater Cairo. Current injection of drugs was reported by 3% of the study population.
The vast majority of children (93%) faced some form of violence or abuse on the street. The most typical forms of abuse were harassment by the police (63%) and by older street children (51%). Physical abuse and violence were reported by 45% of children and sexual abuse by 12%. Harassment and physical abuse tended to be more prevalent among boys, whereas sexual abuse was many fold more common among girls. In fact, over half (53%) of female street children aged 15–17 years in Greater Cairo and 90% of those in Alexandria had a history of sexual abuse.
Ever hearing about HIV/AIDS was high but not universal (79%) in this population of Egyptian street children. Regarding specific knowledge of means of transmission of HIV/AIDS, 59% of respondents mentioned heterosexual relations, 21% mentioned homosexual relations, and 24% mentioned coming in contact with infected blood (e.g., blood transfusion, unclean needles). However, 16% believed that casual physical contact (e.g. shaking hands) could transmit the infection, 2% believed a mosquito bite could transmit the infection, and 14% did not know how HIV is transmitted. One-third (33%) thought a healthy looking person could not have AIDS. When identifying means of preventing HIV, 49% mentioned abstaining from sex, 17% mentioned avoiding skin piercing instruments, 12% mentioned using condoms, and 5% mentioned being faithful to one uninfected partner; 16% of respondents did not know how to avoid contracting the infection.
Among the 15–17-year-olds to whom more extensive sexual behavior questions were addressed, two-thirds (67%) reported having ever engaged in sex with the opposite sex (Table 3). A higher proportion of girls than boys in both Greater Cairo (71 versus 61%, respectively) and in Alexandria (90 and 70%, respectively) reported ever having had sex with someone from the opposite sex. The mean age at first intercourse with someone of the opposite sex was 14 years. Multiple sex partners in the preceding 12 months was reported by more than half (54%) of these older children. Also, more than half (52%) reported never using a condom, with only 20% reporting having used them consistently. Unprotected sex was more common in Greater Cairo (73%) compared with Alexandria (31%) and street girls were more likely to engage in unprotected sex (73%) as compared with street boys (48%).
One-fifth (20%) of boys aged 15–17-year-old in Greater Cairo and more than one-third (37%) in Alexandria reported having sex with another male. The mean age of first intercourse with other males was 13 years. Of boys who had ever engaged in sex with another male, more than half (55% in Greater Cairo, 52% in Alexandria) had sex with more than one partner in the 12 months preceding the survey. Furthermore, condom use with male–male sex was very low, with 90% of boys in Greater Cairo and 70% in Alexandria reporting never using condoms when having sex with other men in the last year. Few girls reported having sex with other females in Greater Cairo; none reported this behavior in Alexandria.
Overall, nearly a fifth (19%) of the 15–17 year old age group reported having sold sex. The proportion was higher for girls than boys in Greater Cairo (26 versus 12%, respectively) as well as in Alexandria (58 versus 21%, respectively). For those who had ever sold sex, the majority (59%) received money, followed by drugs (42%), services (19%), and gifts (13%). Clients buying sex were reported to be other street children (46%), people living or working in the neighborhood (12%), taxi and minibus drivers (11%), shop owners (4%), policemen (3%), in addition to other individuals (25%).
The majority of girls, 88% in Alexandria and 54% in Greater Cairo, experienced being forced to have sex by a male. Forced sex was reported to be perpetrated mainly by street boys (61%), people living or working in the neighborhood (8%), taxi and minibus drivers (8%), shop owners (8%), policemen (2%), in addition to others (12%). For their part, one in five street boys (19% in Alexandria and 22% in Greater Cairo) acknowledged having forced a girl into having sex. Eight percent of boys in Greater Cairo and 6% in Alexandria reported being forced to have sex by another male.
The TLS method employed in our study was successful in systematically recruiting a large number of street children in two cities in Egypt. The sample comprised the sex and age composition of the street children enumerated in the identified locations of each city. The randomized selection from the larger universe of time-location units enumerated confers a probability basis to the sample and strengthens the rigor over a simple convenience sample of children at venues. Moreover, we excluded those children who were already boarding at NGOs within the target areas. We, therefore, conclude that the approach reasonably approximated a representative sample of children living on the streets and serve as a basis for inference on the risks and needs of the larger population of street children in urban Egypt. To our knowledge, this is the first probability sample of street children in the country and its success speaks to its feasibility for expansion and for on-going surveillance in this population.
The risks and needs of these street children are great. Exposures to severe harm were not only prevalent, but the norm for these children. The vast majority, for example, experienced some form of violence or abuse, most often from police and other, typically older, street children. The majority also reported having used drugs, and among the older teens, having sex. Among those having sex, most never used a condom at all and most had multiple sex partners. Most disheartening was the level of sexual abuse experienced by young girls. Over half had been sexually abused; among older adolescent girls in Alexandria, sexual abuse was nearly universal. Our findings on sexual risk behavior and abuse are frequently encountered in street children in other countries. In a study assessing the risk of HIV infection among urban street boys in Tanzania, it was found that almost all street boys were involved in a sexual network in which same sex and heterosexual behavior occurs . In a study among South African youth, it was found that street children between the ages of 11 and 17 years engaged in commercial sex, many had been raped and most reported being sexually active with girlfriends who themselves engaged in transactional sex .
Moreover, our study found that children living on the streets in Egypt overlap the populations severely affected by HIV the world over, namely men who have sex with men (MSM), male and female sex workers (MSW, FSW), and injection drug users (IDU). One in five adolescent boys had sex with other males by age 15–17 years. More than one in four girls and one in eight boys had sold sex to males. We also documented injection drug use in this young population. Although IDU were uncommon, all forms of drug use are instrumental in the transmission of HIV/AIDS. Individuals engage in sex when on drugs , and individuals with a history of drug use are more prone to engage in risky sexual behavior . Also in our study, we found street children provide sex in return for drugs, and likely in return for money used to purchase drugs. Taken together, the connections of street children to other at-risk populations predicts potential rapid spread of HIV, if Egypt's HIV epidemic becomes increasingly concentrated and prevalence elevated among MSM, IDU, and FSW .
Although we believe our data are greatly improved over that obtained through nonprobability samples, we recognize limitations. First, our mapping and enumeration cannot be taken as the entire population of street children in the two cities. By design for reasons of efficiency, the time–location sampling frame comprises numerous locations where the target population is concentrated and visible and omits locations where children are more diffuse or well hidden. On the contrary, the mapping and TLS approaches produced a snapshot of the mobile population of children who circulate in these more accessible locations. Moreover, there currently is no gold standard or complete census of street children. We recognize other potential limitations in the depth of measures used and their appropriate adaptation to the Egyptian setting and Arabic language. Finally, our study did not include biological measures of HIV and other sexually transmitted infections.
Despite limitations, our data tell a compelling story of the need for multiple services for street children in Egypt, including alcohol and substance use treatment, HIV and STI screening and treatment, and programs to train for livelihoods. Structural interventions are also needed, such as the creation of more shelters to limit the time spent on the streets, reunification with families and family counseling, and sensitization of police to mitigate abuse and mistrust of social services. We recognize these challenges are enormous in a resource-limited setting, but the process begins with collecting credible data to use for advocacy.
The study upon which this paper is based was conducted by the Population Council West Asia and North Africa Regional Office, with financing from UNICEF. Both authors were affiliated with the Population Council at the time of conducting the study.
This study was funded by UNICEF.
Conflicts of Interest: None.
1. UNAIDS/WHO Epidemiological Fact Sheet on HIV and AIDS; September 2008; Egypt; 2008.
2. Egypt Ministry of Health and Population National AIDS Program. HIV/AIDS biological and behavioral surveillance survey: Summary report. Cairo, Egypt, 2006.
3. Abu-Raddad LJ, Hilma N, Mumtaz G, Benekirane M, Akala FA, Reidner G, et al
. Epidemiology of HIV infection in the Middle East and North Africa. AIDS
2010. [Epub ahead of print]
4. Marshall BD, Kerr T, Shoveller JA, Montaner JS, Wood E. Structural factors associated with an increased risk of HIV and sexually transmitted infection transmission among street-involved youth. BMC Public Health 2009; 9:7.
5. Anarfi JK. Vulnerability to sexually transmitted disease: street children
in Accra. Health Transit rev 1997; 7(Suppl):281–306.
6. Rapid Situation Assessment of Street Children
in Cairo and Alexandria. 2001. World Food Program, United Nation's Children Fund, and United Nations Office for Drug Control and Crime Prevention.
7. Salem EM and Abd El-Latif. 2002. Sociodemographic characteristics of street children
in Alexandria. East Mediterr Health J 2002; 8
8. Bibars Iman. Street Children
in Egypt: from the home to the street to inappropriate corrective institutions. Environment and Urbanization 1998; 10
9. Magnani R, Sabin K, Saidel T, Heckathorn D. Review of sampling hard-to-reach and hidden populations for HIV surveillance. AIDS 2005; 19(Suppl 2):S67–S72.
10. Burnam MA, Koegel P. Methodology for obtaining a representative sample of homeless persons: the Los Angeles Skid Row Study. Eval Rev 1988; 12:117–152.
11. Robertson MJ, Westerfelt A, and Irving P. 1991. Research note: the impact of sampling strategy on estimated prevalence of major mental disorder among homeless adults in Alameda County, California. Presented at American Public Health Association meeting; November 12, Atlanta, Georgia; 1991.
12. Lockhart C. Kunyenga, ‘real sex,’ and survival: assessing the risk of HIV infection among urban street boys in tanzania. Med Anthropol Q 2002; 16:294–311.
13. Swart-Kruger J, Richter LM. AIDS-related knowledge, attitudes and behaviour among South African street youth: reflections on power, sexuality and the autonomous self. Soc Sci Med 1997; 45:957–966.
14. Parry CD, Carney T, Petersen P, Dewing S, Needle R. HIV-risk behavior among injecting or noninjecting drug users in Cape Town, Pretoria, and Durban, South Africa. Subst Use Misuse 2009; 44:886–904.
15. Kayembe PK, Mapatano MA, Fatuma AB, Nyandwe JK, Mayala GM, Kokolomami JI, Kibungu JP. Knowledge of HIV, sexual behavior and correlates of risky sex among street children
in Kinshasa, Democratic Republic of Congo. East Afr J Public Health 2008; 5:186–192.