image002.jpg

HALIRA Findings

The HALIRA findings are grouped into three areas:

  • general findings on sexual health and behaviour
  • MkV1 process and impact evaluation
  • methodological findings

General HALIRA findings on sexual health and behaviour

  • Reproduction, pregnancy prevention and termination
  • HIV and other sexually transmitted infections
  • Young people’s sexual behaviour

 

HALIRA findings on reproduction, pregnancy prevention and termination

Reproduction in rural Mwanza was generally a source of pride, accomplishment, and future security. Some believed pregnancy desirable in any circumstance. However, pre-marital pregnancy could reduce a woman's brideprice, create financial or legal burdens for her sexual partner, and limit either partner's choice of spouse. Many youths thus wanted to prevent pregnancies, especially school girls, who were expelled if pregnant. Most young women reported traditional pregnancy prevention, such as wearing charms or drinking herbal or ash solutions. A small minority used modern contraceptives, but very few girls did so because of limited access and concerns about confidentiality. Deprovera injections were the most popular modern method, as they were infrequent and easily hidden from partners.

Many villagers perceived condoms negatively for several reasons, for example, believing them to be associated with infection or promiscuity, to reduce male sexual pleasure, or to negatively affect cultural understandings of meaningful sex. Men controlled the terms of sexual encounters, and reported that they would only use condoms with high risk partners. However, few men saw their partners as high risk. Use of condoms appeared to be very low, primarily due to limited demand, although barriers to access also existed.

Abortion was illegal, but it was widely, if infrequently, attempted, by ingestion of laundry detergent, chloroquine, ashes, and/or specific herbs. Some health workers performed illicit abortions, but this method was less accessible, particularly for adolescents. Most women who attempted abortion were young, single and desperate; some succeeded, but they experienced sexual partner opposition, sexual exploitation by practitioners, serious health problems, social ostracism, and/or quasi-legal sanctions. Many informants reported that inopportune pregnancies could alternatively be ‘moved to the back’ and suspended for months or years, using traditional medicine.

 

HALIRA findings on HIV and other sexually transmitted infections

Rural respondents attributed most sexually transmitted infections to natural causes. Treatment for sexually transmitted infections was pluralistic and opportunistic, usually beginning with home remedies (modern or traditional), followed by visits to traditional healers and/or health facilities. Traditional healers were sometimes preferred over health facilities because of familiarity, trust, accessibility, expense, payment plans, and the perceived cause, nature and severity of the illness. For example, only traditional healers were believed to successfully treat bewitchment. Some people, particularly young girls, delayed or avoided seeking treatment for sexually transmitted infections for fear of stigma.

Many people reported belief in both a ‘real' (natural) AIDS, which would lead to certain death, and a similar illness caused by witchcraft, which could be cured using traditional medicine. Very few villagers understood that the HIV asymptomatic period lasted longer than a few months. Locally available biomedical care of people with AIDS-like symptoms consisted of basic treatment of opportunistic infections, as there were no anti-retroviral services available in rural Mwanza at the time of the research. Most individuals with AIDS-like symptoms repeatedly visited traditional healers and health facilities, but many stopped attending health facilities because they came to believe they could not be cured there.

 

HALIRA findings on young people’s sexual behaviour

The majority of youth were sexually active by the age of 15, and most believed that, once sexually active, it was extremely difficult to become abstinent. Youth sexual activity was constrained by clear norms of school pupil abstinence, female sexual respectability and taboos around the discussion of sex. However, these norms were incompatible with several widely held expectations that sexual activity is inevitable unless prevented, that sex is a female resource to be exploited, that restrictions on sexual activity are relaxed at festivals, and that masculine esteem is boosted through sexual experience. Differential commitment to these norms and expectations reflected conflicts between generations and genders.

Sexually active adolescents were severely punished if caught, especially girls and school pupils. Therefore young people usually carefully concealed all sexual relationships, or all but a primary one that might have some social recognition, for example, amongst peers. This meant that couples had little opportunity to develop emotional intimacy through non-sexual contact, while the lack of social recognition of relationships meant little social reinforcement of primary relationships, or discouragement of secondary ones. Both young men and women were opportunistic in developing relationships and engaging in encounters that were logistically feasible within the constraints of gender-segregated village life. Concurrent relationships were thus common, despite a norm of monogamy, particularly amongst women. People were generally unaware of being concurrent partners, and discovery of infidelity usually led to break-up.

Young men’s main motivation to have multiple partners was sexual desire (such as attraction to different partners, or satisfying a sexual urge in a primary partner’s absence), whereas young women were mainly motivated by money and gifts they received in exchange for sex. Transactional sex underlay most non-marital relationships and was not, per se, perceived as immoral. Girls’ and women’s motivations to engage in transactional sex varied, for instance: escaping intense poverty, seeking beauty products or accumulating business capital. There was also strong pressure from peers to engage in transactional sex, in particular to consume like others and to avoid ridicule for inadequate remuneration. Macro-level factors shaping transactional sex (such as economic, kinship and normative factors) overwhelmingly benefited men, but at a micro-level there were different dimensions of power, stemming from individual attributes and immediate circumstances, some of which benefited women. Young women actively used their sexuality as an economic resource, often entering into relationships primarily for economic gain. Transactional sex was likely to increase the risk of HIV by providing a dynamic for partner change, making more affluent, higher risk men more desirable, and creating further barriers to condom use. This was one factor contributing to adolescent girls often having sexual relationships with men several years older than them.

 

HALIRA MkV1 process and impact evaluation

 

HALIRA MkV1 teacher-led, peer-assisted school sexual health programme

The primary intervention within MkV1 was the school-based curriculum, so this was also the main focus of the HALIRA evaluation. In the broader school system, half of Year 7 pupils were 15–17 years old, and few went on to secondary school, suggesting that primary schools were a good venue for adolescent sexual health programmes. However, serious challenges to potential programmes included low enrolment and attendance rates, limited teacher training, little access to teaching resources and official and unofficial practices that alienated some pupils and their parents, such as corporal punishment, pupils being made to do unpaid work, forced pregnancy examinations, and some teachers’ alcohol or sexual abuse.

Within the MkV1 school intervention programme, almost all MkV1 teachers delivered the overwhelming majority of sessions. Similarly, most class peer educators performed well in the drama serial, and sometimes also assisted in other aspects of the intervention. Overall, teachers and class peer educators closely followed the new curriculum with remarkable detail and consistency. As planned, most intervention schools were also visited by a health worker who specifically addressed sexual health issues with MkV pupils; many classes also visited the local health facility as planned. Most MkV1 intervention participants had substantially better sexual health knowledge than their comparison counterparts. These findings suggest that, despite great resource limitations, it is feasible to implement such interventions through existing government structures when they are incorporated into the school curriculum.

MkV1 intervention participants (and particularly class peer educators) were generally more comfortable than their comparison counterparts in talking openly about sexual activity and risk reduction, which is often considered a first step towards positive behaviour change. However, in this case it did not often seem to translate into greater perceived self-efficacy to reduce risk behaviours. A primary influence was the widespread belief that it is too difficult to abstain from sex once young people have already experienced it. Many participants were already sexually active at the start of MkV1, which may argue for an intervention that begins in earlier school years. In addition, MkV1 did not seem successful in increasing participants’ perceived susceptibility to risk, or expectations of negative outcomes from sex. Early adolescents may have limited ability to anticipate behavioural outcomes, particularly long-term outcomes, and their exploratory behaviours may also have an impulsive component. Realistic risk perception may have also been hindered by widespread secrecy about sexual relationships and ignorance about partners’ past and current partners. Possible ways to better address this in intervention programmes is to personalise risk more, for example, through games illustrating transmission of infections, and visits by HIV-infected individuals.

Low expectation of long-term negative outcomes from sex amongst MkV1 participants was further complicated by positive, short-term expectations, such as pleasure, material gain and/or peer esteem. The widespread practice of material exchange for sex may have been a particularly strong impediment to risk reduction, as it had both emotional and financial importance for girls. The MkV1 curriculum acknowledged the temptation and negative consequences of such material exchange in the drama serial, but it may benefit from more attention, both in the existing curriculum and in potential new interventions, such as income-generation projects for girls.

The MkV1 class peer educator component was successful in creating a consistent and entertaining drama serial, contributing towards the goal of modelling desirable behaviours. However, class peer educator ability to informally educate their peers and to be role models for them seemed very limited. Brief class peer educator training was unlikely to overcome (i) their low literacy and education levels; (ii) their subservience within a hierarchical, punitive and didactic school setting and (iii) powerful and contradictory adolescent sexual norms and expectations. The feasibility and effectiveness of adolescent peer education may be limited in this extremely disadvantaged setting, particularly when the programme is intended to go to scale, so the potential for training is limited.

MkV1 teachers developed more positive attitudes about teaching sexual health topics and successfully provided pupils with important information. However, they were less effective in adopting the active-learning and inquiry-based teaching techniques promoted by the WHO and other global leaders in adolescent health education, as such techniques often involved a radical departure from local teaching practices. Further, some undesirable existing practices may have been reduced by the MkV1 training (such as corporal punishment or discouragement of condoms), but it is less clear that others were (such as forced pregnancy examinations or sexual abuse of pupils). Fundamental changes to teacher–pupil relations may require substantial reform in training and supervision within the broader school system.

When MkV1 was introduced, there were widespread misconceptions and negativity towards condoms, both in the school system and the broader community. In that context, the programme was successful in obtaining permission from regional educational authorities and village leaders to discuss condoms in the classroom, in increasing pupil knowledge of the benefits of condom use, in providing pupils with condom demonstrations in health facilities and, to a lesser extent, in increasing pupils’ access to condoms. At the end of the MkV1 trial, MkV1 participants generally seemed more favourable towards condoms than non-participants, although they often remained ambivalent, believing that condom use was unrealistic. In addition, MkV1 participants sometimes could only vaguely describe condoms, or misunderstood specific points related to them, a problem that may have been exacerbated by not being able to actually see condoms during their MkV1 school lessons, because this was prohibited by educational authorities. It seems unlikely that any but a small minority of sexually active MkV participants used a condom during the MkV1 trial.

 

HALIRA MkV1 condom promotion-distribution

The main, immediate reason for low condom use during the MkV1 trial appears to have been lack of demand, rather than lack of supply, or difficulty accessing it. However, even if school pupils had been highly motivated to use condoms, barriers to accessing condoms remained. Concern about the confidentiality of exchanges with shopkeepers, health workers or MkV1 condom promoter-distributors may have inhibited pupils from seeking condoms. The MkV1 condom promotion-distribution intervention was primarily developed to serve young people, but in practice it usually served an adult male population. Increasing condom use among adult men could indirectly reduce young women’s sexual health risk, but the number of clients and condoms sold were too low to have made a substantial impact. Further, some MkV1 condom promoter-distributors were not themselves convinced of the importance of condom use and were poor role models. This intervention was also very costly and labour-intensive in comparison to other interventions in the MkV1 programme, and thus posed great challenges to potential sustainability and scaling up.

For all of these reasons, the MkV1 condom promotion-distribution intervention was discontinued at the end of the MkV1 trial in 2001. Given condom use remains a key MkV intervention message, collaboration with established formal and informal product distribution networks may be a more promising approach in the future, for example, through an expansion of limited rural networks specific to condom social marketing, or more integration within large-scale or small-scale businesses.

 

HALIRA MkV1 community mobilisation

Community mobilisation was a relatively small intervention within the MkV1 intervention programme, largely involving creation of an advisory committee for each ward, and leading a small number of group discussions with parents and local leaders in each village. The MkV1 intervention programme successfully achieved these goals. Many issues addressed within the intervention programme were controversial within communities, such as adolescent sex, or condom use in general. When controversy arose, the intervention team seemed very capable of responding to community members with respect and clarity, successfully communicating the urgent need for intervention, and ultimately winning the support of those most opposed to it. In addition, visibility of the MkV1 name and logo on T-shirts and vehicles seemed to successfully familiarise the general community to the MkV1 name, and the receipt of free materials (such as video shows, T-shirts, sports balls, condoms, and drugs) was extremely well received by implementers, young people and adult community members.

However, HALIRA research found that a high proportion of parents were not well informed about the purpose and content of the MkV1 interventions or the trial, for instance, not knowing about them at all, or having a distorted sense of what was involved. The broader community was generally even less informed than parents or local leaders. In some villages, in addition to positive impressions of the intervention programme, negative rumours developed and persisted over time, particularly that MkV1 taught/encouraged seduction and promiscuity. However, adults involved rarely seemed sufficiently concerned or assertive to voice formal or public opposition, and thus intervention staff did not respond to them directly.

Because many parents, siblings and out-of-school friends and sexual partners were uninformed and/or wary and/or uninformed about the intervention programme content, they were not likely to understand or support participant behaviour change. This may have been particularly challenging for girls with older, out-of-school sexual partners who had not participated in the intervention programme, emphasising the value of sustaining school-based interventions over a longer period, and integrating such in-school interventions with broader, community-wide interventions.

 

HALIRA MkV1 youth-friendly health services

As planned, most MkV1 intervention schools were visited by a health worker who specifically addressed sexual health issues with MkV1 pupils, and many classes also visited the local health facility as planned. Both participant observation and the simulated patients study found that intervention health care workers generally had better attitudes towards young people, particularly in fostering a non-judgemental approach to sexual health, and generally provided more sexual health information to young people when consulted. However, the qualitative research also suggests that health care workers in both arms of the MkV1 trial had similarly low respect for young people’s privacy, and did not promote condoms as desired.

In addition, at the end of MkV1, many intervention participants did seem to be aware of the free condom and sexually transmitted infection services provided by health facilities. Those young people who experienced symptoms of sexually transmitted infection often seemed unable to seek assistance at health facilities, because of distance to facilities and concern about their confidentiality. Historically, teachers in rural Mwanza sometimes forced female pupils to have pregnancy examinations at local health centres en masse, contributing to pupil distrust of teachers and health care workers. This pre-existing practice was not entirely eliminated in MkV1 intervention communities, as there were still reports of it occurring during the MkV1 trial.

 

HALIRA methodological findings

The participant observation method developed by the HALIRA programme was innovative in its attempt to combine the best of qualitative and quantitative methods. Like traditional participant observation, it was in-depth and unstructured in its approach, allowing for the collection of meaningful and sometimes unexpected data. However, unlike the traditional approach of one researcher in one village for 1-2 years, in the HALIRA programme six researchers lived in nine villages for extended and repeated visits, culminating in 158 person-weeks of data that were more representative than traditional qualitative data. The HALIRA programme found that by carefully selecting a moderate number of representative sites, researchers, and informants, qualitative research can collect meaningful and representative data in a practical, effective, and economical way. If such data are collected at systematic intervals over an extended period of time, they may also be valuable in the identification and monitoring of trends.

The HALIRA programme also developed a useful novel approach to group discussions with young people on sensitive sexual issues. Group discussions were conducted with a female and male group in each of three contrasting villages that were not being used for long-term participant observation. The research team first spent three days engaged in participant observation, while also getting to know and recruiting pre-existing friendship groups. Over the subsequent several days, three or four discussions with both a male and a female group were conducted by same-sex researchers. Two to three in-depth interviews were also conducted with individual group members immediately after each discussion, to follow up on important topics.

The assisted self-completion questionnaire method was found to have limited usefulness in assessing the sexual behaviour of African adolescents, showing most promise with older, male and/or educated respondents. However, triangulation with data from other surveys raises questions about the validity of self-reported sexual behaviour in general.

The HALIRA programme assessed the relative strengths and weaknesses of the five main MkV1 research methods in evaluating adolescent sexual behaviour. These methods were: a survey of biological markers (pregnancy, HIV, and other sexually transmitted infections), a face-to-face questionnaire survey, an assisted self-completion questionnaire survey, semi-structured in-depth interviews, and participant observation. In this low prevalence population, biological markers on their own only identified a very small number of adolescents who had had sex. However, in combination with in-depth interviews they may be useful in identifying risk factors for sexually transmitted infections. Self-reported sexual behaviour data were fraught with inconsistencies. In-depth interviews seem to be more effective than assisted self-completion questionnaires and face to face questionnaires, particularly in promoting honest responses among females with sexually transmitted infections. Participant observation was the most useful method for understanding the nature, complexity, and extent of sexual behaviour.