Journal of Perinatology (2008) 28, S46S52

r 2008 Nature Publishing Group All rights reserved. 0743-8346/08 $30

www.nature.com/jp

REVIEW

Early initiation of breast-feeding in Ghana: barriers and

facilitators
C Tawiah-Agyemang1, BR Kirkwood2, K Edmond2, A Bazzano2 and Z Hill2,3
1

Kintampo Health Research Center, Ghana Health Service, Kintampo, Ghana; 2Nutrition and Public Health Intervention Unit,
Department of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London, UK and 3Center for
International Health and Development, Institute of Child Health, University College London, London, UK
To explore why women in Ghana initiate breast-feeding early or late, who
gives advice about initiation and what foods or fluids are given to babies
when breast-feeding initiation is late. Qualitative data were collected
through 52 semistructured interviews with recent mothers, 8 focus group
discussions with women of child-bearing age and 13 semistructured
interviews with health workers, policy makers and implementers. The major
reasons for delaying initiation of breast-feeding were the perception of a
lack of breast milk, performing postbirth activities such as bathing,
perception that the mother and the baby need rest after birth and the baby
not crying for milk. Facilitating factors for early initiation included delivery
in a health facility, where the staff encouraged early breast-feeding, and the
belief in some ethnic groups that putting the baby to the breast encourages
the milk. Policy makers tended to focus on exclusive breast-feeding rather
than early initiation. Most activities for the promotion of early initiation of
breast-feeding were focused on health facilities with very few community
activities. It is important to raise awareness about early initiation of breastfeeding in communities and in the policy arena. Interventions should focus
on addressing barriers to early initiation and should include a community
component.
Journal of Perinatology (2008) 28, S46S52; doi:10.1038/jp.2008.173

Introduction
The World Health Organization (WHO) and United Nations
Childrens Fund (UNICEF) recommend that breast-feeding be
initiated within 1 h of birth because early initiation stimulates
breast milk production, increases uterine activity and may thus
reduce the risk of heavy bleeding and infection. It also fosters
motherchild bonding and increases the duration of breastfeeding.1,2 Recent observational data from Ghana suggest that early
initiation may also have an important impact on neonatal
mortality.3 Infants who were initiated to breast-feeding within 24 h
Correspondence: Dr Z Hill, Institute of Child Health, University College London, 30 Guilford
Street, London WC1N 1EH, UK.
E-mail: [email protected]

of birth were significantly less likely to die in the neonatal period

compared with those who were initiated after 24 h (adjusted odds
ratio 2.44, confidence intervals: 1.733.43). A marked dose
response of an increasing risk of neonatal mortality with an
increasing delay in initiation from 1 h to day 7 was observed, and the
results suggest that if all infants in the study area were breast-fed in
the first hour 22.3% of all neonatal deaths could be prevented. The
percentage of infants breast-fed in the first hour in the study area was
43%,3 similar to the level estimated for the rest of Ghana (46%). Rates
in some sub-Saharan countries are as low as 30%.4
Despite the inclusion of breast-feeding within an hour of birth
in international child feeding recommendations,1,2 much of the
focus of breast-feeding advocacy, and research, has been on
exclusive breast-feeding rather than on early initiation. The recent
finding that early initiation could substantially reduce neonatal
mortality3 should result in more comprehensive efforts to increase
the number of infants breast-fed within an hour of birth. Such
efforts can only be effective if policies and guidelines are based on a
solid knowledge of the barriers and facilitators for early initiation
of breast-feeding. Formative research recognizes that different
population groups view and make sense of the world in diverse
ways, and includes methods that allow us to gain an insiders view
of the barriers and facilitators to behaviors.5,6 This paper describes
the results of formative research to explore why women in Ghana
initiate breast-feeding early or late; who gives advice about
initiation; and what food/fluids are given to babies when initiation
is late. To understand how any future breast-feeding policies could
be implemented, data were also collected on the current breastfeeding policy environment in Ghana.

Methods
The study was conducted in Kintampo, a district in central Ghana.
Kintampo district falls within the forest-savannah transitional
ecological zone and has an estimated population of 147 000. The
district is overwhelmingly rural and farming is the most important
economic activity. Only the district town and 8 of the 149 villages
have electricity and few can be reached by paved roads. There is

Formative research on early initiation of breast-feeding

C Tawiah-Agyemang et al

S47

one district hospital and seven health posts staffed by nurses. There
is one private clinic run by a medical doctor and several private
maternity homes.
Kintampo district falls within the ObaapaVitA study area. The
ObaapaVitA trial is a large community-based, cluster-randomized,
placebo-controlled trial exploring the impact of weekly vitamin A
supplementation on maternal mortality. The trial started at the end
of 2000 and now extends through four contiguous districts in the
Brong Ahafo region of Ghana. All accessible villages in the four
districts are included in the ObaapaVitA trial area and more than
100 000 women aged 1545 years are under active surveillance
and are visited at home every month for capsule delivery and data
collection.
Data for this study were collected through 52 semistructured
interviews with recent mothers, 8 focus groups with women of
child-bearing age and 13 semistructured interviews with health
workers, policy makers and implementers. The focus groups aimed
to collect information on cultural norms related to breast-feeding
and were conducted between March and May 2005 in four villages
selected to reflect district differences in ethnicity, accessibility and
size. The participants were selected by community leaders and the
topics discussed included when breast-feeding is initiated, barriers
to early breast-feeding, beliefs about giving colostrum and who
influences breast-feeding initiation.
The semistructured interviews with recent mothers aimed to
collect information about personal experiences and beliefs related
to breast-feeding, and were conducted in June 2005. Mothers were
selected for interview from the 246 women identified from the
ObaapaVitA trial surveillance system as having given birth in
the last 2 months. The surveillance system also provided
socio-demographic and breast-feeding information, which was used
to purposively select 52 mothers with a range of timing of initiation
of breast-feeding and variations in factors that could be associated
with timing (place of birth, parity and ethnicity). Each mother was
interviewed and asked to describe her pregnancy and the birth.
They were also asked when and what the baby was first fed and
where they got advice about feeding.
The 13 semistructured interviews with health workers, policy
makers and implementers were conducted in August 2006. Health
workers were selected for interview if they had been a source of
antenatal care (ANC) or delivery care for any of the recently
delivered mothers in the semistructured interviews. The health
workers constituted seven nurses and midwives from the district
hospital, two private maternity homes and three health posts. Six
policy makers and implementers were identified at the district,
regional and national level by referrals from the District Health
Management Team and the Nutrition Unit of the Ghana Health
Services. The interviews with the policy implementers explored
current breast-feeding policies and priorities, determined how
policies and programs are developed and supported, and how
information is passed down to the regional and district levels.

The focus groups and semistructured interviews with recent

mothers were conducted in Twi, the local language, by trained
fieldworkers. The focus groups were recorded and transcribed into
English. In the individual interviews, the fieldworkers took field
notes and converted these to detailed English transcripts (fair
notes) at the end of each day. Interviews with health workers,
policy makers and implementers were conducted by the lead
researchers in Twi or English using the same procedure as in the
interviews with recent mothers. Data collection was iterative, with
the findings from the focus groups informing the design of the
interviews with recent mothers and the findings of the interviews
with recent mothers informing the interviews with health workers
and policy implementers/makers. The focus groups and
semistructured interviews were independently analyzed by the two
lead researchers through the systematic identification and coding
of themes. The findings were then discussed and consensus
reached. Quotes presented in the results section are from the
semistructured interviews unless otherwise stated. Quotes in the
third person are from fieldworkers notes and those in the first
person reflect fieldworkers verbatim notes.

Results
The results are presented in two sections. Section one describes
barriers and facilitators to early initiating of breast-feeding. The
data are drawn from all the respondent groups but focus is on the
data from the 52 recent mothers, as these case histories provide
insights into what women report actually doing rather than others
perceptions of their behaviors. The second section outlines the
findings from the interviews with policy makers and implementers.
The 52 recent mothers were purposively selected to ensure a
range of breast-feeding initiation timing and factors associated
with timing were captured. The characteristics of the sample are
shown in Table 1.
Reasons for delaying initiation
Lack of breast milk was the reason given by all 15 recent mothers
who reported delaying initiation for more than 12 h. Seven of the
22 mothers who initiated between 112 h also gave lack of breast
milk as a reason for not initiating earlier. The other reasons given
for initiating between 112 h (illustrated in Box 1) were the
performance of postbirth activities such as bathing of the baby or
the mother, the baby sleeping after the birth or needing rest, the
baby not crying for milk after birth and the mother needing to rest
after the labor.
The reported reasons for delaying initiation were corroborated by
the focus group and health worker interviews; however, women in the
focus groups also reported feeding problems as a reason for delaying.
Women in the focus group discussions and the recently delivered
mothers found it difficult to explain what they meant by not having
enough milk and described many different issues: physical signs that
Journal of Perinatology

Formative research on early initiation of breast-feeding

C Tawiah-Agyemang et al

S48
Table 1 Sample characteristics of the recently delivered women (N 52)

Box 1 Reasons for late initiation

Characteristics

n (%)

Age (in years)

<20
2130
3145

13 (25)
17 (33)
22 (42)

Residence
Urban
Rural

9 (17)
43 (83)

Education
None
Primary school
>Primary school

24 (46)
10 (19)
18 (35)

After the birth the nurse took the child and bathed it and then the child
slept and she also slept. She waited for the child to wake up and then gave
her the breast (24-year-old woman who gave birth at home assisted by a nurse
and breast-fed approximately 4 h after birth).

Ethnicity
Akan
Northern

27 (52)
25 (48)

Breast-feeding initiation
Within an hour
112 h
1324 h
More than 24 h

15
22
6
9

Baby didnt cry for milk (two women)

After the birth the baby didnt cry (for milk), which meant that it was not
hungry so she didnt feed him. She said that if the child had cried she would
have put it to her breast earlier. (30year-old woman who gave birth at home
and breast-fed after approximately 12 h).

Place of delivery
Hospital
Not hospital

19 (37)
33 (63)

Parity
Prima gravida
Multi gravida

14 (27)
38 (73)

(29)
(42)
(12)
(17)

breast milk is absent or is insufficient, beliefs about colostrum and

cultural beliefs about when breast milk arrives. The different
meanings of not having enough milk are illustrated in Box 2. The
belief that breast milk arrives on the third day after birth was
particularly strong in the Northern ethnic groups.
Many of the recently delivered mothers who felt that they did
not have enough milk to start breast-feeding performed actions to
help increase their milk or to bring the milk in. This included
eating specific foods, massaging the breasts with shea butter and
washing the breasts with herbs. Few of the women who
reported insufficient milk could explain why they did not start
breast-feeding with the little breast milk they felt they had; the only
explanation given was that it would be unfair to give the baby an
empty breast or one with not enough milk to satisfy the child.
Who made the decision about when to initiate breast-feeding
varied. The traditional birth attendants (TBAs) or the midwife/
Journal of Perinatology

Postbirth activities (four women)

She gave the breast to the child after she and the baby had bathed and she
had some porridge to eat. (30-year-old woman who gave birth at home and
breast-fed approximately 2 h after birth).
Baby needs rest or was asleep (two women)
It was her first child so it took a long time for her to push the baby out, a
lot of elderly people had to assist her. Immediately the child was born the
women bathed it with hot water, wrapped it and put it to sleep. The child was
very weak and they told her not to touch her because it needed rest. (22-year-old
woman who gave birth at home and breast-fed approximately 5 h after birth).

Women needed rest (seven women)

She started to breast-feed at 10.00 am as this was the time the delivery
pains had reduced and she could sit to breast-feed the baby (38-year-old
woman who gave birth at home and breast-fed after approximately 6 h).
She wanted to give the breast to the child the same day after birth but she
was dizzy and weak from the birth and she needed rest so she waited until
the next day (15-year-old woman who gave birth at the district hospital and
breast-fed after approximately 9 h).

nurse who assisted with the delivery often decided what happened
to the newborn in the first hour after birth. When the TBA or the
midwife took the baby away for bathing or to rest, breast-feeding
initiation was delayed usually by a few hours. Once the baby was
returned to the mother, TBAs or family members usually gave
advice to prima gravidas My mother told me to put the child to
my breast on the third day when the milk started coming. I havent
given birth before so I followed her advice (20-year-old woman).
Women who had delivered before were not usually given any advice
I didnt get any advice this time because I was old enough to
know what to do with my baby (22-year-old woman who
delivered at home and started breast-feeding approximately
54 h after birth).
In total, 35 of the recent mothers gave colostrum, 9 squeezed it
out and 8 could not be classified due to a lack of information. As
shown in Box 2, beliefs about colostrum delayed feeding among

Formative research on early initiation of breast-feeding

C Tawiah-Agyemang et al

S49
Box 2 Not enough milk
Physical signs of absent or insufficient breast milk
She didnt give breast milk because there was none in her breast she
squeezed and realized that nothing at all was coming out. She gave tinned
milk for 2 days. The breast milk came in the evening of the second day. (20year-old woman who gave birth at home and breast-fed approximately 41 h
after birth)
The breast was flat and nothing came out of it (15-year-old woman who gave
birth at home and breast-fed approximately 54 h after birth)
Beliefs about colostrum
She doesnt give the first breast milk to the baby on the first day after
birth. She gives the breast milk to the baby the next day after birth when the
first breast milk has mixed with the second breast milk. (focus group
respondent).
She squeezed the first breast milk away until the white milk came because
the former wasnt good. Then she breast-fed the baby with the white milk.
(28-year-old woman who gave birth at home and breast-fed after approximately
48 h).
Beliefs about the timing of the arrival of breast milk
It is their (Northern ethnic group) belief that there is no milk in the breast
when a woman gives birth, a woman has to eat hot food and drink warm
water for 3 days before the breast milk comes in and that was what she did.
(35-year-old woman who gave birth at home and breast-fed approximately 44 h
after birth).
You have to give the breast to the baby to suck whatever is in the breast
before the real milk will come (focus group respondent).

some mothersFthese were mostly from the Northern tribes. The

length of the delay varied between 10 h and 3 days.
Mothers who initiated breast-feeding between 112 h usually
gave their babies nothing at all or water, and those who initiated
after 12 h gave a variety of liquids including water, evaporated
milk, water with bread, water mixed with shea butter, coconut
water or milo (a malted chocolate drink). Most respondents felt
that a child needed to be breast-fed by day 34 of birth or else the
baby may get sick or even die. Out of the 38 women who had more
than one child, 33 women initiated breast-feeding at similar times
for all their children.
Facilitators for early initiation
A theme that emerged from the semistructured interviews and focus
group discussions with the Bono tribes was that a woman should
give the baby the breast to pull soon after birth to encourage milk
to come although there is no breast milk, you give the breast to
the baby to suck and the milk will come (28-year-old woman who
delivered at home and was advised by aunt). Another potential

Box 3 Health facility delivery facilitates early initiation

The nurses said that she should give her breast to see if the baby will
sucky. if the nurses hadnt asked her to feed the baby she wouldnt have
minded until the baby began to cry. (40-year-old woman who delivered in
the district hospital and breast-fed approximately 30 minutes after birth).
The nurse brought the baby to her and told her to start breast-feeding the
baby, she wanted to have some rest but the nurse insisted that she fed the
baby there and then. The nipple was hard and big so it was difficult for the
baby to take the breast but the nurse stood on her to give the breast to the
childy there was no milk in it as she could not feel anything coming out
when the baby was sucking. She told the nurse but she said that she should
put it in the babys mouth and through that the milk will come. (19 -year old woman who delivered in the district hospital and breast-fed approximately
30 min after birth).

facilitator to early initiation is that among the women who

initiated late, there was a desire among many to breast-feed early
she will breast-feed the baby immediately if only there is breast
milk (28-year-old woman who delivered at home and breast-fed
approximately 9 h after birth).
Delivering at a health facility supported early initiation
(see Box 3). When you deliver at the hospital they ask you to give
the breast to the baby even if there is no breast milk in it
(focus group respondent). Among the interviewed mothers, only 1
of the 15 who initiated more than 12 h after birth delivered at a
health facility, compared with 8 of the 22 who initiated between
112 h and 10 of the 15 who initiated in the first hour.
Advice on early initiation was given to 5 out of the 50 women
who attended ANC and by the family and friends of 2 women, but
was usually given only to first-time mothers. Although three out of
five women who got advice at ANC started feeding within an hour
of birth, only one gave advice from ANC workers as the reason for
initiating early.
The policy environment
Five out of the six policy makers and implementers interviewed
reported that the national policy promotes breast-feeding within
half-an-hour to an hour of birth and all the implementers reported
that early initiation was important. They reported, however, that to
date policy implementation has focused on exclusive breast-feeding
the message has not been balanced up to now and has focused on
exclusive breast-feeding rather than when to start and finish; they
(policy makers) got too excited about exclusivity to think of timing
(national-level breast-feeding policy implementer).
A lack of community-based strategies to promote breast-feeding
was reported by all levels of policy makers. In Kintampo
information about breast-feeding should be provided during ANC,
delivery (if at a health facility), postnatal care and at child welfare
clinics. Respondents reported that communities may also receive
information about breast-feeding through a radio series called
Journal of Perinatology

Formative research on early initiation of breast-feeding

C Tawiah-Agyemang et al

S50

Healthy Happier Home and through radio and television

broadcasts during world breast-feeding week.
The policy implementers reported that policies are usually
generated or modified based on WHO/UNICEF recommendations at
workshops and through discussions in a forum such as the Baby
Friendly Hospital Initiative Authority. These policies are then passed
down to the regions through the twice-yearly regional review
meetings. The national and regional respondents reported that
WHO and UNICEF additionally fund policy implementation in
some regions, but national-level respondents raised some concerns
that for policy making they (WHO/UNICEF) do try to ensure
national ownership, however when it comes to implementation
there are sometimes problems as they (WHO/UNICEF) have
districts they target which are not necessarily where Ghana Health
Service wants to work (national policy implementer).
At the regional level, changes in priority regions were reported
as having a detrimental effect on the continuty of breast-feeding
policy implementation. UNICEF funding for the Brong-Ahafo
region has now stopped, as UNICEF is focusing on other regions;
this has resulted in a shortage of funds for regional activities
(regional-level policy implementer).
The regional policy implementers reported that they send one
copy of the policy documents to each district and hold workshops
outlining the policy for the district-level implementers, who in turn
hold workshops for health workers in their district. When there is
no money for workshops, which is frequently the case, letters are
sent out about the policy change. Dissemination between the
regions and the districts was highlighted as a problem by all levels
of implementers, from the regions down there are no properly
defined dissemination channels (national-level implementer), and
no mechanism exists to check that guidelines reach their
destination.
As the districts and regions devise their own budgets and
priorities, the national implementers reported needing to lobby
district-level officials, to ensure breast-feeding is set as a priority
we can only appeal to the regions to adopt the policies and
prioritiesFyou need to call and be nice to them (national-level
implementer).
Materials such as posters and flip charts are also passed from
national to regional to district to health facility level. At each level,
a request often needs to be made before materials are assigned and
district-level implementers and health workers may then need to
make a trip to either the regional or district office to collect
materials. Respondents reported that this multitiered system does
not always function well (see Box 4). However, a national-level
policy maker highlighted that, in one of the regions, the
storekeeper for the region sends newsletters every month of items
he has in store for the districts to make their request and felt that
this initiative could be emulated by the other regions. None of the
health centers or hospitals visited during the study had any written
guidelines on breast-feeding policies. Few staff had attended
Journal of Perinatology

Box 4 Problems with policy implementation

Sometimes you find all the posters packed up in an office (national-level
policy implementer).
Sometimes a few people are called for workshops but there is no feedback
when they come back from those meetings. Instead of them organizing a
mini-workshop for those of us who could not attend the workshop, they do
not do that claiming that there is no money in the district. Sometimes letters
are sent about new policies but how and when it should be implemented
will not be stated (district-level policy implementer).
There is nothing for us to use. This is a challenge for us. Even posters are not
available to us. I dont remember the last time we received some flip charts
and posters. (health worker).

workshops or trainings, and only one health center had materials

that included a breast-feeding message (this was a safe
motherhood flip chart). Currently, we give verbal messages
without any flip charts or posters to aid the education (districtlevel policy implementer).
Health workers knowledge and practice
Despite the problems reported around the dissemination of breastfeeding information and materials and the focus of policy on
exclusive breast-feeding, health workers knowledge of the
importance of early initiation was high. Health workers reported
that they had been provided with information about initiation
during their nursing or midwifery training and several advantages
of early initiation were mentioned. These included that early
initiation helps the uterus to contract, helps with the expulsion of
the placenta, encourages the maternalchild bond, prevents
postpartum bleeding, helps establish lactation and encourages milk
flow, reduces the chance of the baby receiving prelacteal feeds,
prevents hypoglycemia and softens the first stool.
Although many health workers identified early initiation as
important, those who worked in ANC and postnatal care rarely
reported providing information about initiation. Most health
workers involved in ANC reported that they provide information
based on the 16 topics (one of which is breast-feeding) listed in the
maternal health record card. One topic is covered a month making
it impossible for a woman to receive all the information they
require even if they attend the number of ANC visits recommended
by the Ghana Health Service. Respondents involved in child welfare
clinics reported that because of staff shortages health education is
rarely given.
Health workers involved in delivering babies reported promoting
early initiation if the new mother had no problems related to the
birth. Two of the four health workers involved in delivery reported
asking the mother to breast-feed immediately after birth to
encourage the placenta to detach, when the baby is delivered I cut
the cord and wipe the baby with a dry towel, wrap it and place the

Formative research on early initiation of breast-feeding

C Tawiah-Agyemang et al

S51

baby on the mother to encourage contractions. If the placenta

comes out quickly before the cord is cut I still give the baby to the
mum to breast-feed to help stop the bleeding. The baby breast-feeds
for approximately half-an-hour after which I bath the baby and the
woman goes home to bath (health worker). The other two health
workers reported that they waited for the placenta to be delivered
and then usually let the mother and the baby rest for half-an-hour
to an hour before asking her to breast-feed. They reported delaying
breast-feeding because the mother and baby needed to bath or the
mother needed to rest.
Health workers and policy implementers knowledge of
constraints to early initiation
Most health workers and policy makers had good knowledge of why
women do not initiate breast-feeding early and felt that the
problem was mostly for those who delivered at home. Reported
reasons for delayed initiation included womens beliefs that milk
does not arrive for 3 days or that they do not have enough milk to
start feeding, negative beliefs about colostrum, activities such as
bathing, resting and eating, waiting for the placenta to be delivered
and beliefs that the mother needs to rest after the birth. Health
workers and policy makers also had a good knowledge of what
newborns are given if breast-feeding is delayed.
Although all of the policy makers and implementers reported
that mothers beliefs about not having enough milk are a
perception rather than a reality, all but one health worker reported
that lack of milk was actually a problem. All the health workers
reported that when a woman does not have enough milk she
should still initiate or continue breast-feeding to encourage milk to
come, but should also improve her diet, take a lot of fluids or rest.
One health worker reported that in such cases a mother should
also use formula to supplement the babys diet.

Discussion
This study used formative research to identify some important
barriers and gaps in community and health worker practices and
in policy implementation that need to be tackled if policies
promoting the early initiation of breast-feeding are to achieve the
neonatal mortality reductions suggested by recent studies.3
Any policies to promote breast-feeding initiation must address
the poor knowledge among some women of the importance of the
behavior, community perceptions about milk quantity and time of
arrival and negative beliefs about colostrum. Such barriers have
been identified elsewhere,711 and much information and
experience already exists on how to address these issues. For
example, guidelines, training manuals and counseling cards
already exist to address womens perceptions of a lack of milk.12,13
Although few intervention studies have focused on
breast-feeding initiation, much can be learnt from experiences of
promoting exclusive breast-feeding.1419 Strategies such as

improving hospital practices, counseling, mother support groups

and mass media campaigns are well developed, and the need for a
comprehensive approach that includes both the health service and
the community is recognized.20 As reported during the interviews
with policy makers and implementers, much of the focus of infant
feeding policies in Ghana has been on exclusive breast-feeding and
on health facilities. This focus needs to be broadened to include
early initiation of breast-feeding and community-based approaches.
Data from interventions in Ghana are encouraging. A recent
study in Ghana found that counseling at the prenatal and perinatal
period helps with early initiation and establishment of exclusive
breast-feeding.25 Results from the Linkages project in Northern
Ghana have shown that rates of early initiation can be improved
with a comprehensive and well funded strategy, which included
mass media, health worker training, supporting the Baby Friendly
Hospital Initiative Authority, support groups, mass media and
community mobilization.10,21 However, we found little support of
breast-feeding activities in the study district.
Many health workers are successfully encouraging early
initiation, despite a lack of support. Efforts should be made to
ensure that all health workers present at deliveries are skilled in
supporting mothers in breast-feeding initiation and that they have
the materials they need.
Materials and guidelines need to be provided by the district
and the regional implementers in a more proactive manner
and in sufficient quantities. Health workers beliefs about lack of
breast milk need to be addressed as they may compound womens
beliefs about the importance of having enough milk before starting
breast-feeding, making behavior change more difficult.
As most women attend ANC at least once during pregnancy, ANC
may offer a good education channel but, as in other countries,22
ANC appears to be a missed opportunity for health education.
Problems with counseling during focused ANC have been found
elsewhere in Ghana23 and a study from Tanzania suggests that
improving counseling during ANC requires substantial investment
in human resources.24
Currently most breast-feeding promotion activities are
facility focused and to increase coverage community strategies
are needed. Community strategies to promote the early
initiation of breast-feeding should not only target mothers but also
those who influence behavior such as TBAs, grandmothers and
other female relatives. In this study, these influencers were
important as advice givers to first-time mothers, as breast-feeding
behaviors are learnt from them, and because these influencers
often had control over the baby in the first few hours of life.
Interventions also need to include multiparous women, as
there is sometimes an assumption that such women do not
need support in breast-feeding. Messages on colostrum and the
timing of the arrival of breast milk could be targeted to families
of Northern tribes, as barriers related to these issues appear to be
ethnicity specific.
Journal of Perinatology

Formative research on early initiation of breast-feeding

C Tawiah-Agyemang et al

S52

This study uncovered several facilitators to the early initiation of

breast-feeding. Women want to breast-feed early and beliefs exist
among some ethnic groups that encourage early initiation, such as
a belief that putting the baby to the breast encourages milk to
come. The finding that initiation timing is often similar for all a
womans children means that women who may initiate late can be
easily identified by asking them about their previous practices.
Conclusions and implications
Finding from this study suggest that it is important to raise
awareness of the importance of early initiation of breast-feeding
among mothers and those who assist with deliveries at home. The
national breast-feeding programme has concentrated its efforts in
the health facilities and the main attention has been placed on
exclusive breast-feeding. Involving the community, especially
gatekeepers like TBAs, grandmothers and mother-in-laws, in
interventions to encourage early initiation is important.
Disclosure
Zelee Hill has received consultancy fees from the Gates Foundation. Betty R
Kirkwood has received grant support from DFID, USAID, Saving Newborn Lives
and WHO. Betty R Kirkwood has also participated in the Vikram Patel Welcome
Trust Senior Clinical Research Fellowship in Tropical Medicine and the
Dominique Behague Welcome Trust Research Development Fellowship. The
remaining authors have declared no financial interests.

References
1 WHO. Evidence for the ten steps to successful breastfeeding. WHO/CHD/98.9 World
Health Organization: Geneva, 1998.
2 UNICEF. Facts for Life. 3rd edn United Nations Childrens Fund: New York, 2002.
3 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): 380386.
4 Setty V. Better breastfeeding, healthier lives Population Reports, Series L, No. 14.
Baltimore Johns, Hopkins Bloomberg School of Public Health, The INFO Project,
February 2006.
5 Jones R. Why do qualitative research? BMJ 1995; 311: 2.
6 Patton MQ. How to use qualitative methods in evaluation. Sage: London, 1987;
108143.

Journal of Perinatology

7 Gunnlaugsson G, da Silva MC, Smedman L. Determinants of delayed initiation of

breastfeeding: a community and hospital study from Guinea-Bissau. Int J Epidemiol
1992; 21(5): 935940.
8 Semega-Janneh IJ, Bohler E, Holm H, Matheson I, Holmboe-Ottesen G. Promoting
breastfeeding in rural Gambia: combining traditional and modern knowledge. Health
Policy Plan 2001; 16(2): 199205.
9 Holman DJ, Grimes MA. Patterns for the initiation of breastfeeding in humans. Am J
Hum Biol 2003; 15(6): 765780.
10 Adjei E, Schubert J. Follow-up survey iii: a rapid appraisal of breastfeeding and
complementary feeding knowledge and practices in ghana. Ghs/linkages & partners in
Ghana 2003, Http://www.linkagesproject.org/media/static_pdfs/rap/ghanarap 2003.pdf.
11 Masvie H. The role of Tamang mothers-in-law in promoting breast feeding in
Makwanpur District, Nepal. Midwifery 2006; 22(1): 2331.
12 WHO/UNICEF. Breastfeeding counseling: a training course. World Health
Organization: Geneva, 1993.
13 Linkages. GhanaFBreastfeeding & Infant Feeding Counseling Cards.
http://www.linkagesproject.org/media/publications/Tools/Ghana-counselingcards.pdf
2003.
14 Green CP. Media promotion of breastfeeding: a decades experience. Academy for
Educational Development: Washington DC, 1989.
15 Green CP. Mother Support Groups: A Review of Experience in Developing Countries.
USAID: Arlington, 1998.
16 Green CP. Improving breastfeeding behaviours: evidence from two decades of
intervention research. Academy for Educational Development: Washington DC, 1999.
17 Britton C, McCormick FM, Renfrew MJ, Wade A, King SE. Support for breastfeeding
mothers. Cochrane Database Syst Rev 2007, (1): CD001141.
18 World Health Organization. Community-based strategies for breastfeeding
promotion and support in developing countries. WHO: Geneva, 2003.
19 Hill Z, Kirkwood B, Edmond KM. Family and Community practices that promote child
survival, growth and development: a review of evidence. WHO, Geneva, 2004.
20 Wellstart International. Community-based breastfeeding support: a planning
manual. Wellstart: Washington, DC, 1996.
21 Quinn VJ, Guyon AB, Schubert JW, Stone-Jimenez M, Hainsworth MD, Martin LH.
Improving breastfeeding practices on a broad scale at the community level: success
stories from Africa and Latin America. J Hum Lact 2005; 21(3): 345354.
22 WHO/UNICEF. Antenatal care in developing countries: promises, achievements and
opportunities: an analysis of trends, levels and differentials, 19902001. WHO:
Geneva, 2003.
23 Nyarko P, Birungi H, Armar-Klemesu M, Arhinful D, Deganus S, Odoi-Agyarko H et al.
Acceptability and Feasibility of Introducing the WHO Focused Antenatal Care Package
in Ghana. 2006, http://www.popcouncil.org/pdfs/frontiers/
FR_FinalReports/ghana_who_anc.pdf.
24 von Both C, Flebetaa S, Makuwani A, Mpembeni R, Jahn A. How much time do health
services spend on antenatal care? Implications for the introduction of the focused
antenatal care model in Tanzania. BMC Pregnancy Childbirth 2006; 6(1): 22.
25 Aidam BA, Perez-Escamilla R, Lartey A. Lactation counseling increases exclusive
breastfeeding rates in Ghana. J Nutr 2005; 135(7): 16911695.