Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Midwifery
journal homepage: www.elsevier.com/midw

Breast-feeding intention, initiation and duration among Hong Kong

Chinese women: A prospective longitudinal study
Wenru Wang, PhD, RN (Assistant Professor)a, Ying Lau, PhD, MN, BN (Hon), BSc, IBCLC, RM,
RN (Assistant Professor)a,n, Aloysius Chow, BPsy (Hon) (Research Assistant)a,
Kin Sin Chan, PhD, M. Phil, BSocSci (Hon) (Assistant Professor)b
a
Alice Lee Centre for Nursing Studies, Yong Loo Lin School of Medicine, National University of Singapore, Level 2, Clinical Research Centre, Block MD 11,
10 Medical Drive, Singapore 117597, Singapore
b
Faculty of Social Sciences and Humanities, the University of Macau, Macau Special Administration Region (SAR), China

art ic l e i nf o a b s t r a c t

Article history: Objectives: to (1) estimate the breast-feeding intention, initiation and duration rate; (2) identify the
Received 21 April 2013 reasons to initiate and wean breast feeding and (3) explore predictors of breast-feeding duration.
Received in revised form Design: a prospective longitudinal study
28 June 2013
Setting: antenatal clinics of five regional hospitals from four clusters in Hong Kong
Accepted 15 July 2013
Participants: a population-based sample of 2098 women in the second trimester of pregnancy was
recruited with a systematic sampling method.
Keywords: Measurements: three different sets of self-administered questionnaires were used to measure the breast-
Breast feeding feeding intention, initiation and duration, demographic, socio-economic, obstetric, complications of
Intention
pregnancy and intrapartum variables at three time points. Reasons for initiating and weaning breast
Initiation
feeding, the formal and informal supportive resources of participants during breast feeding were collated
Duration
at the third time points.
Findings: the rates of artificial feeding and breast feeding were 41.1% and 58.9%, whereas breast-feeding
intention and initiation rates were 85.3% and 67.0%, respectively. The breast-feeding duration rates were
11.1%, 10.3%, 10.7% and 26.7%, for the ‘within o1 week’, ‘1–3 weeks’, ‘ 43–6 weeks’ and ‘ 46 weeks’
groups. The common reasons for initiating breast feeding were that breast feeding is beneficial for both
the baby (89.8%) and mother (39.7%). Reasons for weaning breast feeding were insufficient breast milk
(32.7%), tiredness and fatigue (39.7%) and return to work (29.6%). Partner, relatives and nurse midwives
were important supportive resource during breast feeding. Ordinal logistic regression analysis identified
five predictive factors of breast-feeding duration. Participants who were working part-time or were
housewives (p ¼0.037), had monthly family income of o HK$10,000 (p ¼0.034), more than one child
(p¼ 0.001), positive breast-feeding intention (p¼ 0.001) and early breast feeding within the first hour
(po 0.0001) were more likely to have longer breast feeding than their counterparts.
Conclusion: mothers in Hong Kong do not meet the recommendation of the international standards for breast
feeding. Understanding of reasons for initiation, weaning and predictors of breast-feeding duration can help
nurse midwives to develop and customise effective strategies to promote and encourage breast feeding.
Implications for practice: the findings are important for a process-oriented breast-feeding training programme
for nurse midwives. New strategies should be implemented for prolonging breast-feeding duration among
perinatal women.
& 2013 Elsevier Ltd. All rights reserved.

Introduction and appropriate complementary feeding should commence after the

The advantages of breast feeding have been well established
including physical, psychosocial, economic and environmental benefits
(Cornall, 2011). According to the United Nations Fund for Children
(UNICEF), optimal infant breast feeding should be initiated within the
first hour of birth, exclusive breast feeding continue for six months
n
Corresponding author.
E-mail addresses: nurly@nus.edu.sg, lauying37@gmail.com (Y. Lau).
sixth month together with breast feeding for at least two years
(UNICEF, 2009). Some advantages of longer breast-feeding duration
for mothers are improved bone strength (Chapman, 2012) and lower
risk of breast cancer (Nagata et al., 2012); while better academic
achievement (Oddy et al., 2011) and lower risk of obesity (McCrory
and Layte, 2012) are some reported advantages for children who are
breast fed.
However, the global average duration of exclusive breast
feeding is less than six months and the proportion of women
who breast feed until their infant is six months old is dismally low

0266-6138/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.midw.2013.07.015

Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
2 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎
at only 36% (WHO, 2009). Approximately 75% of mother–baby
dyads do not meet the UNICEF recommendations for optimal
breast feeding (Evans et al., 2011; O’Connor et al., 2011; Stuebe
and Bonuck, 2011). In Hong Kong, over 80% of mothers initiate
breast feeding (Baby Friendly Hospital Initiative Hong Kong
Association, 2011), but only 6% of them were exclusively breast
feed during the first three months (Tarrant et al., 2010). A better
understanding of the reasons for breast-feeding initiation (Taleb
et al., 2012; Loof-Johanson et al., 2013) and weaning (Ho and
McGrath, 2011; Olang et al., 2012) will facilitate nurse midwives to
develop effective and customise effective strategies to promote
and encourage breast feeding.
Breast-feeding duration depends on multiple factors associated
with mother, infant, and a supportive environment. Demographic,
socio-economic, obstetric and intrapartum interventions factors
have been linked to breast-feeding duration in the recent papers
(Meedya et al., 2010; Thompson et al., 2011; Inoue et al., 2012).
In addition, it has been found that mothers in hospitals that are
encouraged to breast feed within the first hour of birth (Tarrant
et al., 2011) together with the presence of nurse midwives who
actively promote breast feeding during the perinatal period
(Kervin et al., 2010) will have a longer breast-feeding duration.
Although many studies have been conducted to investigate
the associations between these factors and breast-feeding duration,
inconsistent associations have been reported such as household
income (Mandal et al., 2012; Chen et al., 2013), parity (Akter and
Rahman, 2010; Dennis et al., 2013), mode of delivery (Bezerra et al.,
2012; Mazumder and Hossain, 2012), use of opiate or epidural
analgesia during labour (Woods et al., 2012; Yousefshahi et al.,
2012), and complications of pregnancy (Cordero et al., 2012; Nicklas
et al., in press). Since many factors are still inconclusive and not well
examined in detail, further investigation is required.
A limited numbers of studies about breast feeding have been
conducted in Hong Kong and these studies mainly investigate
intrapartum intervention (Bai et al., 2013) and hospital practices
(Tarrant et al., 2011). The possible predictors of the breast-feeding
duration of mothers in Hong Kong remain unclear. It is therefore
important to identify the factors that may influence breast-feeding
duration, so that effective breast-feeding promotion policies and
counselling activities can be developed and implemented. This
current study examined factors such as demographic, socio-
economic, obstetric, intrapartum characteristics and complications
of pregnancy in relation to breast-feeding duration. These factors
were selected because they are modifiable through healthcare
facility policies, professional and peer support, clinical interven-
tions and education, public health efforts, and community
resources (Taleb et al., 2012; Wen et al., 2012; Bai et al., 2013)
and are of important interest to nursing practice. The current
study aims to address the following research questions:
1. What are the rates of breast-feeding intention, initiation and
duration in five hospitals in Hong Kong?
2. What are the reasons women have to initiate and wean breast
feeding?
3. What are the demographic, socio-economic, obstetric and intra-
partum predictors associated with breast-feeding duration?
Methods
Design and sample
A prospective longitudinal research design was used to collect
data at three different time points. This design has been used to
establish the relationships, direction and magnitude of study
Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
variables (Burns and Grove, 2009). The systematic sampling
method was adopted to enhance the representativeness of the
sample (Kao et al., 2011) and increase the generalisation of the
findings (Kao et al., 2011). It is estimated that every year, 120,000
women attend the antenatal outpatient clinics located in the five
hospitals where the recruitment took place (Hospital Authority,
2010). Assuming a previous prevalence rate of breast-feeding
duration in a local study (Tarrant et al., 2010), a minimum of
2000 participants were required to gain a 95% confidence interval
for the study estimates with a width of 72% (Johnson and
Wichern, 2007). Dividing 120,000 by 2000, a sampling interval
of 60 was obtained and, starting with the random number 19,
every 60th woman thereafter was recruited. Hence the sample
consisted of cases numbered 19, 79, 139, 199 … 120,019 (Kao et al.,
2011). In this study, a total of 2365 Chinese women from five
hospitals in Hong Kong were invited to participate over a period of
12 months. From this group, 2098 women completed the ques-
tionnaire. The inclusion criteria were Chinese childbearing women
with a single fetus attending the antenatal clinics of the five
hospitals in the second trimester of pregnancy. The exclusion
criteria were: (1) twins pregnancy, (2) carrying fetal malforma-
tions, (3) severe personal or family psychiatric or medical history,
(4) not able to provide consent, and (5) breast surgery patient.
Measurement
Three sets of questionnaires were developed based on informa-
tion and data from the literature review (Meedya et al., 2010;
Thompson et al., 2011; Inoue et al., 2012). The first set of ques-
tionnaires recorded the demographic and socio-economic informa-
tion (i.e. age, educational level, marital status, employment status,
monthly family incomes). In addition breast-feeding intention was
measured using a self-administered questionnaire. Two responses
were used to indicate the intention of feeding (i.e. ‘breast feeding
(exclusive breast feeding or/and partial breast feeding)’ versus ‘bottle
feeding’).
The second set of questionnaires recorded complications of
pregnancy (i.e. gestational diabetes mellitus (GDM), gestational
induced hypertension (pre-eclampia), antenatal bleeding (placenta
praviea or abruptio placentae), threaten abortion, nausea and
vomiting, swelling in pregnancy, malposition of fetus (breech or
transverse), abnormal amino fluid (polyhydraminos or oligohydra-
minos), headache and dizziness, congenital abnormality, thalassa-
nemia, sexual transmitted diseases or others).
The third set of questionnaires consisted of obstetric and
intrapartum data (i.e. number of pregnancies, number of babies,
mode of delivery, opioid pain medication or epidural analgesia,
duration of labour and duration of gestation). In addition, breast-
feeding initiation and duration were measured with a self-
administered questionnaire. Whether breast feeding was initiated
within the first hour of birth was reported by the mother (i.e. ‘yes’
versus ‘no’). Duration of breast feeding was divided into five
groups (i.e. ‘never breast feeding’, ‘o1 week’, ‘1–3 weeks’, ‘43–
6 weeks’ and ‘4 6 weeks’). The list of possible reasons and
supportive resources for initiating and weaning breast feeding
were also recorded in the third set of questionnaires.
Data collection
Ethical approvals that complied with the Declaration of
Helsinki were obtained from the Institutional Review Boards from
five hospitals (i.e. from four clusters in Hong Kong). The first set of
questionnaires was collected during the first 12 months of the
study at the antenatal clinics (i.e. 1st–12th month), the second set
of questionnaires was collected during the next 12 months at the
antenatal clinics (i.e. 13th–24th month) and the third set of
 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

questionnaires was collected in the following 12 months by mail
(25th–36th month).
Baseline data consisted of data from the first set of question-
naire and was collected in the second trimester of pregnancy
(i.e. 13th–24th week of gestation).
The second set of questionnaires was administered when
participants were in their third trimester of pregnancy (i.e. 25th–
36th week of gestation). During this time, participants were also
provided with envelopes which were stamped and addressed to
the investigator. This was because the participants had to com-
plete and mail the third set of questionnaires back to the
investigator six weeks after their deliveries. The investigator
would call the participants if they did not mail back their third
set of questionnaires during this time.
Responses for items addressing the independent variables were
dichotomized so they were easy to understand without decreasing
the strength of associations (Farrington and Loeber, 2000). Only
significant variables (po 0.05) of breast feeding duration in
bivariate analysis were selected to build an ordinal logistic
regression model (Hilbe, 2009). The model-fitting statistic, namely
the pseudo R2, measured the success of the model in explaining
the variation in the data (Hilbe, 2009). The Pearson goodness-of-fit
statistic was used to test the estimated values in the fitted model
(Hilbe, 2009). If the model fitted well, the goodness-of-fit mea-
sures would have large observed significance levels (p 40.05). The
test of parallel lines was used to determine the adequacy of the
model (Hilbe, 2009). A non-significant result (p 40.05) suggested
that the model’s assumption of parallel lines was satisfied in the
completed model with the cloglog link.

Data analysis
The IBM SPSS 21.0 was used for descriptive and inferential data
analysis. The five groups (i.e. ‘no breast feeding’, ‘o 1 week’, ‘1–3
weeks’, ‘4 3–6 weeks’ and ‘46 weeks’) were compared using a
Chi-square (χ2) test. In this study, an ordinal logistic regression
analysis was used to model the effects of the potential explanatory
variables on multiple-category feeding duration using the Poly-
tomous Universal Model (PLUM) procedure (Bendall et al., 2011).
Findings
A total of 2365 Chinese women from five hospitals in Hong
Kong were invited to join the study (Fig. 1), of this number 2098
women completed the questionnaire (response rate¼ 88.6%).
Of this group invited, 270 (11.4%) refused to participate in the
study due to the lack of time, fatigue, and reluctance to disclose
information. The retention rate of participants was 72.9% in the

2,365 recruited from Antenatal Outpatient Clinics at five

Hong Kong hospitals

270 (11.4%) refused to

participate

Time One:
2nd Trimester 2,095 (88.6%) finished THE FIRST SET OF QUESTIONNAIRES

(12 – 24 weeks {Demographic, socio-economic, obstetric characteristics and

gestations) Breastfeeding intention [1786 (85.3%)]

568 (27.1%) refused to

participate
Time Two:
3rd Trimester 1,527 (72.9%) completed THE SECOND SET OF QUESTIONNAIRES
(>24 – 36 (complications of pregnancy) at the antenatal clinics
week
gestations) 917 (60.1%) did not return
the questionnaires

Time Three:
610 (39.9%) returned THE THIRD SET OF QUESTIONNAIRES
(6 weeks after
{Breastfeeding initiation [413 (67.7%)] and intrapartum characteristics}
delivery)
by mail and phone reminder

Formula Breastfeeding Breastfeeding Breastfeeding Breastfeeding

Feeding duration: duration: duration: duration:
Group: 1 week 1—3 weeks >3 — 6 weeks > 6weeks
251 (41.1%) 68 (11.1%) 63 (10.3%) 65 (10.7%) 163(26.7%)

Fig. 1. Flowchart for the implementation of the study.

Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
4 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎

second time point. The number of completed questionnaires in the
third time point was 610 (39.9%) which was similar to the typical
response rate for mail surveys – between 20% and 40% (Nachmias
and Nachmias, 2007).
The rate of breast-feeding intention was 85.3% (n ¼1786) and
the breast-feeding initiation rate within the first hour was 67.0%
(n ¼413). It was observed that more women breast fed than those
used artificial feeding (i.e. 58.9% versus. 41.1% respectively). How-
ever, 11.1% (n ¼68) of women weaned breast feeding within the
first week, 10.3% (n¼ 63) of them weaned breast feeding at 41–3
weeks, 10.7% (n ¼65) of them weaned breast feeding at 43–6
weeks and 26.7% (n ¼163) of them continued breast feeding after
6 weeks (Fig. 1). The demographic, socio-economic, obstetric and
intrapartum characteristics of participants in all three different
time points are summarised in Table 1. Over 82% of the partici-
pants were older than 25 years old and completed less than
tertiary level of education. Most of them (92.4–93.8%) were
married and more than half (56.0–57.0%) had full-time jobs with
a monthly family income HK$10,000 or more (68.4–69.9%). Less
than half (43.0–45.1%) were experiencing their first pregnancies
and more than half (63.1–65.6%) were first-time mothers. More
than half of the participants (66.6%) experienced natural birth and
over half of them (51.6%) received opioid pain medication or
epidural analgesia. The majority of the participants (82.0%)
reported less than 15 hours of labour duration and delivered at
37th weeks or later gestations (83.1%). The demographic profile of
the study participants was similar to the general pregnant popula-
tion in Hong Kong (Hospital Authority, 2010).
The reasons for initiating and weaning breast feeding and the
types of supportive resources of participants is summarised in
Table 2. The common reasons for initiating breast feeding were
that breast feeding was beneficial for both the baby (89.8%) and
mother (39.7%). A small proportion of participants cited the
support from nurse midwives during the antenatal period
(18.2%) and postnatal period (12.8%) as reasons for initiating breast
feeding. The common reasons for weaning breast feeding were
insufficient milk supply (32.7%), tiredness and fatigue (30.6%) and
return to work (29.6%). Comparisons of the reasons for weaning
breast feeding at different time points are shown in Table 3. Poor
feeding technique, difficult infant temptation and difficulty feed-
ing were significant reasons to wean breast feeding within the first
week after delivery (po0.05). Tiredness and fatigue were signifi-
cant reasons to wean breast feeding at 41–3 weeks and return to
work was found to be a significant reason to wean breast feeding
at the sixth week (p o0.05). The partners (49.2%) and nurse
midwives in maternal clinics (69.0%) were the most important
informal and formal resources among participants who were
breast feeding. Mothers who weaned breast feeding reported that

Table 1
Demographic, socio-economic, obstetric and intrapartum characteristics of participants in three periods of time.

Demographic and socio-economic characteristics Time one (n ¼2095) Time two (n ¼1527) Time three (n¼ 610)

n % n % n %

Age
≦25 294 14.0 203 13.3 67 11.0
425 1801 86.0 1324 86.7 543 89.0
Educational level
≦Secondary 1722 82.2 1274 83.4 509 83.4
4Secondary 373 17.8 253 16.6 101 16.6
Marital status
Married 1936 92.4 1424 93.3 572 93.8
Unmarried/Cohabiting 159 7.6 103 6.7 38 6.2
Employment of participant
Full-time 1173 56.0 862 56.5 348 57.0
Part-time or unemployed 922 44.0 665 43.5 262 43.0
Total family income
o $10,000 633 30.2 460 30.1 193 31.6
Z $10,000 1462 69.8 1067 69.9 417 68.4

Obstetric characteristics
Number of pregnancy (Gradiva)
First pregnancy 943 45.0 670 43.0 275 45.1
4First pregnancy 1152 55.0 857 56.1 335 54.9
Number of baby (Parity)
First baby 1374 65.6 979 64.1 385 63.1
4First baby 721 34.4 548 35.9 225 36.9
Intention of feeding
Breast feeding 1786 85.3 1300 85.1 514 84.3
Bottle feeding 309 14.7 227 14.9 96 15.7

Intrapartum characteristics
Mode of delivery
Spontaneous delivery 406 66.6
Assisted delivery (vacuum, forceps or section) 204 33.4
Opioid pain medication/Epidural analgesia
No 295 48.4
Yes 315 51.6
Duration of labour
415 hours 110 18.0
≦15 hours 500 82.0
Gestation of babies
Z 37 weeks 507 83.1
o 37 weeks 103 16.9
Early feeding within 60 minutes
Yes 413 67.7
No 197 32.3

Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Table 2
Reasons for initiating, weaning breast feeding and supportive resources.

Reasons for initiating breast feeding (n ¼ 413) n % Reasons for weaning breast feeding (n¼ 196) n %

Beneficial for baby 371 89.8 Tiredness and fatigue 60 30.6

Beneficial for mother 164 39.7 Insufficient breast milk supply 64 32.7
Save money 52 12.6 Return to work 58 29.6
Family support 44 10.7 Poor feeding technique 34 17.3
Support from friends or mass media 51 12.3 Discourage by family members 7 3.6
Professional support from antenatal education session 75 18.2 Difficult infant temperament 8 4.1
Professional support from health care professionals during postnatal period 53 12.8 Baby having difficulty feeding 18 9.2
Others 7 1.7 Others 21 10.7

Supportive resource during breast feeding

Informal resource
Partner 203 49.2 92 46.9
Relative who live with 176 42.6 93 47.4
Relative who do not live with 166 40.2 88 44.9
Friends 178 43.1 96 49.0
Neighbour 55 13.3 34 17.3
Domestic helper 57 13.8 35 17.9
Formal resource
Nurses in maternal clinic 285 69.0 144 73.5
Clinician 121 29.3 66 33.7
Nurses in hospital 228 55.2 124 63.3
Nurses assistants in hospital 65 15.7 39 19.9
Others 33 8.0 20 10.2

Table 3
Comparison of the reasons for weaning breast feeding at the different lengths of breast feeding (n¼ 196).

Reasons for weaning breast feeding Weaned women (n¼ 196) o 1 week (n ¼68) 1–3 weeks (n ¼65) 43–6 weeks (n ¼65) Chi-square test (χ2)

n % n % n % n %

No 136 69.4 45 66.2 38 60.3 53 81.5

Tiredness and fatigue .026n
Yes 60 30.6 23 33.8 25 39.7 12 18.5
No 132 67.3 45 66.2 40 63.5 47 72.3
Insufficient breast milk supply Yes 32.7 33.8 36.5 27.7 .550
64 23 23 18
No 138 70.4 60 88.2 50 79.4 28 43.1
Return to work o .0001nnn†
Yes 58 29.6 8 11.8 13 20.6 37 56.9
No 162 82.7 50 73.5 51 81.0 61 93.8
Poor feeding technique .008nn†
Yes 34 17.3 18 26.5 12 19.0 4 6.2
No 189 96.4 68 100.0 59 93.7 62 95.4
Discourage by family members .126†
Yes 7 3.6 0 0.0 4 6.3 3 4.6
No 188 95.9 62 91.2 61 96.8 65 100.0
Difficult infant temperament Yes 4.1 8.8 3.2 0.0 .033n†
8 6 2 0
No 178 90.8 56 82.4 58 92.1 64 98.5
Baby having difficulty feeding .005nn†
Yes 18 9.2 12 17.6 5 7.9 1 1.5
No 175 89.3 58 85.3 56 88.9 61 93.8
Others .279†
Yes 21 10.7 10 14.7 7 11.1 4 6.2

†
Fisher’s Exact test when the cell count less than 10.
n
p o0.05.
nn
p o 0.01.
nnn
p o 0.001.

relatives (47.4%) and nurse midwives in the maternal clinics
(73.5%) were the most important supportive resources when they
were still breast feeding.
More than a third of participants (37.0%) experienced nausea
and vomiting during the antenatal period (Table 4). Around a
quarter of the participants reported having oedema or swelling
(26.9%) as well as headache and dizziness (22.3%). A few of them
experienced complications of pregnancy such as gestational dia-
betes mellitus (8.2%), gestational hypertension (3.1%), abnormal
amniotic fluid (i.e. polyhydraminos or oligohydramios) (2.1%),
congenital abnormality (0.3%) or others (5.1%) (Table 4). Bivariate
analysis showed that educational level, employment, monthly
family income, number of babies, breast-feeding intention, early
breast feeding within the first hour and complications of preg-
nancy, such as abnormal amniotic fluid, congenital abnormality
and others were found to be significantly different among the
groups (p o0.05) (Table 4).
Table 5 presents the results of an ordinal logistic regression
model of the predictors of breast-feeding duration. Participants
who were working part-time or were housewives (p ¼0.037), had
a family income of less than HK$10,000 (p¼ 0.034), had more than
one child (p¼ 0.001), had positive breast-feeding intention
(p ¼0.001) and started early breast feeding within the first hour
of birth (p o0.0001) were more likely to have longer breast-
feeding duration. The fit of regression to the actual data was
acceptable based on goodness-of-fit with p value of 0.622 (Hilbe,
2009). Pseudo R2 (Cox and Snell) statistics for the regression was
39.8% (0.398) which was reasonable (Hilbe, 2009). Test of parallel
lines indicated that there was no significant difference (p ¼0.441)
for the corresponding regression coefficient across the response

Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
6 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Table 4
Bivariate analysis of demographic, socio-economic, obstetric, intrapartum characteristics and complications of pregnancy on breast-feeding duration (n¼ 610).

Duration of breast feeding Chi-square test (χ2)

Entire sample Never breast feeding o 1 week 1–3 weeks 43–6 weeks 46 weeks

Demographic and socio-economic characteristics p-Value

Age
≦25 67(11.0) 33 (49.3) 9 (13.4) 8 (11.9) 8 (12.3) 9 (13.4) 0.145†
425 543 (89.0) 218 (40.1) 59 (10.9) 55 (10.1) 57 (10.5) 154 (28.4)
Educational level
≦Secondary 509 (83.4) 223 (43.8) 57 (11.2) 54 (10.6) 48 (9.4) 127 (25.0)
0.009nn†
4Secondary 101(16.6) 28 (27.7) 11 (10.9) 9 (8.9) 17 (16.8) 36 (35.6)
Marital status
Unmarried/Cohabiting 38(6.2) 19 (50.0) 4 (10.5) 3 (7.9) 4 (10.5) 8 (21.1)
Married 572(93.8) 232 (40.6) 64 (11.2) 60 (10.5) 61 (10.7) 155 (27.1) 0.826†
Employment of participant
Full-time 348 (57.0) 136 (39.1) 40 (11.5) 52 (14.9) 45 (12.9) 75 (21.6)
o 0.001nnn
Part-time or unemployed 262 (43.0) 115 (43.9) 28 (10.7) 11 (4.2) 20 (7.6) 88 (33.6)
Total monthly family income‡
o$10,000 193 (31.6) 90 (46.6) 17 (8.8) 6 (9.5) 12 (6.2) 68 (35.2)
o 0.001nnn†
Z$10,000 417 (68.4) 161 (38.6) 51 (12.2) 57 (13.7) 53 (12.7) 95 (22.8)

Obstetric characteristics
Number of pregnancy (Gradiva)
4First pregnancy 335 (54.9) 134 (40.0) 34 (10.1) 32 (9.6) 34 (10.1) 101 (30.1)
0.310
First pregnancy 275 (45.1) 117 (42.5) 34 (12.4) 31 (11.3) 31 (11.3) 62 (22.5)
Number of baby (Parity)
4First baby 225 (36.9) 86 (38.2) 19 (8.4) 19 (8.4) 20 (8.9) 81 (36.0)
0.002nn
First baby 385 (63.1) 165 (42.9) 49 (12.7) 44 (11.4) 45 (11.7) 82 (21.3)
Intention of feeding
Bottle feeding 96 (15.7) 70 (72.9) 7 (7.3) 5 (5.2) 5 (5.2) 9 (9.4)
o 0.001nnn†
Breast feeding 514 (84.3) 181 (35.2) 61 (11.9) 58 (11.3) 60 (11.7) 154 (30.0)

Intrapartum characteristics
Mode of delivery
Assisted delivery (Vacuum, forceps or section) 204 (33.4) 85 (41.7) 20 (9.8) 24 (11.8) 20 (9.8) 55 (27.0)
0.854
Spontaneous delivery 406 (66.6) 166 (40.9) 48 (11.8) 39 (9.6) 45 (11.1) 108 (26.6)
Opioid pain medication/Epidural analgesia
Yes 315 (51.6) 111 (37.6) 37 (12.5) 36 (12.2) 33 (11.2) 78 (26.4)
0.305
No 295 (48.4) 140 (44.4) 31 (9.8) 27 (8.6) 32 (10.2) 85 (27.0)
Duration of labour
415 hours 110 (18.0) 45 (40.9) 8 (7.3) 15 (13.6) 12 (10.9) 30 (27.3)
0.516†
≦15 hours 500 (82.0) 206 (41.2) 60 (12.0) 48 (9.6) 53 (10.6) 133 (26.6)
Gestation of babies
o37 weeks 103 (16.9) 43 (41.7) 11 (10.7) 8 (7.8) 14 (13.6) 27 (26.2)
Z37 weeks 507 (83.1) 208 (41.0) 57 (11.2) 55 (10.8) 51 (10.1) 136 (26.8) 0.766†
Early breast feeding within 60 minutes
No 197 (32.3) 170 (86.3) 13 (6.6) 1 (1.6) 5 (2.5) 8 (4.1)
o 0.001nnn†
Yes 413 (67.7) 81 (19.6) 55 (13.3) 62 (15.0) 60 (14.5) 155 (37.5)

Complications of pregnancy
Gestational Diabetes Mellitus
Yes 50 (8.2) 16 (32.0) 4 (8.0) 2 (4.0) 8 (12.3) 20 (40.0)
No 560 (91.8) 235 (42.0) 64 (11.4) 61 (10.9) 57 (10.2) 143 (25.5) 0.067†
Gestational hypertension
Yes 19 (3.1) 9 (47.4) 0 (0.0) 1 (5.3) 4 (21.1) 5 (26.3)
0.303†
No 591 (96.9) 242 (40.9) 68 (11.5) 62 (10.5) 61 (10.3) 158 (26.7)
Oedema or swelling
Yes 164 (26.9) 66 (40.2) 22 (13.4) 20 (12.2) 19 (11.6) 37 (22.6)
0.482†
No 446 (73.1) 185 (41.5) 46 (10.3) 43 (9.6) 46 (10.3) 126 (28.3)
Nausea and vomiting
Yes 226 (37.0) 100 (44.2) 21 (9.3) 22 (9.7) 24 (10.6) 59 (26.1)
0.714
No 384 (63.0) 151 (39.3) 47 (12.2) 41 (10.7) 41 (10.7) 104 (27.1)

Malposition of fetus (Breech/transverse)

Yes 24 (3.9) 8 (33.3) 1 (4.2) 6 (25.0) 3 (12.5) 6 (25.0)
0.148†
No 586 (96.1) 243 (41.5) 67 (11.4) 57 (9.7) 62 (10.6) 157 (26.8)
Polyhydraminos/Oligohydramios
Yes 13 (2.1) 3 (23.1) 5 (38.5) 0 (0.0) 1 (7.7) 4 (30.8)
No 597 (97.9) 248 (41.5) 63 (10.6) 63 (10.6) 64 (10.7) 159 (26.6) 0.021n†
Thalassanemia
Yes 13 (2.1) 6 (46.2) 3 (23.1) 1 (7.7) 1 (7.7) 2 (15.4)
0.624†
No 597 (97.9) 245 (41.0) 65 (10.9) 62 (10.4) 64 (10.7) 161 (27.0)
Sexual transmitted disease
Yes 1 (0.2) 0(0.0) 0 (0.0) 1 (100.00) 0 (0.0) 0 (0.0)
No 609 (99.8) 251 (41.2) 68 (11.2) 62 (10.2) 65 (10.7) 163 (26.8) 0.069†
Threaten abortion
Yes 78 (12.8) 25 (32.1) 12 (15.4) 14 (17.9) 7 (9.0) 20 (25.6)
0.074†
No 532 (87.2) 226 (42.5) 56 (10.5) 49 (9.2) 58 (10.9) 143 (26.9)

Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

Table 4 (continued )

Duration of breast feeding Chi-square test (χ2)

Entire sample Never breast feeding o 1 week 1–3 weeks 43–6 weeks 46 weeks

Headache and dizziness

Yes 136 (22.3) 61 (44.9) 15 (11.0) 12 (8.8) 15 (11.0) 33 (24.3)
0.846†
No 474 (77.7) 190 (40.1) 53 (11.2) 51 (10.8) 50 (10.5) 130 (27.4)
Antenatal bleeding (placenta praviea)
Yes 8 (1.3) 3 (37.5) 2 (25.0) 1 (12.5) 0 (0.0) 2 (25.0)
No 602 (98.7) 248 (41.2) 66 (11.0) 62 (10.3) 65 (10.8) 161 (26.7) 0.675†
Congenital abnormality
Yes 2 (0.3) 0 (0.0) 2 (100.00) 0 (0.0) 0 (0.0) 0 (0.0)
0.003nn†
No 608 (99.7) 251 (41.3) 66 (10.9) 63 (10.4) 65 (10.7) 163 (26.8)
Others
Yes 31 (5.1) 4 (12.9) 6 (19.4) 4 (12.9) 5 (16.1) 12 (38.7)
0.025n†
No 579 (94.9) 247 (42.7) 62 (10.7) 59 (10.2) 60 (10.4) 151 (26.1)

†
Fisher’s Exact test when the cell count less than 10.
‡
Hong Kong $10,000¼ approximately US$1420. It was considered as cut-off because this represents the median monthly domestic household income in Hong Kong.
n
po 0.05.
nn
p o0.01.
nnn
p o 0.001.

Table 5
Results of an ordinal logistic regression model predicting breast-feeding duration (n¼ 610).

Predictors Estimate Standard error p-Value 95% Confidence interval

Lower Upper

Educational level4 Secondary 0.295 0.218 0.176  0.132 0.721

Part-time/Housewife 0.447 0.214 0.037n 0.026 0.867
Total family income o $10,000 0.420 0.198 0.034n 0.032 0.807
4First baby (Multiparity) 0.608 0.176 0.001nn 0.263 0.953
Positive breast-feeding intention 0.898 0.282 0.001nn 0.345 1.452
Polyhydraminos/Oligohydramios 0.167 0.562 0.767  0.935 1.269
Congenital abnormality 0.387 1.465 0.792  2.484 3.259
Other antenatal complication 0.439 0.370 0.235  0.286 1.164
Early breast feeding within 60 minutes 3.264 0.250 o 0.000nnn 2.774 3.754
Goodness-of-fit p ¼ 0.622
Pseudo R2 (Cox and Snell) R2 ¼ 0.398 (39.8%)
Test of parallel lines p ¼ 0.441

n
po 0.05.
nn
p o0.01.
nnn
p o 0.001.

categories, suggesting that the model assumption of parallel lines 88% in Australia (Wen et al., 2012), and breast-feeding duration
was not violated (Hilbe, 2009). was over two years in Nepal (Basnet et al., 2012).

Low-family income and breast-feeding duration

Discussion
Breast-feeding intention, initiation and duration
The health and socio-economic benefits of breast feeding for
infant, mother, and society are well-documented (Cornall, 2011).
However, the protective effect of breast feeding depends on its
duration and exclusivity (Taleb et al., 2012). Although the pre-
valence rate of the breast-feeding intention and initiation rate in
this study were 85.3% and 67.7%, breast feeding sustaining at least
six weeks was only 26.7% (i.e. ‘ 46 weeks’ group) which is similar
to previous local studies (Tarrant et al., 2010; Baby Friendly
Hospital Initiative Hong Kong Association, 2011). Compared with
other countries, Hong Kong still has a long way to achieve the
recommendation of the UNICEF (UNICEF, 2009). In fact, it has been
reported that the breast-feeding intention rate was 91% in main-
land China (Jiang et al., 2012), breast-feeding initiation rate was
Our results indicate that women who had a lower monthly
family income (i.e. oHK$10,000) (Census and Statistics Department,
2013) were more likely to breast feed longer, which was consistent
with a previous study (Mandal et al., 2012). This result may be
explained by the cost effectiveness of breast feeding. Although
breast-feeding mothers consume more food than usual, it is still
relatively less expensive than purchasing milk formula for the infant
(Eidelman and Schanler, 2012). Hence breast feeding saves money
for families while providing the best nourishment possible (Riordan
and Wambach, 2010). In a year, a family can save several hundreds
of dollars that might have been spent on breast milk substitutes.
In contrary, mothers from higher monthly family income had
shorter duration of breast feeding. The reasons might be that
economically affluent families can afford to buy milk formula
(Mazumder and Hossain, 2012). Therefore, economic disadvantage
may be a protective factor linked to longer breast-feeding duration
for Hong Kong women.

Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
8 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Parity and breast-feeding duration
Consistent with a previous study (Dennis et al., 2013), it was
found that mother with more children had a longer duration of
breast feeding. Multiparous women were more likely to repeat
their previous breast feeding experiences and practices from the
preceding children (Phillips et al., 2011a, 2011b). On the other
hand, the lack of awareness of the WHO breast-feeding guidelines
was found among first-time mother (Jiang et al., 2012). And the
use of pacifiers was common (79%) by first-time mothers (Mauch
et al., 2012) which has been shown to be negatively associated
with breast-feeding duration.
Supportive resources
Partner and relatives were important informal supportive
resources for the participants in this study. In a Hong Kong Chinese
family, raising a child is the responsibility of not only the parents
but also the extended family (Lau, 2011). Hence the breast-feeding
intention of mothers is greatly influenced by her family members
(Lau, 2010). Mothers will usually accede to the opinions of the
family because of an eminent Confucian ideal of respect and love
between family members (Lam et al., 2012). Therefore it may be
helpful to involve family members in a breast-feeding promotion
programme to increase the women's breast-feeding duration
(Ho and McGrath, 2011). Family members need more information
about how they can be supportive of breast-feeding mothers,
which may improve breast-feeding duration.
Another important supportive resource is the nurse midwife
in maternal clinics and hospital because they were the common
formal supportive resource reported in this study. A previous
study reported that professional support had the most important
influence on breast-feeding intentions and behaviours (Kervin
et al., 2010). This study found that women with poor feeding
technique, difficult infant temperament and baby having difficulty
feeding were significant reasons to wean breast feeding in the first
week after delivery. Therefore, the role of nurse midwives is
crucial for one-to-one bedside breast feeding educational and
technical support during the first several days after delivery
(Morton, 2012). Antenatal, intrapartum and postnatal breast-
feeding education are important regarding proper feeding techni-
que, feeding cues, perception of baby temperament and helping
babies who experiencing feeding difficulty (Pannu et al., 2011;
Chowdhury et al., 2013)
Nurse midwives must be better trained in the area of breast
feeding so they can provide better guidance to breast-feeding
mothers on how they may increase breast-feeding duration
(Ekstrom et al., 2012; Jiang et al., 2012). To prevent the early
weaning of breast feeding among working mothers, nurse mid-
wives can consider home visits to provide information about how
to pump breast milk, maintain lactation, and store breast milk
while the women are at work (Ho and McGrath, 2011). When
mothers return to work, occupational health nurses can provide
support and educate working mothers to increase the breast-
feeding duration (Riordan and Wambach, 2010).
Intrapartum characteristics and complications of pregnancy
In this study, intrapartum characteristics were not significantly
associated with breast-feeding duration which is in line with the
findings from a recent local study (Bai et al., 2013). Breast feeding
is a complicated process with many influencing factors. Along with
intrapartum characteristics, other factors such as breast-feeding
intention or initiation have been previously demonstrated to
substantially affect the duration of breast feeding (Meedya et al.,
2010; Thompson et al., 2011; Inoue et al., 2012). Unsurprisingly in
Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
this study, intrapartum characteristics were non-significant factors
in consideration of these factors.
Although bivariate analysis showed that abnormal amniotic
fluid (polyhydraminos or oligohydramios), congenital abnormality
and other complications were significant among five groups
of feeding, such variables became non-significant in an ordinal
logistic regression analysis. This may be due to the small number
of participants (0.3–5.1%) who had experienced complications of
pregnancy in this study. Thus, no firm conclusion can be drawn
about those variables in this study. Further investigation with a
larger sample size is necessary.
Insufficient milk supply
Consistent with previous studies (Ho and McGrath, 2011; Olang
et al., 2012), insufficient milk supply was the most common reason
for the weaning of breast feeding. This may not accurately reflect
the reasons for early weaning, because only 1–5% of women have
medically verified problems with milk production and most
women can produce enough breast milk to meet their baby’s
demand (Riordan and Wambach, 2010). This may indicate inade-
quate guidance by nurse midwives on breast-feeding practice to
mothers (Jiang et al., 2012). Nurse midwives can provide this
guidance by educating mothers how to increase the duration and
frequency of breast feeding, improve latch on, offer both breasts in
each breast feeding, allowing the infant to empty the breast
completely, eating a balanced diet, drinking enough fluid and
getting enough rest (Saki et al., 2012; Sultana et al., 2013).
Tiredness and fatigue
Tiredness and fatigue were the next common reasons for
weaning breast feeding, especially in the first to third weeks after
delivery. This finding echoes the challenge of breast feeding from a
previous literature (Heinig, 2010). Emerging research suggests that
tiredness and fatigue are often linked to breast feeding because
fatigue is likely to result in a range of adverse parenting beha-
viours (Giallo et al., 2011; Dunning et al., 2013) and postnatal
depressive symptoms (Wade et al., 2012). Previous studies show
that fatigued mother were less likely to cope with infant crying
and more likely to have parenting stress, which results in further
exhaustion (Giallo et al., 2011; Dunning et al., 2013). Evidence
revealed that postnatal depressive symptoms were a significant
risk factor for weaning breast feeding (Feldens et al., 2012).
The use of supplemental formula may be a solution to their
overwhelming tiredness and fatigue (Heinig, 2010). Therefore,
preparing women for postpartum tiredness and fatigue through
education during pregnancy (Heinig, 2010) and addressing the
mothers’ concerns about infant crying and improving the mothers’
ability to soothe their infants will decrease perceived maternal
tiredness and fatigue (Kurth et al., 2011).
Employment status and breast-feeding duration
Returning to work was also a common reason of early weaning,
especially six weeks after delivery. Mothers who were housewives
or in part-time employment continued breast feeding longer
than those in full-time employment which was consistent with
the finding from a local study (Tarrant et al., 2010). The possible
explanation was more intense postpartum work activity
diminishes breast-feeding activity (Mandal et al., 2012). In Hong
Kong, over 75% of women aged between 25 and 44 years work
full-time for an average of at least 45 hours per week (Census and
Statistics Department, 2013). With a densely populated, urban
environment of more than 7.1 million people emphasising on
economic growth, the working culture in Hong Kong is intense
 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎
(Census and Statistics Department, 2013). Most married couples
are employed, which may explain why more than half of the
women in this study were working full-time. The Hong Kong
government mandated that maternity leave is granted for a
maximum of 10 weeks (Employment Ordinance [Cap. 57], 2010).
Therefore early return to work has frequently been cited as a major
reason for early weaning. Working environments rarely provide
support for lactating women (Tarrant et al., 2010), and there are
few facilities to help employees continue to breast feed their
babies. Companies can support working mothers by providing
facilities such as breast milk pumps and a pumping room, regular
breaks to express milk, and support groups that encourage breast-
feeding duration (Chuang et al., 2010; Riordan and Wambach,
2010). Although breast feeding is widely recognised as the best
nutrition for infants by both professionals and the public, mothers
who choose to breast feed in Hong Kong usually cease upon
returning to work (Tarrant et al., 2010). The Hong Kong govern-
ment may consider a longer maternity leave to promote longer
breast-feeding duration (Jiang et al., 2012; Taleb et al., 2012).
Conclusion
Mothers in Hong Kong do not meet the recommended inter-
national standards for breast feeding. Women who were working
part-time or were housewives, had monthly family income of less
than HK$10,000, had more than one child, had positive breast-
feeding intention and engaging in early breast feeding (i.e. within
the first hour after delivery) were more likely to have longer
breast-feeding duration. The common reasons for weaning breast
feeding were insufficient breast milk, tiredness and fatigue and
return to work. Partners, relatives and nurse midwives were
important supportive resource during breast feeding. Understand-
ing the reasons of initiation, weaning and predictors of breast-
feeding duration can help nurse midwives to develop and custo-
mise effective strategies to promote and encourage breast feeding.
Limitations
Although contributing to the literature, this study has several
limitations. First, this study had been conducted in a city that
might be different from others so the results might be applied only
to similar settings. Second, categorising duration of breast feeding
into five groups may have limited the potential misclassification
bias; mother may have overestimated or underestimated the
duration of breast feeding. Third, there might be recall bias in
the data collected in the third time point because the participants
completed the questionnaire six weeks after their delivery. Finally,
the low response rate in the third time point may limit the
generalisability of the findings.
Implications of this study
The results of this study have significant contributions for
nurse midwives to assist in identification of women who may be
at-risk to wean breast feeding prematurely, such as first time
mothers, full-time employed mothers and mothers from higher
monthly family income. Breast-feeding intention and initiation in
the postpartum period are significant predictors of breast-feeding
duration. The importance of partners and nurse midwives as
significant supportive resources should be a focus of interventions.
The contents of antenatal education may focus on the manage-
ment of insufficient breast milk, tiredness or fatigue and continu-
ing breast feeding upon return to work. The finding is important
for a process-oriented breast-feeding training programme for
nurse midwives (Ekstrom et al., 2012) to promote breast feeding.
Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
9
New strategies may be considered for prolonging breast-feeding
duration among perinatal women such as a community-based
public antenatal education programme (Stuebe and Bonuck, 2011),
internet-based educational programme (O’Connor et al., 2011), a
parenting programme for the management of postpartum fatigue
(Dunning et al., 2013) and two weeks postpartum supportive
programme (Evans et al., 2011).
Conflict of interest
None.
Acknowledgements
Authors gratefully acknowledge the participants and hospital
staffs in five hospitals because the study would not have been
possible without their support.
References
Akter, S., Rahman, M.M., 2010. Duration of breastfeeding and its correlates in
Bangladesh. Journal of Health, Population & Nutrition 28, 595–601.
Baby Friendly Hospital Initiative Hong Kong Association, 2011. World Breastfeeding
Week 2010: Survey Report. Baby Friendly Hospital Initiative Hong Kong
Association, Hong Kong.
Bai, D.L., Wu, K.M., Tarrant, M., 2013. Association between intrapartum interven-
tions and breastfeeding duration. Journal of Midwifery & Women’s Health 58,
25–32.
Basnet, S., Gauchan, E., Malla, K., et al., 2012. Socio-demographic features of
mothers in relation to duration of breastfeeding in Manipal Teaching Hospital,
Pokhara, Nepal. Nepal Journal of Medical Sciences 1, 27–30.
Bendall, J.C., Simpson, P.M., Middleton, P.M., 2011. Prehospital vital sign can predict
pain severity: analysis using oridnal logistic regression. European Journal of
Emergency Medicine 18, 334–339.
Bezerra, V.L.V.A., Nisiyama, A.L., Jorge, A.L., Cardoso, R.M., Silva, E.F., Tristao, R.M.,
2012. Exclusive breastfeeding and factors related to early weaning: a compara-
tive study between 1999 and 2008. Revista Paulista de Pediatria 30, 173–179.
Burns, N., Grove, S.K., 2009. The practice of Nursing Research: Appraisal, Synthesis,
and Generation of Evidence, 6th edn. Saunders/Elsevier, St. Louis, MO.
Census and Statistics Department, 2013. Hong Kong Annual Digest of Statitics, 2013
edn. Census and Statistics Department, Hong Kong.
Chapman, D.J., 2012. Longer cumulative breastfeeding duration associated with
improved bone strength. Journal of Human Lactation 28, 18–19.
Chen, S., Binns, C.W., Zhao, Y., Maycock, B., Liu, Y., 2013. Breastfeeding by Chinese
mothers in Australia and China: the healthy migrant efffect. Journal of Human
Lactation, 1–7.
Chowdhury, Z.T., Henderson, M.A., Watson, R.R., 2013. Breastfeeding and infant
health in the Indian subcontinent: problems and solutions. Nutrition in Infancy
1, 51–66.
Chuang, C.H., Chang, P.J., Chen, Y.C., et al., 2010. Maternal return to work and
breastfeeding: a population-based cohort study. International Journal of
Nursing Studies 47, 461–474.
Cordero, L., Valentine, C.J., Samuels, P., Giannone, P.J., Nankervis, C.A., 2012.
Breastfeeding in women with severe preeclampisa. Breastfeeding Medicine 7,
457–463.
Cornall, D., 2011. A review of the breasfeeding literature relevant to osteopahic
practice. International Journal of Osteopathic Medicine 14, 61–66.
Dennis, C.L., Gagnon, A., Hulst, A.V., Dougherty, G., Wahoush, O., 2013. Prediction of
duration of breastfeeding among migrant and Canadian-born women: results
from a multi-center study. The Journal of Pediatrics 162, 72–79.
Dunning, M., Seymour, M., Cooklin, A., Giallo, R., 2013. Wide awake parenting:
study protocol for a randomised controlled trial of a parenting program for the
management of post-partum fatigue. BMC Public Health 13, 1–9.
Eidelman, A.I., Schanler, R.J., 2012. Breastfeeding and the use of human milk.
Pediatrics 129, e827–e841.
Ekstrom, A., Kylberg, E., Nissen, E., 2012. A process-oriented breastfeeding training
program for healthcare professionals to promote breastfeeding: an intervention
study. Breastfeeding Medicine 7, 85–92.
Employment Ordinance [Cap. 57], 2010. Maternity Protection. Labour Department,
Hong Kong Government, Hong Kong.
Evans, K., Labbok, M., Abrahams, S.W., 2011. WIC and breastfeeding support
services: does the mix of services offered vary with race and ethnity?
Breastfeeding Medicine 6, 401–406.
Farrington, D.P., Loeber, R., 2000. Some benefits of dichotomization in psychiatric
and criminological research. Criminal Behavior and Mental Health 10, 100–122.
10 W. Wang et al. / Midwifery ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Feldens, C.A., Vitolo, M.R., Rauber, F., Cruz, L.N., Hilgert, J.B., 2012. Rsik factors for
discontinuing breastfeeding in South Brazil: a survival analysis. Maternal &
Child Health Journal 16, 1257–1265.
Giallo, R., Rose, N., Vittorino, R., 2011. Fatigue, wellbeing, and parenting in mothers
of infants and toddlers with sleep problems. Journal of Reproductive and Infant
Pschology 29, 236–249.
Heinig, M.J., 2010. Addressing maternal fatigue: a chanllenge to in-hospital
breastfeeding promotion. Journal of Human Lactation 26, 231–232.
Hilbe, J.M., 2009. Texts in Statistical Science: Logistic Regression Models. Chapman
and Hall/CRC Press, London.
Ho, Y.J., McGrath, J.M., 2011. Predicting breastfeeding duration related to maternal
attitudes in a Taiwanese sample. The Journal of Perinatal Education 20,
188–199.
Hospital Authority, 2010. Hospital Authority Statistical Report: 2009–2010. Hospital
Authority, Hong Kong.
Inoue, M., Binns, C.W., Otsuka, K., Jimba, M., Matsubara, M., 2012. Infant feeding
practices and breastfeeding duration in Japan: a review. International Breast-
feeding Journal 7, 1–15.
Jiang, H., Li, M., Yang, D., et al., 2012. Awareness, intention, and needs regarding
breastfeeeding: finding from first-time mothers in Shanghai, China. Breastfeed-
ing Medicine 7, 526–534.
Johnson, R.A., Wichern, D.W., 2007. Applied Multivariate Statistical Analysis,
Pearson Educational International: Pearson, 6th edn. Prentice Hall, New Jersey.
Kao, F.F., Leu, C.H., Ko, C.H., 2011. Remainder mark on systematic sampling. Journal
of Statistical Planning and Inference 141, 3596–3604.
Kervin, B.E., Kemp, L., Pulver, L.J., 2010. Types and timing of breastfeeding support
and its impact on mothers’ behaviors. Journal of Pediatrics and Child Health 46,
85–91.
Kurth, E., Kennedy, H.P., Spichiger, E., Hosli, I., Stutz, E.Z., 2011. Crying babies, tired
mothers: what do we know? A systematic review. Midwifery 27, 187–194.
Lam, W.W.T., Fielding, R., McDowell, I., et al., 2012. Perspectives on family health,
happiness and harmony (3H) among Hong Kong Chinese people: a qualitative
study. Health Education Research 27, 767–779.
Lau, Y., 2010. Breastfeeding intention among pregnant Hong Kong Chinese women.
Maternal & Child Health Journal 14, 790–798.
Lau, Y., 2011. A longitudinal study of family conflicts, social support, and antenatal
derpessive symptoms among Chinese women. Achives of Psychiatric Nursing
25, 206–219.
Loof-Johanson, M., Foldevi, M., Rudebeck, C.E., 2013. Breastfeeding as a specific
value in women’s lives: the experiences and decisions of breastfeeding women.
Breastfeeding Medicine 8, 38–44.
Mandal, B., Roe, B.E., Fein, S.B., 2012. Work and breastfeeding decisions are jointly
determined for higher socioeconomic status US mothers. Review of Economics
of the Household, 1–12.
Mauch, C.E., Scott, J.A., Magarey, A.M., Daniels, L.A., 2012. Predictors of and reasons
for pacifier use in first-time mothers: an observational study. BMC Pediatrics
12, 1–10.
Mazumder, M.S., Hossain, M.K., 2012. Duration of breastfeeding and its determinant
in Banglasdesh. International Journal of Natural Sciences 2, 49–53.
McCrory, C., Layte, R., 2012. Breastfeeding and risk of overweight and obesity at
nine-years of age. Social Sciences & Medicine 75, 323–330.
Meedya, S., Fahy, K., Kable, A., 2010. Factors that positively influence breastfeeding
duration to 6 months: a literature review. Women and Birth 23, 135–145.
Morton, J.A., 2012. Rethniking bedside care: a shared, sustainable, and proactive
model. Journal of Human Lactation 28, 11–13.
Nachmias, C.F., Nachmias, D., 2007. Research Methods in the Social Science, 7th edn.
Worth Publisher, New York.
Nagata, C., Mizoue, T., Tanaka, K., et al., 2012. Breastfeeding and breast cancer risk:
an evalation based on a systmetic review of epidemiologic evidence among the
Japanses population. Japanese Journal of Clinical Oncology 42, 124–130.
Nicklas, J.M., Miller, L.J., Zera, C.A., Davis, R.B., Levkoff, S.E.,Seely, E.W. Factors
associated with depressive symptoms in the early postpartum period among
Please cite this article as: Wang, W., et al., Breast-feeding intention, initiation and duration among Hong Kong Chinese women: A
prospective longitudinal study. Midwifery (2013), http://dx.doi.org/10.1016/j.midw.2013.07.015i
women with recent gestational diabetes mellitus. Maternal & Child Health
Journal, http://dx.doi.org/10.1007/s10995-012-1180-y, in press.
O’Connor, M.E., Brown, E.W., Lewin, L.O., 2011. An internet-based education
program imnproves breastfeeding knowledge of maternal-child healthcare
providers. Breastfeeding Medicine 6, 421–427.
Oddy, W.H., Li, J., Whitehouse, A.J.O., Zubrick, S.R., Malacova, E., 2011. Breastfeeding
duration and achievement at 10 years. Pediatrics 127, e137–e145.
Olang, B., Heidarzadeh, A., Strandvik, B., Yngve, A., 2012. Reasons given by mothers
for discontinuing breastfeeding in Iran. International Breastfeeding Journal 7,
1–7.
Phillips, G., Brett, K., Mendola, P., 2011a. Previous breastfeeding practices and
duration of exclusive breastfeeding in the United States. Maternal & Child
Health Journal 15, 1210–1216.
Pannu, P.K., Giglia, R.C., Binns, C.W., Scott, J.A., Oddy, W.H., 2011. The effectiveness of
health promotion materials and activities on breastfeeding outcomes. Acta
Paediatrica 100, 534–537.
Phillips, G., Brett, K., Mendola, P., 2011b. Previous breastfeeding practices and
duration of exclusive breastfeeding in the United States. Maternal & Child
Health Journal 15, 1210–1216.
Riordan, J., Wambach, K., 2010. Breastfeeding and Human Lactation, 4th edn. Jones
and Bartlett, Sudbury, MA.
Saki, A., Eshraghian, M.R., Tabesh, H., 2012. Patterns of daily duration and frequency
of breastfeeding among exclusively breasfed infants in Dhiranz, Iran, a 6 month
follow-up study ising Bayesian generalized linear mixed models. Global Journal
of Health Science 5, 123–133.
Stuebe, A.M., Bonuck, K., 2011. What predicts intent to breastfeeding exclusively?
Breastfeeding knowledge, attitudes, and beliefs in a diverse urban population.
Breastfeeding Medicine 6, 413–420.
Sultana, A., Rahman, K.U., Manjula, S.M., 2013. Clinical update and treatment of
lactation insufficiency. Medical Journal of Islamic World Academy of Sciences
21, 19–28.
Taleb, S., Oulamara, H., Agli, A.N., 2012. Factors facilitating and forcing the breast-
feeding to Tebessa (East city Algerian). Food and Nutrition Sciences 3, 310–316.
Tarrant, M., Fong, D.Y.T., Wu, K.M., et al., 2010. Breastfeeding and wean practices
among Hong Kong mothers: a prospective study. BMC Pregnancy and Child-
birth 10, 1–12.
Tarrant, M., Wu, K.M., Fong, D.Y.T., et al., 2011. Impact of baby-friendly hospital
practices on breastfeeding in Hong Kong. Birh Issue in Perinatal Care 38,
238–245.
Thompson, R.E., Kildea, S.V., Barclay, L.M., Kruske, S., 2011. An account of significant
events influencing Australian breastfeeding practice over the last 40 years.
Women and Birth 24, 97–104.
UNICEF (The United Nations Fund for Children), 2009. Tracking Progress on Child
and Maternal Malnutrition: A Survival and Development Priority. 〈http://www.
unicef.pt/docs/Progress_on_Child_and_Maternal_Nutrition_EN_110309.pdf〉.
Wade, C., Giallo, R., Cooklin, A., 2012. Maternal fatigue and depression: identifying
vulnerability and relationship to early parenting practices. Advances in Mental
Health 10, 277–291.
Wen, L.M., Simpson, J.M., Rissel, C., Baur, L.A., 2012. Awareness of breastfeeding
recommendations and duration of breastfeeding: finding from the health
beginnings trial. Breastfeeding Medicine 7, 223–229.
WHO (World Health Organization). 2009. World Health Statistics 〈http://www.
who.int/nutrition/topics/exclusive_breastfeeding/en/index.html〉.
Woods, A.B., Crist, B., Kowalewski, S., Carroll, J., Warren, J., Robertson, J., 2012.
A cross-sectional analysis of the effect of patient-controlled epidural analgesia
versus patient controlled analgesia on postcesarean pain and breastfeeding.
JOGNN—Journal of Obstetric, Gynecologic, & Neonatal Nursing 41, 339–346.
Yousefshahi, F., Asadi, M., Rahimi, F., et al., 2012. The effect of meperidine and
peripartum breastfeeding and neonatal weight. Journal of Family and Repro-
ductive Health 7, 29–34.