Scaling up support for children in HIV-affected families by involving early childhood development workers: community views from KwaZulu-Natal, South Africa

Abstract

The United Nations Programme on HIV/AIDS estimates that South Africa has 1.2 million orphans as a result of AIDS. This study investigated the views of communities in KwaZulu-Natal on the potential use of early childhood development (ECD) practitioners for helping children in vulnerable households. It reports on in-depth interviews and focus group discussions with community members and service providers in six different areas of KwaZulu-Natal. Despite the increase in the numbers of deaths, people in affected areas have yet to accept that HIV is the cause. Stigma, denial, myths and witchcraft beliefs were evident, particularly in the rural areas. There was general agreement by the respondents that ECD practitioners with additional training specific to HIV/AIDS could assist families if cost difficulties were overcome. As community members with basic training in childcare they could enhance community responses by linking with community health workers and other service providers to assist children and households in need.

1. Introduction

The southward thrust of the HIV/AIDS epidemic in the latter years of the twentieth century resulted in increasing numbers of orphans and child-headed households in Uganda and Tanzania in the late 1980s, followed by Zambia and Zimbabwe in the 1990s (United Nations Children's Fund, 2003). The impact on South Africa followed approximately a decade later, with HIV/AIDS the largest contributor to the country's burden of disease, resulting in high morbidity and mortality (Bradshaw et al., 2003). As the scale and intensity of the epidemic has escalated, increasing numbers of children have been orphaned (Brookes et al., 2004). Traditionally, as in other countries in Africa, communities responded by absorbing these children into the families of the deceased, but the scale of the epidemic and the high unemployment rate (25–40 per cent in South Africa) has strained this capacity to respond (Statistics South Africa, 2003).

Although South Africa is a middle-income country, it is nevertheless one with conspicuous income inequalities and a Gini index of 59.3 (AfricaFocus, 2004). A decade of democracy has yet to change the major disparities of the apartheid era in the provision of services, including health, education and welfare. The rapid spread of the HIV epidemic has occurred while the country has been engaged in transforming the systems and services needed to serve communities that have been marginalised by so many years of apartheid and service provision has lagged behind owing to the scale of the epidemic (Ntuli & Day, 2005). The numbers of affected people are substantial – 537 000 South Africans were estimated to have AIDS in 2005 (United Nations Programme on HIV/AIDS, 2006). The government's response to HIV/AIDS includes prevention, treatment and care, but the epidemic has overwhelmed its resources in many respects. In KwaZulu-Natal, the largest and one of the poorest of the nine South African provinces, the prevalence of HIV among people who attended public antenatal clinic had risen to 39.5 per cent by 2005 (Department of Health, 2005). Community health workers and home-based carers have been trained to visit households in their areas but the number of these workers is small when compared to the size of the population and the high prevalence of HIV: a total of 2400 community health workers to a population of 9.4 million in KwaZulu-Natal (Taylor & Kvalsvig, 2002).

The mortality from the HIV/AIDS epidemic has left many children without parents. The roll out of anti-retroviral treatment began relatively recently, so coverage is still low and the number of orphans is increasing (Boerma et al., 2006). Most children have been absorbed into extended families, but as aunts and uncles die the burden of care has increasingly shifted to grandparents, many of whom are elderly and unwell themselves and unable to provide children with the optimum care and stimulation (Bower, 2006).

In theory, children in South Africa are protected by the Bill of Rights in the Constitution, which states that they have the right to basic nutrition, shelter, basic health care and social services, to be protected from neglect and abuse and to equality, dignity and a basic education (Republic of South Africa, 1996; United Nations Children's Fund, no date). South Africa has ratified the Convention on the Rights of the Child, which recognises that children should grow up in an atmosphere of happiness, love and understanding, with their best interests considered. The government thus has a statutory obligation towards children infected or affected by HIV/AIDS, to reach these children and to provide them with adequate care, but in practice this is difficult to achieve.

In the past, early childhood development (ECD) initiatives were located mainly in the non-government sector. There are ECD sites or crèches throughout South Africa and many ECD practitioners have received training from non-profit organisations (NPOs) in caring for children of pre-school age (Department of Education, 2001). Increasingly, the government is developing policy with respect to ECD. In 2003, White Paper 5 defined the policy for ECD in South Africa and the purpose of the subsequent Integrated Plan for Early Childhood Development (Department of Education, 1995) is to link services in the education, health and social development departments. The linkage is conceptualised as occurring between government departments and between the government and non-profit sectors.

With these considerations in mind, the aim of the present study was to investigate community understanding of the HIV/AIDS epidemic, to understand what informal as well as institutionalised resources are currently in place to help affected children and families and to explore perceptions about the role or potential role of the ECD worker in providing such help. Ultimately the information was to be used to develop an intervention.

Green and Kreuter's (1991) Precede Model provided a framework for the situational analysis. This model was selected since it provided a methodology for a comprehensive investigation of factors influencing the lives of orphans and vulnerable children and the potential role of ECD workers in addressing such factors. This required investigating social factors affecting community behaviour in respect of orphans, epidemiological factors and behavioural and environmental factors. In addition, educational factors that predispose, enable or reinforce behaviours relating to HIV/AIDS prevention and the treatment of orphans and vulnerable children needed to be investigated, as well as the associated administrative issues (Green & Kreuter, 1991).

2. Study Methods

2.1 Study area and participants

The study used qualitative methods to describe the component factors of Green and Kreuter's (1991) model in selected areas in KwaZulu-Natal using key informant interviews and focus group discussions.

Information was collected through the network of community ECD sites and programmes organised by Training and Resources in Early Education, an NPO working with ECD practitioners throughout the province of KwaZulu-Natal. Four of the 11 districts in the province were included, comprising both urban and rural areas where Training and Resources in Early Education works. The study sample comprised participants from existing projects reaching children, either through ECD sites (where children are sent regularly to an ECD venue, e.g. a crèche) or through Training and Resources in Early Education Siyafundisana projects, which reach out to women looking after children at home. Of the six projects, the urban ones were in Cato Manor (Durban) and Gamalakhe (Port Shepstone) and the rural ones in Bulwer, Centocow, Nkandla and Etafeni, providing a range of sites with trained and experienced ECD staff. The following paragraphs are brief descriptions of the sites.

Cato Manor (population 93 000) is the largest inner-city urban development project in post-apartheid South Africa. Gamalakhe and Etafeni are near Port Shepstone (population 6290): the former is a township and the latter a rural area. Bulwer (population 6979) and Centocow (population 10 962) are small rural villages in southern KwaZulu-Natal (World Map Encarta, no date). Nkandla (population 8191) is a deep rural community in northern KwaZulu-Natal, characterised by high levels of poverty, illiteracy and unemployment (United Nations Children's Fund, 2006).

Interviews were held with key informants (local leaders, church ministers, traditional healers and, where possible, clinic sisters, social workers and NPOs) in each of these six areas. Focus group discussions were held with parents and caregivers, community health workers and ECD practitioners. All interviews and focus group discussion were conducted in IsiZulu.

Information for the development of interview schedules was obtained at a meeting with all 40 Training and Resources in Early Education trainers from around the province. They were asked to identify the most important issues arising from the HIV/AIDS epidemic in their areas. Based on their insights, schedules were developed for the interviews and the focus group discussions. The field team was composed of four experienced research assistants who spent a total of six weeks collecting the data (a week in each area). The Training and Resources in Early Education area coordinators facilitated the research team's entry into each area and obtained permission for the study before the team made contact with potential respondents. The numbers and categories of the study participants are listed in Table 1.

Table 1 Respondents from urban and rural areas in KwaZulu-Natal

Area	Cato Manor	Port Shepstone	Ethafeni	Nkandla	Bulwer	Centocow	Total
Parent/Caregiver	10	10	7	10	11	8	56
Educare practitioner	7	3	2	3	8	10	33
Social worker	–	1	–	1	–	–	2
Local leader	–	1	1	1	1	1	5
Church minister	1	1	1	1	1	1	6
Traditional healer	2	1	–	2	1	1	7
Community health worker	10	10	1	21	7	8	57
Clinic sister	–	–	–	1	1	1	3
Other	1^a	2^b	–	–	3^c	2^d	8
TREE trainers	Group	Group	Group	Group	Group	Group	45
TREE, Training and Resources in Early Education
^aLibrarian.
^bA hospice worker and a street children's project worker.
^cAn orphans' project worker and two volunteer groups.
^dA community project worker and a preschool project worker.

Each interview and focus group discussion was tape-recorded, transcribed and translated into English by the interviewer/moderator and the information was supplemented with interviews with NPO personnel working in the area. The data from the interviews and focus group discussions were analysed and the themes identified.

3. Findings

3.1 Community perceptions about HIV/AIDS: social, epidemiological, behavioural and environmental factors

Many people in the study areas had died and others were critically ill and participants in the focus groups and interviews spoke with sadness about the funerals of young people which occurred regularly in all the areas. There was a general view in all six communities that the numbers of orphans were increasing, but that people in these emotionally charged circumstances had still not come to terms with HIV as the cause. People had difficulty talking directly about HIV/AIDS and used gestures and euphemisms. Even though one might suspect that a person was HIV positive, this was not usually confirmed because most infected people were secretive about the nature of their illness.

Understanding of the issues varied from place to place. A community health worker said there were two distinct groups of people: those who knew that HIV/AIDS kills and others who did not believe it existed. There was more awareness in the urban communities and more open discussion about the epidemic than in the rural areas: rural community leaders did not openly acknowledge and deal with the problem. The participants from rural areas reported that the media (in particular radio) were their main source of information about HIV/AIDS. They realised from the descriptions in radio reports of the symptoms and progression of HIV/AIDS that many of the deaths in their communities resulted from AIDS.

A belief that the deaths were attributable to witchcraft was still evident in some reports. In a focus group discussion in rural Nkandla, the respondents indicated that there was denial of the epidemic in their area and the effects of the epidemic were ascribed to witchcraft: ‘People are becoming enemies because they believe they are bewitched and this is creating animosity’ (community health worker).

Comments on the extent of witchcraft beliefs came from key informants as well. A traditional healer said that very few infected clients acknowledged that the cause of their symptoms was HIV infection: most considered that they had been bewitched. Similarly, an ECD trainer said that only one or two out of ten infected people were willing to admit they were HIV positive: most said they were bewitched or had tuberculosis. The myths and other forms of obfuscation of facts appeared to be a collective denial of the reality of the epidemic. This diffidence and denial had consequences for treatment. Despite regional differences, it was reported from all the areas that HIV/AIDS was not often discussed openly, that most infected people did not disclose their status and many people did not want to take a blood test for HIV. Even people who showed the symptoms spoke instead of other illnesses, either in an attempt to divert unwanted criticism or because they did not want to believe that they were infected: ‘I know of one person who is seriously ill but she says that she is diabetic. The symptoms are just what we suspect for AIDS’ (parent in a focus group discussion).

This secretive attitude was reinforced by the medical profession's strict observance of confidentiality. It was explained that caregivers were only told about a person's HIV/AIDS status by doctors and nurses at the hospitals if the patient was critically ill and would have to be cared for at home. If the person was not seriously ill at the time, the health service providers did not give out any information, even though the caregivers might be able to assist more effectively if they knew what was the matter or might themselves be at risk through not knowing. An ECD trainer noted that, where there had been denial, the children suffered because when they were orphaned they were left not knowing what had really happened to their parents.

It was said that the epidemic was devouring communities, in the sense that it was destroying community life, but because the disease was known to be sexually transmitted and associated with multiple sexual partners there was a stigma to acknowledging that one was infected. Comments about the stigma included ‘Other children laugh at them [orphans] saying that their parents died of AIDS’ (parents in a discussion group) and ‘Some people discriminate against children with parents or deceased parents who are HIV/AIDS victims because they strongly believe that all those with HIV/AIDS have it because of their behaviour’ (community health worker).

The stigma affected orphans. It was said that children listened to their parents gossiping and then talked about what they had heard to other children. In addition, people wrote on rocks and on ruined houses that such and such a person had died of AIDS. In this way affected children were subjected to graffiti and ridicule.

3.2 Services and resources available to communities: educational and administrative factors

3.2.1 Anti-retroviral treatment

Although the anti-retroviral treatment roll out was initiated in 2004 it had not reached many AIDS sufferers in the communities visited. One key informant expressed optimism that the availability of treatment in areas where previously there had been none would be the start of a new era. Community health workers noted that treatment was not yet available for infected children who were sick and that they almost always lacked proper nursing care. In the poor rural communities (particularly Nkandla and Centocow) the cost of transport to clinics was often prohibitive and sometimes the caregivers of infected children did not accept that the child had an HIV infection.

3.2.2 Clinics

At clinics in all the areas, counsellors had been trained to provide pre- and post-test counselling for people testing for HIV. Expectant mothers were advised to test for HIV and this was usually the first time that infected women were made aware of their HIV status. The Prevention of Mother to Child Transmission Programme provided HIV-positive mothers with a single dose of nevirapine for them and a dose for their babies. Counsellors also educated mothers on how to look after themselves and their infants. HIV-positive mothers were advised to talk to their children about HIV/AIDS and to explain to them how to deal with minor injuries at school. They were counselled to love and care for their children and to use the treatment provided and were informed that sick children should be brought to the clinic immediately.

3.2.3 Community health workers

Community health workers are deployed across the province, but not yet in sufficient numbers to serve all communities. In the areas where they were situated they were reported to be very helpful. They visited children and assisted caregivers. They were seen as a resource in the community and were asked to give support to neglected children by intervening with the family. The community health workers advised HIV-positive mothers to eat nutritious food, breastfeed exclusively for six months and take their babies to the clinic to be immunised and weighed. Mothers of HIV-positive children were advised to give them love, to be extra careful with their health and to take them to the clinic when they were ill. Community health workers encouraged mothers to find out their HIV status and to inform their older children if they were sero-positive, because, eventually, in the critical stages, the older children would need to take responsibility. Mothers, however, were often loath to tell their children: ‘Usually mothers feel that when they disclose to their children the children will distance themselves from their mother or they will react negatively’ (community health worker report).

The community health workers expressed the opinion that their communities could not absorb the large numbers of orphans and suggested that the government build orphanages. They also suggested that teachers act as resource people for orphans and vulnerable children by contacting the relevant government department to obtain assistance.

3.2.4 Traditional healers

The interviews with traditional healers revealed that the advice and support they gave patients was similar to that given by other health professionals and they advised clients to go to the clinic for diseases that they could not treat. One of the healers commented that families knew what was wrong with their sick members, but that they did not want to talk about it.

3.2.5 Financial support for children

In the poor communities where HIV is prevalent and the death rate is climbing, access to food parcels and grants is crucial to the health and well-being of the orphans and their caregivers. There were three types of grants which could be of support to AIDS-affected families: the Child Support Grant (US$28 per month), the Foster Care Grant (US$90 per month) and the Disability Grant for AIDS patients (US$120 per month). The Child Support Grant was quicker and easier to obtain than the Foster Care Grant. The conditions for the Foster Care and Disability Grants were more stringent because social workers are required to investigate the family's circumstances. In all the areas studied access to the Child Support Grants was viewed by everyone as difficult to obtain and the research team was told that large numbers of families do not receive the grants to which government policy entitles them, because there are very few social workers to process the applications. Community health workers estimated that between 5 and 45 per cent of families who qualified were receiving the grants and some families had waited as long as three years without success.

3.2.6 Response from the community

Across the study sites there were reports of people and organisations who helped with food and clothing: family and community members helped, as did nurses, teachers, pre-school teachers, community health workers, doctors and local NPOs, but the scale of the need was such that many children still lived in dire poverty. The list of community resources supplied by the respondents confirmed the limitations of the assistance available to affected families and the pervasive poverty and unemployment in the region hampered efforts. Nevertheless, the willingness to help was considerable: neighbours supported children with sick parents and contacted the community health workers. Community workers providing tuberculosis treatment offered food from their own homes to clients where the family lacked adequate resources for food, so that the tablets would have minimal side effects. An ECD trainer reported that, in one area, a support group had been established to help 30 people who had disclosed their HIV status and in some of the cases their children were infected as well.

Churches varied in their response to the crisis. The most comprehensive assistance was in Centocow where the church helped needy people of all denominations and trained community members as home-based carers. This organisation provided school uniforms, blankets, food parcels and clothes to affected families and taught agriculture to orphans to provide them with useful skills. They also invited the youth in the area to attend HIV/AIDS awareness workshops. In some of the study areas churches provided food parcels to families where parents had died from HIV/AIDS, but in others the churches had not offered support.

In Cato Manor, Port Shepstone and Centocow NPOs offered help that included food parcels and clothing. However, in the remaining three areas, where Training and Resources in Early Education was the only NPO working with children, access to help was more limited. Even in areas where NPOs were operating, their outreach was limited to families in their immediate environment.

3.2.7 Potential role of ECD sites: educational and administrative factors

While most orphans were of school-going age, the respondents said that there were increasing numbers of younger orphans. There was general agreement that ECD sites could play an important role in helping orphans and vulnerable children if the cost difficulties were overcome, but that not many orphans were able to attend these sites on a regular basis at present because the caregivers could not afford the fees. The community ECD sites had to generate their own income because few had been registered with the Department of Social Development, although many had applied. None of the practitioners interviewed were receiving subsidies for their sites from the Department of Social Development, so they relied on fees paid by parents to staff to equip the ECD site. Communities were poor and, thus, fees were low (US$3.50 per child per month in some instances and US$5 for babies) resulting in substandard buildings and staff being poorly paid. The ECD practitioners in charge of the sites were therefore not in a position to accept children whose families were unable to pay. Even with the low fees, most families caring for orphans were unable to afford this service. This meant that the children who were most in need of extra care and support were often unable to attend the ECD sites.

3.2.8 Training for the required skills: educational factors

In the interviews and discussions it was agreed that the ECD practitioners were community members who could enhance their communities' response to HIV/AIDS and the care of children on the strength of their training in childcare. By linking them with community health workers and other service providers they could enable development, reinforce health messages, respond to the needs of orphans and vulnerable children (e.g. provide information about access to grants) and advise on suitable care for vulnerable children. To facilitate such a response, their ECD training should include a special focus on the additional skills required to achieve these goals and promote strategies to facilitate inter-sectoral collaboration. In order to be effective the ECD practitioners should be well informed about HIV/AIDS transmission, prevention, treatment and care and should know in detail about the government programmes that can assist, particularly in respect of health and welfare and how to access such programmes. ECD practitioners have a basic training in child development but require further training regarding referral of children with emotional and behavioural problems arising from bereavement.

4. Discussion

Although the epidemic started later in South Africa than in countries further north, the high prevalence resulted in all the study sites in KwaZulu-Natal reporting the existence of many orphans and vulnerable children. Communities have been forced to adapt as the epidemic claims the lives of parents and children have been absorbed into the extended families. This adaptive process has become increasingly difficult, because the surviving adults have too many children to care for and caregivers are often elderly and lack the stamina and resources to provide adequate care and support for children or are too young and lack experience and commitment. In order that the rights of the children are protected the situation demands ‘safety-net initiatives’ (L Biersteker & N Rudolf, unpublished). ECD practitioners who have received basic training in education and care have the potential to fulfil such a role, because they are trained in the holistic care and support of young children and can advise caregivers on addressing children's needs at different stages of their development. This means stepping outside their immediate task of caring only for those children attending an ECD site or participating in an ECD programme to act as a local resource for children and caregivers in the community.

Key informants in this study repeatedly identified the local neighbourhood as the hub for improving ECD service delivery. The World Health Organization (2005) identified three sets of factors that influence ECD and this study has investigated the social factors at each of the three levels. The first set consists of factors within the family and includes stimulation, support and nurturing. The ECD practitioner can help families to understand these factors and fulfil this nurturing role. The second set, which influences the care, development and safety of children, consists of neighbourhoods and communities and the ECD practitioner can encourage community development and promote and facilitate community interest in orphans and vulnerable children. The third set is socio-political, at a more distal level, where the ECD practitioner has less influence. The concept of ECD practitioners as key figures in neighbourhood safety-net initiatives suggests that their training should be developed so that they can contribute adequately in the new role. They should be formally linked to other actors in the community, the financial security of their position should be strengthened and the quality of their work should be monitored.

Community members from all six study areas remarked on the large numbers of orphans. The information gathered about behavioural and environmental factors that have an impact on the well-being of orphans indicated the kind of activities that would enable ECD practitioners to effect improvements in the care of the children and to reinforce health messages to prevent HIV transmission. ECD activities at the local level could include parent and caregiver workshops facilitated by the ECD practitioners that would give practical advice on the care and stimulation of very young children. This would be in addition to their more usual function as the managers of an ECD site.

The Training and Resources in Early Education Siyafundisana stimulation programme is an example of a project where ECD practitioners work with women caring for children at home. ECD practitioners could also be trained to identify and refer children with disabilities and problems that require specialised treatment and care (Training and Resources in Early Education, 2006). Most ECD practitioners in the Siyafundisana programme are women and appropriate training contributes to their own development and the empowerment of women in their communities. Because they are locally based the benefits of the training remain in the community.

ECD practitioners could also provide a link to other development initiatives for families experiencing difficulties. These initiatives could, for example, include advice about nutrition and assistance with food gardening. ECD practitioners are well placed to network with the health services and to link needy families with community health workers who would visit homes to encourage basic health care such as growth monitoring and immunisation. By working together with community health workers to identify and assist needy families, the ECD practitioners could contribute to improved healthcare, hygiene and sanitation in their communities, thus minimising the spread of opportunistic infections. Setting up support groups could assist the caregivers of orphans and the ECD sites could provide a venue for these groups. With appropriate training and informational support, practitioners could identify children with more serious developmental delays and recommend to caregivers that they be taken to appropriate agencies for intervention. An important role for ECD practitioners is the early identification of children in need of anti-retroviral medication. Studies of HIV-infected children suggest that early identification and treatment will prevent or minimise neurodevelopmental disorders (Nozyce et al., 2006). ECD practitioners working in their own communities will probably know which children are HIV exposed and can advise caregivers accordingly.

An example of this kind of local focus is in Uganda, where Action for Children has initiated a ‘Grandparents Action Support’ programme to strengthen the carers' capacity to care for young children through psychosocial support, income generation, improved sanitation and hygiene, immunisation and growth monitoring for children, health care and nutrition support (Nyesigomwe, 2006). In South Africa, a recent initiative uses schools as centres of care and support and ECD sites could participate in such programmes (Department of Education, 2006).

Hemrich & Topouzis (2000) argued that HIV/AIDS is rooted in problems of under-development, such as poverty, food and livelihood insecurity, socio-cultural inequalities and poor support services and infrastructure. These are all constraints that the respondents in this study alluded to and the respondents saw poverty as a limiting factor in what could be achieved. There is an evident need to move from AIDS-specific responses to address this broader issue of improving cooperation across sectors. Hemrich & Topouzis (2000) described the structural, logistical and policy constraints that have hampered the implementation of a multi-sectoral response to the HIV epidemic and impeded South Africa's progress since 1994 (Republic of South Africa, 2004). There is evidence that the child care grant does make a difference to the health of the child (Case et al., 2003) and ECD practitioners should advise caregivers about accessing this resource and how to obtain the birth certificates required for it.

At the grassroots level, as this study suggests, poverty and disempowerment continue to fuel the myths that surround the infection. Issues of ‘bewitchment’ (witchcraft) and stigma influence responses at the community level, especially among rural communities, making the lives of orphans even more difficult. Similar beliefs have been reported elsewhere: Bawa Yamba (1997) explored the link between witch finding and AIDS in Zambia. He described a range of belief systems about HIV/AIDS that caused confusion in people's minds about preventive measures and delay in seeking early treatment and care. Traditional beliefs that infractions of traditional norms result in evil through witchcraft were prevalent and Bawa Yamba (1997) described how these beliefs were considered to be a major factor in the spread of AIDS in rural Zambia.

Witchcraft beliefs can probably be more effectively challenged from within a community. The inclusion of detailed information about HIV/AIDS in the training programmes for ECD practitioners offers an opportunity to use the ECD networks to facilitate improved understanding of the issues, reduce the stigma surrounding HIV/AIDS and reinforce the view that HIV infection has become another treatable chronic disease now that anti-retroviral treatment has become available. Most respondents in this study indicated that their information was obtained by listening to the radio. However, effective communication needs to be two-sided so that people can respond to the information provided and ask questions (Hubley, 1995). Use of local people such as ECD practitioners to expand on information and to advise and reassure may be a useful strategy for reinforcing radio messages so that the information is personalised.

Research has also shown that there is a link between traditional beliefs about HIV/AIDS and the presence of stigma (Kalichman & Simbayi, 2004). The information obtained from the key informants and focus groups in this study suggested a link between accurate information and acceptance of the value of testing, disclosure and anti-retroviral therapy. Denial of infection and bewitchment beliefs flourished when there was inaccurate information about the disease and in places where infected people were subjected to rejection and malicious gossip. Through their interactions with young parents, ECD practitioners may be able to motivate people to know their HIV sero-status by providing information about access to Department of Health programmes such as Voluntary Counselling and Testing and Prevention of Mother to Child Transmission and the provision of anti-retroviral treatment (Department of Health, 2003). This could help to dispel denial and, thus, reduce risky behaviour and increase the uptake of treatment.

The long duration of the AIDS epidemic in other sub-Saharan African countries is a reminder of the need for sustainable interventions. It is clear that there should be multiple strategies that are developmental in nature, building a strong coalition between community workers and professional health, education and social development staff. Green & Kreuter's (1991) Precede Model provided a useful tool for identifying factors to be addressed and could be used for other similar projects. The importance of promoting linkages and coordinating responses between sectors is evident. Using ECD practitioners, community health workers, clinic staff and social workers and working through schools would reinforce health promotion, the prevention of HIV and treatment compliance in those infected. The number of HIV infections has soared since the early 1990s and has continued to increase despite the government's awareness programme (Department of Health, 2000, 2005). The practical difficulties of mounting an inter-sectoral response require attention so that another cadre of community workers can be enlisted in the effort to stem the progress of the epidemic. The development of the ECD practitioners into a stable and effective workforce with relevant training in the care of vulnerable children offers a strategy for limiting the impact of the epidemic.

5. Conclusions and Recommendations

The Bill of Rights in the South African Constitution (1996) promises children adequate care and if children are hungry, alienated and uncared for the repercussions for our society are likely to be severe when they reach adulthood (Masmas et al., 2004). There is an urgent need to scale up support for orphans and vulnerable children and the use of ECD practitioners can contribute to this goal by promoting better care of children and understanding of their social and emotional needs. The evidence from key informants at the community level suggests that the involvement of ECD practitioners could be an important step in building up the continuum of care for children in families affected by HIV/AIDS. This strategy is in line with the government's integrated plan, but to make it a truly effective several aspects of the strategy need attention. The role of ECD workers needs redefining: the term ‘ECD practitioner’ should be extended to include people with a basic training in child care who are running parenting programmes, stimulation programmes or other initiatives that are not confined to a single site. There is value in this use of local resources, because an ECD practitioner lives in the community they serve and are well placed to respond to day-to-day needs. However, more training is needed so that practitioners can address the special needs of orphans and vulnerable children. In order to firm up the quality and sustainability of ECD provision, more financial and informational support is needed for practitioners and close monitoring in order to develop good quality programmes.