Table 2 Included studies 2006–2010

 Doherty et al., 2006

40 HIV+ women; mean age 24y; community health workers.

25 women EBF, 15 EFF. 80% (20) women who had chosen EBF introduced other liquids within the 1st month.

Women who chose to EFF had problems accessing formula milk; women were only able to maintain exclusive feeding practices for a short time due to a variety of constraints; Those who chose FF reported concern about HIV infection was the top reason for their infant-feeding choice.

 Doherty et al., 2006

27 HIV+ women; mean age 25y.

2 (18%) who initiated breastfeeding were still EBF at 12 wks.; 3 stopped between 6 and 12 wks.; Formula-feeders: 88% (14) did not give breast milk to their infants.

HIV-positive women encounter many challenges in maintaining exclusive infant feeding, such as insufficient support from health staff and family pressure.

 Sibeko et al., 2009

11 women (HIV+ and HIV-), mean age 25y

FF – 4 women; MF – 5; EBF – 2 women. 25% (1/4) women who chose to formula feed was able to EFF.

Non-disclosure of HIV status influenced feeding choices, which resulted in mixed feeding. Bad infant feeding advice created confusion and resulted in subpar practices such as mixed feeding.

 Andreson et al., 2013

14 HIV+ women & buddy pairs, > = 18y; data for 12 women

Study end date (2 to 6 months infants), − 6 women were EBF and 6 were FF.

Buddies can provide good support for HIV+ women.

 Bork et al., 2014

751 infants (366 boys, 385 girls)

Ever breastfed: Durban −57.1% (101), KwaZulu 80.9% (72). Breastfeeding (4-6 months): 41, 53% for Durban and KwaZulu-

Natal, respectively.

Not breastfeeding was associated with increased risk of serious infections particularly between 0 and 2.9 months.

 Ijumba et al., 2015

1821 (intervention) and 2136 (control) pregnant women, median age 23y. 30 Com. Health workers (CHW): 15 Intervention and Control group respectively;

76% (1242/1629) women in intervention and 74% (1380/1865) women in control initiated breastfeeding after birth; 34.4% (561) intervention and 32.5% (607) control breast-fed within 1 h of birth.

CHWs had positive effect on EBF, particularly on HIV- women.

 Bork et al., 2013

1028 HIV+ mother–infant pairs

56% (565) women were still breastfeeding at 3 months postpartum: 30% EBF, 18% predominant breastfeeding, and 8% partial. EBF: 90.4, 73.1, 69.9, 56.8, and 80.0% for Bobo-Dioulasso, Mombasa, Nairobi, Durban, and Somkhele, respectively.

Overall, EBF was brief, particularly for boys.

 Cournil et al., 2013

795 HIV+ women

77.7% 618 initiated breastfeeding at birth;

22.3% (177) mothers never breastfed.

By 6 months post-delivery, infants weaned/not breastfed had approximately seven-fold higher risk of dying compared with infants being breastfed.

 Tylleskär et al., 2011

2579 mother-infant pairs

EBF prevalence in South Africa at 12 weeks (24-h recall) in the intervention and control groups: 10% (56/535) 6% (30/485), respectively.

No differences found for the prevalence of diarrhea at age 12 weeks or 24 weeks between the clusters within the countries; No significant change in EBF rates for South Africa which were low at baseline.

 Doherty et al., 2014

964 HIV- mother–infant pairs; mean age 24y in Rietvlei and Umlazi and 25y in Paarl

34% (114/330), 20% (57/283) and 36% (127/351) of HIV-negative women in Paarl, Rietvlei and Umlazi, respectively, stopped breastfeeding before 6 months postpartum.

For HIV-unexposed infants: low birthweight and short rates of breastfeeding increased risk of hospitalization/death during 1st 6 months of birth.

 Doherty et al., 2012

Sub-group analysis of data for 999 women who ever breastfed; median age 22y in Rietvlei and Umlazi; 24y in Paarl.

At 12 weeks postpartum - 20% HIV- and 40% HIV+ women had stopped breastfeeding; 97% HIV- women and 42% HIV+ women ever breastfed.

Less than optimal early feeding practices; Early cessation of breastfeeding occurs among both HIV- and HIV+ women.

Engebretsen et al., 2014

2579 mother-infant pairs in 3 countries: 794 - Burkina Faso, 765 – Uganda, 1020 - South Africa

< 4% women in Burkina Faso and about 50% in South Africa initiated breastfeeding within the 1st hour of birth.

Behavior change may not have occurred in South Africa.

 Ramokolo et al., 2015

641 HIV unexposed children, median age 22 months

Infants using any breast milk changed from 89.3% (3 weeks) to 79.4% (12 weeks); Cereal use at 12 weeks −79.5% who were not breastfed, and 59.6% who were breastfed.

Infant feeding actions in the first 12 weeks can predict the development of childhood overweight and obesity.

 Goga et al., 2009

665 HIV+ & 218 HIV- women (age range 21–30) and infants;

Complete breastfeeding cessation (CBC) HIV+: 43.6% (88) reported CBC by 24 weeks.; HIV-neg: 97% initiated EBF.

Although national guidelines advise HIV+ women to breastfeed, they stop by 24 weeks; 39–44% of women practicing EBF/PBF at

week 3 complied with recommendations.

 Becquet et al., 2009

2190 HIV+ women, aged ≥16 years

90% of infants in both studies continued to be breastfed by 3 months of age; Mixed feed - 22% of infants by 3 months of age.

Breastfeeding duration is the primary contributor to HIV postnatal transmission; risk is quite similar for both exclusive & predominantly fed infants; Risk of postnatal HIV infection was 3.9% for infants breastfed less than 6 months, and 8.7% for those breastfed for at least 6 months.

 Bland et al., 2007

2491 women (1253 HIV+, 1238 HIV-); median age 25

HIV+ women (EBF −78%; replacement feeding 42%; HIV- women: 75% maintained intentions to EBF and < 1% (11) infants were not breastfed.

For most women counselling helped with matching infant feeding intentions based on existing resources for appropriate feeding. Most HIV+ women did not have resources for safe replacement feeding, so they decided to EBF.

 Bland et al., 2008

HIV-women (1219 infants); HIV+ women (1217 infants);

Median duration of EBF: HIV-, 177 days; HIV+, 175 days; EBF at 3 & 5 months: HIV- women (83.1, 76.5%); HIV+ women (72.5, 66.7%) respectively.

Both HIV+ and HIV- women can maintain EBF for 6 months with support in the home from trained lay counselors.

 Chetty et al., 2014

2340 women (1197 HIV-, 1143 HIV+); Median age: HIV- (21.8y); HIV+ (25.1y)

Median duration of EBF: HIV- (179 days); HIV+ (175 days); Birth to 5 months feeding patterns: EBF – HIV- 76.9% (920); HIV+  66.7% (762); Mixed feeding -

HIV- 11.9% (143); HIV+  8.8% (101); No breastfeeding – HIV-10.4% (125); HIV+  24.1% (275).

Breastfeeding did not increase postpartum weight loss; HIV+ women lost less weight during 1st 6 months & 12 months postpartum than HIV- women.

 Coovadia et al., 2007

2722 HIV+ and HIV- pregnant women, median age, 25.1y

83% (1132/1372) HIV-exposed infants initiated EBF from birth; median time for cumulative EBF, 159 days. EBF at 26 weeks – 37% (415/1132).

Mixed breastfeeding increased HIV transmission risk; EBF infants were less likely to get HIV than breastfed infants who used solid foods; HIV+ women can receive support to EBF.

 Doherty et al., 2007

635 HIV+ mother–infant pairs, mean age: Intent to EFF, 25.8y; Intent to EBF, 25.3y

13% who intended to breastfeed were EBF at 12 weeks; 42% mixed feeding; Predominantly breastfeeding (11%); EFF (33.5%).

Risk of getting HIV or death in both breastfeeding groups was high due to low rates of EBF; Inappropriate infant feeding choices were made based on the availability of 3 factors (piped water, electricity, gas or paraffin for cooking fuel, and early disclosure of HIV status).

 Ghuman et al., 2009

168 HIV+ & HIV-women, mean age 24y

97% (163) infants got breast milk as their first feed, 3% (5) were formula-fed; At 14 weeks: EBF was 18%; 52% got water; and 73% solids; 87% (20/23); Week 14: 11% HIV+ women were EBF, and 63% (12/19) mixed feeding.

Most women were not adhering to recommended infant feeding guidelines by 14 weeks of age; Women were unable to maintain EBF. HIV+ mothers breastfed at birth and more likely to formula feed than HIV- women.

 Goga et al., 2012

665 HIV+ and 218 HIV- women; Median age: HIV-pos 25; HIV-neg 23

EBF at 3 weeks: HIV+  42% (130) vs HIV- 17% (33); 47% (271) HIV+ women reported no breastfeeding; HIV- women at weeks 3 and 12: 17% (33) and 3% (5) practiced EBF.

While feeding practices were subpar among both groups of women, HIV-positive women engaged in more safer practices.

 Matji et al., 2009

222 HIV+ and 53 HIV- women;

At 6 weeks: 94% HIV- mothers were breastfeeding, 69% HIV+ mothers were FF; Intro of food by 6 weeks: 14% HIV- mothers; 1 HIV+ mother had stopped breastfeeding by 6 weeks.

Influences in the home environment resulted in changes in infant feeding practices.

 Patil et al., 2015

2053 infants; median age 3–12 days

Breastfeeding initiation within 1 h - 59.7%; EBF at 30 days - 29.5%; Partially Breastfed at 30 days - 36.6%; Completely weaned by 30 days - 2.6%.

Shift from EBF in the first month of life. Liquids and solids were usually given to infants in the first month.

 Rollins et al., 2013

2789 women; Median age: HIV-pos 25.0y; HIV-neg 21.7y

81.4% HIV+ and 92.9% HIV- mothers EBF at 6–8 weeks; 61.8 and 72.6% at 3–4 months; median time for cessation of breastfeeding-171 days.

EBF was associated with less adverse events with mixed feeding or not breastfeeding in both HIV exposed and unexposed infants.

 Rossouw et al., 2016

47 mother–infant pairs, 25 HIV-exposed and 22 HIV-unexposed infants

HIV-exposed infants: One mother initiated breastfeeding and continued up to 18 months; HIV-unexposed infants: all mothers initiated breastfeeding with > 50% within an hour of birth. 62 and 52% of HIV-unexposed mothers were breastfeeding at 12 and 18 months respectively. No infant EBF at 6 months.

Among both groups, there was low compliance with breastfeeding guidelines and dietary diversity.

Breastfeeding rates were low in HIV-exposed infants due to free formula distribution at health facilities.

 Ahmadu-Ali et al., 2013

386 women, mean age 25y

At 6 weeks: 53.1% (205) women were EBF; 26.6% (103) EFF; 20.3% (78) mixed feeding. EBF at 6 weeks: 52.7% (157) – HIV-neg vs 60.6% (43) HIV-pos.

HIV- women reported more counselling during antenatal care than HIV+ women; EBF was lower in HIV+ than HIV- women.

 Faber et al., 2007

505 infants, mean age 9 months; 441 mothers, 64 caregivers, mean age 25y

Breastfeeding initiated - 96% infants; breast milk only - 58%; mixed feeds - 23%; bottle feeds only - 18%. 61% infants had solid foods before 6 months.

Exclusive breastfeeding to age 6 months was rarely practiced. Trained community health workers should help with poor infant feeding practices and micronutrient deficits.

 Kyei et al., 2014

2660 women, 13–50 years

42.5% vs 57.5% women were still breastfeeding; 70.6% of those who stopped breastfeeding breastfed < 24 months.

Study showed duration of breastfeeding in Vhembe district decreased from > 24 months to just 18 months.

 Ladzani et al., 2011

815 HIV-positive women; mean age - 27.7 years

50% EFF, 35.6% EBF, 12.4% mixed feeding. EBF within 1 h of delivery - < 50%; EFF within 1 h of delivery - > 50%.

Knowledge gaps of PMTCT and infant feeding policy contributed to inappropriate feeding choices. Variables associated with mixed feeding: Vaginal delivery; infant hospitalization, and currently pregnant. Variables associated with FF: older age, knowing the HIV status of the infant; and higher HIV transmission/breastfeeding knowledge.

 Ukpe et al., 2009

33 mothers & infants; Mean age women − 30.7y; Mean age infants - 3.5 months.

ERF - 50% (15/30); EBF - 27% (8/30); − Mixed feeding 23% (7/30).

ERF was the most frequent infant feeding practice. Women who FF used different types of commercial milk. Quality of counselling should be strengthened to enhance infant feeding practices.

 Yako and Nzama, 2013

60 women, mean age 26.5y; 53% (32) were HIV-negative and 46.7% (28) were HIV-positive

At six weeks: EBF group – 13.3% (8/19) breastmilk only; mixed feeding – 11.7% (7/19). EFF group – formula only 30% (18/41); Mixed feeding – 15% (9/41).

Educate mothers on best infant feeding practices, including non-introduction of foods/liquids at inappropriate ages.

 Zunza et al., 2011

95 HIV-positive mother-infant pairs, Mean age, 27y

EFF - 97% (62); Formula – 78% (50); EBF – 2 women; Mixed feeding – 19% (12).

Advice needed on breast health during breastfeeding period and optimal infant feeding practices.

 Gbadamosi et al., 2017

186 infants, ages 1 to 12 months

0.6% infants were EBF for > 3 months;

78% breastfeeding at 9 months; 39.5% mixed feeding by end of 1st month; 0.6% EBF > 6 months.

Complimentary foods provided at an early age; EBF was rarely practiced; Interventions needed to support and promote recommended infant feeding guidelines.

 Aku A., 2013

125 HIV-positive mother-infant pairs, mean age, 27.8 years

Replacement feeds −84.3%; Mixed feeding −11.2%; Intro of solids − 10.4%; Mean age for intro of solids - 47 days.

Infant feeding choices influenced by family; SES factors affected growth and nutrition of HIV-exposed infants.

 Abusomwan, Osaigbovo Ebenezer, 2011

395 HIV-pos mothers; 14 to 49 years

EBF - 77.7%, 6 weeks after delivery; Mixed feeding - 3%; replacement feeding - 19.3%.

EBF was the primary infant feeding choice and practice; Hardly any mixed feeding occurred in this group of HIV-positive women.

 Jacobs-Jokhan, D., 2011

200 HIV-pos women, mean age 30y

EFF - 84.5%; EBF −14%; Mixed feed − 1.5%.

Study showed that babies born to mothers who did not receive infant feeding counselling were twice as likely to be HIV positive; Infant feeding counselling is necessary component of antenatal care; HIV-pos women should be counselled soon after diagnosis and throughout care.

 Masters, D., 2006

42 HIV-pos women, <  19 to 49 years

Exclusively formula milk and water - 52% (22); Mixed feeds - 48% (40).

Cultural norms influence infant feeding practices, particularly introduction of solids/liquids. Women EFF wanted to prevent HIV transmission.

 Mushaphi et al., 2008

185 mother-infant pairs; mean age, 25.83y

EBF (0–6 months) - 7.6% of women; 97% - still breastfeeding; 3% had stopped; 43,2% gave solid foods at three months, and 15% < 2 months.

Early introduction of other foods; Although breastfeeding was practiced by many of the mothers, EBF was rare.

 Some et al., 2017

1225 mother-infant pairs- all sites; 222 from East London, South Africa; > 18 years

EBF- first 3 days - 93.4% (199); Mixed feeding first 3 days - 2.3% (5). Breastfeeding initiation within 1 h of birth – 57.7%. Median duration of any breastfeeding was

40.6 weeks.

More mothers in South Africa had to return to work after a few months, stopping them from continuing to breastfeed; Improvements needed in breastfeeding and complementary feeding of children, particularly those who are HIV-exposed.