Infant feeding practices and determinants of poor breastfeeding behavior in Kinshasa, Democratic Republic of Congo: a descriptive study

Millennium Development Goal (MDG) 4 calls for a two-third reduction in the under-five mortality rate by 2015. Between 1990 and 2010, global deaths among children under 5 years of age declined from over 12 million to 7.6 million [1]. However, according to Countdown to 2015, an organization that monitors progress towards reaching MDG 4, of 74 focus countries with available data for 2012 report, only 23 were on track to achieve the goal and 13 had made no progress. All but one (Haiti) of the countries that had made no progress are in Sub-Saharan Africa [2].

The Democratic Republic (DR) of Congo is one of the 13 countries which has seen no progress towards MDG 4. It bears the third largest burden of child deaths worldwide [3] and its under-five mortality rate has remained high: from 180 for every 1000 live births in 1990 to 170 in 2010. Although these deaths are the result of a web of complex determinants [4], there is enough evidence to believe that breastfeeding practices play a major role in the extremely high infant mortality in the DR Congo. First, results from the 2007 Demographic and Health Survey (DHS) [5] show that of the 9.2% of infants who die before their first birthday in DR Congo, 4.2% die during the neonatal period and the remaining 5% between 1 and 12 months. Second, of the 116 out of 1000 babies born alive in 2010 in DR Congo who survived through the first 28 days and subsequently died before their fifth birthday, 20 died from diarrhea and 23 from pneumonia, only malaria claim more under-5 lives (28) while AIDS accounts for only 2 [3]. More deaths in the postnatal period and the predominant role of diarrhea, pneumonia and malaria suggest that factors behind these deaths are to be found among other sources in the feeding practices. In fact, by the age of 6 months, more than 10% of children in DRC are already stunted, virtually 15% are underweight-for-age and approximately the same percentage emaciated [5].

These data have to be understood in the context of relatively high utilization of primary health services and high rates of breastfeeding initiation. Despite the challenges to accessing health care in DR Congo, DHS data showed that 85% of pregnant women attend at least one antenatal visit, 70% of live births occurred in a health facility (97% in Kinshasa) and of children 12 to 23 months of age, 71%, 59%, 45% received the first, second and third doses of DPT immunization administered according to the WHO immunization schedule (20) and the DRC’s Expanded Program of Immunization at 6, 10, and 14 weeks respectively, while 63% had been vaccinated against measles (at 9 months). Moreover, breastfeeding is almost universally accepted (9 out of 10 children are still being breastfed at the age of one) in DR Congo. Yet, recent national surveys [5, 6] showed that only 69% of 0 to 1 month old are exclusively breastfed while 65% of 2 to 3 month old are receiving something other than human milk in an environment where, according to the recent WHO/UNICEF progress report on sanitation and drinking-water [7], only 23% of the urban population have access to improved sanitation facilities and less than 50% to improved drinking-water sources.

Optimal breastfeeding practices, including immediate postpartum initiation of skin to skin contact with breastfeeding within one hour of birth, exclusive breastfeeding (EBF) with no additional fluid or food for 6 months [8], and continuation of breastfeeding thereafter up to 24 months and beyond with age appropriate complementary feeding, have great potential for reducing under five mortality rate [9–11]. If at least 90% of children were exclusively breastfed for the first six months of life, the potential reduction in mortality would be higher than from any other known effective intervention [10]. In most sub-Saharan countries, particularly in those with no or insufficient progress towards MDG 4, the prevalence of EBF among infant 6 months old or younger has not increased substantially and remains generally below 40% [12].

Starting in 1990, global initiatives to improve breastfeeding practices focused on maternity-level policies and practices known as the Ten Steps to Successful Breastfeeding, which serve as the basis for the Baby-friendly Hospital Initiative (BFHI) [13]. A maternity facility can be designated 'baby-friendly' when it has implemented the Ten Steps and has been reviewed using a national assessment approach. These steps include the following: (1) having a written breastfeeding policy that is routinely communicated to all healthcare staff, (2) training all healthcare staff in skills necessary to implement this policy, (3) informing all pregnant women about the benefits and management of breastfeeding, (4) helping mothers initiate breastfeeding within 30 minutes of birth, (5) showing mothers how to breastfeed and maintain lactation, even if they should be separated from their infants, (6) giving newborn infants no food or drink other than breast milk, unless medically indicated and not accepting free or low-cost breast milk substitutes, feeding bottles or teats, (7) allowing mothers and infants to remain together 24 hours a day, (8) encouraging breastfeeding on demand, (9) giving no artificial teats or pacifiers to breastfeeding infants, and (10) fostering the establishment of breastfeeding support groups and referring mothers to them upon discharge from the hospital or clinic. Implementation of the Ten Steps is associated with improvement in the rates of EBF [14–16].

BFHI is not being implemented to any extent today in DR Congo. The main attempt to implement BFHI steps in the country was led by UNICEF in the early 2000s as part of a national campaign of breastfeeding promotion, Overall 25 health facilities including 13 in Kinshasa out of more than 6,000 eligible facilities were certified through this effort. The last hospital certified was in 2004 in Katanga province when the funding stopped. Just two years after the peace deal that ended the deadliest war since World War II, which, in addition to decades of gross mis-management, have left the country infrastructures in shambles [17, 18], the country was simply not ready to take over the initial UNICEF efforts.

Beside the fact that BFHI is not systematically implemented in DR Congo, to our knowledge, there has been no published report on breastfeeding practices and factors that are behind the persistent high rate of early supplementation. Data on West Africa as a whole indicates that water supplementation was a major contributor to the high rates of less than exclusive breastfeeding in the early 1990s, with most neighboring countries reporting more than 50% offering water in the first months of life. In 2000, only about 35% of women in DR Congo were still providing water to their infants in the first months of life [19, 20], possibly accounting for the increase in EBF rates noted above.

This study was carried out as part of a preliminary phase of a cluster randomized trial designed to test whether support for the implementation the Ten Steps, coupled with provision of breastfeeding support in well-child and immunization clinics as a unique approach to the Step Ten (establishment of breastfeeding support groups in community), would improve the rate of EBF at six months in Kinshasa (NCT01428232). The main objective was to describe breastfeeding practices and identify the main determinants of the widespread early supplementation in Kinshasa, DR Congo.

A cross-sectional survey was carried out in six selected health facilities offering maternal and child health services in Kinshasa, DR Congo. The six health facilities were selected from a network of 44 maternity clinics that receive support from The University of North Carolina at Chapel Hill in partnership with the Kinshasa School of Public Health for implementation activities to prevent transmission of HIV from mother-to-child. To inform selection, study staff (two medical doctors) visited all 44 facilities between October and November 2011 and collected information on factors that might affect the quality of care provided in each facility. Health facilities were then stratified by location (urban or periurban) and type of management. Within each stratum, facilities were sorted by the number of births and the proportion of mothers returning for one week postpartum visit and matched across strata for the average workload (number of births/number of personnel). Six facilities with relatively comparable workload across strata, the largest number of births per month, and the highest return for the one week visits within each stratum were selected. This selection scheme was adopted to ensure that health care facilities at the periphery of the city, which might have lower patient volumes with lower socio-economic status than facilities in urban areas, were included in the final sample while limiting the overall heterogeneity among the selected facilities.

In each facility, all healthcare personnel of the maternity and well-child clinics and a convenience sample of mothers of healthy infants six months or younger presenting consecutively at the well-child clinic who agreed to participate in the study, were recruited and interviewed. Interviews were administered by two trained and bilingual (French and Lingala, the main language used in Kinshasa) interviewers using structured questionnaires developed and adapted from Infant Feeding Practice Study II’s [21] and the Demographic and Health Survey’s [22] questionnaires. The mothers’ questionnaire was available in French and Lingala and contained questions on socio-demographic characteristics, maternity experience, breastfeeding support received, infant feeding practices (including a 24-hour recall), and other questions related to knowledge, beliefs and attitudes regarding breastfeeding. Healthcare worker’s questionnaire was available only in French (the official working language) and contained questions on professional experience, past training on breastfeeding, knowledge, attitude and practice of breastfeeding support, and questions related to the implementation of Ten Steps in their facility. Each questionnaire contained both open and closed ended questions and took about 30 to 45 minutes to administer.

Information collected with questionnaires was double-entered into an EPI-Info database. All responses to open ended questions that were given in Lingala were immediately translated into French prior to data entry by our bilingual study staff. The data was summarized using median and interquartile range (IQR) for continuous variables and proportions for categorical variables. All responses to open ended questions were reviewed for themes, and similar responses were grouped together in simple counts. All analyzes were conducted using SAS 9.2 (SAS Institute, Cary NC).

The study was approved by the Institutional Review Board of the University North Carolina at Chapel Hill (study # 11–1332) and the Ethical Committee of the Kinshasa School of Public Health. All participants provided signed informed consent.

The goal of this study was to describe breastfeeding practices and identify potential determinants of the persistent low rate of EBF in Kinshasa. Using the Social Ecological framework [23], a conceptual model that addresses the importance of interventions directed at changing intrapersonal, interpersonal, organizational, community, and public policy factors which support and maintain unhealthy behaviors, our results suggest that the most important determinants of unnecessary liquid supplementation and early introduction of complementary food are mainly to be found in the first three levels of the framework: 1) the individual level characteristics such as knowledge, attitudes, beliefs, and skills; 2) the interpersonal factors, particularly the support from family, friends and healthcare providers; and 3) community factors such as community support groups or breastfeeding friendly hospital policy. The need for comprehensive support was also reflected in a recent review of community-based support for breastfeeding that concluded that interventions appear to have the greatest impact when they are part of multi-level, comprehensive interventions [24].

At the core of the widespread use of supplemental feedings is the limited knowledge among healthcare providers and mothers about human milk production, composition, and its adequacy for infants’ needs. This combined with the fact that over 73% of healthcare providers interviewed believed that the quantity of milk produced in the first 24 to 48 hours after birth is not sufficient for an infant’s needs with their expressed concern about low blood sugar, it is easy to understand why such a high number of infants are supplemented (in particular with sugar water) even before initiation of breastfeeding. Although concern about low blood sugar in newborns is justifiable, as it can cause brain injury [25], the best way to prevent low blood sugar in newborns is early initiation of breastfeeding with frequent feedings [26–28]. Moreover, recommendation or provision of sugar water to infants by healthcare workers contradicts the message on EBF and reinforces the cultural misbeliefs that “an infant needs to drink water”. In addition, of mothers who reported having received some breastfeeding support after birth or during well-child visits, the only somewhat consistent advice was on positioning, and nipple care. But it is unclear what nipple care might include that would not be potentially damaging.

Furthermore, it is of interest that despite the relatively high level of education among this group of urban mothers, up to 45% of them did not know the recommended duration of EBF. The fact that most mothers who gave solid food to their infants did so because of their belief that the child had reached the age to start complementary food, suggests that social desirability bias may be associated with either of the two responses. However, whether mothers knowingly gave a wrong recommended duration of EBF as a way of justifying their own introduction of complementary food early or because they did not know the recommended duration of EBF, may suggest that the knowledge and subsequent behaviors are modifiable with proper education and support.

Most mothers reported having someone at home who helps them take care of their infants and that this person has consider influence on how the infant is fed. Interventions aiming at changing some of the less than optimal breastfeeding behaviors should include an effort to reach and educate family members about EBF and its recommended duration. Half of the participating mothers reported receiving some breastfeeding support during well-child clinic visits suggesting well-child clinics could be used as potential alternative or supplement to community support for breastfeeding mother after discharge from maternity unit. But the effectiveness of well-child clinics as a unique approach for Step 10 merits further study.

Formula supplementation was observed mostly in the first three to four months of life. The primary reason given by mothers for formula supplementation was the impression that the baby was not getting enough milk. It is well documented that most mothers who stop exclusively breastfeeding do so because of the perception that their infant is not getting enough or is not satisfied with breast milk alone [29]. However, it is recognized that this ‘reason’ is often a proxy for other problems and constraints in the mother’s life [30]. The lack of readily available source of breastfeeding information and support following hospital discharge may explain, in part, the high frequency of formula supplementation [31, 32]. Moreover, in addition to the non-implementation of the 10 steps to successful breastfeeding, the lack of training of health workers combined with the absence of a written policy on breastfeeding in health facilities, limits the impact of the observed efforts to implement other steps of the BFHI.

Finally, it is of interest in this setting, where nearly all women breastfeed initially, that there was strong emphasis on positioning rather that patterns of breastfeeding. This may be due to the fact that this is the emphasis in many training materials developed in the industrialized countries, where few women have the opportunity to observe breastfeeding or have help from their family or friends who are successful breastfeeders. It may be important to adjust the content of breastfeeding training for health professionals in these settings to be less about latch on and more about patterns over time.

Despite the systematic approach used to describe breastfeeding behaviors and determinants, this study has limitations. Our data came from a small convenience sample of mothers and healthcare providers from a very limited number of health facilities in Kinshasa, DR Congo. Educated mothers who exclusively breastfeed their infants for the recommended duration may cluster in certain health facilities that might be seen as offering better quality of services. Even with the careful considerations that went into the selection of facilities that served as study sites, the small number of facilities (six) selected may not represent the average health facility in Kinshasa. Moreover, the small sample size of healthcare workers and mothers interviewed does not allow statistical testing, In addition, information on breastfeeding practices and reasons for supplementation were all collected through face-to-face interviews and are susceptible to recall or social desirability bias. However, the fact that the rate of EBF in our sample is comparable to the one reported in national representative samples [5, 6] suggests a limited impact of these potential biases and offers some confidence for applicability beyond the study sample.

This work supports the understanding that comprehensive approaches, especially including Step 2, training of health workers, and Step 10 Community support, may lead to changes in poor hospital practices that reinforce negative cultural beliefs and sub-optimal practices among mothers. This, coupled with the lack of community support groups, may account for much of the persistent low rate of EBF in Kinshasa, DR Congo.