Sexuality is an area that has long attracted sociologists and anthropologists as well as demographers because of the identified relationships between sexual behaviors and certain reproductive health problems. In Sub-Saharan Africa, the first studies on sexual behavior were conducted to determine the causes of widespread infertility and sterility, especially in Central Africa. Most concluded that these problems resulted from decreased control of elders over young people, economic hardship and the presence of societies with permissive sexual customs.1 These factors were considered to have contributed to the spread of sexually transmitted diseases (STDs) that cause infertility (gonorrhea, chlamydia and mycoplasma) or disrupt pregnancy (syphilis). After the beginning of the AIDS epidemic in the early 1980s, some STDs were found to increase the probability of sexual transmission of HIV.2

Because the majority of African women do not use modern contraceptive methods, those who are sexually active are at high risk of conception. Thus, in this region of the world, there is a high risk of maternal and infant mortality because of early pregnancy, induced abortion (which, in the many countries where it is illegal, is often carried out in unsafe circumstances), short interpregnancy intervals, and high and prolonged fertility.3

Despite these risks, little research has been done in Sub-Saharan Africa to determine the factors affecting sexual behavior, especially in young people. The few studies that have been done have been essentially quantitative and descriptive.

Among health problems related to sexuality, AIDS is currently the main concern of the public authorities in Africa; its prevalence continues to increase in most of the countries in the region, with disastrous psychosocial, demographic and socioeconomic repercussions.

In Cameroon, as everywhere else in Africa, sensitization and information programs on AIDS have been implemented along with efforts to promote the use of condoms. However, these programs have had limited results. In fact, in certain social strata, an appreciable increase in the level of awareness about AIDS and prevention methods has not translated into adoption of healthy sexual behaviors.4

The prevalence of AIDS and its rate of increase are higher in Northwest Province than in the other provinces of Cameroon.5 Bamenda, the capital of the province, had a population estimated at about 111,000 in the 1987 census; a large proportion of its inhabitants are younger than 25. The city is home to several ethnic groups. Some, like the Makon-Banyague, have permissive sexual customs. Others, including the Metta, the Bali-Ngemba, the Tikari-Nsoh and the Bamiléké, impose strict control of sexual activity.

This article examines socioeconomic and demographic characteristics in a sample of 671 young people in Bamenda to identify the factors associated with risky sexual activity. It focuses on young people because they engage in more frequent sex and have a long period of sexual activity before them; thus, the future prevalence of AIDS depends heavily on the direction that prevention programs take with respect to young people.

Methodology

Definition of Concepts

For the purposes of this study, adolescence is considered to encompass approximately ages 12-25, a period of transition between childhood and adulthood during which the individual assumes his position as an active member of society. It is divided into four overlapping stages: sexual awakening (12-15), first sexual intercourse (14-17), gender role definition (16-19) and social role definition (18-25).

The first two stages correspond to biological and physiological adolescence and the latter two to psychological and social adolescence. In this study, first sexual intercourse occurring during physiological adolescence (i.e., before age 16) is considered precocious.

The sexual risk behaviors that we examine are those that expose young people to a greater risk of infection by the AIDS virus. They include having multiple concurrent sex partners, having had sex with a casual partner and failing to use a condom during sexual intercourse.

Hypotheses

This study is based on the following hypotheses:

•Adolescents usually do not obtain information about sex from their parents or guardians.

•For the most part, they are informed about HIV, its main means of transmission and its methods of prevention.

•Large proportions have precocious sexual intercourse and engage in sexual behavior that may expose them to HIV infection.

•Sociocultural and socioeconomic factors influence their sexual behavior.

Data

To test these hypotheses, the study used data from the Survey on the Sexual Behavior of the Young People of Bamenda, which was conducted in August 1995. In the first phase of this two-phase cluster survey, five neighborhoods were randomly selected in proportion to the size of their populations. In the second phase, interviewers followed a preset itinerary from household to household until they reached the sample size of young people for a given neighborhood.

The survey collected information on the participants' social, economic and demographic characteristics, on the characteristics of the households in which they lived and on their family environment. It also asked questions about the young people's sex education by their parents or guardians, their awareness of AIDS, their knowledge about methods of preventing infection with HIV, their sources of information about AIDS, the measures they had taken to prevent the disease and their sexual behaviors.

Analysis

Simple cross-tabulation was used to determine the proportion of young people who had obtained information about sex and sexuality from their parents or guardians, the level of awareness about AIDS and the use of methods of AIDS prevention. Multivariate classification analysis was then used to examine the characteristics of young people who had had precocious first sexual intercourse, and logistic regression was used to look at the characteristics of those who had engaged in risky sexual behaviors.

Results

Sample Characteristics

Males were more heavily represented than females in the sample (59% vs. 41%). Nearly all (96%) of the young people were unmarried, so those who were married were excluded from further analysis, leaving 646 in the final sample. At the time of the survey, 31% of the unmarrried young people interviewed had had primary schooling, 43% had reached middle school, 16% had reached high school and 5% had continued their education beyond high school; only 5% had had no schooling (Table 1).

Sex Education

Overall, approximately three of five young people said they had discussed one or more sexual topics with their schoolmates or friends. On the other hand, fewer than two of five had discussed such issues in a family setting. Most of those who said that they had had such discussions with family members had had them either with their elders or with their mother. On average, the young people in our sample thought that the ideal age for starting sex education was 13.8 years for girls and 15.0 years for boys.

Awareness of AIDS

Overall, 97% of the young people had heard of AIDS (Table 2). About half had obtained information about AIDS from radio (54%) or television (50%), while smaller proportions had heard about the syndrome from friends or school (17% each), newspapers (14%) and health centers or family (9% each, not shown). The level of AIDS awareness varied very little by gender.

Sexual intercourse was identified by most of the young people (94%) as a way in which HIV is transmitted, followed by infected blood (51%), shared needles (39%) and maternal transmission (14%). Prostitutes (66%) and individuals who are unfaithful to their sexual partners (40%) were cited as the groups at greatest risk of contracting the virus. The most commonly mentioned methods of preventing AIDS were use of condoms (89%), abstinence (24%) and fidelity (15%). Again, there were few differences by gender in these responses.

The main measures that these young people had taken to prevent AIDS (not shown) were abstaining from sex (64%), being faithful to one partner (55%), using condoms with casual partners (19%) and limiting the number of partners (13%). Females were more likely than males to have adopted abstinence (71% vs. 60%) and fidelity (61% vs. 45%), while males were more likely to have adopted condom use (22% vs. 15%).

First Sexual Intercourse

For the most part, males and females had begun having sexual intercourse before they were 18 (Table 3). However, three males out of 10 (versus two females out of 10) had already had sex by age 14. Consequently, the mean age at first sexual intercourse was slightly lower for males than for females (15.6 vs. 15.8). On average, the young people surveyed thought that the ideal age to begin having sexual intercourse was 18—approximately two and one-half years later than their actual mean age at first sexual activity.

Most females (81%) had first had sex with someone older than themselves, while most males had done so with someone of the same age (40%) or younger than themselves (34%). Some 37% of females and 30% of males said their first sexual experience had not been voluntary. Force aside, the main reason cited for initiating sexual intercourse was curiosity (53% of males and 42% of females), followed by the influence of friends (12% and 19%, respectively).

A multiple classification analysis (results not shown) indicates that in this group of young people, having had intercourse before age 16 was strongly correlated with the father's ethnicity, with school and youth-group attendance and with educational level. All other things being equal, early sexual intercourse was most likely to occur among the Tikari-Nsoh and the Bamiléké: Some 54-55% of the young people belonging to these groups had engaged in sexual activity before age 16, compared with only 32% of Bali-Ngemba young people. Precocious sexual intercourse was least likely among young people whose father belonged to ethnic groups that were not native to Northwest Province (30%).

School attendance was positively correlated with precocious entry into sexual activity. However, the relationship between educational level and early first intercourse was negative, especially after middle school. All else being equal, 49% of young people with a primary or middle-school education had had intercourse before age 16, compared with 29% of those with a high school education. Contrary to the result expected, youth-group attendance was negatively correlated with early initiation of sexual activity.

The relationship between gender and age at first intercourse became highly significant after the effects of other factors were taken into account, with 50% of males and 35% of females having had sex before age 16.

Other Sexual Behaviors

Among young people who had already had sexual intercourse, 27% had multiple sexual partners at the time of the survey, while 16% said that they had no current partner (data not shown). Forty-one percent of sexually experienced young people had had casual sexual intercourse during the previous 12 months. Only 25% of the young people who were sexually active at the time of the survey were using condoms. At every age, the proportion of males who had several concurrent sexual partners or who had had casual sex was higher than that of females.

These indicators of sexual behavior also correlated with characteristics of the young people and their families (Table 4). Compared with sexually experienced young people whose father's ethnic group was not native to Northwest Province, for example, those whose father belonged to the Bali-Ngemba, Bamiléké, Makon-Banyague or "other Northwest" groups were significantly more likely to be sexually active at the time of the survey (odds ratios of 1.1 each). Sexual activity was significantly less common among youths whose father was a farmer or fisherman than among those whose father was an office worker (0.6). It was more common among those whose parents did not live together (1.5), those whose parents had a polygamous marriage (1.8) and those living in a poor household (1.4) than among youths without those characteristics. In addition, compared with young people who lived with both parents, those who lived with their grandparents were significantly less likely to be sexually active (0.3), while those who lived alone, with a brother or sister or with one parent were more likely to be sexually active (1.2-1.6). Young people who had had intercourse before they were 16 were less likely than those who had not to be sexually active at the time of the survey (0.7).

Compared with young people whose father's ethnic group was foreign to Northwest Province, those whose father belonged to the Bamiléké were more likely to have more than one sexual partner at the time of the survey (odds ratio of 1.2), while those whose father belonged to the Metta or "other Northwest" groups were less likely to do so (0.7-0.8). The likelihood of having multiple partners was significantly higher among youths whose father was unemployed and those living in a household with an average standard of living (1.3-1.5); it was significantly lower among those whose father worked as a farmer or fisherman (0.6).

Compared with those who lived with both parents, youths who lived with their grandparents were significantly less likely to have multiple partners (odds ratio of 0.0), while those who lived alone, with one parent, with a sibling or with other persons were more likely to have more than one partner (1.9-4.0). Females were significantly less likely than males to have multiple partners at the time of the survey (0.8), and young people who had had sex before age 16 were more likely than those who had not to have more than one partner (1.5).

Youths whose father belonged to the Metta were less likely than those whose father's ethnic group was not native to Northwest Province to have had a casual partner in the 12 months preceding the survey (odds ratio of 0.8). Neither the father's occupation nor the parents' type of marriage had any effect on this behavior. However, young people whose parents did not live together were significantly more likely to have had a casual partner in the previous year (1.2).

Youths from a family with a poor or average standard of living were significantly more likely than those from a family with a high living standard to have had casual sex (odds ratios of 1.3-1.4). In comparison with adolescents who had lived with their parents during childhood, those who had lived with other persons were more likely to have had a recent casual partner (1.3). Compared with those who lived with both parents at the time of the survey, young people who lived with their grandparents or with other persons were significantly less likely to have had casual sex in the last year (0.5-0.8), while those who lived with only one parent, with a sibling or alone were more likely to have done so (1.2-2.3).

Casual sex was also more likely among youths who did not have sufficient resources to meet their needs than among those who did (odds ratio of 1.2). Compared with young people with a high school education, those who had a primary or middle-school education were more likely to have had a casual partner in the year before the survey (1.3-1.5). Females were less likely than males to have had a casual partner (0.6), and young people who had had sex before they were 16 were more likely than other youths to have done so (1.5).

Compared with youths whose father's ethnic group was foreign to Northwest Province, those whose father belonged to the Makon-Banyague or the Bamiléké were more likely not to be using condoms at the time of the survey (odds ratios of 1.2-1.3), while those belonging to the Metta were less likely not to be using them (0.9). Of all sexual risk behaviors examined, nonuse of condoms was the one most influenced by the father's occupation: Compared with youths whose father was an office worker, those whose father was unemployed were more likely not to be using condoms (2.0), while those whose father worked in commerce or was a skilled worker were less likely not to be using them (0.5 each).

Living in a poor household was associated with a reduced likelihood of condom nonuse (odds ratio of 0.8). In addition, compared with youths who had lived with their parents during childhood, youths who had lived with other persons were less likely not to be using condoms (0.8). Furthermore, in comparison with young people who were living with both parents at the time of the survey, those who were living alone or with a brother or sister were less likely not to be using condoms (0.5-0.7), while those who were living with one parent or with other persons were more likely not to be using them (1.1-1.7). Failure to use condoms was also more likely among youths who did not have sufficient resources than among those who did (1.1). Finally, young people with a middle-school education were more likely not to be using condoms at the time of the survey than were those with a high school education (1.3).

Conclusions

Like research conducted in other regions of Cameroon and in other African countries,6 this study found that communications between adolescents and their parents or guardians on the subject of sex were poor. According to one observer, the low level of sex education in the family setting appears to result both from a lack of adequate parental knowledge about the subject and because the traditional education that the parents received placed little emphasis on explanation and verbalization.7 These results point to the need for education programs to improve parents' knowledge of the reproductive health of young people, to prepare them for establishing good and open relationships with their children and to give appropriate advice in the area of sex education. These results also bear witness to the need for introducing sex education programs in schools and youth centers.

The young people surveyed were well informed about AIDS, its main means of transmission and methods of prevention, but they continued to have sexual relations that could expose them to infection. Thus, informing and educating young people about sex and AIDS does not seem to be sufficient to motivate them to change their sexual behaviors.

The results of this study reveal that nearly half of the young people surveyed had had sexual intercourse before age 16, that most were sexually active at the time of the survey and that about one-third had several sex partners, but that only one-fourth were using condoms. Moreover, two out of five had had casual sex during the 12 months preceding the survey. Consequently, AIDS prevention programs should place a greater emphasis on fidelity, reducing the number of sex partners and using condoms.

The fact that the percentage of young people using condoms in Bamenda was far higher than what has been found at the national level in other studies suggests that AIDS awareness may have positively influenced condom use in Bamenda. This awareness may also explain at least part of the difference observed in Bamenda between the percentage of young people who are sexually active and the percentages who have several partners or engage casual sex. These differences support recommendations for AIDS prevention programs.

Our differential analyses of the sexual behavior of young people showed that the age at which first sexual intercourse occurs is correlated with social and cultural factors, while sexual risk behaviors are related both to those factors and to economic factors.

The differences in sexual behavior by ethnicity lend support to the theories of Murdock,8 whose pertinence has long been recognized by other researchers searching for the causes of infertility in Central Africa.9 Consequently, these differences may reflect differences in sexual permissiveness among ethnic groups in Northwest Province. This explanation, however, seems insufficient, because the positive relationship observed between membership in the Bamiléké or Tikari-Nsoh group—the two groups that impose strict control of sexual mores—and the behaviors involved can only be explained by a tendency to rebel against traditional sexual norms.

The relationship between precocious sexual intercourse and school attendance may reflect the fact that school promotes the meeting of young people of the opposite sex in an environment that is not controlled by their family. Indeed, curiosity and the influence of schoolmates or friends were mentioned by many of them as reasons for their first sexual intercourse. The negative relationship between educational level and precocious first sexual intercourse, which confirms findings of studies done in Benin10 and in Cameroon,11 may reflect the fact that remaining in school for a long time delays social maturation.

As the results of this study have revealed a strong positive relationship between early first intercourse and most of the other sexual behaviors being studied, actions aimed at delaying entry into sexual activity by young people should also be part of the content of the AIDS prevention programs in Africa. Parents should monitor their adolescent children's choice of friends. The finding that almost four in 10 young women said that their first sexual experience had not been voluntary suggests that parents need to mobilize their communities against sexual violence toward girls.

Young peoples' sexual risk behaviors are influenced by both cultural factors and economic factors; this means that their environment and living conditions are extremely important. In addition, it underscores the importance of the living arrangements of parents and their cohabitation with their young children. As noted in a study set in Uganda, "most of the students' sexual activities occurred when parents were away or when the students were staying with relatives. Also, although numbers were small, generally orphans and children living with single parents were more likely to engage in sexual activities...."12 Young people who live with their parents are less likely to have emotional problems,13 and their behavior is more likely to be under their parents' control.

The results of this study suggest that noncohabitation by parents, conjugal instability and noncohabitation with children may raise the risk that adolescents will engage in risky sexual activity. Consequently, sensitization of parents about these problems should be included in AIDS prevention programs intended to motivate adolescents to avoid risky sexual behaviors.

As young people who have unemployed fathers, live in poor households or have insufficient resources to meet their needs are at greater risk than their peers of having several sex partners and of engaging in casual sexual intercourse, poverty appears to contribute to risky sexual activity. To satisfy their material needs, they may engage in sexual activity with multiple partners or casual partners or agree to have sex without a condom. Data derived from qualitative research conducted along with our survey confirm these results, as most of the young women who participated in individual interviews declared that they had sexual intercourse in exchange for money, gifts or clothing:

"Because I have children, I cannot have sexual relations with a man if he does not give me something."—A single 21-year-old woman

"I got involved in this sexual life because I lost my parents, I have two children and I don't have a husband to help me resolve my problems and those of my children....This life provides me the wherewithal I need to solve my little family's problems, but if I get married or find a job, I'll give it up, because it exposes me to a lot of risks."—A single 19-year-old woman

"I can't have sexual relations with boys at all because they don't have the wherewithal. However, responsible men almost all help me when I have money problems."—A single 21-year-old woman

Men, on the other hand, frequently adopt risky sexual behavior to procure the maximum gratification of their sexual desires. However, adolescent males claim that some of their schoolmates and friends do so for economic reasons. This is particularly true of boys from poor families, who may receive aid from girls in the form of work (such as domestic help) or money, or assistance from married women who clandestinely keep single boys in exchange for sexual gratification. Still, girls are particularly likely to engage in risky sex for economic reasons, which negatively influences their power to require use of condoms during sexual intercourse. Improvement of the living conditions of families in Bamenda, particularly those of single mothers, might well reduce the transmission of HIV over the long term.