Share this article on:

HIV, infant feeding, and survival: old wine in new bottles, but brimming with promise

Coovadia, Hoosen Ma; Coutsoudis, Annab

doi: 10.1097/QAD.0b013e32826fb731

From the aVictor Daitz Professor of HIV Research, South Africa

bDepartment of Paediatrics and Child Health, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Congella, South Africa.

Received 8 March, 2007

Accepted 15 March, 2007

Correspondence to H.M. Coovadia, Victor Daitz Professor of HIV/AIDS Research, Nelson Mandela School of Medicine, University of Kwazulu/Natal, 719 Umbilo Road, Congella 4013, South Africa. E-mail:

A string of new studies, most of which have been presented at international meetings and are as yet unpublished, have resulted in a gathering momentum of interest in HIV and infant feeding. In rich and some middle-income countries, transmission of HIV antenatally, and during labour and delivery, has been substantially reduced by antiretrovirals and other interventions. Postnatal transmission was prevented by advising against breastfeeding; the alternative was replacement feeding with formula milks as these could be safely prepared, were affordable, and were culturally acceptable. The discourse on formula feeding or breastfeeding by HIV-positive mothers in developing countries has been mired in confusion or locked in robust disagreement. WHO/UNICEF/UNAIDS guidelines [1] offered a framework for making choices based on socio-economic conditions and household capability to prepare formula hygienically. The difficulty in the field was to offer an optimum balance of choices between the two types of milk feeding. Exclusive breastfeeding for 6 months is the recommendation for those choosing breastfeeding. Recent data from ongoing and completed studies in Africa have suggested that the effects of avoidance of breastfeeding or cessation even at 6 months by HIV-positive mothers can be disastrous. This finding repeats what has been and remains the bedrock of public health policy for infants and children well before the HIV epidemic, and recognized through centuries of human experience. The new findings have led to the WHO refining its Guidelines on HIV and Infant Feeding in late 2006.

Back to Top | Article Outline

New data on HIV and infant feeding

Results of recent studies and programmes to prevent mother-to-child transmission of HIV (PMTCT) allow examination of decisions which impact on infants receiving either no breastmilk at all or receiving breastmilk for short periods (4–6 months).

The strongest data comes from randomized controlled trials (RCTs) [2,3]. In Botswana [2] infants were randomized to receive either breastfeeding or formula milk for 6 months. Replacement feeding did indeed result in fewer HIV infections but it also led to increased infectious disease mortality; at 18 months, the combination of HIV infection and mortality (the ‘HIV-free’ survival) in the two arms was similar. In Zambia [3], the effects of early cessation of breastfeeding at 4 months, compared to continued breastfeeding, was investigated. By 24 months, the HIV-free survival in the two groups was similar: 17% of 329 infants who stopped breastfeeding early had died, compared with 19% of 331 infants who continued receiving breastmilk. Therefore these two RCTs have shown that neither ‘no breastfeeding’ nor a short period of breastfeeding, holds any overall advantage over continued breastfeeding, as the decrease in HIV transmission is countered by an increase in infectious disease mortality.

There are additional data. A study from Cote d'Ivoire [4] allowed women to choose either formula feeding from birth or exclusive breastfeeding from birth. HIV-free survival was similar in replacement fed and breastfed infants. Further, programmatic evidence is available from Botswana, a stable, democratic, middle-income, high-growth-rate country which initiated a PMTCT programme in 1999 which provided free formula to all HIV-infected mothers. An explosive outbreak of diarrhoea in the first quarter of 2006 resulted in large numbers of fatalities and gave a sobering vision of the dangers of formula feeding among populations in which breastfeeding had been the norm [5]. The largest risk factor accounting for these unprecedented diarrhoea deaths was ‘not breastfeeding’ (odds ratio, 50.0; 95% confidence interval, 4.5–100).

Early results are available from two ongoing trials in Kenya and Malawi, in which the efficacy of maternal HAART for 6 months postpartum, followed by early cessation of breastfeeding, is being assessed. The KiBS study from Kisumu, Kenya [6] reported increased growth faltering, diarrhoea incidence, and related hospitalizations, immediately following breastfeeding cessation. The BAN study in Malawi [7] also reported a significant increase in diarrhoea incidence and hospital admissions following early breastfeeding cessation around 6 months of age. The benefits of HAART to maternal health in these two studies did not, as expected, show some downstream health improvements in the infants. The PEPI study in Malawi [8] is testing three different infant antiretroviral prophylactic regimens administered during breastfeeding. Mothers are counselled to stop all breastfeeding at 6 months of age. The early reports indicate an unexpectedly high incidence of diarrhoea and diarrhoea-associated mortality in the period immediately following breastfeeding cessation, when compared to a historical control with continued breastfeeding [8]. Finally in the HIVIGLOB trial in Kampala, Uganda [9], which is testing the efficacy of a perinatal vaccine to prevent HIV transmission, mothers are also counselled to discontinue breastfeeding at 6 months. Early results suggest that breastfeeding cessation was associated with an increased risk of serious gastroenteritis.

Can infants who are already HIV infected also benefit from breastfeeding? Reliance on the adult model, where continued or new exposure to HIV results in a worse outcome, tended to favour cessation of breastfeeding. The Zambian study [3] reported that among 153 infants with HIV infection there was a significant benefit for continued breastfeeding; mortality at 12 months was higher in the group that stopped breastfeeding at 4 months (57%) compared with that in the group that continued breastfeeding (29%, P = 0.01). The evidence from this site corroborates data from an earlier study [10] and has, inter alia, been used to refine the earlier WHO guidelines [1] to recommend that ‘HIV-infected infants should continue receiving breastmilk for at least 2 years’ [11].

The adverse effects associated with early cessation (at or before 6 months) of breastfeeding (in HIV-exposed, HIV-uninfected infants and in HIV-infected infants) are not surprising given that, in infants above 6 months of age, breastfeeding continues to supply about 50% of the infant's energy requirements. Early cessation of breastfeeding may pose significant problems in developing countries where suitable high-energy, safe breastmilk substitutes may not be available [12].

Back to Top | Article Outline

Footprints of infant feeding and child health from yesterday

The foregoing are exactly the consequences of formula feeding in poor populations that were recorded repeatedly in the last century, and which have been known for hundreds of years [13–15]. Many contemporary studies have taken care to adjust for confounding variables and to account for reverse causality [16,17]. The most consistent effect of the benefit of breastfeeding has been in terms of protecting against diarrhoea. Feachem and Koblinsky [18] reviewed 35 studies on diarrhoeal incidence from 14 countries and found consistent and clear evidence of the protective effect of breastfeeding. A further review of the literature by Victora [19] showed that in addition to breastfeeding reducing the incidence of diarrhoea, breastfeeding reduces the severity of diarrhoeal episodes, and case-fatality rates. Importantly, Victora's review highlighted the point that recently weaned infants may be more vulnerable. The child survival benefits of breastfeeding in developing countries have recently been reinforced by reliable data [20,21].

Back to Top | Article Outline

Policy implications of new data

The implications of the data available on feeding for infants born to HIV-positive mothers globally, are listed here.

  • For the overwhelming majority of women in developing countries: make breastfeeding safe by minimizing HIV transmission and maximizing health and survival in infants and children.
  • For a minority of women in developing countries: make formula feeding safe by minimizing morbidity and mortality and maximizing health and survival in infants and children.
Back to Top | Article Outline

Priorities for research

Accordingly, the priorities for research are to investigate new ways of making breastfeeding safer for HIV-positive women as well as scaling up known positive interventions (see Table 1). There is also a need for making formula feeding safer.

Table 1

Table 1

Back to Top | Article Outline

Making formula feeding safe for HIV-infected women

One such attempt involves providing formula together with household goods to all women affected by HIV in very poor communities, to minimize contamination of the milk. The programme envisages provision of formula together with stoves, kerosene, bottles, and ‘when possible food aid and housing assistance’ [22]. Findings from a similar attempt to reduce mother-to-child-transmission of HIV in Haiti [23], suggest that such an approach is unlikely to be successful. The latter project encouraged formula feeding within a comprehensive programme at several levels of support; however, the Infant Mortality Rate remained extremely high (217 per 1000 live births). Such interventions come at a cost; infant morbidity and mortality costs, maternal effort and inconvenience costs, economic opportunity costs, equity costs of providing resources to some and not to others, and leakages due to sale of capital goods [24]. Therefore, though the intentions are noble, the solutions are impractical, and given the alternative of breastfeeding, they are unscientific.

The basic problem for such schemes is one of the scale and intractability of poverty and the unlikelihood that such interventions at individual household level can make a public health impact in African countries devastated by the AIDS epidemic. The recent UN Development Programme Report [25] and 2007 UNICEF Report on the State of the World's Children [26] indicated the massive social needs of the poor. There are more than 1.1 billion people and 2.6 billion globally without adequate access to clean water and sanitation, respectively. This leads to 1.8 million child deaths from diarrhoea and other diseases caused by unclean water and poor sanitation. In sub-Saharan Africa diarrhoea due to lack of water kills five times more children than HIV/AIDS. The world suffers from a ‘water crisis’ and this is not just due to absolute physical shortages of water supply, but is rooted in ‘poverty, inequality and unequal power relationships, as well as flawed water management policies’ [25]. Reversal of poverty will clearly take a long time – too long for HIV-infected mothers to wait. We need a strategy on safe infant feeding for infants of HIV-infected women now, while at the same time we work on medium and longer-term strategies to decrease poverty.

This is not to say that the restrictions on use of formula for poor populations necessarily apply without qualification to all groups in all developing countries. It is necessary, in some resource-constrained settings and middle-income countries affected by HIV/AIDS, to test the hypotheses that measures to empower the poor (for example through microfinancing or a basic income grant), provision of basic hygiene and food packages, and prophylaxis with affordable antibiotics, may allow the safe use of formula.

Back to Top | Article Outline


New studies among HIV-infected women in poor populations, have demonstrated the impact on child health when formula milks replace breastfeeding, either from birth or during infancy. Diarrhoeal diseases are frequent, hospitalizations become necessary, growth and development falter, overt malnutrition supervenes, and mortality rates increase; socio-economic costs may also be incurred. Measures to make breastfeeding by HIV-infected women safe are urgently required. These findings confirm the historical experience in the pre-HIV era and require a response similar to that period, when international mobilization led to codification of policies for the protection, promotion and support of breastfeeding.

Back to Top | Article Outline


1. WHO, UNICEF, UNAIDS. New Data on the Prevention of Mother to Child Transmission of HIV and their Policy Implications. Geneva: WHO/UNICEF/UNAIDS, 2000.
2. Thior I, Lockman S, Smeaton LM, Shapiro RL, Wester C, Heymann SJ, et al. Breastfeeding plus infant zidovudine prophylaxis for 6 months vs formula feeding plus infant zidovudine for 1 month to reduce mother-to-child HIV transmission in Botswana: a randomized trial: the Mashi Study. JAMA 2006; 296:794–805.
3. Sinkala M, Kuhn L, Kankasa C, Kasonde P, Vwalika C, Mwiya M, et al. No benefit of early cessation of breastfeeding at 4 months on HIV-free survival of infants born to HIV-infected mothers in Zambia: The Zambia Exclusive Breastfeeding Study. CROI 2007, 14th Conference of Retroviruses and Opportunistic Infections, Los Angeles, California, US, 2007 [abstract 74LB].
4. Becquet R, Bequet L, Ekouevi DK, Viho I, Sakarovitch C, Fassinou P, et al. Two-year morbidity-mortality and alternatives to prolonged breast-feeding among children born to HIV-infected mothers in Cote d'Ivoire. PLoS Medicine 2007; 4:1–12.
5. Creek T, Arvelo W, Kim A, Lu L, Bowen A, Finkbeiner T, et al. Role of infant feeding and HIV in a severe outbreak of diarrhea and malnutrition among young children, Botswana, 2006. CROI 2007, 14th Conference of Retroviruses and Opportunistic Infections, Los Angeles, California, US, 2007 [abstract 770].
6. Thomas T, Masaba R, van Eijk A, Ndivo R, Nasokho P, Thigpen M, Fowler M. Rates of diarrhea associated with early weaning among infants in Kisumu, Kenya. CROI 2007, 14th Conference of Retroviruses and Opportunistic Infections, Los Angeles, California, US, 2007 [abstract 774].
7. Kourtis A, Fitzgerald D, Hyde L, Tien HC, Chavula C, Mumba N, et al. Diarrhea in uninfected infants of HIV-infected mothers who stop breastfeeding at 6 months: The Ban Study Experience. CROI 2007, 14th Conference of Retroviruses and Opportunistic Infections, Los Angeles, California, US, 2007 [abstract 772].
8. Kafulafula G, Thigpen M, Hoover D, Li Q, Kumwenda N, Mipando L, et al. Postweaning gastroenteritis and mortality in HIV-uninfected African infants receiving ARV prophylaxis to prevent MTCT of HIV-1. CROI 2007, 14th Conference of Retroviruses and Opportunistic Infections, Los Angeles, California, US, 2007 [abstract 773].
9. Onyango C, Mmiro F, Bagenda D, Mubiru M, Musoke P, Fowler M, et al. Early breastfeeding cessation among HIV-exposed negative infants and risk of serious gastroenteritis: findings from a perinatal prevention trial in Kampala, Uganda. CROI 2007, 14th Conference of Retroviruses and Opportunistic Infections, Los Angeles, California, US, 2007 [abstract 775].
10. Coutsoudis A, Pillay K, Spooner E, Coovadia HM, Pembrey L, Newell M-L. Morbidity in children born to women infected with human immunodeficiency virus in South Africa: does mode of feeding matter? Acta Paediatr 2003; 92:890–895.
11. WHO. WHO HIV and Infant Feeding Technical Consultation held on behalf of the inter-agency task team on prevention of HIV infections in pregnant women, mothers and their infants. Geneva, October 2006. Accessed: 26 June 2007
12. Dewey KG, Brown KH. Update on technical issues concerning complementary feeding of young children in developing countries and implications for intervention programmes. Food Nutr Bull 2003; 24:5–28.
13. Lawrence RA. Breastfeeding: A guide for the medical profession. 4th ed. St Louis, Missouri: Mosby; 1994.
14. Cunningham AS, Jelliffe DB, Jelliffe EFP. Breastfeeding and Health in the 1098s: A global epidemiological review. J Pediatr 1991; 118:659–666.
15. Jelliffe DB, Jelliffe EFP. Human milk in the modern world. Oxford: Oxford University Press; 1978.
16. Habicht JP, DaVanzo J, Butz WP. Does breastfeeding really save lives, or are potential benefits due to biases? Am J Epidemiol 1986; 123:279–290.
17. Bauchner H, Leventhal JM, Shapiro ED. Studies of breast-feeding and infection: How good is the evidence? JAMA 1986; 256:887–892.
18. Feachem RG, Koblinski MA. Interventions for the control of diarrhoea diseases among young children: Promotion of breastfeeding. Bull WHO 1984; 62:271–291.
19. Victora CG. Infection and disease: The impact of early weaning. Bull WHO 1996; 4:390–396.
20. WHO Collaborative Study Team on the Role of Breastfeeding on the Prevention of Infant Mortality. Effect of breastfeeding on infant and child mortality due to infectious diseases in less developed countries: a pooled analysis. Lancet 2000; 355:451–455.
21. Bahl R, Frost C, Kirkwood BR, Edmond K, Martines J, Bhandari N, et al. Infant feeding patterns and risks of death and hospitalization in the first half of infancy: multicentre cohort study. Bulletin of WHO 2005; 83:418–426.
22. Farmer PE, Nizeye B, Stulac S, Keshavjee S. Structural violence and clinical medicine. PLOS Medicine 2006; 3:1686–1691.
23. Noel F, Wright PF, Bois P, Deschamps M-M, de Matteis P, Cassangnol R, et al. Contribution of bacterial sepsis to morbidity in infants born to HIV-infected Haitian mothers. J Acquir Immune Defic Syndr 2006; 43:313–319.
24. Coutsoudis A, Goga AE, Rollins N, Coovadia HM. Free formula milk for infants of HIV-infected women: blessing or curse? Health Policy and Planning 2002; 17:154–160.
25. Anon. Human development report 2006, beyond scarcity: power, poverty and the global water crisis. New York: United Nations Development Programme; 2006.
26. UNICEF. The state of the world's children 2007. Women and children: the double dividend of gender equality. New York: UNICEF House; 2006.
© 2007 Lippincott Williams & Wilkins, Inc.