Education And Debate

Public policies and the orphans of AIDS in Africa

BMJ 2003; 326 doi: (Published 19 June 2003) Cite this as: BMJ 2003;326:1387
  1. Alok Bhargava, professor (bhargava{at},
  2. Betty Bigombe, consultant
  1. Department of Economics, University of Houston, Houston TX 77204, USA
  2. World Bank, 1818 H Street NW, Washington, DC 20433, USA
  1. Correspondence to: A Bhargava
  • Accepted 17 March 2003

International help to care for Africa's orphans is essential not only for their immediate welfare but also to protect the long term prosperity of these countries. A researcher in child health and former Ugandan government peace minister assess how to make the best use of resources

The AIDS epidemic is wreaking havoc in sub-Saharan Africa. The HIV seroprevalence among young adults is nearly 40% in some countries,1 and millions of children have lost their parents. Although the extended family can alleviate these children's plight, it is unrealistic to assume that the children can escape from poverty without massive support from agencies such as the World Bank and the United States Agency for International Development. We visited Ethiopia, Malawi, and Tanzania in March 2002 as consultants to the World Bank to assess the ongoing programmes and to suggest strategies for improving child welfare. This article outlines our findings from visiting over 20 non-governmental organisations and national ministries responsible for caring for orphans of AIDS.

Embedded Image

Grandparents (and other relatives) often need help with funding education of orphans


Maternal and infant health

Although maternal nutrition, access to antenatal care, and vaccination programmes are important for improving infant health,24 the high prevalence of HIV among women in sub-Saharan Africa is a more urgent problem. The median survival time for HIV positive infants in Rwanda was 12.4 months.5 Antiretroviral drugs can reduce transmission from mother to infant, but so far only a tiny proportion of African women in pilot programmes have had access to these drugs.6

Several approaches are important for reducing the birth of HIV positive infants. The first is counselling about size of families. Demographic surveys in Ethiopia found that the ideal number of children was 5.6.7 However, couples' preferences depend on factors such as the need for children to generate income. When there is a drastic shock, such as parental death, to the households, parental attitudes may change. The five mothers receiving supplemental foods for their infants that we interviewed at the Abebetche Gobena orphanage in Addis Ababa, for example, wanted only two healthy children. This was presumably a result of clinical staff talking to the mothers about fertility and healthcare issues. The annual cost of caring for an infant in Abebetche Gobena was $471 (£314); the corresponding costs in Malawi ranged from $250 to $1700. These are very high for countries with gross domestic products of $110-$250 per capita.8

Secondly, the use of condoms has been emphasised in campaigns to reduce HIV transmission.9 In situations where women's low negotiating power makes using condoms unfeasible, it is important to provide other methods of contraception, especially if couples have surpassed their fertility goals. Improvements in the quality of family planning services will encourage their use. Recent figures show that the percentages of married women using standard Western family planning methods are 14% in Ethiopia, 45% in Malawi, and 33% in Tanzania.7 10 11 These figures are low and would increase with greater investments in the healthcare infrastructure.

Fostering of orphaned infants by the extended family is important, but relatives often cannot afford to support them. Without subsidies, many of the infants will end up in orphanages. The non-governmental organisation Wamata in Dar es Salaam provides food and supplies worth $9 a month to households fostering orphans. However, it can afford to help only 40 families. An annual subsidy of $100 to fostering households would cover the additional expenses of caring for infants and be cheaper than care in an orphanage.

Distribution of such funds is another issue. The Tanzanian Commission for AIDS estimates that the country has 1.5 million orphans. Funds need to be directed to fostering households through governmental agencies as well as non-governmental organisations, and these should also monitor children's welfare.

Preschool and school aged children

The effect of parental death on education is likely to vary. School enrolment rates in sub-Saharan Africa are already low, often below 50%. When a parent dies, older children may be expected to take up paid employment and care for younger siblings. The ability of bereaved children to continue in school depends on households' resources and the public support for education. Enrolment increased by 30% after school fees were abolished in Uganda,12 showing that poverty is an important factor.

Economic incentives are needed to achieve universal primary education. In Ethiopia, Malawi, and Tanzania, about $40 per year per child would be sufficient to cover the costs of tuition, uniform, supplies, and transport. Thus, fostering households and households where a parent is dying of AIDS should be guaranteed an annual subsidy of $40 per child provided that they send the child to school.

Currently, much of the care for school aged children in sub-Saharan Africa is delivered by non-governmental organisations. As there are so many orphaned children, the operational costs of these organisations need to be taken into account. For example, the annual administrative cost per child was $64 for SOS Kinderdorf International Village in Lilongwe, Malawi, and this was more than the $53 spent on food and shelter. Transferring funds directly from donor agencies to schools or the responsible ministries would reduce administrative costs.

Educational subsidies could be funded from debt relief programmes. The “estimated total nominal debt service relief” in 2002 was $1.93bn for Ethiopia and $1bn for Malawi.13 Because there are about 1.2 million orphans in Ethiopia, an annual subsidy of $40 per child would amount to $48m a year.

Lastly, the educational infrastructure is critical for learning, and teacher training is an important component. Teaching about strategies for reducing HIV transmission, such as promoting abstinence, should begin when children are young.14 As the first step, HIV positive school teachers should be given priority for antiretroviral drugs. Many African children would benefit from lower absenteeism among school teachers.


The policies for adolescents merit attention since HIV prevalence increases rapidly in this group. For example, young people in Zambia and Kenya experience peer pressure to seek multiple sex partners.15 16 Similarly, girls from low income groups in Mozambique may be forced into relationships to receive material benefits.17 Pilot studies by the International Labour Organization in Addis Ababa suggest vocational training programmes such as sewing, cooking, and knitting can help young women quit commercial sex. In two non-governmental organisations in Addis Ababa, the annual costs of providing such skills to adolescents were about $100 per person; the corresponding cost in Tanzania was $327. Organisations providing training could also promote use of condoms.

Premature death can result in shortage of skilled labour. This is important for productivity because life expectancy affects economic growth.18 Vocational training programmes should ideally be matched to demand in the labour markets. The World Bank has sponsored surveys of firms in Africa to investigate the determinants of productivity.19 Surveys of businesses employing blue collar workers such as electricians, carpenters, and plumbers are essential to assess the demand for labour. The non-governmental organisation Kiwohede in Dar es Salaam trains around 100 adolescents a year for mechanical, carpentry, and electrical jobs. Adolescents stay in the programme for 3–6 months at a cost of about $500.

Loans for vocational training programmes are likely to bring quick returns if the programmes are guided by the demand for labour. The World Bank has experience in lending for vocational training.20 However, small scale programmes run by nongovernmental organisations are hampered by factors such as high licensing fees. Because AIDS is reducing the labour force, organisations providing vocational training merit subsidies of $100-$500 per adolescent. It would be useful for the World Bank to set up national coordinating centres that facilitate access to and evaluation of vocational training programmes.


The AIDS epidemic in sub-Saharan Africa cannot be tackled without a concerted effort from international and national agencies. It requires a combination of pragmatic approaches. We advocate channelling funds from debt relief programmes directly to schools to increase enrolment and minimise transaction costs. Moreover, it would be cost effective to direct funds from the newly established Global Fund to Fight AIDS, Tuberculosis, and Malaria to maternal and infant healthcare programmes and to education. Young children without both parents should be given the highest priority. If children are cared for by their adolescent siblings, then such adolescents should be the first to qualify for vocational training. In the absence of a cure for AIDS, international agencies have an important role in providing technical advice as well as funds for maintaining the current levels of economic activity in sub-Saharan African countries.

Summary points

Maternal and infant healthcare programmes need scaling up and households fostering infants should be given annual subsidies of $100

Annual subsidies of $40 per fostered child are needed to increase attendance at school

Subsidies of $100-$500 per adolescent could provide vocational training and increase productivity

Subsidies could be largely funded by channelling money from debt relief programmes and other sources


  • See also editorial by Ammann

  • Funding The authors' field trip was supported by the World Bank. The opinions expressed here are those of the authors and should not be attributed to any organisation.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
View Abstract

Log in

Log in through your institution


* For online subscription