Several studies have shown that support from HPs improves breastfeeding rates in general and EBF in particular [15–17, 20]. However, according to our observations, this support is non-existent or inadequate in some Niamey HFs. EBF promotion does not seem to be a priority for the HPs concerned.
Context and resources of the healthcare facilities
Our observations took place during various types of consultation where pregnant women, women giving birth and mothers of infants up to six months of age are seen, and also consultations involving vaccinations, healthy and sick infants.
Generally speaking, the internal organization and certain practices in the HFs were not conducive to promoting EBF effectively. Access to the premises was not controlled. We witnessed purely commercial activities unrelated to health care, such as auction sales. These activities went on during consultations, which distracted the mothers and increased the noise level, making discussions between mothers and professionals difficult . Often mothers entered the children's consulting room in groups of two or three and there were no baby changing tables or chairs for them to use. Such conditions were not at all conducive to being able to promote EBF effectively. In-depth one-on-one interviews to give advice to each mother were difficult to do and often done hastily. On the other hand, there were certain practices that should be maintained and encouraged in the HFs, such as the absence of nurseries, as recommended by the WHO . Babies stayed in the same room as their mother, which encourages mother-child interaction and the initiation and practice of on-demand feeding. In addition, HPs took away from the mothers all visible equipment used for mixed feeding, such as cups and baby bottles. However, HPs did not explain the reasons for these actions to the mothers and particularly the dangers of not exclusively breastfeeding, including the fact that the infant would not benefit from the antibodies in the mother's milk and be more vulnerable to infections such as acute respiratory infections and diarrhea diseases [1, 3, 5], which are some of the leading causes of infant mortality in Niger .
Human and material resources
Niamey HFs have to deal with staff shortages and difficult physical conditions. HP workforce levels in Niger do not meet WHO standards. As one midwife explained: "We are overwhelmed to the point of asking volunteers and trainees to help because sometimes we have as many as 450 deliveries per month." She added that: "Working conditions are very difficult. There is absolutely no disposable equipment. Everything is reusable and used again after disinfecting it with bleach. The facility does not have a generator so every midwife has to carry a flashlight to use during power cuts. If the water is cut off, the guard is the only person with a key to access the water tank".
Even though HPs followed certain practices recommended to encourage breastfeeding, such as immediate skin-to-skin contact after delivery, they did not offer any assistance with initiating breastfeeding in the first half hour after birth. They were more concerned with disinfecting the equipment and caring for the mothers (sutures, for example) and babies (e.g., measuring them) as well as women in labour. During all our observations, we never witnessed HPs providing practical sessions for mothers, such as showing them how to put the baby to the breast. The only situation we observed where EBF was explained in detail and putting the baby to the breast was checked and corrected involved a "femme relais" (FR) [femmes relais are women residing in the districts where the healthcare facility is. They are usually older women who have experience with childbirth and infant health and know most of the women of child-bearing age in the area they cover. They are volunteers who are trained by health professionals to educate the population on certain health-related matters. Their number varies from district to district]. Some HPs maintained that they did not have enough time to promote EBF. They said that each day they see "between 30 and 50 pregnant women just for prenatal checkups, not counting those who come for family planning" [midwife]. Also, in cases involving infants less than six months old suffering from diarrhea, the HPs were concerned with treating the diarrhea without thinking to ask about infant feeding method.
A valuable resource: "femmes relais"
The "femmes relais" are an important resource in promoting EBF, as was stressed by the HPs in the HFs observed and the discussion group participants. According to one nurse, four FRs were available to her HF, but did not really do much except on national vaccination days (NVD) for which they receive a bonus. Normally, they did things like cooking demonstrations, breastfeeding awareness, neighbourhood population mobilization during vaccination campaigns, and looking for women who stopped going to the HF. This HP added that: "they are very effective and know how to mobilize the women for all aspects, but they need a little bonus so they are more motivated" [nurse]. In fact, during our observations, the only time we witnessed detailed reasons given for recommending EBF was during a health education session conducted by an FR. She explained some of the benefits and practical aspects. She advised against bottle feeding and stressed its many disadvantages. This FR also demonstrated the correct position for breastfeeding, then asked each mother to put her baby to her breast so that she could identify and correct incorrect positioning. As the discussion group participants pointed out, this example suggests that it would be beneficial to organize and train FRs as a resource to do promotional activities and help reduce the HPs' workload. In an informal discussion, one FR said: "We currently participate in the final health education session [before mothers leave the maternity ward] but not on a regular basis; we do not generally do home visits except in rare cases where we find pregnant women who do not come to prenatal checkups. However, we are frequently mobilized during NVD campaigns."
Promotion of infant formula
Contrary to the International Code of Marketing of Breastmilk Substitutes, infant formula promoters [generally called medical delegates, they are pharmaceutical laboratories representatives who are in charge of doing the promotion of their products, which include medicines and breast milk substitutes] try to convince HPs of the need to use these products to feed infants. For example, concerning so-called "starter" milk designed for infants up to six months, one promoter explained to HPs that: "Even if we stress breastfeeding, it is impossible to do without canned milk [this refers to tins of powdered infant formula], because some mothers have difficulty breastfeeding and others don't have enough milk, some can afford canned milk, the process of expressing breast milk is unpleasant; and furthermore, Healthmilk [fictitious name] contains essential amino acids that strengthen the baby's immune system; it is very similar to mother's milk" [medical delegate]. A midwife supported these arguments, saying that "it is important to help mothers feed their babies" and adding "even if our maternity ward is Baby-Friendly" [midwife]. Another medical delegate came to promote "drugs to stimulate appetite and control fever, vomiting and diarrhea." He told HPs that "they were designed for infants from one month old," gave them samples and recommended that they prescribe these products. Promoting and prescribing these types of products encourage mothers to introduce foods as early as one month. Few posters showing breastfeeding mothers and recommending EBF up to 4-6 months could be seen in two of the three HFs we observed. However, there were others advertising infant formula, which inevitably discourages EBF. In one consulting room visited by about 70 mother/child dyads per day, there were no breastfeeding posters although a medical delegate had persuaded HPs to put up stickers advertising infant formula visible from the examination table. For example, when the mother of a healthy three-month-old who was growing well told one HP: "I want to start giving him canned milk because I want to start going to the sewing room," the professional did not talk to her about the option of expressing and preserving her breast milk, but prescribed baby "Healthmilk", showing the mother the sticker and saying: "This is good milk, you see the beautiful baby" [social worker]. The HP told the mother to bring in the powdered infant formula so she could show her how to use it. With practices like these, HPs are encouraging mixed feeding and giving mothers the message that breast milk substitutes are also a good choice for infant nutrition and health.
Niger has subscribed to the international initiatives aiming at promoting, protecting and supporting breastfeeding which are mainly: the Code of Marketing of Breast-milk Substitutes, the Innocenti Declaration and more recently the Global Strategy for Infant and Young Child Feeding. Despite all this, the participants stated, and we also noticed during our observation sessions, that the Code of Marketing of Breast-milk Substitutes, to which Niger has subscribed, isn't taken into account because the promotion of these products is still done in the healthcare facilities since there is no strict control. This constitutes, according to them, an important obstacle to the promotion of exclusive breastfeeding. There were, in the same facility, contradictory messages which created confusion among mothers and even among certain HPs. Obviously, in this particular healthcare facility, it is difficult to achieve optimal food practices such as exclusive breastfeeding. Moreover, according to the evaluation made by UNICEF and the Ministry of Health of Niger, between 1996 and 2005, 36 healthcare facilities received accreditation as Baby Friendly Hospitals. However, in 2007 only 13 retained BFHI status; the others lost it because they didn't respect all the Ten Steps for Successful Breastfeeding and the International Code of Marketing of Breast-milk Substitutes .
Practices of health professionals
Indifference or ignorance?
Health education group sessions given by the HFs on topics such as malaria, vaccination, prenatal checkups, prevention of mother-to-child transmission (MTCT) of HIV, etc., were usually organized by the HPs before the day's other activities. On average they lasted 15 minutes and did not include a question period. Some women did not understand the language in which they were given and there were no translation services. And EBF was rarely mentioned during these health education sessions or individual consultations. For example, out of 25 women seen in a day in one prenatal consulting room, breastfeeding was not discussed with any of them. Often HPs gave advice only to mothers who asked for it or had a particular problem, such as a low-weight baby. Our findings are similar to others reported in the literature . During our observations, we noted that the most of the time (at least 60%) spent with each woman during prenatal checkups was devoted to MTCT counseling. Aspects related to mother and child nutrition were rarely mentioned. Mothers-to-be were not given any preparation, advice or explanations about feeding options. They were only encouraged to talk about their health problems and were not given the opportunity to ask questions about nutrition. In addition, depending on funding, HPs were motivated to stress one topic more than another. This source of motivation was confirmed by some of the discussion group participants; as one nurse said: "For example MTCT is the biggest topic right now because that program is very well funded. There is very little funding for breastfeeding." Funding is a very important aspect, because in Africa in general and in Niger in particular, HPs are paid to attend continuing education and retraining sessions.
With regard to practical help with breastfeeding, we did not see any HP provide this service during all our observations. Although some mothers used incorrect feeding positions in front of the HPs, the HPs did not intervene to help them. These HP behaviours might be attributable to certain obstacles that were brought up by the discussion group participants . Lack of training was mentioned as the main obstacle to EBF promotion. During informal conversations, some HPs told us that during their academic training, they had only a two hour course about breastfeeding in general (not exclusive breastfeeding specifically) whereas the WHO recommends at least 18 hours of training including the practical and clinical aspects . And, in the syllabus of the main schools/institutes/faculty of HPs' training that we consulted, there is no indication of a course about breastfeeding. In fact, other studies have found that HPs have limited knowledge of EBF [23–25]. The following conversation illustrates this: To the mother of a four-month-old who said she was bottle-feeding her baby powdered infant formula, a HP did not advise her against mixed or bottle feeding but said: "Since he is four months old, you can also give him thin cereal and at six months you can start him on thick cereal" [social worker]. To some mothers with infants under six months, the same HP advised giving orange juice. She told other mothers to give nothing but breast milk, but did not tell them why or until what age. With some mothers, she never asked how they were feeding their infants at all. Contradictory advice could be confusing, especially if the mothers talk to each other afterwards.
We also observed another factor discouraging EBF promotion, particularly in the maternity unit where mothers had their first contact with their newborns and were probably more willing to listen to advice about good practices for the baby's health. Their stay in the unit was very short, only 24 hours after delivery, which limits the contact between HPs and mothers and the time to promote EBF. During these 24 hours, women only saw the HPs if mother or baby has a specific health problem. On the other hand, advantage was not taken of other good opportunities to promote EBF, such as when mothers returned for the baby's TB vaccination (which usually occurs in the first week after childbirth), and could have been told that their milk contains antibodies that act like a vaccine.
During one observation session, some HPs told us informally that "it will be hard to practice EBF in Niger because of people's beliefs and especially during the hot season" [social worker]. They themselves thought that "water at least must be given" [social worker]. These beliefs are detrimental to the promotion of EBF and also explain why HPs encouraged mothers to practice mixed feeding, as in the case of a mother of a three-month-old who said that she "gives Nutrimilk [fictitious name] in addition to breast milk because the baby regurgitates; I think it's the breast milk that he can't keep down and that Nutrimilk stays in his stomach longer." One HP asked a colleague to recommend Healthmilk (the medical delegate had visited that morning) or prescribe a special anti-regurgitation milk sold in the pharmacy. Rather than taking the opportunity to explain to the mother why this belief is unfounded and stop it from being repeated, she encouraged her to use infant formula. This lack of conviction on the part of the HPs seemed to predispose them not to promote EBF, and certainly not to act as leaders in their community.
Messages given to mothers: late, incomplete and contradictory
The verbal support given to mothers was often incomplete and sometimes the opposite of what is recommended. In the rare cases where HPs talked about EBF, they did not give any convincing explanation, as is shown in this advice given by an HP during a health education session: "Don't give anything to the baby, not even water, until six months because mother's milk is 90% water" [midwife]. The reasons for recommending EBF, its benefits and the disadvantages of mixed feeding or completely avoiding breastfeeding were not explained. Colostrum was also rarely mentioned; when it was, mothers were simply told not to throw it away, without any explanation of its role, its importance and the mistaken beliefs about it. Because mothers are usually very motivated to comply with the vaccination schedule, this could be used as an opportunity to explain that colostrum acts as the first vaccine and contains substances that protect the baby. For example, during a one-on-one consultation, one HP counseled a pregnant woman as follows: "You must eat until you are satiated so that you are strong enough to push during the delivery; you must also give your first milk to the baby to protect it against disease and not give anything except breast milk for the first six months" [midwife]. However, advising a mother to practice EBF without giving reasons, especially in view of the beliefs prevalent in Niger, is not enough to convince her and does not give her enough information to pass on to her family. The nutritional advice HPs gave was more concerned with supplementary foods that should only be started at the age of six months [37–40]. Besides, HPs often asked the question: "What does the baby eat?" to which the great majority of mothers replied, "cereal" and HPs added: "With breast milk, right?" This suggests that introducing cereal is a good idea. In addition, even mothers who may not be breastfeeding felt obliged to say they were. Also, some HPs continued to repeat the advice given in the child's health booklet, which completely contradicts the EBF recommendation. This official document for monitoring the infant's health and development is out-of-date but is still used in HFs and is a major source of confusion in the information it purveys, especially for mothers who can read.
Our study has several limitations since it does not cover all possible factors for EBF rate stagnation in Niger. First, the recruitment of the participants was on a voluntary basis and the health professionals may have changed their practices since they were being observed. Second, we have not explored in depth some dimensions such as: HPs' knowledge, their beliefs and the mothers' opinions about the EBF promotion, the attitudes and practices of HPs. Third, the results have not been validated with the participants and cannot be generalized as they could be different in the other regions of Niger. Finally, there could be some limitations of, and criticisms leveled against content analysis in qualitative research, including the subjectivity of the researcher .