- Research article
- Open Access
- Open Peer Review
Need to optimise infant feeding counselling: A cross-sectional survey among HIV-positive mothers in Eastern Uganda
BMC Pediatricsvolume 9, Article number: 2 (2009)
The choice of infant feeding method is important for HIV-positive mothers in order to optimise the chance of survival of their infants and to minimise the risk of HIV transmission. The aim of this study was to investigate feeding practices, including breastfeeding, in the context of PMTCT for infants and children under two years of age born to HIV-positive mothers in Uganda.
In collaboration with The Aids Support Organisation Mbale, we conducted a cross-sectional survey involving 235 HIV-positive mothers in Uganda. Infant feeding practices, reasons for stopping breastfeeding, and breast health problems were studied. Breastfeeding duration was analysed using the Kaplan-Meier method based on retrospective recall.
Breastfeeding was initiated by most of the mothers, but 20 of them (8.5%) opted exclusively for replacement feeding. Pre-lacteal feeding was given to 150 (64%) infants and 65 (28%) practised exclusive breastfeeding during the first three days. One-fifth of the infants less than 6 months old were exclusively breastfed, the majority being complementary fed including breast milk. The median duration of breastfeeding was 12 months (95% confidence interval [CI] 11.5 to 12.5). Adjusted Cox regression analysis indicated that a mother's education, socio-economic status, participation in the PMTCT-program and her positive attitude to breastfeeding exclusively, were all associated with a reduction in breastfeeding duration. Median duration was 3 months (95% CI 0–10.2) among the most educated mothers, and 18 months (95% CI 15.0–21.0) among uneducated mothers. Participation in the PMTCT program and being socio-economically better-off were also associated with earlier cessation of breastfeeding (9 months [95% CI 7.2–10.8] vs. 14 months [95% CI 10.8–17.2] and 8 months [95% CI 5.9–10.1] vs. 17 months [95% CI 15.2–18.8], respectively). The main reasons for stopping breastfeeding were reported as: advice from health workers, maternal illness, and the HIV-positive status of the mother.
Exclusive breastfeeding was uncommon. Exclusive replacement feeding was practised by few HIV-positive mothers. Well-educated mothers, mothers who were socio-economically better-off and PMTCT-attendees had the shortest durations of breastfeeding. Further efforts are needed to optimise infant feeding counselling and to increase the feasibility of the recommendations.
Widespread promotion of exclusive breastfeeding could prevent child mortality by 8% . However, transmission of HIV through breast milk has made breastfeeding counselling more complicated in low-income countries where HIV is prevalent.
Every year, more than half a million infants become infected with HIV. These infection rates are disproportionately distributed geographically; mother-to-child-transmission, in the context of antiretroviral prophylaxis is below 1% in Europe and the USA, but exceeds 30% in many poorly resourced countries, with Sub-Saharan Africa carrying the highest burden [2, 3]. The incidence of HIV infection among children has fallen in many areas, but this seems to be more related to a reduction of the HIV prevalence among mothers than gains in the PMTCT program .
In many countries where HIV is prevalent, the infant mortality rate is high. Considering the risks of both infant mortality and HIV transmission, breastfeeding is strongly correlated with a higher HIV-free child survival rate compared to formula feeding where the infant mortality rate is above 4% [5, 6]. Exclusive breastfeeding can be associated with higher HIV-free survival at 6 months than mixed feeding [6, 7]. Post-natal vertical HIV transmissions increase with a longer breastfeeding duration . Cessation of breastfeeding at the age of 6 months has consequently been recommended wherever replacement feeding at that age is acceptable, feasible, affordable, sustainable and safe [8, 9]. Failure to sustain replacement feeding and re-introduction of breastfeeding after initial cessation is particularly risky in terms of mother-to-child transmission of HIV . Infant feeding recommendations for HIV-positive mothers are confusing and have resulted in disadvantageous feeding patterns and mixed feeding in particular . Mixed feeding is associated with a higher morbidity and mortality risk than exclusive breastfeeding for infants of both HIV-positive and HIV-negative mothers, and with increased HIV transmission from HIV-positive mothers [6, 7, 12–16]. Exclusive breastfeeding is also associated with a reduced risk of breast health problems .
The frequencies with which exclusive breastfeeding, mixed feeding, prelacteal feeding and replacement feeding are practised differ widely throughout Africa . Region-specific measurements are essential to develop regional-specific recommendations.
We have investigated infant feeding practices, including breastfeeding, for infants and children under 2 years of age born to HIV-positive mothers in Mbale, Eastern Uganda.
The study was conducted during 2005 in Eastern Uganda in collaboration with The Aids Support Organisation (TASO). TASO is the largest national-based non-governmental organisation working with HIV-positive people in Africa. It is a grassroots movement providing counselling, information, support and medical treatment.
The study area included Mbale district, together with areas accessed through TASO Outreach Clinics in adjacent regions in the districts of Sironko, Pallisa and Kumi. Mbale district has a population of 720,000 of predominantly Bagisu people, with 90% living as subsistence farmers in rural environments. The overall literacy rate is 64% for men and 49% for women . Uganda has an HIV prevalence of 7.5% in women aged 15–49 years (2005) . In the period of the study, introduction of routine HIV counselling and testing was starting in Ugandan hospitals . The acceptance of testing had increased substantially. The national PMTCT program was introduced in Uganda first as a pilot in 1998 and more widely in 2001 .
This cross-sectional study collected information from 240 HIV-positive mothers with children aged 0–23 months. All mothers were recruited from TASO by consecutive sampling, and they participated voluntarily with informed consent. No mother contacted as a potential interviewee refused to participate. Five mother-infant pairs were excluded from the study because of missing information or the child was over 23 months old. Accordingly, 235 HIV-positive mothers were included in the study. Three pairs of data collectors who were fluent in Lumasaba (the local language), Luganda (the language of the central region) and English conducted the interviews with the mothers.
To check the reliability, 15 mothers were re-interviewed by another pair of data collectors 2–4 weeks after the initial interview. The answers showed only minor discrepancies and a high degree of consistency.
The structured interview contained topics concerning breastfeeding and feeding habits, feeding knowledge, mother's and father's education, occupation, household assets, time of HIV diagnosis, self-rated health , mother-rated health of infant and PMTCT program participation. A list of 30 liquid, semi-solid and solid foods was utilized with 24-hour recall, 1-week recall and recall since birth. Using a symptom-based semi-quantitative approach, we examined breastfeeding problems and reasons for stopping breastfeeding. The questionnaire was pre-tested to ensure correct understanding of the questions.
Double entry was done in EpiData 3.1, and SPSS 14 was used for data analysis.
Feeding information was based on World Health Organisation (WHO) definitions and recommendations , as follows. Exclusive breastfeeding: giving breast milk only, except for medicines and vitamin or mineral supplements; predominant breastfeeding: breast milk is nutritionally dominant, but with the possible addition of water-based fluids, fruit juices, tea without milk or oral rehydration salts; complementary feeding including breast milk (often referred to as mixed feeding): non-human milk, semi-solids or other solids given in addition to breast milk; replacement feeding: breastfeeding stopped or never being given any breast milk. Exclusive replacement feeding was defined as never having given any breast milk. Pre-lacteal feeding was defined as any food item or liquid other than breast milk given to the infant during the first 3 days after delivery.
Ethical approval was obtained from Makerere University, Faculty of Medicine Ethics and Research Committee, and the Uganda National Council for Science and Technology. Informed consent was obtained from each participant.
Baseline characteristics were examined with frequency tables. Feeding patterns were compared using χ2 statistics. Consistency was checked by Cohen's Kappa statistic.
The participants were grouped socio-economically into quintiles based on wealth assessment using principal component factor analysis . Housing characteristics and assets including toilet facilities, number of rooms and beds, roof material, lantern, radio, television, bicycle and vehicles were included in the model. The quintiles were based on the first principal component, a recognised method as a good proxy for household wealth . Breastfeeding duration in this cross-sectional study was estimated using Kaplan-Meier survival analysis with a Mantel-Cox log rank test to compare the estimates. Self-reported breastfeeding duration is used in the model. Cox regression analysis was used to estimate the independent impact of each factor on breastfeeding duration. Co-linearity, hazard plots and residual plots were checked. A backward "conditional" regression model was used in the multivariate analysis with removal set at a 0.1 level of significance.
The median maternal age was 30 years (inter-quartile range [IQR] 28–35) (Table 1). The age distribution of the infants and children was: 0–5 months: 37 infants; 6–11 months: 53 infants; 12–17 months: 65 children; 18–23 months: 80 children [see Additional file 1]. Median maternal education was 5 years of schooling (IQR 3–7). Fathers were more educated than mothers with a median of 7 years of education (IQR 5–10). Half the mothers were widowed.
Of 235 HIV-positive mothers, 215 (91.5%) initiated breastfeeding while 20 (8.5%) never breastfed their infants. Among 128 mothers who attended the PMTCT program, 18 (14%) avoided breastfeeding completely, while 2 of those not participating did not breastfeed (p < 0.001). Among the attendees and non-attendees in the PMTCT program, the proportions opting for exclusive breastfeeding during the first three days were not significantly different, 28.1% versus 27.1%, respectively. Ten among the 46 mothers in the better-off quintile did not initiate breastfeeding within the first three days. Nine of these 10 (90%) mothers continued with exclusive replacement feeding. Among the poorer 189 mothers, 11 (42%) out of 26 mothers who did not initiate breastfeeding within the first three days continued with exclusive replacement feeding, whereas 15 mothers introduced breastfeeding later on (p < 0.05).
Within the first two hours after delivery, 131 (56%) had initiated breastfeeding, with 178 (76%) having done so within the first day. Pre-lacteal feeding was given by 150 (64%) while 65 (28%) practised exclusive breastfeeding during the first three days. Replacement feeding was practised by only one of the 85 mothers diagnosed with HIV after delivery.
One-week recall and 24-hour recall gave similar results for the infant feeding patterns, giving a Spearman correlation coefficient of 0.96 among infants below 6 months of age and 1.0 above 6 months of age (Table 2). Among the infants less than 6 months old, one-fifth were exclusively breastfed and most were fed complementary including breast milk. Two-thirds of the infants older than 6 months were fed complementary including breast milk and the remaining third were replacement fed. The rate of replacement feeding increased with age. Among children aged 12–17 months, 37 (58%) were replacement fed, while 75 (93%) of children aged 18–23 months were replacement fed.
The median duration of breastfeeding was 12 months (95% confidence interval 11.5–12.5). Education was associated with a marked reduction in duration of breastfeeding, with a median duration of 3 months (95% CI 0–10.2) among mothers with more than 12 years of schooling, and a median of 18 months (95% CI 15.0–21.0) among mothers lacking education (Figure 1). This effect was seen both with and without adjusting for other factors with Cox regression with a 4.5 and 6.4-fold increase in hazard ratio for breastfeeding cessation, respectively (Table 3). The level of the father's education, whether the mother was a farmer, and the timing of HIV-diagnosis in relation to birth had similar effects on the crude analysis, but not in the adjusted Cox regression analysis. Breastfeeding duration differed substantially among mothers in the poorest and least poor quintiles. The median duration was 8 months (95% CI 5.9–10.1) among the least poor and 17 months (95% CI 15.2–18.8) among the poorest (Figure 2). Mother's age, marital status, rural or urban life, self-rated health of the mother and mother-rated health of the infant were not significantly associated with breastfeeding duration. Mothers counselled in the PMTCT program stopped breastfeeding earlier than those who did not attend the program. Duration of breastfeeding was shorter among mothers who had discussed breastfeeding with someone compared to those who had not. Those who considered exclusive breastfeeding to be beneficial for the infant, stopped breastfeeding earlier compared to those who considered it harmful. In a restricted analysis including children above 18 months of age, the median duration of breastfeeding was 12 months (95% CI 11.4–12.6).
Reasons for cessation of breastfeeding
Breastfeeding had been stopped by 116 mothers at the time of the interview. Advice from health workers, illness of the mother, and the fact that the mother was HIV-positive were reported as the main reasons for stopping breastfeeding (Table 4). Other reasons for stopping breastfeeding were: breastfeeding difficulties, perceived insufficient milk production, the notion that the child was "old enough" or "big enough", and that the child could eat without help. Family pressure, work and new pregnancies were reported by only a few mothers. Those stopping breastfeeding before the infant was six months old gave similar reasons to all the mothers stopping breastfeeding.
Fewer than half the breastfeeding mothers experienced problems relating to breastfeeding [see Additional file 2]. A fifth had problems with breastfeeding related to illnesses, such as generalised pain, frequent fever and a feeling of weakness. Breast pain, sore and cracked nipples, and swelling of the breast also burdened 40 (19%) mothers. Three mothers (1%) were diagnosed with mastitis or breast abscess.
This study shows that exclusive breastfeeding was uncommon among the HIV-positive women who opted to breastfeed. Thus, most infants received complementary feeding including breast milk from a very early age, which is an unfavourable situation. Our findings in Uganda agree with studies in other parts of Africa among both HIV-positive mothers and the general population [18, 27]. A positive finding is that breastfeeding duration is shortened by many HIV-positive mothers, especially among the well educated, the socio-economically better-off, and those who have attended the PMTCT program or discussed infant feeding with someone. Well-educated mothers breastfed for ~1 year less than their uneducated peers. Whether the shortened breastfeeding duration had any negative effects on the children was not assessed in our study. A recently published randomised study from Zambia indicates that early abrupt weaning of breastfeeding does not significantly reduce HIV-free mortality rates . In addition, prolonged breastfeeding gave a higher survival rate for HIV-positive children compared to those weaned early. HIV screening using a dried blood spot from infants was a feasible approach to the early identification of HIV-positive infants who may benefit from prolonged breastfeeding .
Exclusive breastfeeding of infants under 6 months old was less commonly practised among HIV-positive mothers than among the general population reported in the DHS-study in Uganda – 24% of HIV-positive mothers and 63.2% of the general population mothers according to the 24-hour recall data . Similarly, a study of the general population in the same area also reported higher rates of exclusive breastfeeding . Is this an effect of information about the risk of HIV transmission through breastfeeding reaching the HIV-positive mothers? Counselling on infant feeding in many African countries, including Uganda, has been reported to be suboptimal and may be one of the important reasons for the widespread practice of complementary feeding including breast milk . Health workers often overestimated the risk of HIV transmission through breastfeeding and many gave the impression that HIV transmission from mother-to-child is nearly universal [22, 31]. Based on key informant interviews, our impression was that replacement feeding was promoted more strongly in the national PMTCT program, whereas exclusive breastfeeding seemed to be more counselled in the non-governmental organisations working with HIV (unpublished data). Although breastfeeding duration was shorter among participants in the national PMTCT program than those not participating, we did not see differences in the rates of exclusive breastfeeding. Another explanation for the dominance of complementary feeding including breast milk may be the fear of making the infant totally reliant on breast milk, which could be particularly true for HIV-positive mothers .
In rural and semi-urban HIV-positive mothers in Eastern Uganda, replacement feeding was uncommon. This agrees with a Tanzanian study where replacement feeding with infant formula or cow's milk was seen as unacceptable or infeasible . Introduction of breastfeeding after initial replacement feeding was common except among the socio-economically better-off. Initiating breastfeeding after abrupt weaning is associated with increased viral loads of breast milk, and consequently could be hazardous .
Only 56% of the mothers initiated breastfeeding within the first two hours after delivery and 76% initiated it within the first day. A study in Ghana indicated a 2.4-fold increase in risk of neonatal death among infants for whom breastfeeding was not begun within the first day compared to those for whom it was . The authors of that study calculated that 16.3% of neonatal deaths could have been prevented if all neonates had been breastfed within the first hour. Pre-lacteal feeding was given by 64% of the HIV-positive mothers in our study. Not breastfeeding exclusively during the first days has also been shown to increase neonatal mortality [15, 33].
Although being HIV-positive was a major reason for stopping breastfeeding by 64% of the mothers, the median breastfeeding duration was 12 months. Breastfeeding duration among HIV-positive mothers was clearly shorter than among the general population, which in the DHS-study in Uganda was 19.9 months . It may seem counterintuitive that mothers perceiving exclusive breastfeeding to be beneficial were breastfeeding for a shorter time than mothers considering exclusive breastfeeding to be harmful. We interpret this as an indication that counselling had some impact, both in terms of increasing knowledge of infant feeding and of influencing behaviour.
Self-reported breastfeeding problems were similar to those given by mothers in the general population in the same area of Uganda, and were slightly more common than among HIV-positive and general population mothers in South Africa [17, 34]. The low rate of exclusive breastfeeding probably contributed to the higher proportion of breastfeeding problems. Mothers with breast health problems have a greatly increased risk of infecting their children with HIV [17, 35]. Data from South Africa indicate a greater than threefold risk of transmitting HIV from mother to infant when the mother had a serious breast health problem. Similarly, any breast health problems show an increased hazard ratio for HIV transmission compared to the absence of such problems .
The cross-sectional design of the study inherently left out diseased children. A similar cross-sectional study in 2003 provided a comparative group in the general population . A limitation of this study is that Kaplan-Meier and Cox regression analyses were conducted in a population where half the infants were censored at the last time-point due to ongoing breastfeeding, which may limit the precision of our estimates and could introduce bias. However, we observed similar associations in an analysis restricted to children with at least 18 month follow-up, and with much lower censoring rates. A third of the HIV-positive mothers acquired their HIV-status after delivery, which might influence the Cox regression analysis compared to the situation where all mothers were diagnosed HIV-positive before delivery. A restricted analysis excluding the mothers acquiring their HIV-diagnosis after delivery gave similar results in the Cox regression compared to when all mothers were included (not published). The use of anti-retroviral medicines was not recorded, but was not common even if roll-out started approximately at the time as the study in Mbale. The recall setting in this study was not optimal, and there may also have been socially desirable answers. It has been suggested that dietary recall once a week has high sensitivity and specificity for exclusive breastfeeding and other feeding patterns to a given age . In our study, recall periods of 24 hours and 1 week yielded similar results. Some studies have indicated that breastfeeding duration is overestimated to an escalating degree with increasing age [37, 38], while others have stated that breastfeeding duration is accurately reported . The fact that there was full agreement about breastfeeding duration between the initial and re-interviews reduces the likelihood that this measurement was significantly biased. The fact that all mothers were recruited through TASO may have caused a socio-economically skewed selection. We still feel confident that the data are representative of a large proportion of HIV-positive women in the region in Uganda.
There was a wide difference between the infant feeding practices in this group and WHO recommendations. Infant feeding recommendations for HIV-positive mothers have been confusing , which might explain the shortcomings of the practices. More beneficial practices among the well educated is a reason to increase the level of education, while also putting more efforts into counselling of less well-educated mothers.
Well-educated mothers breastfed for a substantially shorter time than their less well-educated peers. Mothers who were socio-economically better-off or had participated in the PMTCT program also breastfed for shorter durations.
Except among a limited group in this population, replacement feeding was not considered a realistic option in this rural setting. Complementary feeding including breast milk was the dominant practice for infants under 6 months old among the HIV-positive mothers.
There still seems to be many obstacles to optimal infant feeding. Further efforts are needed to optimise counselling on infant feeding and increase the implementation of the recommendations.
Demographic and Health Surveys
human immunodeficiency virus
Prevention of Mother-to-Child Transmission
Statistical Package for the Social Sciences
The Aids Support Organisation
World Health Organisation.
Bhutta ZA, Ahmed T, Black RE, Cousens S, Dewey K, Giugliani E, Haider BA, Kirkwood B, Morris SS, Sachdev HP, Shekar M: What works? Interventions for maternal and child undernutrition and survival. Lancet. 2008, 371 (9610): 417-440. 10.1016/S0140-6736(07)61693-6.
Lehman DA, Farquhar C: Biological mechanisms of vertical human immunodeficiency virus (HIV-1) transmission. Rev Med Virol. 2007, 17 (6): 381-403. 10.1002/rmv.543.
John GC, Nduati RW, Mbori-Ngacha DA, Richardson BA, Panteleeff D, Mwatha A, Overbaugh J, Bwayo J, Ndinya-Achola JO, Kreiss JK: Correlates of mother-to-child human immunodeficiency virus type 1 (HIV-1) transmission: association with maternal plasma HIV-1 RNA load, genital HIV-1 DNA shedding, and breast infections. J Infect Dis. 2001, 183 (2): 206-212. 10.1086/317918.
Dube S, Boily MC, Mugurungi O, Mahomva A, Chikhata F, Gregson S: Estimating vertically acquired HIV infections and the impact of the prevention of mother-to-child transmission program in Zimbabwe: insights from decision analysis models. J Acquir Immune Defic Syndr. 2008, 48 (1): 72-81. 10.1097/QAI.0b013e31816bcdbb.
Kuhn L, Stein Z, Susser M: Preventing mother-to-child HIV transmission in the new millennium: the challenge of breast feeding. Paediatr Perinat Epidemiol. 2004, 18 (1): 10-16. 10.1111/j.1365-3016.2003.00528.x.
Coovadia H, Kindra G: Breastfeeding to prevent HIV transmission in infants: balancing pros and cons. Curr Opin Infect Dis. 2008, 21 (1): 11-15.
Coovadia HM, Rollins NC, Bland RM, Little K, Coutsoudis A, Bennish ML, Newell ML: Mother-to-child transmission of HIV-1 infection during exclusive breastfeeding in the first 6 months of life: an intervention cohort study. Lancet. 2007, 369 (9567): 1107-1116. 10.1016/S0140-6736(07)60283-9.
Iliff PJ, Piwoz EG, Tavengwa NV, Zunguza CD, Marinda ET, Nathoo KJ, Moulton LH, Ward BJ, Humphrey JH: Early exclusive breastfeeding reduces the risk of postnatal HIV-1 transmission and increases HIV-free survival. Aids. 2005, 19 (7): 699-708. 10.1097/01.aids.0000166093.16446.c9.
WHO, UNICEF, UNFPA, UNAIDS: HIV and infant feeding: new evidence and programmatic experience: Report of a technical consultation held on behalf of the Inter-agency Task Team (IATT) on Prevention of HIV infections in pregnant women, Mother and their Infants, Geneva, Switzerland, 25–27 October 2006. [http://www.who.int/child_adolescent_health/documents/9789241595971/en/]
Thea DM, Aldrovandi G, Kankasa C, Kasonde P, Decker WD, Semrau K, Sinkala M, Kuhn L: Post-weaning breast milk HIV-1 viral load, blood prolactin levels and breast milk volume. Aids. 2006, 20 (11): 1539-1547. 10.1097/01.aids.0000237370.49241.dc.
Leshabari SC, Blystad A, de Paoli M, Moland KM: HIV and infant feeding counselling: challenges faced by nurse-counsellors in northern Tanzania. Hum Resour Health. 2007, 5: 18-10.1186/1478-4491-5-18.
Arifeen S, Black RE, Antelman G, Baqui A, Caulfield L, Becker S: Exclusive breastfeeding reduces acute respiratory infection and diarrhea deaths among infants in Dhaka slums. Pediatrics. 2001, 108 (4): E67-10.1542/peds.108.4.e67.
Bhandari N, Bahl R, Mazumdar S, Martines J, Black RE, Bhan MK: Effect of community-based promotion of exclusive breastfeeding on diarrhoeal illness and growth: a cluster randomised controlled trial. Lancet. 2003, 361 (9367): 1418-1423. 10.1016/S0140-6736(03)13134-0.
Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, Mathers C, Rivera J: Maternal and child undernutrition: global and regional exposures and health consequences. Lancet. 2008, 371 (9608): 243-260. 10.1016/S0140-6736(07)61690-0.
Edmond KM, Zandoh C, Quigley MA, Amenga-Etego S, Owusu-Agyei S, Kirkwood BR: Delayed breastfeeding initiation increases risk of neonatal mortality. Pediatrics. 2006, 117 (3): e380-386. 10.1542/peds.2005-1496.
Kramer MS, Kakuma R: Optimal duration of exclusive breastfeeding. Cochrane Database Syst Rev. 2002, CD003517-1
Bland RM, Becquet R, Rollins NC, Coutsoudis A, Coovadia HM, Newell ML: Breast health problems are rare in both HIV-infected and HIV-uninfected women who receive counseling and support for breast-feeding in South Africa. Clin Infect Dis. 2007, 45 (11): 1502-1510. 10.1086/523320.
Mukuria A, Kothari M, Abderrahim N: Infant and Young Child Feeding Update. 2006
Uganda Bureau of Statistics: Population and Housing Census Report for Uganda 2002. Entebbe: Ministry of Finance, Planning and Economic Development, Uganda. 2003
Hladik W, Musinguzi J, Kirungi W, Opio A, Stover J, Kaharuza F, Bunnell R, Kafuko J, Mermin J: The estimated burden of HIV/AIDS in Uganda, 2005–2010. Aids. 2008, 22 (4): 503-510.
Homsy J, Kalamya JN, Obonyo J, Ojwang J, Mugumya R, Opio C, Mermin J: Routine intrapartum HIV counseling and testing for prevention of mother-to-child transmission of HIV in a rural Ugandan hospital. J Acquir Immune Defic Syndr. 2006, 42 (2): 149-154. 10.1097/01.qai.0000225032.52766.c2.
Harms G, Schulze K, Moneta I, Baryomunsi C, Mbezi P, Poggensee G: Mother-to-child transmission of HIV and its prevention: awareness and knowledge in Uganda and Tanzania. SAHARA J. 2005, 2 (2): 258-266.
Chandola T, Jenkinson C: Validating self-rated health in different ethnic groups. Ethn Health. 2000, 5 (2): 151-159. 10.1080/713667451.
Gaillard P, Piwoz E, Farley TM: Collection of standardized information on infant feeding in the context of mother-to-child transmission of HIV. Stat Med. 2001, 20 (23): 3525-3537. 10.1002/sim.1092.
Wamani H, Tylleskar T, Astrom AN, Tumwine JK, Peterson S: Mothers' education but not fathers' education, household assets or land ownership is the best predictor of child health inequalities in rural Uganda. Int J Equity Health. 2004, 3 (1): 9-10.1186/1475-9276-3-9.
Filmer D, Pritchett LH: Estimating wealth effects without expenditure data – or tears: an application to educational enrollments in states of India. Demography. 2001, 38 (1): 115-132.
Coovadia HM, Bland RM: Preserving breastfeeding practice through the HIV pandemic. Trop Med Int Health. 2007, 12 (9): 1116-1133.
Kuhn L, Aldrovandi GM, Sinkala M, Kankasa C, Semrau K, Mwiya M, Kasonde P, Scott N, Vwalika C, Walter J, Bulterys M, Tsai WY, Thea DM: Effects of early, abrupt weaning on HIV-free survival of children in Zambia. N Engl J Med. 2008, 359 (2): 130-141. 10.1056/NEJMoa073788.
Rollins N, Little K, Mzolo S, Horwood C, Newell ML: Surveillance of mother-to-child transmission prevention programmes at immunization clinics: the case for universal screening. AIDS. 2007, 21 (10): 1341-1347. 10.1097/QAD.0b013e32814db7d4.
Engebretsen IM, Wamani H, Karamagi C, Semiyaga N, Tumwine J, Tylleskar T: Low adherence to exclusive breastfeeding in Eastern Uganda: a community-based cross-sectional study comparing dietary recall since birth with 24-hour recall. BMC Pediatr. 2007, 7: 10-10.1186/1471-2431-7-10.
Chopra M, Rollins N: Infant feeding in the time of HIV: rapid assessment of infant feeding policy and programmes in four African countries scaling up prevention of mother to child transmission programmes. Arch Dis Child. 2008, 93 (4): 288-291. 10.1136/adc.2006.096321.
Fjeld E, Siziya S, Katepa-Bwalya M, Kankasa C, Moland KM, Tylleskar T: 'No sister, the breast alone is not enough for my baby' A qualitative assessment of potentials and barriers in the promotion of exclusive breastfeeding in southern Zambia. Int Breastfeed J. 2008, 3 (1): 26-10.1186/1746-4358-3-26.
Leach A, McArdle TF, Banya WA, Krubally O, Greenwood AM, Rands C, Adegbola R, de Francisco A, Greenwood BM: Neonatal mortality in a rural area of The Gambia. Ann Trop Paediatr. 1999, 19 (1): 33-43. 10.1080/02724939992617.
Engebretsen IMS, Wamani H, Semyaga N, Karamagi C, Tumwine J, Tylleskär T: Cultural obstacles to exclusive breastfeeding in eastern Uganda: experiences and attitudes. 18th ICN (International Congress of Nutrition), Durban Sept 05. 2005
Embree JE, Njenga S, Datta P, Nagelkerke NJ, Ndinya-Achola JO, Mohammed Z, Ramdahin S, Bwayo JJ, Plummer FA: Risk factors for postnatal mother-child transmission of HIV-1. Aids. 2000, 14 (16): 2535-2541. 10.1097/00002030-200011100-00016.
Bland RM, Rollins NC, Solarsh G, Broeck Van den J, Coovadia HM: Maternal recall of exclusive breast feeding duration. Arch Dis Child. 2003, 88 (9): 778-783. 10.1136/adc.88.9.778.
Huttly SR, Barros FC, Victora CG, Beria JU, Vaughan JP: Do mothers overestimate breast feeding duration? An example of recall bias from a study in southern Brazil. Am J Epidemiol. 1990, 132 (3): 572-575.
Gillespie B, d'Arcy H, Schwartz K, Bobo JK, Foxman B: Recall of age of weaning and other breastfeeding variables. Int Breastfeed J. 2006, 1: 4-10.1186/1746-4358-1-4.
Launer LJ, Forman MR, Hundt GL, Sarov B, Chang D, Berendes HW, Naggan L: Maternal recall of infant feeding events is accurate. J Epidemiol Community Health. 1992, 46 (3): 203-206. 10.1136/jech.46.3.203.
The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2431/9/2/prepub
We thank Nulu Semiyaga, Herbert Mugooda, Harriet Mukiibi, Faith Kakai, Edward Kutusi, Sarah Wayero and Philip Kabiri for their efforts in collecting the data; TASO-Mbale including counsellors and administrative staff for their collaboration; and all the mothers and children who contributed to this study.
The study was funded by The Norwegian Programme for Development, Research and Education (NUFU) by grant no 43/2002 "Essential nutrition and child Health in Uganda." LTF, IE and TT were employed and funded by the University of Bergen. JKT was employed and funded by Makerere University. JW was employed by TASO. HW was funded by the Norwegian Quota Programme – Scholarship for Studies in Norway. The funding bodies had no influence on the study design, data collection, analysis and interpretation of data, writing of the manuscript or the decision to submit the manuscript for publication.
The authors declare that they have no competing interests.
LTF: design, implementation, analysis and writing. IMSE: design, analysis and co-writing. HW: analysis and co-writing. JW: implementation of the study and co-writing. JKT: initiation of the study and co-writing. TT: initiation of the study, design, implementation, analysis and co-writing.